Datasets:

TaylorAI

/

pubmed_author_manuscripts

like 3

Follow

Taylor 16

PMC004xxxxxx/PMC4396961.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101313252 34584 J Vis Exp J Vis Exp Journal of visualized experiments : JoVE 1940-087X 25549203 4396961 10.3791/52432 NIHMS831093 Article Macrophage Cholesterol Depletion and Its Effect on the Phagocytosis of Cryptococcus neoformans Bryan Arielle M. 1* Farnoud Amir M. *1 Mor Visesato 1 Del Poeta Maurizio 1 1 Department of Molecular Genetics and Microbiology, Stony Brook University Correspondence to: Maurizio Del Poeta at maurizio.delpoeta@stonybrook.edu * These authors contributed equally 21 11 2016 19 12 2014 19 12 2014 28 11 2016 94 10.3791/52432This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Cryptococcosis is a life-threatening infection caused by pathogenic fungi of the genus Cryptococcus. Infection occurs upon inhalation of spores, which are able to replicate in the deep lung. Phagocytosis of Cryptococcus by macrophages is one of the ways that the disease is able to spread into the central nervous system to cause lethal meningoencephalitis. Therefore, study of the association between Cryptococcus and macrophages is important to understanding the progression of the infection. The present study describes a step-by-step protocol to study macrophage infectivity by C. neoformansin vitro. Using this protocol, the role of host sterols on host-pathogen interactions is studied. Different concentrations of methyl--cyclodextrin (MCD) were used to deplete cholesterol from murine reticulum sarcoma macrophage-like cell line J774A.1. Cholesterol depletion was confirmed and quantified using both a commercially available cholesterol quantification kit and thin layer chromatography. Cholesterol depleted cells were activated using Lipopolysacharide (LPS) and Interferon gamma (IFNγ) and infected with antibody-opsonized Cryptococcus neoformans wild-type H99 cells at an effector-to-target ratio of 1:1. Infected cells were monitored after 2 hr of incubation with C. neoformans and their phagocytic index was calculated. Cholesterol depletion resulted in a significant reduction in the phagocytic index. The presented protocols offer a convenient method to mimic the initiation of the infection process in a laboratory environment and study the role of host lipid composition on infectivity. Immunology Issue 94 Infection phagocytosis Cryptococcus cholesterol cyclodextrin macrophages Introduction Phagocytosis is a process by which extracellular entities are internalized by host cells. It is a key weapon in the immune system’s arsenal to defend against pathogens, but the process may often be subverted by pathogens to allow for internalization and spreading throughout the body1. Phagocytosis is mediated by several signaling events that result in attachment and engulfment via rearrangements of the host cell’s cytoskeleton. ‘Professional’ phagocytes are able to recognize and bind to opsonins on the surface of the invading pathogen to signal for attachment and the formation of lamellipodia, which engulf the pathogen and form a phagosome2. Among the so-called ‘professional’ phagocytes are macrophages. Macrophages are highly specialized cells that carry out protective functions that include seeking out and eliminating disease causing agents, repairing damaged tissues, and mediating inflammation, most of these through the process of phagocytosis1,2. Cryptococcus neoformans is a species of pathogenic yeast that causes a serious disease known as Cryptococcosis. Cryptococcus spores are inhaled by the host and result in a pulmonary infection that is usually asymptomatic. It is thought that exposure is extremely prevalent; a sample of 61 children from the Pediatric Infectious Diseases Clinic at the Bronx-Lebanon Hospital Center found that all those surveyed had antibodies to the cryptococcal polysaccharide glucuronoxylomannan and other studies have shown prevalence in both human immunodeficiency virus (HIV) uninfected and infected adults3,4. Alveolar macrophages are the first line of response to the pulmonary infection and in most cases successfully clear the pathogen. However, in immunocompromised individuals (e.g., HIV and AIDS patients) the yeast is able to survive within the macrophages. In these cases, the macrophages can serve as a niche for the replication of the pathogen and may facilitate its dissemination to the central nervous system (CNS) where the disease becomes fatal5–8. It is thought that macrophages may even deliver the yeast directly into the meninges, helping the yeast to cross the blood brain barrier via the “Trojan horse” model3,9–11. Thus, it is important to understand the process of phagocytosis and the factors that affect it, especially in cryptococcal infections. Previous work in other pathogen systems point to cholesterol and lipid rafts formed by cholesterol as having an important role to play in phagocytosis12–15. Cholesterol is the most abundant lipid species in mammalian cells and comprises 25 – 50% of the mammalian cell membrane16. It has been found to play a role in modulating the biophysical properties of membranes by changing their rigidity17. Cholesterol and sphingolipids together form lipid microdomains within the membrane known as lipid rafts. Lipid rafts have been found to be involved in the formation of caveolae, as well as providing an isolated domain for certain types of signaling16–18. Due to their small size, it is difficult to study lipid rafts in vivo. One useful way to study the role of lipid rafts is to alter their constituents. Methyl-β-cyclodextrin (MβCD) is a compound that has been found to deplete cholesterol from mammalian membranes and is commonly used to study the role of lipid rafts18. In this protocol, we present a method to deplete cholesterol from host cell membranes and quantify the effect of the depletion on the ability of the host cells to phagocytose C. neoformans in vitro. This procedure makes use of cell culture techniques on an immortalized macrophage like cell line (J774A.1) as a model for infection. Cholesterol depletion was accomplished by exposure to MβCD, which has a hydrophobic core specific to the size of sterols and is able to act as a sink for cholesterol to draw it out of the membrane19. Cholesterol depletion was measured quantitatively using a commercially available kit and qualitatively using a modified Bligh-Dyer lipid extraction followed by thin layer chromatography (TLC)20. Phagocytosis was measured by infecting the cell line with a culture of opsonized yeast mixed with a cocktail of interferon-γ and lipopolysaccharide for activating the macrophages. Cryptococcus was opsonized using a glucuronoxylomannan (GXM) antibody21–23. Staining and microscopy experiments allowed for visualization of the cells and calculation of the phagocytic index to assess the degree of phagocytosis. Taken together, this protocol describes a basic method that integrates the alteration of lipid composition with a physiological process. Protocol 1. Cholesterol Depletion of J774A.1 Cells with MβCD In a sterile biosafety cabinet, seed 105 J774A.1 macrophage-like cells per well on a 96-well cell culture plate in 200 μl of Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S). Incubate at 37 °C and 5% CO2 O/N. Remove media from the cell monolayer and wash the cells twice with 1x phosphate buffered saline (PBS) that has been filtered or autoclaved. Add 200 μl of MβCD solution at the desired concentration (10 mM or 30 mM in PBS) or 1x PBS as a control and incubate for 30 min at 37 °C with shaking. Remove supernatant and reserve at RT for quantitative analysis with commercially available kit immediately following the procedure. Wash the cells two to three times with 1x PBS or serum free DMEM and continue with infection or lyse cells by pipetting two to three times with deionized H2O for analysis with thin layer chromatography or with a kit. NOTE: Following cholesterol depletion a commercially available cholesterol quantification kit can be used. See materials section for details. Follow manufacturer’s instructions as written. 2. Observation of Cholesterol Content by Thin Layer Chromatography (TLC) Wash a TLC tank twice with acetone and once with a solution of petroleum ether:diethyl ether:acetic acid (65:30:1 by volume). Saturate the tank with the petroleum ether:diethyl ether:acetic acid (65:30:1 by volume) solution and leave O/N. NOTE: Organic solvents should always be used under a fume hood to prevent inhalation of vapor. Gloves and lab coat should be worn at all times. Acetic acid is a strong acid and should be used with caution. In a sterile biosafety cabinet, seed 106 J774A.1 macrophage-like cells per well on a 6-well cell culture plate in a volume of 5 ml of warm DMEM supplemented with 10% FBS and 1% P/S. Incubate at 37 °C, 5% CO2 O/N. Deplete macrophages of cholesterol by following steps 1.2 – 1.4, substituting 1 ml for 200 μl where applicable to account for the larger well size. Add 500 μl of Trypsin-EDTA to each well, incubate for 3 min at 37 °C, and gently scrape cells with a cell scraper. Transfer into a microfuge tube and add an additional 500 μl of warm DMEM supplemented with 10% FBS and 1% P/S. Spin the cells for 5 min at 300 × g and remove the supernatant. Add an additional 500 μl of warm DMEM supplemented with 10% FBS and 1% P/S to the cell pellet and resuspend carefully by pipetting up and down. Remove 10 μl of cells and count cells on a hemocytometer. Normalize the cell concentrations from each sample and add an equal number of cells to glass tubes. NOTE: At this step, one may choose to combine cells from the same treatment groups to obtain a more concentrated final lipid extract. Centrifuge cells at 300 × g for 5 min at RT and remove media. Add 2 ml of methanol and vortex. Then, add 1 ml of chloroform and vortex. Check the phase status to make sure the solution in monophasic. NOTE: Tubes can be stored at 4 °C O/N. Centrifuge at 1,700 × g for 10 min at RT and transfer the supernatant to a new tube Add an additional 1 ml of chloroform followed by 1 ml of dH2O. Vortex twice for 30 sec. Centrifuge at 1,700 × g for 5 min at RT. Weigh a glass tube on a sensitive balance and use a glass Pasteur pipette to transfer the lower phase into the glass tube of known weight. Dry down lipids in a centrifugal evaporator until dry (approximately 2 hr). Weigh tube with dried lipids and calculate the dry lipid weight. NOTE: Dry lipids can be stored in −20 °C until ready to perform TLC. Dilute dried lipids in enough chloroform to normalize the concentration of lipid (usually 20 – 50 μl) and load 20 μl of the diluted lipid on a silica TLC plate. Load 20 μg of cholesterol diluted in 20 μl of chloroform as a standard. Add dried TLC plate to the saturated TLC tank and allow solvent to migrate up to 1 cm before plate edge. Remove TLC plate from tank and allow it to dry about 5 min. Visualize lipids by placing in an iodine vapor tank to check migration. Remove and allow spots to fade for about 10 – 15 min under the hood. NOTE: Iodine is an inhalation hazard. Always use under a fume hood. Prepare a solution for neutral lipid staining by combining 60 ml of methanol with 60 ml of deionized H2O, 4 ml of sulfuric acid, and 630 mg of manganese chloride. Carefully and slowly dip the TLC plate into the neutral lipid staining solution in a tray and remove without sloughing off the silica layer. NOTE: The neutral lipid staining solution can be reused several times and so it can be retrieved from the tray and placed back in a bottle for later use. Allow plate to dry under the hood at RT until all bubbles has disappeared. Heat the TLC plate on a heat block set to 160 °C and char to the desired color. NOTE: A densitometry program such as Vision Works LS can be used to quantify the charred bands to compare lipid samples. 3. Infection of Macrophages with C. neoformans (H99) In a sterile biosafety cabinet, seed 105 macrophage-like cells per well on a 96-well cell culture plate in 200 μl of DMEM supplemented with 10% FBS and 1% P/S. Incubate at 37 °C, 5% CO2, 5% CO2 O/N. NOTE: Infection can also be done in glass bottomed confocal dishes for easier imaging; all amounts remain the same. Grow a culture of C. neoformans (H99) by inoculating 10 ml of YNB with one colony obtained from a struck plate and incubating it O/N at 30 °C with shaking. Wash and count C. neoformans (H99) cells. Centrifuge C. neoformans O/N culture at 1,700 × g for 10 min at 4 °C. Remove media and discard. Wash cells with 5 ml of 1x PBS. Centrifuge at 1,700 × g for 10 min at 4 °C. Remove PBS and wash with 5 ml of filtered 1x PBS. Centrifuge at 1,700 × g for 10 min at 4 °C. Repeat this step 2 more times. Remove PBS and resuspend in 5 ml of 1x PBS. Make a serial dilution in PBS to obtain a 1:500 dilution of the washed culture. Add 100 μl of the original sample to 900 μl of 1x PBS to obtain a 1:10 dilution. Add 100 μl of 1:10 diluted sample to 900 μl of 1x PBS to obtain a 1:100 dilution. Add 200 μl of the 1:100 diluted sample to 800 μl of 1x PBS to obtain a 1:500 dilution Take 10 μl of 1:500 dilution and count on hemocytometer to calculate the number of cells. Prepare working solution for activating macrophages and opsonizing C. neoformans. Dilute LPS and IFNγ 100x from stock solutions by adding 10 μl to 990 μl. NOTE: LPS and IFNγ are used to enhance phagocytic uptake but are not required for phagocytosis. If there is an interest in fungicidal activity of macrophages, perform activation O/N at 37 °C with shaking prior to infection. Per sample combine 7.5 μl of diluted LPS, 1.25 μl of diluted IFNγ, 1.25 μl of GXM antibody and the volume of the C. neoformans culture that gives 1.25 × 105 cells. Bring volume up to 250 μl multiplied by number of samples with DMEM supplemented with 10% FBS and 1% P/S. Vortex and incubate solution for 20 min at 37 °C with shaking. NOTE: Cholesterol depletion (steps 1.2 – 1.4) can be done concurrently with the opsonization step. Be sure to treat macrophages prior to combining the working solution, as the opsonized cells should optimally be used no longer than 20 min following the step 3.4.3 incubation. Infect Macrophages Wash macrophages twice with serum free DMEM and add 200 μl of opsonized C. neoformans working solution to each well. Incubate for 2 hr at 37 °C. Fix and Stain Cells Remove media and wash cells 2 times with DMEM. Air dry the cell monolayer for 10 min and add 200 μl of ice cold methanol to fix the cells. Incubate for 15 min at RT and remove any remaining methanol. Add 200 μl of 10x Giemsa and incubate for 5 min at RT. Wash 2 – 3 times with deionized water and dry O/N with cap off. NOTE: Imaging can be done the next day or up to a week following staining. Visualize and Count Using a microscope, count 300 cells per data point (if there are 2 of the same treatment count 150 per well) and note the number of infected macrophages and number of engulfed Cryptococcus cells. NOTE: To ensure even sampling per data point use 2 plates per condition, choose 3 non-overlapping areas per plate and count 50 cells per area. Calculate phagocytic index by multiplying the percentage of infected macrophages by the mean number of C. neoformans per macrophage. Normalize phagocytic index by expressing values as a percentage of the 1x PBS treated control. After several trials calculate the mean value and the standard deviation of the mean to determine trends in phagocytic index. Use student t-test to determine significance. Take micrographs of cells at 1,000X or 400X magnification. 4. Trypan Blue Assay In a sterile biosafety cabinet, seed 106 macrophage-like cells per well on a 6-well cell culture plate in a volume of 5 ml of Dulbecco’s minimal essential medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Incubate at 37 °C, 5% CO2 O/N. Deplete macrophages of cholesterol by following steps 1.2 – 1.4 substituting 2 ml for 200 μl where applicable to account for the larger well size. NOTE: Be sure to include a control treated with 1x PBS as well as a control that is scraped prior to any treatment. Add 500 μl of 1x PBS to each well and gently scrape cells with a cell scraper. Transfer into a microfuge tube and suspend the cells by gently pipetting up and down. Remove 10 μl of cells and stain with 1 μl of 4% Trypan blue. Count cells on a hemocytometer and calculate viability using the following equation: % viability = [1 − (Blue cells/total cells)] × 100. Normalize values to the control that was untreated. After several trials calculate the mean value and the standard deviation to determine trends in viability. Use student t-test to determine significance. Representative Results Cholesterol Depletion Analysis of the supernatant reserved in step 1.3 of the protocol by following the manufacturer’s instructions in the Amplex Red Cholesterol Assay kit yields an elevated concentration of cholesterol in MβCD treated sample as compared to the 1x PBS control. Depending on cell type and MβCD concentration used cholesterol depletion may vary. For J774 treated with 10 mM MβCD, a depletion of approximately 50% was observed. Depletion can be calculated using values obtained from the supernatant and cell lysate collected in step 1.4 (Figure 1). Cell lysate analyzed using TLC shows a marked decrease in staining of cholesterol in cells treated with increasing concentration of MβCD (Figure 2A). Densitometry analysis of the TLC shows a similar trend to the quantitative assay (Figure 2B). The Bligh-Dyer method gives a crude extract of total lipids and it is essential to allow for adequate separation of lipids in order to identify the correct band utilizing the cholesterol standard. Infection After following the infection procedure, cells remain adhered and intact. Cell morphology remains unchanged between treatment groups. A control group that has not been exposed to C. neoformans serves as a checkpoint (Figure 3). It is possible to obtain suboptimal results and may manifest as lysis of cells and other abnormal morphologies. The most likely cause is contamination of the cell line or reagents used in the procedure. Micrographs of optimally infected cells clearly show C. neoformans engulfed within the mammalian cells. Differences in number of phagocytized yeast may be noted by observation between treatment groups (Figure 4). After calculating phagocytic index from 300 macrophage cells per treatment group a reduction in phagocytic index is found in cholesterol depleted cells (Figure 5). The reduction in the phagocytic index does not appear to be dependent on potential differences in macrophage activation, although they may occur. Performing the infection in the absence of macrophage activators, but after treatment with MCD results in a similar reduction of phagocytic index (data not shown). Trypan Blue Trypan Blue staining is used to assess the viability of cells after cholesterol depletion. No change in viability is observed between PBS treated and 10mM MCD treated cells. Viability appears to drop off slightly after treatment with 30 mM MCD, which may be expected due to the approximately 75% depletion in cholesterol (an essential lipid) observed in the densitometry analysis (Figure 6 and Figure 2B). Discussion In working with this protocol it is important to obtain accurate cell counts when plating mammalian cells and opsonizing C. neoformans cells. This minimizes variation between trials and ensures an accurate 1:1 target to effector ratio throughout the study. It is also critical to coordinate the timing of the cholesterol depletion and infection to prevent the opsonized yeast cells or treated macrophage cells from resting at RT in between the procedures. Long waiting periods could lead to loss of antibody opsonization or the replenishing of depleted cholesterol before infection can begin. If experiments are done with precision the data analysis allows for conclusions to be discerned about the role of cholesterol in phagocytosis. The limitations of the technique prevent any conclusions as to the specific mechanism by which cholesterol depletion lowers the phagocytic index of macrophage like cells, and it is unclear whether the effect is directly due to cholesterol or due to a secondary mechanism. Further work along this vein investigates other constituents of lipid rafts such as sphingolipids or proteins known to function in phagocytosis such as the Fcγ receptor and the complement receptor 3 2. Modifying this technique to use either antibody opsonization or complement alone could help distinguish a role for cholesterol in one or both of these known pathways. It is also important to remember that MβCD extracts cholesterol based on its hydrophobicity and size, thus sterols of a similar size may also be depleted and will migrate at a similar rate as cholesterol on a TLC. It is also important to note that cholesterol depletion can partially affect macrophage activation, this is unlikely to be responsible for the difference observed in uptake as performing the infection in the absence of IFN- and LPS shows the same reduction in the uptake of C. neoformans (data not shown), but it is of interest when modifying this technique to study the anti-fungal activities of macrophages and the role of activation. This method also does not allow us to discern whether cholesterol depletion has any therapeutic implications in fungal infection. Further work in vivo with cholesterol-lowering drugs and epidemiological studies of patients using cholesterol-lowering drugs could further elucidate a role for cholesterol depletion in the treatment of the disease and may offer a more selective way to inhibit cholesterol accumulation. This procedure could easily be used to study uptake of other pathogens or solid particles (i.e., glass beads) being phagocytized and allow for the study of basic biology of phagocytosis. Modifications could allow for the study of other aspects of phagocytosis by treating the macrophage cells with enzymes to selectively degrade other membrane components or various drugs and inhibitors which may be of interest. It should also be noted that flow cytometry may present a more accurate and quantitative way to characterize phagocytosis and could be used to replace the direct microscopic count24. Altogether, this is a fairly simple technique that can be used as a starting point for more in depth studies that answer questions about how lipids may play an important part in infection and immune response. This work was supported by NIH grants AI56168, AI71142, AI87541 and AI100631 to MDP. Maurizio Del Poeta is Burroughs Wellcome Investigator in Infectious Diseases. Figure 1 Cholesterol content of supernatant after treatment Quantification of cholesterol in the supernatant collected from treated cells shows enrichment in MCD when compared to 1x PBS. Cholesterol depletion is 50 ± 5% calculated from total cholesterol in 1x PBS (supernatant + cell lysate). Error bars show standard deviation (n = 5). Figure 2 TLC of cholesterol in cell lysate and densitometry Image of developed TLC plate visualized with MnCl2 charring. A marked decrease in cholesterol is seen after MβCD treatment (A). Densitometry analysis of bands as compared to the PBS treated control (shown as 100%) confirms trend found in Cholesterol quantification assay (B). Please click here to view a larger version of this figure. Figure 3 Uninfected control micrographs of treated J774 macrophages Images of uninfected J774 cells taken at 200X magnification. Scale bar is 50 μm. 1x PBS (A), 10 mM MβCD (B), and 30 mM MβCD (C) treated cells show no change. Please click here to view a larger version of this figure. Figure 4 Infection of J774 macrophages with C. neoformans. Images of infected J774 cells taken at 400X (top row A.1 – C.1) and 1,000X (bottom row A.2 – C.2) magnification are shown. Internalized C. neoformans cells appear as blue-violet spheres with a lighter ring surrounding them. Cells treated with 1x PBS (A), 10 mM MβCD (B), and 30 mM MβCD (C) show differences in C. neoformans uptake. Please click here to view a larger version of this figure. Figure 5 Phagocytic index Phagocytic index is shown with respect to the control group that was treated with PBS (Marked at 100 for comparison). Phagocytic index was reduced by 25% by 10 mM MβCD treatment and by almost 55% by 30 mM treatment. Error bars show standard deviation of the mean (n = 4). Figure 6 Cell viability Variations in cell viability by trypan blue assay show little variation when comparing all three treatment groups. There is a slight drop off in viability in the 30 mM MβCD treatment group, which can be expected from depletion of such a major component of the membrane. Error bars show standard deviation (n = 4). Video Link The video component of this article can be found at http://www.jove.com/video/52432/ Disclosures The authors have nothing to disclose. 1 Sarantis H Grinstein S Subversion of phagocytosis for pathogen survival Cell Host & Microbe 12 4 419 31 10.1016/J.Chom.2012.09.001 2012 23084912 2 Rougerie P Miskolci V Cox D Generation Of Membrane Structures During Phagocytosis And Chemotaxis Of Macrophages: Role And Regulation Of The Actin Cytoskeleton Immunological Reviews 256 1 222 39 10.1111/Imr.12118 2013 24117824 3 Liu T Perlin DS Xue C Molecular Mechanisms Of Cryptococcal Meningitis Virulence 3 173 181 10.4161/Viru.18685 2012 22460646 4 Abadi J Pirofski L A Antibodies Reactive With The Cryptococcal Capsular Polysaccharide Glucuronoxylomannan Are Present In Sera From Children With And Without Human Immunodeficiency Virus Infection The Journal Of Infectious Diseases 180 3 915 9 10.1086/314953 1999 10438394 5 Kechichian TB Shea J Del Poeta M Depletion Of Alveolar Macrophages Decreases The Dissemination Of A Glucosylceramide-Deficient Mutant Of Cryptococcus Neoformans In Immunodeficient Mice Infection And Immunity 75 10 4792 8 10.1128/IAI.00587-07 2007 17664261 6 Casadevall A Cryptococci At The Brain Gate: Break And Enter Or Use A Trojan Horse? The Journal Of Clinical Investigation 120 5 1389 92 10.1172/JCI42949 2010 20424319 7 Chrétien F Pathogenesis Of Cerebral Cryptococcus Neoformans Infection After Fungemia The Journal Of Infectious Diseases 186 4 522 30 10.1086/341564 2002 12195380 8 Luberto C Martiñez-Marino B Identification Of App1 As A Regulator Of Phagocytosis And Virulence Of Cryptococcus Neoformans The Journal Of Clinical Investigation 112 7 1080 94 10.1172/JCI18309 2003 14523045 9 Mcquiston TJ Williamson PR Paradoxical Roles Of Alveolar Macrophages In The Host Response To Cryptococcus Neoformans Journal Of Infection And Chemotherapy: Official Journal Of The Japan Society Of Chemotherapy 18 1 1 9 10.1007/S10156-011-0306-2 2012 22045161 10 García-Rodas R Zaragoza O Catch Me If You Can: Phagocytosis And Killing Avoidance By Cryptococcus Neoformans. FEMS Immunology And Medical Microbiology 64 2 147 61 10.1111/J.1574-695X.2011.00871.X 2012 22029633 11 Coelho C Bocca AL Casadevall A The Intracellular Life Of Cryptococcus Neoformans Annual Review Of Pathology 9 219 38 10.1146/Annurev-Pathol-012513-104653 2014 12 Rao M Peachman KK Alving CR Rothwell SW Depletion Of Cellular Cholesterol Interferes With Intracellular Trafficking Of Liposome-Encapsulated Ovalbumin Immunology And Cell Biology 81 415 423 10.1046/J.1440-1711.2003.01192.X 2003 14636238 13 Sein KK Aikawa M The Prime Role Of Plasma Membrane Cholesterol In The Pathogenesis Of Immune Evasion And Clinical Manifestations Of Falciparum Malaria Medical Hypotheses 51 2 105 110 10.1016/S0306-9877(98)90102-5 1998 9881815 14 Pucadyil TJ Tewary P Madhubala R Chattopadhyay A Cholesterol Is Required For Leishmania Donovani Infection: Implications In Leishmaniasis Molecular And Biochemical Parasitology 133 145 152 10.1016/J.Molbiopara.2003.10.002 2004 14698427 15 Tagliari L Membrane Microdomain Components Of Histoplasma Capsulatum Yeast Forms, And Their Role In Alveolar Macrophage Infectivity Biochimica Et Biophysica Acta 1818 3 458 66 10.1016/J.Bbamem.2011.12.008 2012 22197503 16 Crane JM Tamm LK Role Of Cholesterol In The Formation And Nature Of Lipid Rafts In Planar And Spherical Model Membranes Biophysical Journal 86 5 2965 79 10.1016/S0006-3495(04)74347-7 2004 15111412 17 Brown DA London E Functions Of Lipid Rafts In Biological Membranes Annual Review Of Cell And Developmental Biology 14 111 36 10.1146/Annurev.Cellbio.14.1.111 1998 18 Simons K Toomre D Lipid Rafts And Signal Transduction Nature Reviews Molecular Cell Biology 1 1 31 9 10.1038/35036052 2000 11413487 19 Ilangumaran S Hoessli DC Effects Of Cholesterol Depletion By Cyclodextrin On The Sphingolipid Microdomains Of The Plasma Membrane The Biochemical Journal 335 Pt 2 433 40 1998 9761744 20 Bligh EG Dyer WJ A Rapid Method Of Total Lipid Extraction And Purification Canadian Journal Of Biochemistry And Physiology 37 8 911 917 10.1139/O59-099 1959 13671378 21 Mukherjee S Lee S Casadevall A Antibodies To Cryptococcus Neoformans Glucuronoxylomannan Enhance Antifungal Activity Of Murine Macrophages Infect Immun 63 2 573 579 1995 7822024 22 Deshaw M Pirofski LA Antibodies To The Cryptococcus Neoformans Capsular Glucuronoxylomannan Are Ubiquitous In Serum From HIV+ And HIV− Individuals Clinical And Experimental Immunology 99 3 425 32 1995 7882565 23 Tripathi K Mor V Bairwa NK Del Poeta M Mohanty BK Hydroxyurea Treatment Inhibits Proliferation Of Cryptococcus Neoformans In Mice Frontiers In Microbiology 3 5 187 10.3389/Fmicb.2012.00187 2012 22783238 24 Chaka W Quantitative Analysis Of Phagocytosis And Killing Of Cryptococcus Neoformans By Human Peripheral Blood Mononuclear Cells By Flow Cytometry Clin Diagn Lab Immunol 2 6 753 759 1995 8574842

PMC004xxxxxx/PMC4969178.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101242342 32406 Contemp Clin Trials Contemp Clin Trials Contemporary clinical trials 1551-7144 1559-2030 27282117 4969178 10.1016/j.cct.2016.06.002 NIHMS794281 Article Interim Methadone and Patient Navigation in Jail: Rationale and Design of a Randomized Clinical Trial Schwartz Robert P. M.D. 1* Kelly Sharon M. Ph.D. 1 Mitchell Shannon G. Ph.D. 1 Dunlap Laura Ph.D. 2 Zarkin Gary A. Ph.D. 2 Sharma Anjalee M.S.W. 1 O’Grady Kevin E. Ph.D. 3 Jaffe Jerome H. M.D. 14 1* Friends Research Institute, Baltimore, MD USA 2 RTI International, Research Triangle Park, NC USA 3 Department of Psychology, University of Maryland, College Park, College Park, MD USA 4 Department of Psychiatry, University of Maryland School of Medicine, Baltimore, MD USA * Please address correspondence to Robert P. Schwartz, M.D., Friends Research Institute, Inc., 1040 Park Avenue, Suite 103, Baltimore, MD 21201 USA; Voice: 410-837-3977 x276; Fax: 410-752-4218; rschwartz@friendsresearch.org 14 6 2016 07 6 2016 7 2016 01 7 2017 49 2128 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background Methadone maintenance is an effective treatment for opioid dependence but is rarely initiated in US jails. Patient navigation is a promising approach to improve continuity of care but has not been tested in bridging the gap between jail- and community-based drug treatment programs. Methods This is an open-label randomized clinical trial among 300 adult opioid dependent newly-arrested detainees that will compare three treatment conditions: methadone maintenance without routine counseling (termed Interim Methadone; IM) initiated in jail v. IM and patient navigation v. enhanced treatment-as-usual. The two primary outcomes will be: (1) the rate of entry into treatment for opioid use disorder within 30 days from release and (2) frequency of opioid positive urine tests over the 12-month follow-up period. An economic analysis will examine the costs, cost-effectiveness, and cost-benefit ratio of the study interventions. Results We describe the background and rationale for the study, its aims, hypotheses, and study design. Conclusions Given the large number of opioid dependent detainees in the US and elsewhere, initiating IM at the time of incarceration could be a significant public health and clinical approach to reducing relapse, recidivism, HIV-risk behavior, and criminal behavior. An economic analysis will be conducted to assist policy makers in determining the utility of adopting this approach. methadone treatment interim methadone patient navigation criminal justice jail Introduction and background Opioid use disorder is a serious problem among the millions of annual arrestees throughout much of the developed world [1, 2]. In the US, individuals with heroin and/or other opioid misuse are overrepresented in jails compared to the general population with 8% of the US sentenced jail population reporting such misuse [3]. Although methadone maintenance treatment (MMT) in the community is an effective treatment for opioid use disorder [4, 5], such treatment is rarely initiated in US jails [6]. Given the increased risk of relapse, re-arrest, and overdose death after brief periods of imposed abstinence [7–9], there is considerable public health and economic interest in demonstrating cost-effective approaches to engage arrestees in effective treatment. Jails house arrestees awaiting trial and those serving sentences up to one year. There are about 3,163 jails in the US [10]. While some US jails provide methadone to pregnant women [11] and a few (e.g., in Albuquerque, NM, Baltimore, MD, New Haven, CT, and New York, NY) continue MMT for patients receiving treatment in the community at the time of arrest [12], the jail at Rikers Island in New York City is one of the only US jails that has reported initiating MMT for those of its inmates who request it [13]. Barriers to providing MMT in jails include security concerns related to methadone storage, stigma, unfamiliarity or philosophical opposition to opioid maintenance, lack of sufficient medical staff, and logistical and cost issues [14, 15]. An additional barrier is the cost of counseling needed under Federal and State methadone regulations. This barrier could be overcome by providing methadone without routine counseling. Such care is delivered throughout many European and Australian communities [16, 17]. In the US, methadone treatment without routine counseling (termed Interim Methadone [IM]) has been shown to be as effective as methadone with routine counseling in suppressing illicit opioid use [18–20]. These findings provide support for offering methadone without routine counseling to out-of-treatment arrestees who are opioid-dependent at the time of incarceration. Reports from the Rikers Island MMT indicate that many inmates started on MMT do not enter treatment upon release [21]. This led to the suggestion of providing help to patients, such as obtaining tokens for public transportation, ID cards, Medicaid, and negotiating the treatment entry process [6]. Patient navigation (PN), is a form of strengths-based case management that has been shown to be beneficial in increasing cancer screening and follow-up rates, improving entry and adherence to HIV treatment, and increasing the likelihood of drug treatment entry [22–26]. It is reasonable to assume that patient navigation might be useful to assist inmates in maintaining continuity of care in MMT upon release from incarceration. Because patient navigation was originally developed to assist women to receive medical services [27] and women seeking to enter substance abuse treatment face particular barriers including social stigma, childcare concerns [28], gender might be a moderating factor in response to navigation. The present study will compare the effectiveness and cost-effectiveness of three conditions: methadone maintenance without routine counseling (termed Interim Methadone; IM) initiated in jail v. IM and patient navigation v. enhanced treatment-as-usual for newly-arrested adults in the Baltimore City Detention Center. This study is part of the National Institute on Drug Abuse-funded SOMATICS cooperative study that is examining approaches to delivering FDA-approved pharmacotherapies to recently arrested adults with opioid dependence. The other two studies, which are examining extended-release naltrexone (XR-NTX), are led by University of California Los Angeles and New York University and are described elsewhere in this journal. 2. Research Design and Study Population 2.1 Study Design This study is a parallel three-group randomized effectiveness trial that examines: (1) interim methadone maintenance (IM) initiated in jail v. (2) IM and patient navigation (PN) v. (3) enhanced treatment-as-usual (ETAU) that includes a brief methadone detoxification. The participants will be 300 opioid-dependent newly-arrested detainees receiving treatment for opioid withdrawal in the Baltimore City Detention Center. 2.2 Research questions and hypotheses The primary research aims are to determine the relative effectiveness and IM+PN v. IM alone v. ETAU delivered to a newly-arrested adult population with an opioid use disorder in regard to: (1) increasing the likelihood of post-release treatment entry and retention in community-based methadone treatment; (2) reducing the likelihood of post-release opioid and cocaine use; and, (3) reducing HIV-risk and criminal behavior, arrest, and days of incarceration. We hypothesize that IM+PN and IM Alone Conditions will have superior outcomes compared to ETAU, and IM+PN will have superior outcomes to IM alone. Secondary research aims are to determine the relationship between gender and effectiveness and to examine the cost, cost effectiveness, and cost-benefit of the three study Conditions from a societal perspective. 2.3 Study site The study is being conducted by Friends Research Institute in Baltimore. The recruitment and during-detention methadone treatment site is in the Baltimore City Pretrial and Detention Services, a division of Maryland’s Department of Public Safety and Correctional Services (DPSCS). DPSCS operates a methadone treatment program that provides methadone detoxification and maintains methadone treatment for arrestees who were enrolled in a methadone program at the time of arrest. Prior to the present study, this program did not initiate methadone maintenance treatment for arrestees with an opioid use disorder with the exception of pregnant women. The site does not provide buprenorphine or extended-release naltrexone to arrestees. The four community Opioid Treatment Programs agreed to receive new patients who were released from the Detention Center on methadone because these programs did not have waiting lists. 2.4 Inclusion/Exclusion Criteria To be included in the study, detainees must: (1) meet DSM-5 criteria for opioid use disorder; (2) be detained for at least 48 hours (because those detainees who are released quickly are most often released within 48 hours and hence would not have time to receive services provided through the study); (3) be receiving opioid withdrawal treatment at the men’s and women’s Pre-trial Facilities; (4) be able and willing to provide informed consent in English; (5) be detained for a charge that, if found guilty, will likely result in a sentence of less than 1 year; (6) plan to reside in Baltimore (city or county) upon release; and (7) be 18 years of age or older. Individuals are excluded if they: (1) are enrolled in opioid agonist treatment (methadone or buprenorphine treatment) in the community at the time of arrest; (2) have a medical or psychiatric condition that would make participation unsafe in the judgment of the medical staff or the PI; (3) are pregnant; (4) are allergic to methadone; or (5) require treatment for moderate or severe alcohol or sedative hypnotic withdrawal. 2.5 Recruitment The Medical Staff of the Detention Center refers newly-detained adults who are receiving opioid detoxification to speak with the Friends Research Institute’s Research Assistant (RA) about the study. Nursing coordinates the RA’s visit for informed consent and baseline assessment and the pre-screening medical eligibility visit with the methadone program physician. 2.6 Informed Consent During the research visit, the RA describes the study, reviews the informed consent form and reviews the risks and benefits of participation. The RA underlines the points that participation is completely voluntary and will not impact their criminal justice status. Individuals who decline participation are informed about alternative options including completing the detoxification that they are receiving and attending drug abuse treatment in the community. The RA administers a consent quiz on which individuals must receive a perfect score within three attempts to be deemed eligible. 2.7 Screening, Randomization, and Follow-up Procedures After the individual provides written informed consent, the RA administers the baseline instruments described below. The RA completes the eligibility checklist and obtains the prescreen eligibility checklist and physical exam from the physician. The PI reviews the eligibility checklists and source documents (RA eligibility checklists, consent quiz, informed consent, physician pre-screen eligibility checklist and physical exam and pregnancy test results [which are collected for all women upon detention]) and then enrolls eligible individuals. Following enrollment, the RA meets with the participant to provide standardized information about drug abuse, HIV prevention, overdose prevention, and information on how to contact Baltimore’s substance abuse treatment helpline for an intake and referral to treatment in the community. The RA opens the next randomization envelope and informs the participant of his/her assigned study Condition. Participants are assigned to Conditions using a random permutation procedure, such that, within gender, for each block of 3, 6, or 9 participants, one-third will be assigned at random to the IM+PN Condition, one-third to the IM alone Condition, and one-third to the ETAU Condition. Random block sizes are used to thwart any attempt by the RA or others to deduce the random assignment procedure. The Project Manager provides the RA with sealed opaque envelopes based on this random permutation procedure. Participants do not receive compensation for the baseline interview during incarceration in order not to have the appearance of monetary coercion for study enrollment, but they will receive $30 for each of the four follow-up interviews regardless of whether they are in the community or re-incarcerated. 2.8 Data Management RAs complete baseline study assessments on paper teleform (including only the participant’s study ID number and no other identifiers) and upload PDFs of the teleforms to the UCLA Data Management Center (DMC). Paper forms are necessary because the RAs do not have internet access in the Detention Center. Data from follow-up visits are entered by the RA using UCLA DMC’s web-based data entry system. 3.0 Approvals and Data and Safety Monitoring 3.1 Approvals The Friends Research Institute (FRI) Institutional Review Board (IRB) approved the study. The US Office of Human Research Protections approved the study protocol and agreed with the IRB that it met the conditions for ethical conduct of research among prisoners under 45 CFR 46.306(a)(2)(iv). The study was registered at ClinicalTrials.gov (NCT 02334215). A federal Certificate of Confidentiality was obtained to protect the confidentiality of the participants’ data. 3.2 Data and Safety Monitoring The study is being monitored by a Data and Safety Monitoring Board (DSMB) at UCLA. The FRI IRB, the UCLA DSMB, and NIDA (the study sponsor) monitor recruitment, retention, and study safety. All Serious Adverse Events are reported to the IRB, DSMB, and NIDA medical monitor regardless of their possible relationship to study procedures. 4.0 Interventions 4.1 Enhanced Treatment-as-Usual (ETAU) Individuals assigned to ETAU receive opioid withdrawal treatment with a methadone detoxification over about one week. In addition, Research Assistants (RAs) provided these participants with standardized information used in all three SOMATICS studies on the harms associated with drug abuse, referral information to substance abuse treatment in the community, and HIV and overdose prevention information. 4.2 Interim Methadone Maintenance Alone (IM Alone) Individuals assigned to IM Alone receive an individualized gradual dose induction and administered methadone under direct observation through the Pretrial and Detention Service’s Methadone Treatment Program. They do not receive routine counseling but are able to receive mental health treatment at their request (as would any other inmate). They remain on methadone during their sojourn in the Detention Center until their release, unless they request to discontinue treatment or are transferred to another facility (e.g., due to sentencing or for serious rule infractions such as attempted diversion of methadone). In these cases, whenever possible, the participant undergoes a gradual dose reduction under medical supervision. The RA asks the participant at enrollment which of four participating methadone maintenance treatment (MMT) programs he/she would like to attend in the community. The Pre-trial and Detention Services Methadone Program nursing staff will arrange for transfer to the community programs upon release. Participants are told by the RAs to report to their MMT program on the day following release (or within three days, at the most) to ensure admission and continuity of care. The nursing staff arranges “courtesy dosing” at one of the four cooperating community-based MMT programs (as would occur on a trip to another state) until the participant can be seen by the receiving MMT program for intake and admission. 4.3 Interim Methadone plus Patient Navigation (IM+PN) In addition to interim methadone described above, participants assigned to the IM+PN Condition are seen by the study’s Patient Navigator once while detained for an assessment of community reentry needs (focused on barriers to MMT entry). The navigator is available to the participant approximately weekly for up to three months post-release. During this time, following a patient navigation manual developed for the project based on the work of Sorensen and colleagues [6], the navigator seeks to meet the participant at the MMT program to increase the likelihood of a smooth admission process and reaches out to those participants who do not attend their first visit or who subsequently drop out of treatment in the community following release. Using strengths-based case management and motivational techniques, the navigator helps the participant obtain needed services such as an ID card, reduced fare bus passes, health insurance, and medical or psychiatric appointments. The navigator has a small amount of funds available (an average of approximately $40 per participant) to assist the participant in obtaining ID cards, bus passes, and other related items. 5.0 Assessments Assessments are conducted by trained RAs. The RAs are blind to study Condition at baseline but it will not be possible to blind RAs at follow-up visits conducted at 1, 3, 6, and 12 months post-release because follow-up interviews for those in treatment are conducted at the MMT. All participants will be sought for follow-up interviews regardless of whether they remained in treatment. Addiction Severity Index (ASI) - Lite is a 30–45 minute self-report interview covering seven domains over the participant’s past 30 days, including days of heroin and cocaine use, and days committing illegal activities [29]. The ASI will be administered at baseline and all follow-up points. Urine Drug Screening Urine samples will be collected by project staff at 1, 3, 6, and 12 months post-release and tested by an approved rapid drug screen test card for opiates, oxycodone, methadone, buprenorphine, cocaine, marijuana, amphetamine, and benzodiazepines. Methadone or buprenorphine positive tests will not be treated as “illicit” drugs for the purposes of analysis if the participants are enrolled in a treatment program. Modified Composite International Diagnostic Interview, version 2 (CIDI-2) for Substance Use Disorders The modified CIDI-2 [30] will be used to determine whether individuals met the DSM-5 criteria for opioid and cocaine use disorders or remission in the 12 months prior to baseline and during the one-month period prior to the 3, 6, and 12 month study assessments. The CIDI-2 was recommended by an expert panel of the NIDA Clinical Trials Network (CTN) for gauging DSM-IV remission in clinical studies [31]. It has been shown to have excellent reliability in diagnosing individuals with drug dependence [32]. Arrests and Incarcerations In addition to self-reports, the official arrest and incarceration records will be obtained for 1 year prior and 1 year post-study enrollment from the Maryland Department of Public Safety and Correctional Services. Risk Assessment Battery (RAB) is a 45-item questionnaire covering substance use and sexual HIV-risk behaviors that has been extensively used with drug-dependent populations [33]. It will be administered at baseline, 6 and 12 month follow-up. The scale’s drug- and sex-risk scores will be used as secondary outcome measures. World Health Organization Quality of Life (WHOQOL-BREF) is a brief 32-item instrument developed by the World Health Organization that has been used in a wide variety of populations internationally [34–36] and has been found to have strong psychometric properties [37, 38]. The WHOQOL-BREF produces scores in four QoL domains: physical, psychological, social, and environmental. The WHOQOL-BREF also contains a single item, which is not incorporated into any of the four scale scores, asking participants to rate their overall QoL on a 5-point Likert-type scale from very poor to very good. It will be administered at baseline, 1, 3, 6 and 12 month follow-up. Methadone Dose Higher methadone maintenance doses in jail have been shown to be associated with higher rates of treatment entry following release from incarceration [39] and improved retention rates in community-based treatment [5]. As such, methadone dose will be recorded at release from the Detention Center and at each follow-up to permit an examination of the relationship between methadone dose and outcomes by treatment Condition. Methadone Treatment Exposure Questionnaire This brief 7 item questionnaire was devised for the present study and inquires whether and when the participant entered methadone treatment following release as well as the number of days of methadone treatment in the community. It will be administered at baseline, 1, 3, 6 and 12 month follow-up. We will validate the self-report with available Opioid Treatment Program records. Economic Form 90 survey was originally designed to collect alcohol use and economic outcome data for alcohol treatment studies [40, 41]. It will be modified to collect data on patients’ residential drug treatment, outpatient, emergency room, and inpatient hospital utilization; and criminal behavior, including number of arrests, severity of offense, and nights incarcerated. Also collected will be labor market information, including employment status, current wage, average hours worked per week, and amount of money received from government sources (e.g., Social Security) and off-the-books earnings. It will be administered at baseline, 3, 6 and 12 month follow-up. Overdose Adverse Event Form A brief questionnaire devised for this study will be administered at 1,3, 6 and 12 month follow-up to determine the number and type of non-fatal overdoses that occur. Substance Abuse Services Cost Analysis Program (SASCAP) Provider costs for each of the study conditions will be estimated using an activity-based costing approach which will allow cost estimation at the service level for study participants. To collect activity-level resource use and cost data, we will modify the SASCAP [42] to collect resource use and cost data for identified activities that are performed within each study condition. Activities will comprise clinically relevant treatment activities (including related support activities), and we will exclude research-related activities from the cost estimation. The SASCAP consists of a provider questionnaire administered once during the intervention phase to collect activity-level resource use and cost data for provider staff and non-labor resources such as contracted services, building space, supplies and materials, and other miscellaneous resources (e.g., utilities). It has been used to reliably estimate the costs of specific treatment activities and the total cost per patient [43–46]. 5.1 Primary Outcomes The study has two primary outcomes: (1) the rate of entry into treatment for opioid use disorder within 30 days from release from incarceration measured by self-report on the methadone treatment exposure questionnaire and (2) frequency of opioid positive urine test results over the 12-month follow-up period, determined from study administered urine screening at each of the four follow-up points. 5.2 Secondary Outcomes The study has a number of secondary outcomes measured as change over time across the four follow-up interviews. These include the presence of opioid use and cocaine use disorder, the drug- and sex-risk scores on the RAB, the specific domain and overall scores on the WHOQOL-BREF [35], the number of days in treatment for opioid use disorder, self-reported illicit opioid and cocaine use and criminal behavior, the number of arrests and incarcerations, and health care utilization. The cost of substance use services, health care utilization, and incarceration will be calculated over the 12 month post-release follow-up period. 5.3 Hypotheses There are several study hypotheses. Hypothesis 1A is that the IM+PN and IM Alone Conditions will have higher rates of treatment entry and greater retention, and concomitant lower rates of illicit opioid and cocaine use and of meeting DSM-5 criteria for opioid and cocaine use disorders, HIV-risk behavior, criminal behavior, arrests, and days of incarcerations than the ETAU Condition. We are not aware of any randomized trial comparing methadone treatment entry rates for jail-based methadone maintenance, with or without PN, compared to treatment as usual. The rationale for this hypothesis stems from a longitudinal non-random assignment study from the Rikers Island MMT Program that has shown better treatment entry rates post-release for inmates started on MMT rather than provided with detoxification in jail [6]. Hypothesis 1B is that the IM+PN condition will have superior outcomes compared to the other study conditions. This hypothesis is supported by research showing that PN is associated with higher rates of treatment entry than no PN in non-jail and non-prison samples of opioid-addicted adults [22]. Hypothesis 2A is that women in the IM+PN condition will have superior outcomes to men in the IM+PN condition, while hypothesis 2B is that women in the IM alone and ETAU will have poorer outcomes than men in those two conditions. These hypotheses stem from the few studies of reentry from jail for men and women treated with methadone that have shown that women are less likely to enter and remain in treatment [6]. In contrast, the PN literature shows that women respond well to this intervention [47, 48]. Therefore, we hypothesize that women will respond better than men to IM+PN. 6.0 Statistical Analysis 6.1 Explanatory Variables There will be a single treatment variable with three conditions: Intervention Condition [IM+PN v. IM Alone v. ETAU); a single moderator variable: Participant Gender, and three predictor variables – participant age, prior methadone maintenance treatment (yes v. no), and self-reported cocaine use at baseline. The predictor variables were chosen because of their association with outcomes in prior research in community based methadone treatment [49–53]. Finally, the “repeated factor” for all outcome variables measured repeatedly (see Table 1) will be assessment Time point, allowing for evaluation of both differential course and impact of the interventions. 6.2 Planned Contrasts It is possible to construct two orthogonal, single degree of freedom planned contrasts that directly test Hypothesis 1A and 1B, respectively. Contrast 1A will compare the two IM treatment conditions (IM+PN and IM Alone, pooled) to the ETAU condition, directly addressing the question of the differential effectiveness of some form of IM treatment in comparison to ETAU. Contrast 1B will compare the IM+PN condition to the IM alone condition, directly addressing the question of the relative effectiveness of adding PN to IM treatment. Similarly, it is possible to construct two orthogonal, single df planned contrasts that directly test Hypothesis 2A and 2B, respectively. Contrast 2A will compare males and females in the IM+PN condition, directly addressing the question of the differential effectiveness of IM+PN treatment for males and females. Contrast 2B will compare males and females in the IM Alone and ETAU conditions, pooled, directly addressing the question of the differential effectiveness of non-PN treatments for males and females. Given that both the course and impact of treatment are important to evaluate, planned contrasts can be examined both as each interacts with the “repeated factor” of Time (when there is a repeated factor), and as a “simple effect” at a given time point, in the event either Planned Contrast X Time interaction subeffect proves to be significant. 6.3 Economic Evaluation Using the collected cost data, we will derive total costs and costs per participant, and costs for specific services for each intervention. The total provider cost for each of the interventions will be the sum across each activity of: (1) staff labor costs (e.g., time spent performing intervention activities); (2) costs of building space; (3) costs of any equipment; (4) costs of medication; (5) costs of any supplies or materials; and (6) costs of any other miscellaneous resources used in the intervention. Taking the mean across participants for a given intervention will yield the mean cost per participant of that intervention. Following our cost estimation, we will conduct a cost-effectiveness analysis of the interventions from the provider perspective. We will combine the estimated provider costs of delivering the interventions with selected intervention effects. Our cost-effectiveness method will follow the approach described in the literature (e.g.,[44, 54]). Starting with the intervention with the smallest cost (or effectiveness), cost-effectiveness ratios will be computed for each intervention relative to the next most expensive option after eliminating intervention options that are dominated by other interventions [54]. An intervention may be either strictly dominated (higher cost and lower effectiveness than another option) or weakly dominated (higher cost-effectiveness ratio than a more effective option). Separate cost-effectiveness analyses will be performed for each selected outcome. Primary outcomes for the cost-effectiveness analysis include: (1) percentage of participants without opioid use disorder diagnosis at 12-month follow-up (based on the DSM-5 Opioid Use Disorder Diagnosis; and (2) number of days of opioid use in the past 30 days assessed at the 12-month follow-up. Secondary outcomes that we will also examine include: (1) number of days incarcerated; and (2) percentage of participants not engaging in HIV-related risky behavior (as measured by the RAB [33]). Finally, we will conduct a cost-benefit analysis to estimate the economic benefits associated with reductions in criminal activity and criminal justice system costs, improved employment, and reduced health care use (e.g., ER visits, hospitalizations). The difference between the monetized economic benefits and intervention costs represents the net economic benefits of the interventions (or cost savings). 6.4 Sample Size, Power, and Effect Size Power was estimated according to a procedure for general linear models outlined by Stroup [55, 56] as well as by the set correlation method [57, 58], assuming the primary outcome measures follow a normal distribution. Assuming a Type I error rate of .01 and a sample size of 300, the general linear model power estimates (1 – β, where β is the Type II error rate) associated with the hypotheses exceeded .81 in all cases, assuming “small” (.2 of a standard deviation [57]) differences associated with each such effect, even assuming an unstructured covariance matrix for those outcomes measured repeatedly. For the set correlation approach, assuming a Type I error rate of .01 and a sample size of 270 due to attrition in order to remain conservative, an effect size f2=.045 for a Planned Contrast effect for an outcome measured only once, and effect sizes f2=.045, .059, and .064 with a Planned Contrast X Time subeffect for outcomes measured two, four, or five times, respectively, for the hypotheses would yield a power of .9 for that subeffect, where f2 is defined as the ratio of the variance of the means relative to the variance of the observations [57]. These effect sizes, f2, all fall in the “small-to-medium” range, with f2=.02 considered a “small” effect and f2=.15 a “medium” effect [57]. In other words, and imprecisely, under the assumption that the effect in the population was ≥ .045 for a Planned Contrast or ≥ .045, .059, or .064 for the Treatment Condition X Time subeffects, for outcomes measured two, four, or five times, respectively, there is an 90% chance of concluding that effect is significant if α is set to .01 and 270 participants are assessed at 12-month follow-up. 8.0 Discussion Despite the large number of adults with opioid use disorder (OUD) arrested in the US every year, pharmacotherapy for OUD is rarely provided in the over 3,000 US detention centers. After almost 50 years of experience with methadone, only one US jail (in New York City) has reported initiating opioid agonist treatment pharmacotherapy for arrestees with OUD. As mentioned briefly in the introduction, there are numerous barriers to initiating pharmacotherapy for OUD in jail settings. Methadone treatment, unlike buprenorphine or extended release naltrexone, requires special federal and state program licenses. A critical mass of patients is desired to make operating a methadone program in a correctional institution practical, thus making it less likely to open such programs in smaller jails or those with small numbers of opioid dependent inmates. There are also philosophic barriers to providing opioid agonist medications to inmates. Additionally, budgetary considerations are barriers to providing any of these treatments. Corrections budgets may not reap most of the potential benefit of a post-release reduction in criminal behavior and health care costs. The lack of strong cost-benefit economic data of these programs makes it somewhat difficult to convince some policymakers of the utility of these programs. Developing a research database for these treatments with prisoners has been difficult, although not impossible, because of the legacy of concerns regarding human subject protection in vulnerable populations and that it requires approval from the federal Office of Human Research Protection. Despite all of these challenges, the number of correctional institutions providing opioid agonist treatment is increasing throughout the world. To date, prisons in most countries of the European Union, Australia, Canada, China, Iran, Indonesia, Moldova, and Kyrgyzstan have such programs [59]. The rationale for these programs has been the principle of making treatment in the community available to incarcerated individuals, the goal of reducing the spread of HIV infection, and the hopes of reducing recidivism. The present study, by examining the use of methadone without counseling, will provide a potentially practical solution for detention centers that may not have the funds to provide counseling but that already have existing medical infrastructure to administer methadone. Unlike the other FDA-approved medications for treating opioid dependence, methadone itself costs pennies per day. Thus, these cost-considerations might provide some support for the use of interim methadone treatment in jails. Because the low range of reported rates of entry into community-based MMT programs by inmates initiated on methadone at the Rikers Island program (ranging from 14% to 50%) leaves considerable room for improvement, we have chosen to compare IM alone to IM+PN. We expect that patient navigation, by reducing barriers to continued treatment in the community and re-engaging those who drop out over the first 3 months of MMT, will result in the IM+PN condition having higher rates of treatment entry, longer retention in treatment, and thereby have superior treatment outcomes. In designing the present trial, we considered alternative designs. First, we considered using buprenorphine or extended-release naltrexone. Both of these medications have some advantages because they can be used with less regulatory and storage security burdens than methadone. However, they both are considerably more expensive to purchase than methadone. Buprenorphine appears to be much easier to divert during treatment than methadone because it must be administered sublingually, and if diversion is to be minimized, the patient must be observed for 5–10 minutes until the dose is absorbed (in tablet or film form). A study using buprenorphine in a Maryland prison with sentenced prisoners expected to be released within several months found that more than 10% of participants attempted to divert the medication and hence were discontinued from receiving the medication [60]. This rate of attempted buprenorphine diversion was similar to that found by Magura and colleagues in a study at Rikers Island comparing methadone to buprenorphine treatment for sentenced jail inmates [21]. Although extended-release naltrexone has the advantage of being a non-controlled substance, it is relatively expensive and not yet widely available in the community. Nevertheless, there are certainly advantages and disadvantages of these three medications and conceptually it would be advantageous to patients to have all three as available in jails as they are in the community. We decided to include an enhanced treatment-as-usual (ETAU) arm as a comparison because virtually all jails in the US (including in Baltimore) use detoxification with non-opioid medications (or less often with methadone) as their “treatment as usual.” ETAU also includes substance use, overdose, and HIV-prevention information, and a referral to community treatment services, thus exceeding the normal provision of service to this population. These latter elements of ETAU are shared by the other two SOMATICS studies. The economic analysis may be of benefit to policy makers and correctional and public health officials. We anticipate that IM alone will have greater treatment costs than ETAU and that IM+PN will have greater costs than IM alone. However, the greater costs for the two experimental conditions may yield superior outcomes and thereby prove more cost-effective than ETAU. Similarly, cost offset in terms of reduced hospital and criminal justice costs associated with superior outcomes for the two experimental conditions may yield greater net benefits for these conditions compared with ETAU. There are a number of limitations to the study design. First, because the efficacy of methadone treatment was not in question, the use of a placebo was deemed inappropriate, and therefore the Patient Navigation intervention is being compared to an enhanced version of treatment as usual. Second, the study is being conducted in a "real world" jail in an urban community with both a substantial number of arrests and a relatively high prevalence of illicit opioid use. Therefore, findings may not generalize to other jail systems in other localities where the number of arrestees or the prevalence of illicit opioid use or the availability of maintenance treatment in the community could make the utilization of methadone maintenance impractical. For the same reason, the economic analysis may not generalize to other localities. However, in jail facilities where there is a lower prevalence of illicit opioid use among arrestees there may be sufficient time to supervise the use of buprenorphine. Initiating buprenorphine maintenance for opioid dependent arrestees might result in fewer deaths and continued treatment following discharge. We would like to thank the medical team at Wexford Health: Kelly Blizzard, RN, Tabethia Pardoe, RN, Kara Hope, RN and Drs. Woreta, Goodwin, Tessias, Getachew and Luka and the Maryland Department of Public Safety and Correctional Services including Drs. Sharon Baucom and Adaora Odunze; and the Daybreak, Glenwood Life, Man Alive, and REACH Opioid Treatment Programs, for their assistance in making this study possible. This study is being funded by the National Institute on Drug Abuse grant number 2U01DA013636 (PI: Schwartz). ClinicalTrials.gov: NCT02334215 Table 1 Data Collection Schedule and Measures Measures Baseline 1-month 3-month 6-month 12-month Treatment Entry (Methadone Treatment Exposure Form) ♦ Days in Treatment (self-report from EF-90) ♦ ♦ ♦ Illicit Opioid and Cocaine Use (ASI and urine drug testing) ♦ ♦ ♦ ♦ ♦ DSM-5 Criteria of opioid and cocaine use disorder (modified CIDI-2 SAM) ♦ ♦ ♦ ♦ Criminal Behavior (Addiction Severity Index) ♦ ♦ ♦ ♦ ♦ Arrests and Incarcerations: (Self report) ♦ ♦ ♦ ♦ HIV Risk Behavior (Risk Assessment Battery) ♦ ♦ ♦ Overdose Adverse Event Form ♦ ♦ ♦ ♦ Methadone Treatment Exposure Questionnaire ♦ ♦ ♦ ♦ Quality of Life (WHOQOL-BREF) ♦ ♦ ♦ ♦ ♦ Economic Form 90 (Health care utilization) ♦ ♦ ♦ ♦ Substance Abuse Services Cost Analysis Program (SASCAP) † Arrest and incarceration (Criminal Justice Records) ♦ ♦ Note: ASI = Addiction Severity Index; CIDI-2 SAM = Modified Composite International Diagnostic Interview 2 Substance Abuse Module; WHOQOL-BREF = World Health Organization Quality of Life-Brief Form * Participants transitioning directly to other community Methadone Maintenance Programs or buprenorphine treatment will be considered still in treatment. † Consistent with prior research, the SASCAP will be administered once during the treatment phase. Table 2 Consent Quiz INSTRUCTIONS: Please circle T for True or F for False. T or F 1. This study is comparing three ways to treat opiate addiction in the Detention Center. T or F 2. If I am in the study, I can choose which study group I am in. T or F 3. Methadone has no side effects. T or F 4. If I get the methadone detox group I may have a greater chance of overdosing if I relapse     to heroin use. T or F 5. I can drop out of the study once I start. T or F 6. The study does not provide drug abuse counseling services while I am in the Detention     Center. T or F 7. The patient navigator will help every participant in the study. T or F 8. Because I am in the Detention Center, I have to be in the study. T or F 9. Possible side effects of methadone include drowsiness. T or F 10. There are risks to being in the study. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Reference 1 Dolan K Khoei EM Brentari C Stevens A Prisons and drugs: A global review of incarceration, drug use and drug services 2007 Oxford Beckley Foundation 2 Fazel S Bains P Doll H Substance abuse and dependence in prisoners: a systematic review Addiction (Abingdon, England) 2006 101 2 181 191 3 James DJ Profile of Jail Inmates, 2002. Special Report. US Department of Justice Programs Bureau of Statistics 2004 accessed May 12, 2016 http://www.bjs.gov/content/pub/ascii/pji02.txt 4 Johnson RE Chutuape MA Strain EC Walsh SL Stitzer ML Bigelow GE A comparison of levomethadyl acetate, buprenorphine, and methadone for opioid dependence The New Engl. J Med 2000 343 18 1290 1297 11058673 5 Strain EC Bigelow GE Liebson IA Stitzer ML Moderate- vs high-dose methadone in the treatment of opioid dependence: a randomized trial JAMA 1999 281 11 1000 1005 10086434 6 Magura S Rosenblum A Lewis C Joseph H The effectiveness of in-jail methadone maintenance J Drug Issues 1993 23 75 99 7 Office of National Drug Control Policy 2013 Annual report, arrestee drug abuse monitoring program II, Executive Office of the President 2014 Washington, DC 8 Bird SM Hutchinson SJ Male drugs-related deaths in the fortnight after release from prison: Scotland, 1996–99 Addiction (Abingdon, England) 2003 98 2 185 190 9 Farrell M Marsden J Acute risk of drug-related death among newly released prisoners in England and Wales Addiction (Abingdon, England) 2008 103 2 251 255 10 A.J. Assocation Statistics of Note accessed March 10, 2015 https://members.aja.org/About/StatisticsOfNote.aspx 11 Nunn A Zaller N Dickman S Trimbur C Nijhawan A Rich JD Methadone and buprenorphine prescribing and referral practices in US prison systems: results from a nationwide survey Drug Alcohol Depend 2009 105 1–2 83 88 19625142 12 Trigg BG Dickman SL Medication-assisted therapy for opioid-dependent incarcerated populations in New Mexico: statewide efforts to increase access Substance abuse : official publication of the Association for Medical Education and Research in Substance Abuse 2012 33 1 76 84 13 Center for Substance Abuse Treatment, Substance Abuse Treatment for Adults in the Criminal Justice System, Substance Abuse and Mental Health Services Administration (Treatment Improvement Protocol [TIP] Series, No. 44) DHHS Publication No. (SMA) 05-4056 2005 Rockville, MD 14 McKenzie M Nunn A Zaller ND Bazazi AR Rich JD Overcoming obstacles to implementing methadone maintenance therapy for prisoners: implications for policy and practice J Opioid Manag 2009 5 4 219 227 19736902 15 Rich JD McKenzie M Shield DC Wolf FA Key RG Poshkus M Clarke J Linkage with methadone treatment upon release from incarceration: a promising opportunity J Addict. Disease 2005 24 3 49 59 16186082 16 European Monitoring Centre for Drugs and Drug Addiction (EMCDDA), Annual report: The State of the drugs problem in Europe 2010 Portugal EMCDDA accessed March 17, 2015 http://www.emcdda.europa.eu/online/annual-report/2010 archived by Webcite at http://www.webcitation.org/5yG72bp1G 17 Farrell M Ward J Mattick R Hall W Stimson GV des Jarlais D Gossop M Strang J Methadone maintenance treatment in opiate dependence: a review BMJ 1994 309 6960 997 1001 7950725 18 Schwartz RP Kelly SM O'Grady KE Gandhi D Jaffe JH Interim methadone treatment compared to standard methadone treatment: 4-Month findings J Subst. Abuse Treat 2011 41 1 21 29 [PMCID: PMC3110526] 21353445 19 Schwartz RP Kelly SM O'Grady KE Gandhi D Jaffe JH Randomized trial of standard methadone treatment compared to initiating methadone without counseling: 12-month findings Addiction (Abingdon, England) 2012 107 5 943 952 PMCID: PMC3319854 20 Gruber VA Delucchi KL Kielstein A Batki SL A randomized trial of 6-month methadone maintenance with standard or minimal counseling versus 21-day methadone detoxification Drug Alcohol Depend 2008 94 1–3 199 206 18243585 21 Magura S Lee JD Hershberger J Joseph H Marsch L Shropshire C Rosenblum A Buprenorphine and methadone maintenance in jail and post-release: a randomized clinical trial Drug Alcohol Depend 2009 99 1–3 222 230 18930603 22 Gardner LI Metsch LR Anderson-Mahoney P Loughlin AM del Rio C Strathdee S Sansom SL Siegal HA Greenberg AE Holmberg SD A.T.A.A.S.S. Group Efficacy of a brief case management intervention to link recently diagnosed HIV-infected persons to care AIDS (London, England) 2005 19 4 423 431 23 Sorensen JL Masson CL Delucchi K Sporer K Barnett PG Mitsuishi F Lin C Song Y Chen T Hall SM Randomized trial of drug abuse treatment-linkage strategies J. Consult. Clin. Psychol 2005 73 6 1026 1035 16392976 24 Bradford JB Coleman S Cunningham W HIV System Navigation: an emerging model to improve HIV care access AIDS patient care and STDs 21 Suppl 2007 1 S49 S58 25 Mejta C Bokos PJ Mickenberg J Maslar ME Senay E Improving substance abuse treatment access and retention using a case management approach J Drug Issues 1997 27 2 329 340 26 Scott CK Dennis ML Foss MA Utilizing Recovery Management Checkups to shorten the cycle of relapse, treatment reentry, and recovery Drug Alcohol Depend 2005 78 3 325 338 15893164 27 Freeman HP Muth BJ Kerner JF Expanding access to cancer screening and clinical follow-up among the medically underserved Cancer practice 1995 3 1 19 30 7704057 28 Copeland J A qualitative study of barriers to formal treatment among women who self-managed change in addictive behaviours J Subst Abuse Treat 1997 14 2 183 190 9258863 29 McLellan AT Kushner H Metzger D Peters R Smith I Grissom G Pettinati H Argeriou M The Fifth Edition of the Addiction Severity Index J Subst Abuse Treat 1992 9 3 199 213 1334156 30 Andrews G Peters L The psychometric properties of the Composite International Diagnostic Interview Social psychiatry and psychiatric epidemiology 1998 33 2 80 88 9503991 31 Forman RF Svikis D Montoya ID Blaine J Selection of a substance use disorder diagnostic instrument by the National Drug Abuse Treatment Clinical Trials Network J Subst Abuse Treat 2004 27 1 1 8 15223087 32 Horton J Compton W Cottler LB Reliability of substance use disorder diagnoses among African-Americans and Caucasians Drug Alcohol Depend 2000 57 3 203 209 10661671 33 Metzer DS Navaline HA Woody GE Assessments of substance abuse: HIV risk assessment battery (RAB), Encyclopedia of Drugs Alcohol, and Addictive Behavior 2001 34 Noerholm V Groenvold M Watt T Bjorner JB Rasmussen NA Bech P Quality of life in the Danish general population--normative data and validity of WHOQOL-BREF using Rasch and item response theory models Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation 2004 13 2 531 540 35 The World Health Organization Development of the World Health Organization WHOQOL-BREF quality of life assessment The WHOQOL Group, Psychological medicine 1998 28 3 551 558 9626712 36 The World Health Organization Quality of Life (WHOQOL) - Bref 2004 Geneva, Switzerland World Health Organization 37 O'Carroll RE Smith K Couston M Cossar JA Hayes PC A comparison of the WHOQOL-100 and the WHOQOL-BREF in detecting change in quality of life following liver transplantation Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation 2000 9 1 121 124 38 Skevington SM Lotfy M O'Connell KA W. Group The World Health Organization's WHOQOL-BREF quality of life assessment: psychometric properties and results of the international field trial. A report from the WHOQOL group Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation 2004 13 2 299 310 39 Wickersham JA Zahari MM Azar MM Kamarulzaman A Altice FL Methadone dose at the time of release from prison significantly influences retention in treatment: implications from a pilot study of HIV-infected prisoners transitioning to the community in Malaysia Drug Alcohol Depend 2013 132 1–2 378 382 23414931 40 Miller WR Del Boca FK Measurement of drinking behavior using the Form 90 family of instruments J. Stud. Alcohol 1994 12 112 118 41 Scheurich A Muller MJ Anghelescu I Lorch B Dreher M Hautzinger M Szegedi A Reliability and validity of the Form 90 interview European Addict Res 2005 11 1 50 56 15608472 42 Zarkin GA Dunlap LJ Homsi G The Substance Abuse Services Cost Analysis Program (SASCAP): A new method for estimating drug treatment services costs Eval Program Plann 2004 27 1 35 43 43 Zarkin GA Bray JW Mitra D Cisler RA Kivlahan DR Cost methodology of COMBINE J Stud Alcohol 2005 15 50 55 discussion 33 44 Zarkin GA Bray JW Aldridge A Mitra D Mills MJ Couper DJ Cisler RA Cost and cost-effectiveness of the COMBINE study in alcohol-dependent patients Arch Gen Psychiatry 2008 65 10 1214 1221 18838638 45 Zarkin GA Bray JW Davis KL Babor TF Higgins-Biddle JC The costs of screening and brief intervention for risky alcohol use J Stud Alcohol 2003 64 6 849 857 14743949 46 Dunlap LJ Zarkin GA Bray JW Mills M Kivlahan DR McKay JR Latham P Tonigan JS Revisiting the cost-effectiveness of the COMBINE study for alcohol dependent patients: the patient perspective Medical care 2010 48 4 306 313 20355261 47 Burhansstipanov L Dignan MB Schumacher A Krebs LU Alfonsi G Apodaca CC Breast screening navigator programs within three settings that assist undeserved women J Cancer Educat 2010 25 2 247 252 48 Han HR Lee H Kim MT Kim KB Tailored lay health worker intervention improves breast cancer screening outcomes in non-adherent Korean-American women Health Educ Res 2009 24 2 318 329 18463411 49 Schwartz RP Kelly SM O'Grady KE Mitchell SG Brown BS Antecedents and correlates of methadone treatment entry: a comparison of out-of-treatment and in-treatment cohorts Drug Alcohol Depend 2011 115 1–2 23 29 PMCID: PMC3059350 21126830 50 Saxon AJ Wells EA Fleming C Jackson TR Calsyn DA Pre-treatment characteristics, program philosophy and level of ancillary services as predictors of methadone maintenance treatment outcome Addiction (Abingdon, England) 1996 91 8 1197 1209 51 Deck D Carlson MJ Retention in publicly funded methadone maintenance treatment in two Western States J Behav Health Serv Res 2005 32 1 43 60 15632797 52 Shah NG Celentano DD Vlahov D Stambolis V Johnson L Nelson KE Strathdee SA Correlates of enrollment in methadone maintenance treatment programs differ by HIV-serostatus AIDS (London, England) 2000 14 13 2035 2043 53 Booth RE Corsi KF Mikulich SK Improving entry to methadone maintenance among out-of-treatment injection drug users J Subst. Abuse Treat 2003 24 4 305 311 12867204 54 Drummond MF Sculpher MJ Torrance GW O'Brien BJ Stoddart GL Methods for the Economic Evaluation of Health Care Programmes 2005 Third New York Oxford University Press 55 Stroup WW Mixed model procedures to assess power, precision, and sample size in the design of experiments 1999 Lincoln, NE University of Nebraska, American Statistical Association 19 24 56 Littell RC Millike GA Stroup WW Wolfinger RD Schabenberger O SAS for mixed models 2006 Cary, NC SAS Institute, Inc. 57 Cohen J Statistical power analysis for the behavioral sciences 1988 Second Hillsdale, NJ LEA 58 Borenstein M Cohen J Statistical power analysis 1988 Hillsdale, NY Erlbaum 59 Kastelic A Pont J Stöver H Opioid Substitution Treatment in Custodial Settings A Practical Guide 2008 Oldenburg, BIS-Verlag 60 Gordon MS Kinlock TW Schwartz RP Fitzgerald TT O'Grady KE Vocci FJ A randomized controlled trial of prison-initiated buprenorphine: prison outcomes and community treatment entry Drug Alcohol Depend 2014 142 33 40 24962326

PMC005xxxxxx/PMC5068189.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101489144 35623 Circ Cardiovasc Genet Circ Cardiovasc Genet Circulation. Cardiovascular genetics 1942-325X 1942-3268 27625337 5068189 10.1161/CIRCGENETICS.116.001431 EMS69925 Article Combination of Whole Genome Sequencing, Linkage and Functional Studies Implicates a Missense Mutation in Titin as a Cause of Autosomal Dominant Cardiomyopathy with Features of Left Ventricular Non-Compaction Hastings Robert MBChB DPhil MRCP 1 de Villiers Carin PhD 1 Hooper Charlotte PhD 1 Ormondroyd Liz PhD, MSc 1 Pagnamenta Alistair PhD 23 Lise Stefano PhD 3 Salatino Silvia PhD 3 Knight Samantha JL PhD, CBiol, MSB, FRCPath 23 Taylor Jenny C. PhD 23 Thomson Kate L. BSc, FRCPath 14 Arnold Linda MSc 1 Chatziefthimiou Spyros D. PhD 5 Konarev Petr V. PhD 56 Wilmanns Matthias PhD 5 Ehler Elisabeth PhD 7 Ghisleni Andrea MSc 7 Gautel Mathias MD, PhD 7 Blair Edward BMSc, MBChB 4 Watkins Hugh MD, PhD, FRCP 1 Gehmlich Katja PhD 1 1 Division of Cardiovascular Medicine in the Radcliffe Department of Medicine, University of Oxford; BHF Centre of Research Excellence 2 NIHR Biomedical Research Centre Oxford, University of Oxford, Oxford, United Kingdom 3 Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, United Kingdom 4 Department of Clinical Genetics, Churchill Hospital, Oxford University NHS Trust, Oxford, United Kingdom 5 European Molecular Biology Laboratory, Hamburg, Germany 6 Laboratory of Reflectometry and Small-Angle Scattering, A.V.Shubnikov Institute of Crystallography, Russian Academy of Sciences, Moscow, Russian Federation 7 Randall Division of Cell and Molecular Biophysics and Cardiovascular Division, King’s College London BHF Centre of Research Excellence, London, United Kingdom Correspondence: Katja Gehmlich, Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, Level 6, West Wing John Radcliffe Hospital, Headley Way, Oxford OX3 9DU, United Kingdom, Tel: ++44 1865 234902, Fax: ++44 1865 234681, katja.gehmlich@cardiov.ox.ac.uk 3 10 2016 13 9 2016 10 2016 20 10 2016 9 5 426435 This file is available to download for the purposes of text mining, consistent with the principles of UK copyright law. Background High throughput next generation sequencing techniques have made whole genome sequencing accessible in clinical practice, however, the abundance of variation in the human genomes makes the identification of a disease-causing mutation on a background of benign rare variants challenging. Methods and Results Here we combine whole genome sequencing with linkage analysis in a three-generation family affected by cardiomyopathy with features of autosomal dominant left-ventricular non-compaction cardiomyopathy. A missense mutation in the giant protein titin is the only plausible disease-causing variant that segregates with disease amongst the eight surviving affected individuals, with interrogation of the entire genome excluding other potential causes. This A178D missense mutation, affecting a conserved residue in the second immunoglobulin-like domain of titin, was introduced in a bacterially expressed recombinant protein fragment and biophysically characterised in comparison to its wild-type counterpart. Multiple experiments, including size exclusion chromatography, small angle X-ray scattering and circular dichroism spectroscopy suggest partial unfolding and domain destabilisation in the presence of the mutation. Moreover, binding experiments in mammalian cells show that the mutation markedly impairs binding to the titin ligand telethonin. Conclusions Here we present genetic and functional evidence implicating the novel A178D missense mutation in titin as the cause of a highly penetrant familial cardiomyopathy with features of left-ventricular non-compaction. This expands the spectrum of titin’s roles in cardiomyopathies. It furthermore highlights that rare titin missense variants, currently often ignored or left un-interpreted, should be considered to be relevant for cardiomyopathies and can be identified by the approach presented here. genetics basic science research left ventricular noncompaction cardiomyopathy whole genome sequencing titin telethonin missense mutation Introduction Cardiomyopathies (CM) are a diverse group of diseases affecting the heart muscle 1; many of them are inherited and transmitted in autosomal dominant patterns. The first cardiomyopathy genes were identified by genome-wide linkage analysis in large families 2. In practice, however, the small size of most families, or even the availability of members of larger families, often limits the power of linkage analysis. Recently, high throughput next generation sequencing (NGS) techniques have become widely accessible, making whole genome sequencing (WGS) cost- and time-effective. However, the abundance of variation in the human genome 3 makes it difficult to distinguish rare benign variants from rare disease-causing mutations in an isolated individual, even with growing knowledge of variants in population cohorts (e.g. >60,000 sequenced exomes in the ExAC database, http://exac.broadinstitute.org/). NGS poses, therefore, a significant clinical challenge: the capability to assess variants as pathogenic lags significantly behind variant identification, especially for non-synonymous point mutations 4, 5. Algorithmic predictors are currently unable to accurately assess their exact impact on protein-protein interactions or even on protein folding. Experimental validation of genetic variants is therefore an increasingly indispensable component of NGS discoveries. In the current study, we combine WGS with linkage analysis in a medium-sized family affected by cardiomyopathy with features of left-ventricular non-compaction cardiomyopathy (LVNC). By performing WGS in two family members, filtering against variants seen in normal population cohorts and using linkage information derived from single nucleotide polymorphism (SNP) arrays of 13 family members, we could identify a missense variant in the titin gene (TTN) as the most plausible cause of disease in the family. Functional data, generated from biophysical and protein binding experiments on this titin missense variant provide further support of a causative role in cardiomyopathy through domain misfolding and destabilisation, resulting in impaired binding to the ligand telethonin (also known as t-cap). Methods Clinical Evaluation The study was approved by the Oxfordshire Research Ethics Committee B (REC Ref 09/H0605/3) and all subjects gave informed consent. A three-generational family with history of cardiomyopathy was recruited. Clinical assessment and genetic studies were performed in available family members, who had clinical examination, ECG, echocardiography (with contrast agent where appropriate) and cardiac MRI if possible. Diagnosis of cardiomyopathy was based on established criteria. The diagnosis of LVNC was based on published criteria from echocardiographic and/or cardiac MR imaging 6, 7: the compaction ratio (CR), i.e. the ratio of the thickness of non-compacted to compacted myocardium >2.3 measured on MRI in diastole, or >2.0 on echocardiography in systole was used to diagnose LVNC. Genetic Studies SNP array genotyping was performed using the Illumina HumanCytoSNP-12v1 BeadChip (Illumina, San Diego, CA), containing nearly 300,000 genetic markers, according to the manufacturer’s protocols. A refined subset of roughly 24,000 SNPs in approximate linkage equilibrium was generated using the software PLINK v1.07 8 and the HapMap genotype file available from the PLINK website (http://pngu.mgh.harvard.edu/purcell/plink/). Linkage analysis of the SNP subset was performed using MERLIN v1.1.2 9, specifying an autosomal dominant disease model. Genomic intervals with LOD scores > 0, compatible with segregation of variants in these regions, were selected for downstream analyses. WGS was performed on genomic DNA extracted from peripheral blood as part of the WGS500 project as described previously 10. Sequence reads from the affected individuals were mapped to the human reference genome (hs37d5 version of build 37) using STAMPY 11. Duplicate reads were removed with PICARD (http://broadinstitute.github.io/picard/). The software Platypus (version 0.8.1, default parameters) 12 was used jointly on the two .bam files in order to call SNPs and short (< 50 bp) indels across both samples. All the 5,946,161 identified variants were annotated with an in-house pipeline based on the Variant Effect Predictor (VEP) Ensembl framework (version 77) 13. A number of additional databases were used to integrate the information provided by VEP (Table S1). Known associations with diseases were screened using HGMD (http://www.hgmd.cf.ac.uk/ac/index.php) and ClinVar 14. Variants were filtered by in-house Python scripts based on criteria outlined in Table S1 (steps 1-10), followed by manual inspection (steps 11-13). The variants remaining after step 10 are documented in Results and in Tables S2, S3. Confirmatory Sanger sequencing was performed with the primers listed in Table S4. Both SNP and WGS data were interrogated also for clinically relevant copy number variants (CNVs) using Nexus Copy Number 7.5.2 Discovery Edition (BioDiscovery, Hawthorne, CA; see Supplementary methods). Functional characterisation of the titin missense variant The mutation was introduced into human titin Z1Z2 constructs (amino acids 1-196, accession no. ACN81321.1) for bacterial and mammalian expression using Quikchange II XL (Agilent) with primers given in Table S4. Bacterial expression and purification was performed as previously described 15. Size exclusion chromatography - Tridetector analysis (light scattering, refractive index, and UV absorbance), small angle X-ray scattering (SAXS) experiments, circular dichroism spectroscopy, and thermolysin digests were essentially performed as described 15–18 and experimental details are given in Suppl. Material. NRC cultures were established and transfected 16 using hemagglutinin(HA)-tagged expression constructs and counter-stained for titin T12 epitope 19 or telethonin (mouse monoclonal antibody, Santa Cruz) 48 hrs post transfection and analysed by confocal microscopy. GST pulldown assays were performed as described 20 using mammalian expression constructs for telethonin amino acids 1-90 and 1-167 fused to GST, and titin Z1Z2 fused to GFP (pEGFP-N1, Clontech) in transfected COS-1 cells. Förster Resonance Energy Transfer (FRET) experiments from transfected COS-1 cells and the assessment of reduced protein stability in NRC and COS-1 cells are described in the Suppl. Material. Results The proband was a 20yr old male (II-3 in Figure 1A) who died suddenly in hospital in 1970 having presented with rapidly decompensating congestive heart failure; at post mortem his heart (680 g) had evidence of dilatation and both macroscopic and microscopic hypertrophy but no myocyte disarray. His brother (II-4) was later found to have an enlarged heart with wall thickness at the upper limit of normal and marked hypertrabeculation. The proband’s sister (II-2) presented with a non ST-elevation myocardial infarct due to coronary embolus at the age of 61. LVNC with mild LV dilatation and apical hypertrophy was diagnosed at this time (Figure 1B, C). Cascade screening identified the same condition in further family members with consistent clinical features of adult onset cardiomyopathy with features of LVNC. Five affected family members had sufficient non-compaction to meet the diagnostic criteria for LVNC while three others with early or mild disease had lesser extent of hypertrabeculation but clear evidence of cardiomyopathy with LV dilatation and/or systolic dysfunction (Figures 1A, S1, Table 1). Aside from the proband who had advanced congestive failure, there were no arrhythmic features in any affected family member, nor were there any extra-cardiac (e.g. neuromuscular) manifestations. Identification of TTN mutation A178D segregating with disease Affected first cousins III-1 and III-4 were selected for WGS. Sequencing was performed by Illumina Cambridge as 100bp paired-end reads to a mean coverage of 56.9x and 52.0x respectively, such that 99 % of the genome was covered at 20x or more in both samples, identifying 5,946,161 variants shared by the two individuals. In addition, SNP arrays were performed on all individuals of the family (except II-3 and III-2, Figure 1A). Neither the SNP array nor WGS data revealed likely causative CNVs. Genomic regions identical by descent were identified through linkage analysis (see Methods, Figure S2) and out of the 100,789 candidate variants within the three linkage regions (on chromosomes 2, 9 and 16), potentially pathogenic ones were selected based on an autosomal dominant model, caused by a rare heterozygous mutation. Variants were filtered accordingly by in-house Python scripts and the remaining six variants were manually inspected (Table S2). Four of them were excluded: one is assumed to be an artefact due to an incorrect transcript being present in Ensembl and another variant did not segregate with disease in the family; two splice variants were predicted to be silent (at positions -5 and -3 of a 3' splice junction, respectively, for details see Table S3). Only two final candidate variants were considered conceivably linked to the phenotype: missense changes in PDP2 and TTN, respectively (Table S2). PDP2 codes for pyruvate dehyrogenase phosphatase catalytic subunit 2 and has low expression levels in the heart. Although the change E316K is predicted to be damaging by Polyphen and SIFT algorithms (Table S2), a heterozygous loss-of-function in this enzyme would not be expected to produce a phenotype, and indeed heterozygous loss-of-function mutations in PDP1 are clinically silent 21. The variant is not plausible as a cause of a penetrant dominant disorder because it is found six times in 121,412 alleles in the ExAC database. Six instances would equal at least 10 % of all expected LVNC cases in ExAC, assuming a maximal prevalence of 1:1,000 for the disease 22. This appears an implausibly high percentage for a novel, unpublished disease-causing variant. In support, in the two largest clinical cardiomyopathy cohorts published to date, the most common reported pathogenic variant (MYBPC3, p.Arg502Trp) detected in 104 out of 6179 HCM cases (1.7%, 95CI 1.4-2.0%), was only observed 3 times in ExAC (3/120,674) with all other pathogenic variants for HCM or DCM being present 0 or 1 times only 23. The second variant is found in TTN, the gene which codes for titin, an abundant skeletal muscle and heart specific protein with crucial functions 24, 25 (and reviewed in 26). Mutations in titin have been associated with CM and skeletal myopathy (reviewed in 27). The identified missense variant c.533C>A in TTN, which codes for a p.A178D change at the amino-acid level, is absent in ExAC. Sanger sequencing confirmed the co-segregation of the heterozygous mutation with disease in all affected individuals of the family (Figures 1, S3A, LOD score 2.1). Thus comprehensive whole genome analysis reveals this as the most plausible causative mutation in the family. Functional studies Prediction of deleterious effects of the mutation Each single molecule of the giant protein titin spans half a sarcomere from the Z-disk to the M-band 28. The first two Immunoglobulin-like (Ig) domains (Z1Z2) of titin are located in the Z-disk and form a super-stable complex with telethonin 29. The A178 position is evolutionarily very well conserved back to zebrafish and lamprey. Additionally, A178 is located in a highly conserved structural section (Figure S3B), the β-strand F of the second Ig-domain of Z1Z2, neighbouring the β-strand G of titin Z2 which forms a strong and extended interaction with the β-strands of telethonin (30, Figure 2A). The A178D mutation is predicted to directly affect the β-strands B and C as well as the loop connecting the β-strands B and C due to steric hindrance of D178 with V127 and P133 respectively (Figure 2B). Thus, the insertion of a charged residue in this position is likely to have significant impact on the secondary structure of this domain and could potentially cause misfolding of the protein. Altered protein characteristics of purified titin Z1Z2 A178D recombinant fragment To assess how the A178D mutation affects the folding and stability of the protein, recombinant titin Z1Z2 WT and A178D were expressed in E. coli and purified under native conditions. Of note, the yield of the soluble protein fraction was consistently lower for A178D compared to WT preparations, despite equal total expression levels (data not shown). Circular dichroism (CD) spectroscopy demonstrated a typical β-sheet signature for WT Z1Z2 (Figure 3A). In contrast, the spectrum for Z1Z2 A178D differs significantly: Although the characteristic negative band at 216 nm is still present, but slightly shifted, there was no significant positive band at around 200 nm. The absence of this band, associated with β-sheet conformation, and the presence of a negative peak at around 198 nm, characteristic of random coil structures, indicate that the Z1Z2 A178D mutant is partially unfolded. In support, thermal denaturation experiments for Z1Z2 A178D showed high fluorescence signal already at low temperatures, suggesting solvent exposed hydrophobic residues due to partial unfolding. No melting temperature can be deducted for titin Z1Z2 A178D, in contrast to the WT protein, which has a melting temperature of 62 ºC, typical for Ig domains (Figure S4). SAXS experiments confirmed the presence of unfolded parts/flexible domains in Z1Z2 A178D, as shown by the Kratky plot (Figure S5A), whereas Z1Z2 WT displays a typical profile for folded structures. The domain destabilisation as a consequence of partial unfolding is evidenced by the formation of higher oligomers (approx. 20-mers) for the Z1Z2 A178D mutant in vitro: Size exclusion chromatography and Tridetector analysis revealed that in contrast to the monomeric Z1Z2 WT, the A178D mutant eluted in two peaks, corresponding predominantly to higher molecular aggregates and to a lesser extent to dimeric protein (Figure 3B, Table 2). SAXS measurements also confirmed that Z1Z2 WT is monomeric, whereas Z1Z2 A178D is found in a higher oligomeric state (Figure S5B, Table 2). In conclusion, the mutation A178D leads to partial misfolding of bacterially expressed Z1Z2 protein fragment. Reduced stability of titin Z1Z2 A178D as a consequence of the partial misfolding When performing denaturing gel electrophoresis, a degradation product was observed exclusively for Z1Z2 A178D preparations (arrowhead in Figure 4A) and upon thermolysin treatment, only Z1Z2 A178D showed rapid degradation, while Z1Z2 WT was resistant to the protease treatment (Figure 4B). In addition, Z1Z2 A178D showed reduced stability when expressed in neonatal rat cardiomyocytes and COS-1 cells (Figures 4C, S6), suggesting that the mutation destabilises Z1Z2 also in a physiological, cellular environment. However, formation of large aggregates was not observed in transfected cells expressing Z1Z2 A178D (Figures 4D, S7). Impaired binding to telethonin Localisation of transfected Z1Z2 was not altered in the presence of the A178D mutation (Figure 4D). To assess the consequences of the mutation on binding telethonin, semi-quantitative GST-pulldown assays were performed with titin Z1Z2 and telethonin co-expressed in mammalian cells. Z1Z2 A178D showed impaired binding to two telethonin constructs (Figure 5A, B). The interaction between titin and telethonin was further quantified in FRET experiments, where close proximity of proteins in a complex allows energy transfer from Cyan Florescent Protein (CFP) to Yellow Fluorescent Protein (YFP) between two fusion protein constructs 31. By introducing the A178D mutation into a Z1Zr3-CFP construct, FRET efficiency to telethonin-YFP was almost abolished (Figure 5C, D), validating and quantifying the observation that A178D impairs binding to telethonin in the cellular context. Taken together, our functional data suggest that the A178D mutant may affect protein folding, stability and impairs binding to telethonin, thus supporting its pathogenic potential. Discussion In this study, we present a three-generation family with multiple individuals affected by cardiomyopathy with features of LVNC, systolic impairment and an autosomal dominant inheritance pattern. Of note, the affected family members show a consistent phenotype with prominent hypertrabeculation as the main abnormality in the majority; this is relatively unusual as it is more typical to see LVNC in individual members of families with other forms of cardiomyopathy. We employed a combination of WGS in two affected individuals and linkage analysis in 13 family members; this approach identified only two rare candidate variants across the whole genome that segregated with the autosomal dominant cardiomyopathy. Since one of the identified genes (PDP2) is barely expressed in the heart, and the variant appears in implausible high numbers in the ExAC database, it is extremely unlikely to be disease-causative. In contrast, titin, the gene affected by the other missense variant (TTN p.A178D), has crucial functions in the heart and is a known disease gene for cardiomyopathies (see below). Despite the fact that the family is too small for traditional genome-wide linkage analysis to identify the genetic cause of the disease (the LOD score of 2.1, i.e. odds ratio 1:125, is well below the threshold of 3.0, i.e. odds ratio 1:1000), interrogation of the entire genome adds substantial weight to a likely causative role of the titin missense mutation for disease: no other plausible mutations, including larger genomic re-organisations (CNVs), were detected in any other genes in the linkage regions and the remainder of the genome is excluded by negative LOD scores. Titin has been implicated in cardiac and skeletal muscle disease, occasionally involving a combination of both. Mutations in this gene have been described in various forms of CM, such as Dilated CM, Arrhythmogenic Right Ventricular CM, Hypertrophic CM and Restrictive CM (reviewed in 27). Truncating variants in titin (TTNtv) are the most frequent genetic finding in idiopathic Dilated CM, being present in 15-25 % of the cases 32 and are also frequent in Peripartum CM (15 %) 33. However, penetrance appears to be low, as TTNtv are also found in approx. 1 % of normal populations and hence the large majority of carriers do not manifest with disease 34. More recent work 35 showed that Dilated CM causing TTNtv are enriched in the sarcomeric A-band region, whereas TTNtv found in control cohorts tend to spare the A-band region and are in exons with low usage in cardiac transcripts. An internal promotor in titin rescuing TTNtv N-terminally of the A-band region may explain this phenomenon 36. Titin missense mutations have been identified in Dilated and Hypertrophic CM cohorts 4, 37, 38. A causative role for TTN p. W976R in Dilated CM is well supported by co-segregation within a large family and functional data 39, 40. However, generally, titin missense mutations are challenging to interpret, as rare benign variants are common in normal population cohorts. In the ExAC database, more than a third of the individuals carry a rare missense variant in titin (21,939 missense variants with <0.01 % allelic frequency in 58,687 exomes), and although a proportion of these may represent recessive pathogenic alleles 27, only a very small fraction will be disease-causing with dominant inheritance. Hence, clinical practitioners require co-segregation information to assign causality as bioinformatic prediction tools can only give probabilistic data 4, 37. As we document here, interrogation of the entire genome combined with linkage analysis can help to narrow down lists of potential causative variants, even in small families. Our finding of TTN p.A178D in a family with features of LVNC expands the spectrum of titinopathies: to our knowledge, this is the first report of a titin missense mutation implicated in cardiomyopathy with predominant features of LVNC and one of the first titin missense mutations supported by robust genome-wide genetics and detailed functional data. The latter suggests a likely pathogenic role of titin A178D by a) evidence of protein degradation, partial unfolding and domain destabilisation in vitro, b) protein destabilisation in two cellular systems and c) altered binding properties to the ligand telethonin. Although extrapolations from such in vitro experiments on isolated domains to the full length giant protein are not without uncertainty, such parameters will be useful complements in the future studies of other TTN missense variants. It is currently unclear how this particular mutation leads to this distinct phenotype, and more insight into the biology of Z-disk titin is needed to understand the underlying disease pathways. This will be addressed with the help of model organisms 36, 41 or patient-derived induced pluripotent stem cell derived cardiomyocytes 40, focussing on the titin-telethonin complex 29 and its downstream signalling targets 42 in future work. Supplementary Material Supplemental Material Acknowledgments” We thank Stephan Lange (UCSD) for a titin Z1Z2 expression constructs. SAXS data were collected at the beamline P12, operated by EMBL, Hamburg unit, at the PETRA III storage ring (DESY, Hamburg, Germany). We gratefully thank Dmitry Svergun and his group for help with the SAXS data, the SPC facility at EMBL Hamburg for technical support and Annabel Parret for her help with the Tridetector Analysis. Sources of Funding: KG is supported by British Heart Foundation Grants (FS/12/40/29712, PG/15/113/31944). KG, RH and HW acknowledge support from the BHF Centre of Research Excellence, Oxford (grant codes HSRNWBY, HSRNWB11 and RE/13/1/30181). KLT is the recipient of a National Institute for Health Research (NIHR) doctoral fellowship (NIHR-HCS-D13-04-006). This publication includes independent research supported also by the NIHR Biomedical Research Centre, Oxford. The work was supported also by funding from the Wellcome Trust Core Award Grant Number 090532/Z/09/Z. The views expressed are those of the authors and not necessarily those of the Department of Health or Wellcome Trust. MG and AG were supported by the EU MUZIC network, the MRC and the Leducq Foundation. MG holds the BHF Chair of Molecular Cardiology. Clinical Perspective High throughput next generation sequencing techniques have made whole genome sequencing accessible and are increasingly applied in clinical practice. However, the abundance of variation in the human genomes makes the identification of a disease-causing mutation on a background of benign rare variants challenging. To illustrate, more than one third of individuals in normal population cohorts carry a rare missense variant in the giant protein titin (coded by the gene TTN), but only a very small fraction of these will be disease-causing with dominant inheritance. Hence titin missense variants are currently often ignored or left un-interpreted when found in cardiomyopathy patients. Here we combine whole genome sequencing with linkage analysis in a three-generation family affected by cardiomyopathy with features of autosomal dominant left-ventricular non-compaction cardiomyopathy. A missense mutation in titin (TTN p. A178D) is the only plausible disease-causing variant that segregates with disease amongst affected individuals of the family, with interrogation of the entire genome excluding other potential causes. Functional studies on this missense mutation demonstrate domain misfolding and destabilisation, resulting in paired binding to the ligand telethonin/t-cap,and hence supporting its highly likely causative role. Our report expands the spectrum of titin’s roles in cardiomyopathies and furthermore highlights that rare titin missense variants should be considered to be relevant for cardiomyopathies and can be identified by combining whole genome sequencing with linkage analysis in medium-sized cardiomyopathy families. Figure 1 A – Pedigree of the family, males depicted as squares, females as circles, slanted symbols deceased individuals. Clinically affected individuals are marked in grey, unaffected are shown in white, “?” means unclassified clinical status. The presence of the TTN p.A178D mutation is indicated (“+”present, “-“absent, ND not determined.) Individuals selected for WGS are marked with thicker symbols (III-1 and III-4). B – Echocardiogram images showing the characteristic 'spongy' appearance of non-compaction in individual II-2 with and without contrast. C – Echocardiogram image from individual II-4 showing significant dilatation, but maintaining a thickened myocardium and preserved ejection fraction. Figure 2 A – Position of the TTN p. A178D on a structural model (pdb: 1YA5) of the titin Z1Z2 domains (purple) in complex with telethonin (pink). B – Close-up of the site of mutation. The red discs show van der Waals overlaps or steric clashing that A178D is predicted to cause with valine127 and proline133. The figures of the crystal structure were generated by pymol (http://www.pymol.org). Figure 3 A – CD spectroscopy of purified titin Z1Z2 fragments (WT solid line and A178D dashed line). B – Size exclusion chromatography for titin Z1Z2 fragments (WT solid line and A178D dashed line). Z1Z2 WT elutes as monomeric protein ($), whereas peaks corresponding to dimer (*) and higher molecular aggregates (#) are observed for Z1Z2 A178D. Figure 4 Destabilisation of the titin Z1Z2 fragment in the presence of the A178D mutation: A – Denaturing gel-electrophoresis of purified titin Z1Z2 fragments (WT and A178D) expressed in E. coli. The WT fragment is detected as a single band of 23 kD (arrow). Only for Z1Z2 A178D a degradation product (arrowhead) is observed. The position of marker proteins and their size (in kD) is indicated. B – Titin Z1Z2 protein fragments (WT – left, A178D – right) were incubated with protease thermolysin for the length indicated. Control: titin Z1Z2 protein sample without thermolysin. A stable degradation fragment of approx. 15 kD is observed for the mutant titin Z1Z2. The position of marker proteins and their size (in kD) is indicated. C – Decreased stability of titin Z1Z2 A178D in NRC: NRC were infected in duplicates with adenoviral particles for HA-tagged titin Z1Z2 (WT or A178D, MOI 5). Infection with parental empty vector (GFP) and non-infected cells (NI) served as controls. Steady-state titin fragment protein amount was assayed by Western blotting for the HA-tag. Probing for hrGFP served as infection control and probing for endogenous GAPDH served as loading control. Despite equal infection rates (for confirmatory control experiments see Figure S6A), less titin A178D protein fragment was detected, indicating reduced stability in NRC. D – Localisation of titin Z1Z2 in NRC: Cells were transfected with constructs coding for HA-tagged titin Z1Z2 WT (top, first row) or titin Z1Z2 A178D (bottom, first row) mutant protein fragment and counterstained for endogenous titin with T12 antibody (middle row, Z-disk proximal epitope, but not recognising the transfected titin Z1Z2 protein fragment). Merged images are shown in the third row, HA shown in red, endogenous titin in green. Scale bar represents 10 microns. Figure 5 Functional implications of the titin Z1Z2 A178D mutation. A – Semi-quantitative GST-pulldown assays using telethonin fragments fused to GST (left aa 1-90, right full length) and titin Z1Z2 fragments (WT and A178D as indicated) as GFP fusions expressed in COS-1 cells. Bound titin-GFP fragments are detected by Western blotting (top row), input lysate controls are shown in the second row. Pulled down GST-telethonin fragments are shown in row three, as well as lysate controls (bottom row). B – Quantification of GST-pulldown experiments from panel A visualises reduced binding of titin Z1Z2 to telethonin in the presence of the A178D mutation, values are expressed as bound protein relative to lysate, with the first WT experiment set to 100 % (n = 2 per group, values expressed as mean with standard deviation for error bars; one representative experiment of three independent ones is shown). Due to the semi-quantitative nature of the experiments, no statistical test was performed. C – FRET experiments using COS-1 cells co-transfected with telethonin (aa 1-90) fused to YFP (first row) and titin Z1Zr3 fused to CFP (second row). Images pre- and post-bleach are shown, FRET ratios are shown in the third row. Inserts show magnification of indicated area. Scale bar represents 10 microns. D – Quantification of FRET efficiency for titin Z1Zr3 WT/A178D and telethonin pairs. The A178D mutation reduces the FRET efficiency from approx. 0.15 to 0.01 indicating a dramatic loss of binding ability (WT n=20 and A178D n=26 cells; * p < 0.0001 unpaired Student’s t-test). Table 1 Summary of clinical findings. Symbol Sex Age* Clinical status† Genetic status Trabeculation compaction ratio (CR)‡ IVS D PW D LVED D LVES D EF ECG Comments I-1 M 87 affected TTN A178D 2.5 (echo) 24 13 40 35 42 left axis deviation, anteroseptal Q wave Marked ASH with low EF, CVA, hypertension; meets diagnostic criteria for LVNC (echo). I-2 F 72 unaffected WT 1.8 (echo) 12 11 48 30 76 AF, paced LVNC excluded (echo), structurally normal heart at age 70, moderate concentric LVH by age 85 II-2 F 61 affected TTN A178D 2.0 (echo) 9 10 53 36 47 left axis deviation Mildly thickened apical segments, cardiac embolus at 61y; meets diagnostic criteria for LVNC (echo). II-3 M 20 assumed affected no DNA Rapidly progressive HF with sudden death at 20y (1970); hypertrophy and dilatation at post mortem. II-4 M 37 affected TTN A178D <2 (MRI) 12 12 64 44 57 normal Dilated LV, late Gd on MRI, hypertension. II-6 M 42 unaffected WT 11 9 50 30 78 normal III-1 M 27 affected TTN A178D 2.5 (echo) 12 11 64 42 63 QRS 120 msec Mild regional systolic dysfunction; meets diagnostic criteria for LVNC (echo). III-2 M 21 unclassified no DNA 0.8 (echo) 13 13 54 37 68 normal Mild concentric LVH III-3 M 32 affected TTN A178D 2.8 (MRI) 11 13 52 35 68 normal Hypertension; meets diagnostic criteria for LVNC (MRI) III-4 F 23 affected TTN A178D 2.6 (MRI) 8 9 46 32 58 normal Meets diagnostic criteria for LVNC (MRI) III-5 M 25 affected TTN A178D 2.0-2.5 (MRI) 10 10 46 27 51 inferior T-wave inversion Hypokinesia apical LV incl. septum; borderline for diagnostic criteria for LVNC (MRI) III-6 M 23 affected TTN A178D 1.5 (MRI) 8 9 53 34 48 Q wave & T-wave inversion in lead III Mild DCM, faint late Gd, borderline dilated LV with mildly impaired function, inferior hypokinesia III-7 M 21 unclassified WT 1.6 (MRI) 9 7 55 38 59 normal Documented myocarditis at 21y (MRI) Cardiac dimensions (IVSD – Interventricular Septal Thickness at Diastole, PWD –Posterior Wall Thickness at Diastole, LVEDD – Left Ventricular End Diastolic Diameter, LVESD – Left Ventricular End Systolic Diameter) are given in mm Abbreviations: EF – Ejection Fraction (in %). ASH – Asymmetric Septal Hypertrophy, CVA – Cerebrovascular Accident, LVH – Left Ventricular Hypertrophy, Gd – Gadolinium, HF – Heart Failure, AF – Atrial Fibrillation, DCM – Dilated CM. Blank cells indicate no data available. * Age of diagnosis or first clinical assessment, however parameters of most recent cardiac assessment are given (with exception of I-2 where data at first assessment aged 70 are given and II-7, where the last assessment before myocarditis is shown). † Clinical status ‘affected’ means affected by cardiomyopathy. Whether individuals meet diagnostic criteria for LVNC is shown in the Comments column (in brackets shown whether MRI or echo criteria have been used). ‡ For the definition of trabeculation compaction ratio (CR) see Methods section, in brackets the mode of imaging is indicated. Representative MRI images are shown in Figure S1. Table 2 Biophysical characterisation of recombinant Z1Z2 WT and A178D protein fragments Titin Z1Z2 WT Titin Z1Z2 A178D Calculated molecular weight (kDa) 22.7 22.8 Size exclusion chromatography Retention time (mL) 15.7 9.9 (1st peak) 14.4 (2nd peak) Static Light Scattering Molecular weight (kDa) 21 ± 2 452 ± 45 (1st peak) 45 ± 4 (2nd peak) Small Angle X-ray Scattering Vp excluded volume of the hydrated particle (nm3) 40 ± 5 305 ± 20 Rg radius of gyration (nm) 3.10 ± 0.05 6.8 ± 0.1 Dmax maximum particle size (nm) 10.5 ± 0.5 25.0 ± 1.0 Normalized Kratky plot folded partly unfolded/ flexible domains Disclosures: None. 1 Watkins H Ashrafian H Redwood C Inherited cardiomyopathies N Engl J Med 2011 364 1643 1656 21524215 2 Jarcho JA McKenna W Pare JA Solomon SD Holcombe RF Dickie S Mapping a gene for familial hypertrophic cardiomyopathy to chromosome 14q1 N Engl J Med 1989 321 1372 1378 2811944 3 1000 Genomes Project Consortium Abecasis GR Auton A Brooks LD DePristo MA Durbin RM An integrated map of genetic variation from 1,092 human genomes Nature 2012 491 56 65 23128226 4 Haas J Frese KS Peil B Kloos W Keller A Nietsch R Atlas of the clinical genetics of human dilated cardiomyopathy Eur Heart J 2015 36 1123 1135a 25163546 5 Watkins H Assigning a causal role to genetic variants in hypertrophic cardiomyopathy Circ Cardiovasc Genet 2013 6 2 4 23424253 6 Jenni R Oechslin E Schneider J Attenhofer Jost C Kaufmann PA Echocardiographic and pathoanatomical characteristics of isolated left ventricular non-compaction: a step towards classification as a distinct cardiomyopathy Heart 2001 86 666 671 11711464 7 Petersen SE Selvanayagam JB Wiesmann F Robson MD Francis JM Anderson RH Left ventricular non-compaction: insights from cardiovascular magnetic resonance imaging J Am Coll Cardiol 2005 46 101 105 15992642 8 Purcell S Neale B Todd-Brown K Thomas L Ferreira MA Bender D PLINK: a tool set for whole-genome association and population-based linkage analyses Am J Hum Genet 2007 81 559 575 17701901 9 Abecasis GR Cherny SS Cookson WO Cardon LR Merlin--rapid analysis of dense genetic maps using sparse gene flow trees Nat Genet 2002 30 97 101 11731797 10 Taylor JC Martin HC Lise S Broxholme J Cazier JB Rimmer A Factors influencing success of clinical genome sequencing across a broad spectrum of disorders Nat Genet 2015 47 717 726 25985138 11 Lunter G Goodson M Stampy: a statistical algorithm for sensitive and fast mapping of Illumina sequence reads Genome Res 2011 21 936 939 20980556 12 Rimmer A Phan H Mathieson I Iqbal Z Twigg SR Consortium WGS Integrating mapping-, assembly- and haplotype-based approaches for calling variants in clinical sequencing applications Nat Genet 2014 46 912 918 25017105 13 McLaren W Pritchard B Rios D Chen Y Flicek P Cunningham F Deriving the consequences of genomic variants with the Ensembl API and SNP Effect Predictor Bioinformatics 2010 26 2069 2070 20562413 14 Landrum MJ Lee JM Riley GR Jang W Rubinstein WS Church DM ClinVar: public archive of relationships among sequence variation and human phenotype Nucleic Acids Res 2014 42 D980 985 24234437 15 Zou P Gautel M Geerlof A Wilmanns M Koch MH Svergun DI Solution scattering suggests cross-linking function of telethonin in the complex with titin J Biol Chem 2003 278 2636 2644 12446666 16 Geier C Gehmlich K Ehler E Hassfeld S Perrot A Hayess K Beyond the sarcomere: CSRP3 mutations cause hypertrophic cardiomyopathy Hum Mol Genet 2008 17 2753 2765 18505755 17 Reinhard L Mayerhofer H Geerlof A Mueller-Dieckmann J Weiss MS Optimization of protein buffer cocktails using Thermofluor Acta Crystallogr Sect F Struct Biol Cryst Commun 2013 69 209 214 18 Shaya D Kreir M Robbins RA Wong S Hammon J Bruggemann A Voltage-gated sodium channel (NaV) protein dissection creates a set of functional pore-only proteins Proc Natl Acad Sci U S A 2011 108 12313 12318 21746903 19 Furst DO Osborn M Nave R Weber K The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line J Cell Biol 1988 106 1563 1572 2453516 20 Gehmlich K Asimaki A Cahill TJ Ehler E Syrris P Zachara E Novel missense mutations in exon 15 of desmoglein-2: role of the intracellular cadherin segment in arrhythmogenic right ventricular cardiomyopathy? Heart Rhythm 2010 7 1446 1453 20708101 21 Cameron JM Maj M Levandovskiy V Barnett CP Blaser S Mackay N Pyruvate dehydrogenase phosphatase 1 (PDP1) null mutation produces a lethal infantile phenotype Hum Genet 2009 125 319 326 19184109 22 Carrilho-Ferreira P Almeida AG Pinto FJ Non-compaction cardiomyopathy: prevalence, prognosis, pathoetiology, genetics, and risk of cardioembolism Curr Heart Fail Rep 2014 11 393 403 25239435 23 Walsh R Thomson KL Ware JS Funke BH Woodley J McGuire KJ Reassessment Of Mendelian Gene Pathogenicity Using 7,855 Cardiomyopathy Cases And 60,706 Reference Samples Genet Med 2016 8 17 10.1038/gim.2016.90 [Epub ahead of print] 24 Labeit S Kolmerer B Linke WA The giant protein titin. Emerging roles in physiology and pathophysiology Circ Res 1997 80 290 294 9012751 25 Wang K McClure J Tu A Titin: major myofibrillar components of striated muscle Proc Natl Acad Sci U S A 1979 76 3698 3702 291034 26 Gerull B The Rapidly Evolving Role of Titin in Cardiac Physiology and Cardiomyopathy Can J Cardiol 2015 31 1351 1359 26518445 27 Chauveau C Rowell J Ferreiro A A rising titan: TTN review and mutation update Hum Mutat 2014 35 1046 1059 24980681 28 Hidalgo C Granzier H Tuning the molecular giant titin through phosphorylation: role in health and disease Trends Cardiovasc Med 2013 23 165 171 23295080 29 Bertz M Wilmanns M Rief M The titin-telethonin complex is a directed, superstable molecular bond in the muscle Z-disk Proc Natl Acad Sci U S A 2009 106 13307 133310 19622741 30 Zou P Pinotsis N Lange S Song YH Popov A Mavridis I Palindromic assembly of the giant muscle protein titin in the sarcomeric Z-disk Nature 2006 439 229 233 16407954 31 Truong K Ikura M The use of FRET imaging microscopy to detect protein-protein interactions and protein conformational changes in vivo Curr Opin Struct Biol 2001 11 573 578 11785758 32 Herman DS Lam L Taylor MR Wang L Teekakirikul P Christodoulou D Truncations of titin causing dilated cardiomyopathy N Engl J Med 2012 366 619 628 22335739 33 Ware JS Li J Mazaika E Yasso CM DeSouza T Cappola TP Shared Genetic Predisposition in Peripartum and Dilated Cardiomyopathies N Engl J Med 2016 374 233 241 26735901 34 Watkins H Tackling the achilles' heel of genetic testing Sci Transl Med 2015 7 270fs1 35 Roberts AM Ware JS Herman DS Schafer S Baksi J Bick AG Integrated allelic, transcriptional, and phenomic dissection of the cardiac effects of titin truncations in health and disease Sci Transl Med 2015 7 270ra6 36 Zou J Tran D Baalbaki M Tang LF Poon A Pelonero A An internal promoter underlies the difference in disease severity between N- and C-terminal truncation mutations of Titin in zebrafish Elife 2015 4 e09406 26473617 37 Begay RL Graw S Sinagra G Merlo M Slavov D Gowan K Role of Titin Missense Variants in Dilated Cardiomyopathy J Am Heart Assoc 2015 4 pii: e002645 38 Lopes LR Zekavati A Syrris P Hubank M Giambartolomei C Dalageorgou C Genetic complexity in hypertrophic cardiomyopathy revealed by high-throughput sequencing J Med Genet 2013 50 228 239 23396983 39 Gerull B Gramlich M Atherton J McNabb M Trombitas K Sasse-Klaassen S Mutations of TTN, encoding the giant muscle filament titin, cause familial dilated cardiomyopathy Nat Genet 2002 30 201 204 11788824 40 Hinson JT Chopra A Nafissi N Polacheck WJ Benson CC Swist S HEART DISEASE. Titin mutations in iPS cells define sarcomere insufficiency as a cause of dilated cardiomyopathy Science 2015 349 982 986 26315439 41 Gramlich M Pane LS Zhou Q Chen Z Murgia M Schotterl S Antisense-mediated exon skipping: a therapeutic strategy for titin-based dilated cardiomyopathy EMBO Mol Med 2015 7 562 576 25759365 42 Knoll R Linke WA Zou P Miocic S Kostin S Buyandelger B Telethonin deficiency is associated with maladaptation to biomechanical stress in the mammalian heart Circ Res 2011 109 758 769 21799151

PMC005xxxxxx/PMC5111163.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8907828 20554 J Psychopharmacol J. Psychopharmacol. (Oxford) Journal of psychopharmacology (Oxford, England) 0269-8811 1461-7285 24327452 5111163 10.1177/0269881113515061 NIHMS828453 Article Psychiatric profiles of mothers who take Ecstasy/MDMA during pregnancy: Reduced depression 1 year after giving birth and quitting Ecstasy Turner John JD 1 Parrott Andrew C 2 Goodwin Julia 1 Moore Derek G 1 Fulton Sarah 3 Min Meeyoung O 3 Singer Lynn T 3 1 University of East London, London, UK 2 Swansea University, Swansea, UK 3 Case Western Reserve University, Cleveland, USA Corresponding author: John JD Turner, School of Psychology, University of East London, London E15 4LZ, UK. j.j.d.turner@uel.ac.uk 9 11 2016 10 12 2013 1 2014 16 11 2016 28 1 5561 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background The recreational drug MDMA (3,4-methylenedioxymethamphetamine) or ‘Ecstasy’ is associated with heightened psychiatric distress and feelings of depression. The Drugs and Infancy Study (DAISY) monitored the psychiatric symptom profiles of mothers who used Ecstasy/MDMA while pregnant, and followed them over the first year post-partum. Methods We compared 28 young women whom took MDMA during their pregnancy with a polydrug control group of 68 women who took other psychoactive drugs while pregnant. The Brief Symptom Inventory (BSI) was completed for several periods: The first trimester of pregnancy; and 1, 4 and 12 months after childbirth. Recreational drug use was monitored at each time point. Results During the first trimester of pregnancy, MDMA-using mothers reported higher depression scores than the polydrug controls. At 1 year after childbirth, their BSI depression scores were significantly lower, now closer to the control group values. At the same time point, their self-reported use of MDMA became nearly zero, in contrast to their continued use of Cannabis/marijuana, nicotine and alcohol. We found significant symptom reductions in those with BSI obsessive-compulsive and interpersonal sensitivity, following Ecstasy/MDMA cessation. Conclusions The findings from this unique prospective study of young recreational drug-using mothers are consistent with previous reports of improved psychiatric health after quitting MDMA. Cessation depression drug addiction Ecstasy MDMA middle class mother post-partum pregnancy quitting recreational drugs Introduction ‘Ecstasy’ or 3,4-methylenedioxymethamphetamine (MDMA) is used as an illicit drug by subgroups of adolescents and young adults. Its recreational use is mainly associated with dance clubs, all-night ‘raves’ and house parties (Parrott et al., 2008; Winstock et al., 2001). Population surveys in the US reveal usage levels as high as 9.5% in college students (Johnston et al., 2005; Singer et al., 2004). In the American National Survey on drug use and health, Ecstasy/MDMA is found to be used more by young women than men (Wu et al., 2010). Neuroimaging studies of abstinent MDMA users reveal significantly lower levels of the serotonin transporter (SERT) (Erritizoe et al., 2011; Kish et al., 2010), and are widely interpreted as suggesting serotonergic neurotoxicity (Benningfield and Cowan, 2013; Parrott, 2013a; Puerta et al., 2009; Ricaurte et al., 2000). Recreational use of MDMA is also associated with various neuropsychobiological problems, including memory deficits (Montgomery et al., 2010; Rogers et al., 2009; Zakzanis and Campbell, 2006), impairments in higher cognitive processing (Fox et al, 2002; Parrott, 2012, 2013b; Reay et al., 2006), sleep apnea (McCann et al., 2009), raised cortisol levels (Parrott, 2009; Parrott et al., 2012), psychosocial impairment (Topp et al., 1999) and various psychiatric problems (Briere et al., 2012; MacInnes et al., 2000; Milani et al., 2004; Morgan et al., 2002; Schifano et al., 1998; Singer et al., 2004; Verheyden et al., 2003). Laboratory animal studies show adverse effects of MDMA upon the developing foetus (Adori et al., 2010; Skelton et al., 2008), raising concerns about potentially damaging effects when taken by female recreational users during pregnancy. To date, there has been no controlled empirical data addressing this question, although there is some evidence of adverse birth consequences (McElhatton et al., 1999; Singer et al., 2012b). To investigate the potential effects of foetal MDMA exposure on development, the US National Institute on Drug Abuse (NIDA) funded the Drugs and Infancy Study (DAISY). This prospective study monitored a group of mothers whom took recreational Ecstasy/MDMA while pregnant, and a control group of pregnant females, the other ‘polydrug’ users. The two groups were followed over time, in order to monitor the physical development and psychobiological well-being of their children. Over the first year of life, the children of MDMA-using mothers displayed significantly poorer gross psychomotor skills than control group children (Singer et al., 2012a, 2012b). The DAISY study also assessed maternal well-being, using the Brief Symptom Inventory, a self-reporting measure of psychiatric health for non-clinical populations, derived from the earlier Symptom Check List-90 (Derogatis and Nelisaratos, 1983). This psychiatric measure was included, because previous research shows higher symptom profiles in abstinent Ecstasy/MDMA users. Soar et al. (2001) reviews the medical case study literature, which indicated an increased risk of several psychiatric disorders, including depression and psychosis, in MDMA users. Schifano et al. (1998) noted that regular Ecstasy/MDMA users are at increased risk of developing various psychiatric problems, the most frequent being depression. MacInnes et al. (2000) found significantly raised Beck Depression Inventory (BDI) scores in a non-clinical sample of abstinent regular Ecstasy/MDMA users. Singer et al. (2004) found that abstinent Ecstasy users reported significantly higher BSI scores for anxiety, depression and obsessive-compulsive disorder than non-user controls. Milani et al. (2004) reported significant gender effects, with female Ecstasy/MDMA users reporting higher levels of BSI anxiety, depression and somatization scores. Verheyden et al. (2003) investigated the reasons for quitting Ecstasy/MDMA: They found that most users in their large survey reported improved mental health after drug cessation. In the current DAISY, the BSI allowed us to prospectively monitor the psychiatric health of our pregnant mothers and to investigate how any changes in drug usage were associated with their report of psychological distress on the BSI. Based on previous findings, it might be predicted that elevated psychiatric symptoms would be evident in mothers who are continuing MDMA users, whilst those who discontinue use may show improvements; however, given its uniqueness, and the additional biopsychosocial changes associated with pregnancy and motherhood, the aims of the study were largely exploratory. Methods Experimental design The data in the current report were collected as part of the maternal assessment component of DAISY, a prospective study primarily exploring the effects of recreational drug use, notably MDMA/Ecstasy, on infant social and cognitive development (Moore et al., 2010; Singer et al., 2012a, 2012b). In a mixed design, mothers who used MDMA/Ecstasy during pregnancy (MDMA/Ecstasy users) were compared with those who used other drugs, but not MDMA/Ecstasy (Polydrug user controls), across measures of drug use and symptoms of mental distress, at four distinct time periods: the first trimester of pregnancy and at 1, 4 and 12 months post-partum. Participants We prospectively recruited 96 pregnant women from the UK through midwife referrals, leaflets describing the study at prenatal clinics and advertisements in commercial pregnancy magazines. We sought pregnant women whom were using recreational drugs during pregnancy, listing ecstasy, tobacco, Cannabis, alcohol and cocaine as examples. The majority of participants were therefore recreational ‘polydrug’ users. Exclusionary factors included: positive HIV status, moderate or severe intellectual disability, chronic medical disorder or psychiatric diagnosis. In total, there were 28 mothers in the MDMA-exposed group, who used MDMA (and other substances) during pregnancy, and 68 non-MDMA controls (some of whom used substances during pregnancy, but not MDMA). The majority of the sample were white, married or with a partner, and educated to a UK degree level. Their mean ages at the birth of their infants were 30.3 (SD 6.4) years of age in the MDMA-exposed group and 28.4 (SD 6.2) in the controls. The groups did not differ on basic demographic profiles. Participants were informed of data confidentiality and they gave written informed consent. The study protocol was approved by ethics committees from the University of East London, UK; Case Western Reserve University, US; and the National Health Service, UK. For a fuller description of the participant sample and screening procedures, see Singer et al. (2012a). Drug usage All women were individually interviewed about their substance use by fully trained female research assistants. The interview was an adaptation of the Maternal Post-Partum Interview, which was developed for earlier studies of maternal cocaine exposure (Singer et al., 2002). Interview questions covered substances commonly used in the UK and were based on the University of East London Recreational Drug Usage Questionnaire (Parrott et al., 2001). The list of drugs included tobacco/cigarettes, alcohol, Cannabis, Ecstasy/MDMA, amphetamine, cocaine, LSD, benzodiazepines, hallucinogenic mushrooms, ketamine and opiates. It may be noted that mephedrone (m-cathinone or ‘m-cat’) was not on this list, since the DAISY study was undertaken before ‘m-cat’ was used as a recreational drug (Schifano et al., 2011). Mean usage for each drug per week was calculated by multiplying the frequency of use with the amount taken per occasion. The MDMA user group comprised women who reported taking MDMA during pregnancy or in the month prior to pregnancy. Those who reported MDMA use prior to this time were categorized as non-users, because the study was designed to assess foetal drug exposure. Assessment battery The study included a comprehensive battery of assessment measures, covering various aspects of child behaviour and physical health indices, maternal activities and psychological well-being (Singer et al., 2012a,2012b). This report describes the findings from the Brief Symptom Inventory (BSI) (Derogatis and Nelisaratos, 1983). This questionnaire comprises 53 self-rating questions across nine psychiatric subscales, for: depression, anxiety, phobic anxiety, hostility, somatic complaints, obsessive-compulsive behavior, interpersonal sensitivity, paranoid ideation and psychosis/schizophrenia. The summary measure, the General Severity Index (GSI), provided a general index for overall psychiatric distress. The assessments covered four occasions: first trimester of pregnancy, 1 month post-partum, 4 months postpartum and 12 months post-partum. Statistical analyses Data that were positively skewed were transformed using natural logarithm, prior to analysis; however, the means and SDs are reported for the untransformed scores. Bivariate correlations were employed to calculate the inter-relationships between variables. Multicollinearity was assessed using tolerance and variance inflation factor. We implemented repeated measures Analysis of Variance (ANOVA), using a mixed model approach, by SAS Proc Mixed with maximum estimation method, to compare the substance use for both groups, MDMA-users during pregnancy (n = 28) and non-users of MDMA during pregnancy (n = 68), at the four different assessment times (during pregnancy, 4 weeks after birth, 12 weeks and 52 weeks). As noted earlier, both groups contained polydrug users of various substances, both legal (tobacco and/or alcohol), and illegal (Cannabis, amphetamine and/or cocaine) (Moore et al., 2010). Because the dependent variables were repeated measures and correlated within subjects, we used an unstructured covariance matrix to account for these correlated responses. We included interaction terms between drug groups and time, to test for homogeneity of MDMA effects over time. For all BSI outcome measures, we employed repeated measures Analysis of Covariance (ANCOVA). The covariates included other substance usage that differed by MDMA status at p < 0.10, and were correlated with the given outcome at p < 0.10 on at least two time points: They were then entered into the longitudinal model. Different sets of covariates were adjusted on each psychological outcome, and included demographic variables and use of all other drugs. Results The socioeconomic and educational profiles of mothers enrolled in the study are described more fully elsewhere (Singer et al., 2012a). In brief, the cohort was primarily white, married or in a stable relationship, and represented a wide range of socioeconomic backgrounds that included many from middle and higher psychosocial groupings. The MDMA-using mothers and polydrug control mothers were well matched on most variables (Singer et al., 2012a). Table 1 describes the group mean weekly rates of usage for the five main types of drug used: alcohol, nicotine/cigarettes, Cannabis/marijuana, cocaine and Ecstasy/MDMA. Other psychoactive drugs were taken by a few individuals, and those data are described more fully elsewhere (Moore et al., 2010). A mixed ANOVA was conducted with group as the between-conditions factor and time as the within-conditions factor. The between-groups ANOVA revealed that the two groups did not differ in overall use of alcohol, cigarettes, Cannabis nor cocaine; although the cocaine group effect was statistically borderline (Table 1, where the group effect for Ecstasy was not calculated, because it was used to define these two groups). The ANOVA for the time factor was significant for all five drugs (all p = 0.01 or smaller), with lower rates of usage during the weeks after giving birth. The ANOVA grouped by time interactions were significant for alcohol and cigarettes (F[3,88] = 4.06; p < 0.005 and F[3,88] = 3.61; p < 0.02 respectively), with the MDMA mothers using slightly more than the controls during the first trimester of pregnancy, but slightly less than controls across all the other time periods (Table 1). The group × time interaction was not significant for Cannabis, though Ecstasy/MDMA-using mothers appeared to be taking slightly more Cannabis than controls, across all time points (Table 1). The group × time interaction was significant for cocaine (F[3,88] = 3.48; p < 0.05), with the most usage during the first session by Ecstasy users (Table 1). The Ecstasy-using mothers reported taking an average of 0.84 Ecstasy tablets/week during the first trimester of pregnancy. In terms of previous lifetime usage (Singer et al., 2012a), they reported first using Ecstasy at a mean age of 20.2 years (range 14 – 29 years), had taken it on an average of 171 times/lifetime (range 6 – 936 times), and typically ingested an average of 3 tablets per occasion (range 1 – 8 tablets), with an average maximum usage per occasion of 7.4 Ecstasy tablets (range 2 – 20 tablets). Turning to their usage around the time of pregnancy, the mean total amount of MDMA used during pregnancy and in the month prior was 25 tablets (range 0.45 – 180 tablets). Within the polydrug control group, several women had used ecstasy/MDMA previously, but were currently non-users (Singer et al., 2012a). The Brief Symptom Inventory findings are summarized in Table 2. The main focus of interest here is the difference in psychiatric well-being between the first and last sessions. Over that time period, the control group mothers showed a significant decline in BSI symptoms for somatization (p < 0.001) and anxiety (p < 0.05). Over the same period, the Ecstasy/MDMA subgroup mothers showed significant declines in BSI symptoms for (Table 2): somatization (p < 0.001), depression (p < 0.05), interpersonal sensitivity (p < 0.05) and obsessive-compulsive disorder (p < 0.05). Discussion The young mothers in the DAISY study provided a unique cohort in several respects. Although recreational polydrug users, they were predominantly middle class with middle socioeconomic status, and in stable interpersonal relationships; hence, unlike many studies of illicit drug users, they were not socially disadvantaged. The study covered an extended time period of nearly 2 years, and is to our knowledge the first study of pregnant Ecstasy/MDMA users. The cohort of almost 100 mothers was comparatively large, especially for a prospective study with repeated assessments. One of the main aims of the DAISY study was to investigate the effects of recreational Ecstasy/MDMA usage during pregnancy on subsequent child development. The main findings were that the children of Ecstasy/MDMA using mothers displayed significant psychomotor problems in comparison to control group children, as described elsewhere (Singer et al., 2012a, 2012b). The study design allowed us to monitor changes in maternal reports of psychological well-being over time, in particular any alterations in their psychiatric status from the first to the last assessment. In this respect, both groups of mothers showed significantly higher somatization scores during the first trimester of pregnancy, when compared to 12 months post-partum (Table 2). The control group mothers also showed a significant reduction in BSI symptoms of anxiety, while the MDMA subgroup showed a very similar trend (Table 2). The first trimester of pregnancy is a period of pronounced somatic body changes, and so intuitively explains the higher somatization scores in both groups of women. Thus, the reduced BSI somatization scores 1 year post-partum may reflect a return to physical normality in both groups of women. The first trimester of pregnancy is also a period of general anxiety, with natural concerns and worries over becoming pregnant. This may help to explain the comparatively higher BSI anxiety scores during the first trimester, and the reduced scores at the final session (Table 2). The Ecstasy/MDMA-using mothers showed a different pattern of change, compared to the controls, on three BSI subscales, for: depression, obsessive compulsive disorder and interpersonal sensitivity (Table 2). The Ecstasy subgroup mothers reported feeling more depressed than control mothers at the first time point, with a statistically borderline between-group difference (p = 0.058, two-tail). At 1 year post-partum, the depression scores for the MDMA group had reduced significantly (p < 0.05), to become almost identical to the control group (Figure 1). The BSI depression scores for the control group mothers remained broadly unchanged over this period. The MDMA group also showed significant BSI reductions for interpersonal sensitivity and obsessive-compulsive disorder (Table 2). In order to examine the potential reasons for these changes, the changing patterns of drug usage over time should be noted. The Ecstasy/MDMA-group mothers had reduced their usage of Ecstasy to near-zero after giving birth (Table 1); hence, 1 year post-partum they had become former MDMA users. Their BSI improvement may reflect this cessation of Ecstasy/MDMA use. There is extensive empirical literature demonstrating higher rates of psychiatric distress in current Ecstasy/MDMA users and psychiatric gains following drug cessation. Schifano et al. (1998) gave structured psychiatric interviews to young Ecstasy/MDMA users at an addiction centre in Italy, reporting that around one-half the sample reported symptoms of psychiatric distress, especially depression, but also psychotic disorder, impulse control disorder, bulimia and panic disorder. MacInnes et al. (2000) compared young Ecstasy/MDMA users and polydrug controls, with participants screened to exclude anyone with a prior psychiatric history. On the BDI, Ecstasy users displayed significantly higher depression scores than the non-MDMA-user controls. In a survey of over 700 young people from the UK and Italy, the SCL-90 symptom profiles of the Ecstasy polydrug users were significantly higher than the non-MDMA-user controls (Parrott et al., 2001). In a US study of abstinent MDMA users compared to non-user controls who visited raves (Singer et al., 2004), the Ecstasy/MDMA users reported significantly higher BSI depression, anxiety and obsessive-compulsive disorder than the controls. Brière et al. (2012) prospectively found that taking up recreational Ecstasy/MDMA in Canadian schoolchildren led to increased depression 1 year later. There are also indications that psychiatric health can improve after quitting. Morgan et al. (2002) report that current Ecstasy/MDMA users have elevated scores on many SCL-90 subscales, whereas former Ecstasy users have scores intermediate between the current Ecstasy users and the non-user controls. Verheyden et al. (2003) interviewed former users about their reasons for quitting Ecstasy/MDMA. Over one-half reported that ‘mental health problems due to MDMA’ were the main reason for quitting drug use: That using Ecstasy led to feelings of anxiety and depression, and that they feared for their mental health in the longer-term. Over 70% of those participants report ‘improved mental health’ after quitting. An important potential confounder for Ecstasy/MDMA research is the use of other recreational drugs, because many Ecstasy users take a range of psychoactive drugs (Parrott et al., 2001; Parrott et al., 2007; Sala and Braida, 2004; Scholey et al., 2004). In the DAISY study, we collected systematic drug usage data at all four time points. As noted above, the use of Ecstasy/MDMA was largely restricted to the first trimester of pregnancy. In contrast, the use of alcohol, tobacco and Cannabis continued throughout the study. There is some indication of a decline in all drug use in the Ecstasy/MDMA group, with significant group/time interactions for alcohol and cigarettes, especially. As such, it could be argued that the depression effect in the Ecstasy/MDMA users was in part due to changes in alcohol and/or cigarette use, as both have been linked to higher depression scores (Munafo and Araya, 2010; Raimo and Schuckit, 1998); however, usage rates at baseline were broadly similar to 1 year post-partum, in both groups (Table 1). Hence, the changes in psychiatric status noted here (Table 2) cannot easily be attributed to alcohol, tobacco, nor Cannabis usage; however, the usage pattern for cocaine was very similar to Ecstasy/MDMA, with almost total cessation after the first trimester (Table 1). Thus, the selective reductions in particular psychiatric symptoms may reflect the cessation of Ecstasy/MDMA and/or cocaine usage. There are several ways in which central nervous system (CNS) stimulant drugs like MDMA can enhance psychiatric distress. In acute terms, MDMA is a powerful mood intensifier, but it can boost positive and negative-feeling states; thus, increased levels of happiness and euphoria are often accompanied by emotional tension. This intensification of both positive and negative moods is reported in studies of recreational users and in placebo-controlled laboratory studies (Kirkpatrick et al., 2012; Parrott et al., 2011). It is also noted in the psychotherapeutic situation: Two clients undergoing ‘MDMA-assisted psychotherapy’ experienced a resurgence of previous psychiatric problems following acute MDMA administration, with one client needing psychotherapy for a year afterwards, to resolve the MDMA-induced problems (Greer and Tolbert, 1986; Parrott, 2007). In sub-acute terms, MDMA use is typically followed by a period of neurochemical recovery, when low moods and feelings of depression predominate; indeed, the ‘mid-week blues’ can often last for several days and may reach clinical levels in some individuals (Curran and Travill, 1997). Because the positive mood intensification under MDMA is brief (several hours), and the post-MDMA period of mood recovery is more prolonged (several days), the average weekly mood of Ecstasy users will often be lower than in non-users (Parrott and Lasky, 1998). Such effects are supported in the animal literature by the acute and subacute impact of MDMA on 5-HT, notably, delays in recovery of this transmitter in brain regions regulating emotion (Colado et al., 1999); and similar pattern reductions in other functional serotonergic factors, such as SERT and tryptophan hydroxylase (Adori et al., 2011). In addition, in chronic terms, abstinent Ecstasy/MDMA users report higher levels of stress and lower levels of happiness than non-user controls (Scholey et al., 2011). When used repeatedly, sympathomimetic drugs such as amphetamine, cocaine and MDMA can adversely affect the hypothalamic pituitary adrenal (HPA) axis and impair homeostatic control via the stress hormone cortisol (Seyle, 1955). Indeed, acute MDMA use can increase cortisol levels by 800% in young dance club attendees (Parrott et al., 2008). While sub-chronically, recent Ecstasy/MDMA users display a 400% increase of cortisol in 3-month hair samples (Parrott et al., 2012); hence, recreational MDMA is both an acute and chronic stressor for the HPA axis (Parrott, 2009). There is also evidence that premorbid factors may heighten the likelihood of clinical problems in disadvantaged individuals; this interactive ‘diathesis-stress’ model for recreational Ecstasy/MDMA is described more fully elsewhere (Parrott, 2006). The possible causative factors (including neurotoxicity, recovery and/or HPA axis changes) for the effects observed here in the current data, and in much of the literature, still need considerable further empirical investigation. There are several limitations to the DAISY study. We relied on self-reported drug use and cannot therefore be certain that ‘Ecstasy’ comprised ‘MDMA’; however, data collection occurred during 2003 – 2006, which corresponded with a period of high MDMA purity in the UK. This was apparent in another study we undertook during 2006, which shows very high concordance between self-rated Ecstasy and MDMA use as detected in saliva samples (Parrott et al., 2008). The second weakness was the absence of a non-user control group, because many studies have found that polydrug users are more impaired than non-users (Morgan et al., 2002; Parrot et al., 2001). Thirdly, although the DAISY study was designed as a prospective study, this was only partially achieved (Moore et al., 2010); hence, missing data point’s retrospective ratings were sometimes required (Singer et al., 2012a). Finally, the overall BSI difference scores were not large (Table 2); however, we were not expecting strong drug effects, because our participants were psychiatrically normal and their use of most drugs was similar at the first and last time points. Furthermore, although the group mean reduction of 0.2 on the BSI depression subscale may have been comparatively slight, it would still be beneficial for the individual user. It would also reduce the likelihood of individuals with prior vulnerability factors from developing more severe psychiatric problems (Parrott, 2006). In summary, recreational stimulant drugs such as MDMA, cocaine and amphetamine, are well-known to be associated with enhanced psychiatric distress. The DAISY study found that women who took Ecstasy/MDMA during their first trimester of pregnancy reported slightly higher psychiatric symptom profiles than a control group of polydrug-using mothers. One year after giving birth, their psychiatric symptom profiles improved to values near the control group (Table 2 and Figure 1). The main explanatory factor proposed for this gain in psychiatric well-being was the cessation of Ecstasy/MDMA usage, coupled with the parallel reduction in cocaine use. Hence, this study confirmed that a reduction in stimulant drug usage can have beneficial effects on well-being. Finally, we should also note that the DAISY study investigated the effects of MDMA use during pregnancy on the child’s subsequent development. It reveals that the children of MDMA-using mothers have various impairments in gross psychomotor skill (Singer et al., 2012a,2012b); hence, an important message for young females and their partners is to stop taking MDMA before pregnancy. This will protect the developing child and enhance maternal well-being. Acknowledgements We would like to thank all the mothers who gave of their time and patience. Many thanks also to Fleur Braddick, Emma Axelsson, Stephanie Lynch, Helena Ribeiro, Caroline Frostick, Alice Toplis and Helen Fox, for undertaking the data collection and scoring. Funding This work (DAISY study) was funded by the National Institute on Drug Abuse in America (grant number DA-14910-05). Figure 1 Brief Symptom Inventory ratings of depression during the first trimester and at 12 months post-partum, in women reporting MDMA/Ecstasy use during pregnancy, and in control women taking other recreational drugs during pregnancy. *p < 0.05; Error bars indicate ±1 SE. MDMA: ‘Ecstasy’ or 3,4-methylenedioxymethamphetamine. Table 1 Ecstasy/MDMA, alcohol, cigarettes, marijuana/Cannabis and cocaine usage patterns for 28 mothers whom took Ecstasy/MDMA during pregnancy and a control group of 68 polydrug users during pregnancy. Drug values represent mean weekly rates of usage, during first trimester of pregnancy and three times up to 1 year post-partum. Drug type Maternal group First trimester of pregnancy 1 month post-partum 4 months post- partum 12 months post-partum ANOVA Group Time (G×T) Ecstasy (tablets) Polydrug controls 0.00 +/− 0.00 0.00 +/− 0.00 0.02 +/− 0.16 0.01 +/− 0.02 No between- group analysis < .0001 - Ecstasy users 0.82 +/− 1.57 0.01 +/− 0.03 0.03 +/− 0.13 0.06 +/− 0.09 Alcohol (units) Polydrug controls 6.94 +/− 16.90 3.11 +/− 10.66 6.48 +/−10.89 13.75 +/−24.02 n.s .0001 .02 Ecstasy users 12.07 +/− 16.62 1.33 +/− 1.80 5.30 +/− 5.70 6.01 +/− 5.99 Cigarette (numbers) Polydrug controls 28.15 +/−48.10 23.45 +/− 50.13 27.27 +/− 40.02 32.88 +/−48.14 n.s .0001 .003 Ecstasy users 44.78 +/− 49.50 17.88 +/−30.79 17.59 +/− 22.23 28.68 +/−34.37 Cannabis (joints) Polydrug controls 7.44 +/− 19.24 3.36 +/− 7.87 3.12 +/−7.51 5.26 +/−12.95 n.s .0001 n.s. Ecstasy users 10.28 +/− 20.81 6.86 +/− 17.36 6.20 +/− 16.12 7.35 +/−15.46 Cocaine (grams) Polydrug controls 0.02 +/−0.18 0.001 +/− 0.01 0.01 +/− 0.07 0.02 +/− 0.14 .057 .013 .03 Ecstasy users 0.23 +/− 0.85 0.01 +/− 0.04 0.02 +/− 0.06 0.02 +/− 0.05 MDMA: ‘Ecstasy’ or 3,4-methylenedioxymethamphetamine. Note: Units refer to UK units of alcohol (1 unit = 10ml or 7.9 grams of alcohol); n.s.= non-significant. Table 2 Psychiatric symptoms on the Brief Symptom Inventory during and after pregnancy for 28 mothers whom took Ecstasy/MDMA during pregnancy and for a non-user control group of 68 mothers whom took other drugs during pregnancy (polydrug controls). Group Time 1: Early-mid Pregnancy Time 2: Postpartum 1 month Time 3: Postpartum 4 months Time 4: Postpartum 12 months Paired comparison, Time 1 vs. 4 General symptoms Polydrug controls 0.61 0.51 0.54 0.50 - Ecstasy users 0.79 0.71 0.81 0.56 - Depression Polydrug controls 0.50 0.45 0.57 0.50 - Ecstasy users 0.87 0.74 0.80 0.51 p < 0.05 Anxiety Polydrug controls 0.55 0.46 0.48 0.34 p < 0.05 Ecstasy users 0.74 0.68 0.69 0.56 - Hostility Polydrug controls 0.71 0.65 0.66 0.59 - Ecstasy users 0.74 0.80 1.24 0.55 - Psychoticism Polydrug controls 0.31 0.25 0.36 0.30 - Ecstasy users 0.52 0.51 0.62 0.42 - Somatization Polydrug controls 0.58 0.36 0.27 0.32 p < 0.001 Ecstasy users 0.78 0.50 0.51 0.39 p < 0.01 Paranoid ideation Polydrug controls 0.61 0.48 0.64 0.66 - Ecstasy users 0.75 0.70 066 0.73 - Obsessive-compulsive Polydrug controls 1.08 1.10 0.98 0.89 - Ecstasy users 1.20 1.23 1.31 0.82 p < 0.05 Interpersonal sensitivity Polydrug controls 0.74 0.64 0.64 0.73 - Ecstasy users 0.92 0.85 0.96 0.57 p < 0.05 Phobic anxiety Polydrug controls 0.27 0.20 0.30 0.20 - Ecstasy users 0.47 0.39 0.52 0.37 - MDMA: ‘Ecstasy’ or 3,4-methylenedioxymethamphetamine. Conflict of interest The authors declare no conflict of interest. References Ádori C Andó RD Szekeres M Recovery and aging of serotonergic fibers after single and intermittent MDMA treatment in dark agouti rat J Comp Neurol 2011 519 2353 2378 21456018 Ádori C Zelena D Tímár J Intermittent prenatal MDMA exposure alters physiological but not mood related parameters in adult rat offspring Behav Brain Res 2010 206 299 309 19782105 Benningfield MM Cowan RL Brain serotonin function in MDMA (ecstasy) users: Evidence for persisting neurotoxicity Neuropsychopharmacol 2013 38 253 255 Brière FN Fallu JS Janosz M Prospective associations between meth/amphetamine (speed) and MDMA (ecstasy) use and depressive symptoms in secondary school students J Epidemiol Commun Health 2012 66 990 994 Colado MI Granados R O’Shea E The acute effect in rats of 3, 4-methylenedioxyethamphetamine (MDEA, ‘Eve’) on body temperature and long term degeneration of 5-HT neurones in brain: A comparison with MDMA (‘Ecstasy’) Pharmacol Toxicol 1999 84 261 266 10401727 Curran HV Travill RA Mood and cognitive effects of 3,4-methylenedioxymethamphetamine (MDMA, ‘ecstasy’): weekend ‘high’ followed by mid-week ‘low’ Addiction 1997 92 821 831 9293041 Derogatis L Nelisaratos N The Brief Symptom Inventory: An introductory report Psycholog Med 1983 13 595 605 Erritizoe D Frokjaer VG Holst KK In vivo imaging of cerebral serotonin transporter and serotonin (2A) receptor binding in 3,4-methylenedioxymethamphetamine (MDMA or ‘ecstasy’) and hallucinogen users Arch Gen Psychiatr 2011 68 562 576 21646575 Fox HC McLean A Turner JJD Neuropsychological evidence of a relatively selective profile of temporal dysfunction in drug-free MDMA (‘ecstasy’) polydrug users Psychopharmacol 2002 162 203 214 Greer G Tolbert R Subjective Reports of the Effects of MDMA in a Clinical Setting Journal of Psychoactive Substances 1986 18 319 327 Johnston LD O’Malley PM Brackman JG Monitoring the future national survey on drug abuse 1975 – 2004: Volume 2; College students and adults aged 19 – 45 Report for the US National Institute of Health, no. 05–5728 2005 National Institute on Drug Abuse Bethesda Kirkpatrick MG Gunderson EW Perez AY A direct comparison of the behavioral and physiological effects of methamphetamine and 3,4-methylenedioxymethamphetamine (MDMA) in humans Psychopharmacol 2012 219 109 122 Kish SJ Lerch J Furukawa Y Decreased cerebral cortical serotonin transporter binding in ecstasy users: A positron emission tomography/[(11)C]DASB and structural brain imaging study Brain 2010 133 1779 1797 20483717 McCann UD Sgambati FP Schwartz AR Sleep apnea in young abstinent recreational MDMA (‘ecstasy’) consumers Neurology 2009 73 2011 2017 19955499 McElhatton PR Bateman DN Evans C Congenital abnomalies after prenatal ecstasy exposure Lancet 1999 354 1441 1442 10543673 MacInnes N Handley SL Harding GFA Former chronic methylenedioxymethamphetamine (MDMA or ecstasy) users report mild depressive symptoms J Psychopharmacol 2001 15 181 186 11565625 Milani RM Parrott AC Turner JJD Gender differences in self-reported anxiety, depression and somatization among ecstasy/MDMA polydrug users, alcohol/tobacco users and nondrug users Addict Behav 2004 29 965 971 15219343 Montgomery C Hatton NP Fisk JE Assessing the functional significance of ecstasy-related memory deficits using a virtual reality paradigm Hum Psychopharmacol 2010 25 318 325 20521322 Moore DG Turner JD Parrott AC During pregnancy, recreational drug-using women stop taking ecstasy (3,4-methylenedioxy-N-methylamphetamine) and reduce alcohol consumption, but continue to smoke tobacco and Cannabis: initial findings from the Development and Infancy Study J Psychopharmacol 2010 24 1403 1410 19939863 Morgan MJ McFie L Fleetwood LH Ecstasy (MDMA): Are the psychological problems associated with its use reversed by prolonged abstinence? Psychopharmacol 2002 159 294 303 Munafo MR Araya R Cigarette smoking and depression: A question of causation Brit J Psychiatry 2010 196 425 426 20513848 Parrott AC MDMA in humans: Factors which affect the neuropsychobiological profiles of recreational Ecstasy users, the integrative role of bio-energetic stress J Psychopharmacol 2006 20 147 163 16510474 Parrott AC The psychotherapeutic potential of MDMA (3,4-methylenedioxymethamphetamine): an evidence-based review Psychopharmacology 2007 191 181 193 17297639 Parrott AC Cortisol and MDMA (3,4-methylenedioxymethamphetamine): Neurohormonal aspects of bioenergetic-stress in Ecstasy users Neuropsychobiol 2009 60 148 158 Parrott AC MDMA and serotonergic neurotoxicity: Empirical evidence for adverse effects in humans - no need for translation Brit J Pharmacol 2012 166 1518 1520 22404300 Parrott AC MDMA neurotoxicity: the functional implications of serotonin loss in recreational ecstasy users Neurosci Biobehav Revs 2013a 37 1466 1486 23660456 Parrott AC Human psychobiology of MDMA or ‘Ecstasy’: An overview of 25 years of empirical research Hum Psychopharmacol 2013b 28 289 307 23881877 Parrott AC Gibbs A Scholey AB MDMA and methamphetamine: some paradoxical negative and positive mood changes in an acute dose laboratory study Psychopharmacology 2011 215 527 36 21318566 Parrott AC Lasky J Ecstasy (MDMA) effects upon mood and cognition; before, during, and after a Saturday night dance Psychopharmacol 1998 139 261 268 Parrott AC Jones L Sands HR High cortisol levels in recent Ecstasy/MDMA users: Preliminary findings from the Swansea, Westminster and Dresden collaborative study British Psychological Society Annual Psychobiology Conference UK 3–5 September 2012 2012 Conference Abstract p.16 Parrott AC Lock J Conner AC Dance clubbing on-MDMA and during abstinence from MDMA: Prospective neuroendocrine and psychobiological changes Neuropsychobiol 2008 57 165 180 Parrott AC Milani RM Gouzoulis-Mayfrank E Cannabis and Ecstasy/MDMA (3,4-methylenedioxymethamphetamine): An analysis of their neuropsychobiological interactions in recreational users J Neural Transmiss 2007 114 959 968 Parrott AC Milani RM Parmar R Recreational Ecstasy/MDMA and other drug users from the UK and Italy: Psychiatric symptoms and psychobiological problems Psychopharmacol 2001 159 77 82 Puerta E Hervias I Aguirre N On the mechanisms underlying 3,4-methylenedioxymethamphetamine toxicity: the dilemma of the chicken and the egg Neuropsychobiology 2009 60 119 129 19893329 Raimo EB Schuckit MA Alcohol dependence and mood disorders Addict Behav 1998 23 933 946 9801727 Reay JL Hamilton C Kennedy DO MDMA polydrug users show process-specific central executive impairments coupled with impaired social and emotional judgment processes J Psychopharmacol 2006 20 385 388 16574712 Ricaurte GA McCann UD Szaboc Z Scheffelc U Toxicodynamics and long-term toxicity of the recreational drug, 3,4-methylenedioxymethamphetamine (MDMA, ‘Ecstasy’) Toxicology Letters 2000 112–113 143 146 Rogers G Elston J Garside R The harmful health effects of recreational ecstasy: A systematic review of observational evidence Health Technol Assess 2009 13 1 315 Sala M Braida D Endocannabinoids and 3,4-methylenedioxymethamphetamine (MDMA) interaction Pharmacol Biochem Behav 2005 81 407 416 15927242 Schifano F Albanese A Fergus S Mephedrone (4-methylmethcathinone; ‘meow meow’): chemical, pharmacological and clinical issues Psychopharmacol 2011 214 593 602 Schifano F Di Furia L Forza G MDMA (‘ecstasy’) consumption in the context of polydrug abuse: A report on 150 patients Drug Alc Depend 1998 52 85 90 Scholey AB Owen L Gates J Hair MDMA samples are consistent with reported Ecstasy use: Findings from a study investigating effects of Ecstasy on mood and memory Neuropsychobiology 2011 63 15 21 20962543 Scholey AB Parrott AC Buchanan T Increased intensity of Ecstasy and polydrug usage in the more experienced recreational Ecstasy/MDMA users: a WWW study Addict Behav 2004 29 743 52 15135556 Singer LT Salvator A Arendt RE Effects of cocaine/polydrug exposure and maternal psychological distress on infant birth outcomes Neurotoxicol Teratol 2002 24 127 135 11943500 Singer LT Linares TJ Ntiri S Psychosocial profiles of older adolescent MDMA users Drug Alc Depend 2004 74 245 252 Singer LT Moore DG Fulton S Neurobehavioral outcomes of infants exposed to MDMA (Ecstasy) and other recreational drugs during pregnancy Neurotoxicol Teratol 2012a 34 303 310 22387807 Singer LT Moore DG Min MO One-year outcomes of prenatal exposure to MDMA and other recreational drugs Pediatrics 2012b 130 407 413 22908109 Skelton MR Williams MT Vorhees CV Developmental effects of 3,4-methylenedioxymethamphetamine: A review Behav Pharmacol 2008 19 91 111 18332674 Soar K Turner JJD Parrott AC Psychiatric disorders in recreational Ecstasy (MDMA) users: A literature review focusing upon personal predisposition factors and drug histories Hum Psychopharmacol 2001 16 641 646 12404545 Topp L Hando J Dillon P Ecstasy use in Australia: Patterns of use and associated harm Drug Alc Depend 1999 55 105 115 Verheyden SL Maidment R Curran HV Quitting ecstasy: An investigation of why people stop taking the drug and their subsequent mental health J Psychopharmacol 2003 17 371 378 14870948 Winstock AR Griffiths P Stewart D Drugs and the dance music scene: A survey of current drug use patterns among a sample of dance music enthusiasts in the UK Drug Alc Depend 2001 64 9 17 Wu P Liu X Pham TH Ecstasy use among US adolescents from 1999 to 2008 Drug Alc Depend 2010 112 33 38 Zakzanis KK Campbell Z Memory impairment in now abstinent MDMA users and continued users: A longitudinal follow-up Neurology 2006 66 740 741 16534114

PMC005xxxxxx/PMC5111169.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101234237 32207 J Biomed Mater Res A J Biomed Mater Res A Journal of biomedical materials research. Part A 1549-3296 1552-4965 26841263 5111169 10.1002/jbm.a.35675 NIHMS757124 Article Persistence, Distribution, and Impact of Distinctly Segmented Microparticles on Cochlear Health following In Vivo Infusion3* Ross Astin M. 12 Rahmani Sahar 15 Prieskorn Diane M. 2 Dishman Acacia F 35 Miller Josef M. 2 Lahann Joerg 145 Altschuler Richard A. 2* 1 Department of Biomedical Engineering, University of Michigan, Ann Arbor 48109, USA 2 Kresge Hearing Research Institute, University of Michigan, Ann Arbor 48109, USA 3 Department of Biophysics, University of Michigan, Ann Arbor 48109, USA 4 Department of Chemical Engineering, University of Michigan, Ann Arbor 48109, USA 5 Biointerfaces Institute, University of Michigan, Ann Arbor 48109, USA * Corresponding author: astinr@umich.edu (Astin Ross) 11 2 2016 02 3 2016 6 2016 01 6 2017 104 6 15101522 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Delivery of pharmaceuticals to the cochleae of patients with auditory dysfunction could potentially have many benefits from enhancing auditory nerve survival to protecting remaining sensory cells and their neuronal connections. Treatment would require platforms to enable drug delivery directly to the cochlea and increase the potential efficacy of intervention. Cochlear implant recipients are a specific patient subset that could benefit from local drug delivery as more candidates have residual hearing; and since residual hearing directly contributes to post-implantation hearing outcomes, it requires protection from implant insertion-induced trauma. This study assessed the feasibility of utilizing microparticles for drug delivery into cochlear fluids, testing persistence, distribution, biocompatibility, and drug release characteristics. To allow for delivery of multiple therapeutics, particles were composed of two distinct compartments; one containing polylactide-co-glycolide (PLGA), and one composed of acetal-modified dextran and PLGA. Following in vivo infusion, image analysis revealed microparticle persistence in the cochlea for at least 7 days post-infusion, primarily in the first and second turns. The majority of subjects maintained or had only slight elevation in auditory brainstem response thresholds at 7 days post-infusion compared to pre-infusion baselines. There was only minor to limited loss of cochlear hair cells and negligible immune response based on CD45+ immunolabling. When Piribedil-loaded microparticles were infused, Piribedil was detectable within the cochlear fluids at 7 days post-infusion. These results indicate that segmented microparticles are relatively inert, can persist, release their contents, and be functionally and biologically compatible with cochlear function and therefore are promising vehicles for cochlear drug delivery. segmented microparticle electrohydrodynamic co-jetting cochlear drug delivery biocompatibility 1. Introduction There is increasing need for local drug delivery to the cochleae of patients with various cochlear pathologies, including cochlear implants. Local drug delivery could be used to increase survival and function of remaining auditory neurons following trauma such as implantation1,2. With the increasing number of implantees with remaining sensory cells and hearing, there is also growing interest in drug treatment for preservation and protection of remaining hair cells and their synaptic connections with the auditory nerve3. There are many potential strategies for providing local drug delivery to the cochlea (Table 1) including hydrogels, microfluidics, osmotic pumps, gene therapy/viral vectors, and micro/nanoparticles. Although multiple local delivery strategies are under development, systemic drug delivery, either orally or intravenously4, has historically been the primary mode of pharmaceutical delivery to the cochlea because it is relatively safe and non-invasive. However, because of the existence of the blood-labyrinth barrier, one of the principal limitations of systemic delivery is that therapeutic levels of the drug may never reach the cochlea. Further, systemic delivery is more likely to lead to unwanted side effects, as there is increased opportunity for the drug to act on other organ systems. Local delivery directly into the cochlea is therefore becoming a preferred strategy for the delivery of pharmaceuticals to the inner ear3,5. One option is intratympanic or transtympanic delivery that involves the placement of pharmaceuticals into the middle ear and relies on diffusion across the round window membrane (RWM) for agents to reach the inner ear. Such delivery is relatively non-invasive and exploits the permeability of the RWM. A drug can be injected through the tympanic membrane directly into the middle ear or more commonly, it is placed in a release medium (i.e. biodegradable hydrogel) on or near the RWM6. A disadvantage for use of intratympanic delivery is that it can lead to uncertainty as to the amount of drug that reaches the cochlea due to drug losses through the Eustachian tube in the middle ear, the size and charge of delivered particles, and variations in the thickness and composition of the RWM itself4. Another local option is intracochlear delivery directly into the cochlear fluids via surgical intervention. Unfortunately, the half-life of drugs directly injected into the cochlea is relatively short (minutes or hours) which limits the utility of this approach. Further, when a drug is injected, it is likely to accumulate at the site of injection rather than being distributed along the cochlear spiral. An additional intracochlear delivery method utilizes microfluidics, including osmotic pumps, reciprocating microfluidic systems, and incorporation of microfluidic channels within the cochlear implant4. Disadvantages of this method are that the pumps/channels must be refilled and their fill ports are susceptible to biofilm formation. A more recent strategy for intracochlear delivery is the use of viral vectors, or gene therapy to deliver molecular therapeutics to the cochlea wherein most recent studies have focused on the use of adenovirus (Ad) and adeno-associated viruses (AAV) as cochlear delivery vehicles7. Though promising, this approach is not without its concerns. Specifically, the use of a virus poses significant toxicity concerns related to immunogenicity and some of the vectors can be difficult to generate2. Further, because of lack of cell targeting vectors are randomly dispersed within the cochlea; thereby limiting the number of genes that can be efficiently delivered to auditory neurons. Effective delivery to the intended target is critical in order to have the desired therapeutic impact. Intracochlear drug delivery via a micro- or nanocarrier overcomes many of the limitations encountered with the aforementioned drug delivery techniques. In particular when placed in the cochlea, a micro-/nano-carrier enables a sustained release of drug over time (days or weeks) and could provide greater distribution and accumulation of the drug along the cochlear spiral therefore providing enhanced therapeutic benefit. As a result, polymeric micro- and nanoparticles have gained prominence as drug delivery vehicles. Several particle types have been utilized to facilitate therapeutic delivery to the inner ear including hydroxyapatite, silica, and polymeric nanoparticles8. Because of their wide-ranging and tunable properties, polymeric micro/nanoparticles are garnering increased interest for cochlear drug delivery. Specifically, polymeric materials such as poly-l-lysine (PLL), polylactic-co-glycolic acid (PLGA), and poly(L- caprolactone) (PCL) have been investigated as potential drug carriers to attenuate pathological conditions of the inner ear6,9. The characteristics of PLGA particles are particularly promising as they can be designed to meet many of the desired criteria for a drug delivery vehicle. These criteria include biocompatibility (PLGA is an FDA approved material and its degradation products are naturally eliminated from the body), controlled drug release (polymer degradation rates of weeks to months can be tuned to meet therapeutic needs by changing the ratio of lactic to glycolic acid), tailored size and shape enhance persistence and distribution of particles in the cochlea (resulting from control of fabrication process parameters), and potential for targeted delivery (via particle surface functionalization using the free chemical groups on the polymer surface). Moreover, polymer particles fabricated with electrohydrodynamic co-jetting (EHD) can be compartmentalized allowing loading of multiple pharmaceuticals into a single particle. Loading of multiple therapeutic agents is important to facilitate the delivery of drugs capable of attenuating the acute impact (swelling, etc.) of trauma and providing the sustained release of agents needed to potentially protect remaining hearing. Further, discrete compartments also enable pharmaceuticals with distinct pharmacokinetic profiles to be delivered from the same platform. This flexibility expands the combination of agents that can be simultaneously screened to identify the best drug combinations to ameliorate hearing loss following cochlear trauma; thereby ultimately improving the utility of the intervention. Our study tested intracochlear release of the dopamine agonist Piribedil from segmented microparticles. This glutamate inhibitor/anti-excitotoxic agent was chosen for three reasons: previous efficacy in protection from noise, current clinical use and potential for re-purposing, and inherent fluorescence10. The use of EHD also allows for fine control of particle shape and size, characteristics that have previously been demonstrated to impact drug delivery, macrophage uptake, and cell uptake of particles11. Cell binding is size and shape limited because test articles must be sufficiently small compared to a target cell. The size of target cells in the inner ear range from 8–20 µm in length, therefore a candidate microparticle with a diameter of approximately 8 µm was selected to encourage cell binding and limit particle phagocytosis by macrophages. Use of material that forms acidic, albeit natural byproducts, in the pH sensitive fluid environment of the cochlea, was controlled by modulation of particle characteristics such as size and porosity12. By selecting a particle size of approximately 8 µm for assessment of cochlear drug delivery, we are well below the size range of concern with respect to acid accumulation within particles13. Further, use of acetal dextran enables the fabrication of a porous particle, thereby inhibiting the development of an acidic particle interior because of increased exchange between the particle and the incubation medium in which it is contained. As porosity is increased, the ability to form extremely acidic microenvironments decreases because degradation products escape more readily from the interior and surrounding medium. In this study, we demonstrate the ability of dual carrier polymeric microparticles to persist and release their contents in a biocompatible manner within the cochlea. 2. Materials & Methods 2.1 Particle Materials Dextran, chloroform, dimethyl sulfoxide (DMSO), dimethylformamide (DMF), pyridinium p-toluenesulfonate, 2-methoxypropane triethylamine, acetic acid, sodium acetate, phosphate buffered saline (PBS), poly[(m-phenylenevinylene)-alt-(2,5-dihexyloxy-p-phenylenevinylene)] (MEHPV) as the blue marker for confocal imaging, and tween 20 were used as purchased from Sigma Aldrich, USA. Polylactide-co-glycolide (PLGA) with a molecular weight of 44 kDa and a ratio of 50:50 lactide to glycolide was purchased from Lactel Corporation. Piribedil was purchased from Ontario Chemicals. 2.2 Particle Fabrication Particles were made from polylactide-co-glycolide and dextran acetal (PLGA/dex). Dextran acetal was chemically modified from dextran according to previously published work by Bachnelder, et al14. A light-emitting polymer, poly[(m-phenylenevinylene)-alt-(2,5-dihexyloxy-p-phenylenevinylene (MEHPV), was also incorporated into the PLGA and PLGA/dex compartments to facilitate particle visualization via confocal microscopy. PLGA (MW: 44 kDa) with a lactic to glycolic acid ratio of 50:50 was used and PLGA/dex compartments contained 1:3 PLGA and 2:3 Dextran Acetal. All segmented particles were created via electrohydrodynamic co-jetting15, a process that involves a side-by-side capillary needle system containing polymer solutions and the application of an electric field to the system. The interface between the solutions is stabilized by the field enabling the formation of an electrified polymer jet of particles with multiple distinct compartments. Additional details on processing of the particles used in this study for cochlear delivery can be found in Supplemental information and Rahmani, et al16,17. The jetted microparticles were imaged via confocal laser scanning microscopy (CLSM) using an Olympus confocal microscope at the Microscopy and Imaging Laboratory facilities at the University of Michigan. Prior to an experiment, a known mass of particles was suspended in artificial perilymph (AP; 118 mM NaCl, 30 mM KCl, 2.0 mM MgSO4, 1.2 mM CaCl2, 5.0 mM HEPES; pH = 7.35–7.40, osmolality = 285–294 mOsm) or artificial perilymph with guinea pig serum albumin (GPSA) to create a 15 mg/mL infusion solution. 2.3 In vivo infusion Surgeries were conducted in a sterile environment and utilized aseptic technique. For in vivo infusions, Hartley guinea pigs (Charles River Laboratory, Wilmington, MA) were anesthetized and a post auricular approach was used to provide access to the middle ear. The temporal bone was drilled to visualize the cochlea and a fine pick was used to create a small hole in the basal turn of the cochlea near the round window. A microcannula with a silastic ball was inserted 0.5 mm into the basal turn of the scala tympani and cyanoacrylate was used to seal the cannula in place as outlined previously19. The microcannula was made from polyethylene 10 tubing and polyimide (I.D. = 0.12 mm, O.D. =0.16 mm). The silastic ball was made from Sylgard. A syringe infusion pump was used to deliver either a particle solution or a vehicle solution consisting of artificial perilymph and guinea pig serum albumin into the scala tympani of the guinea pigs at a flow rate of 1 µl/minute over 5 minutes. Infusions were always performed on the left ear and the right ear was used as needed for a contralateral control. NIH guidelines for the care and use of laboratory animals have been observed. 2.4 Harvesting, cryoprotection, and decalcification of cochlear specimens Guinea pigs were anesthetized and euthanized by injection of sodium pentobarbital. In all cases, secondary euthanasia was performed by transecting the aorta and ventricle. Animals were then decapitated and the temporal bones that encase the cochleae were detached. Excess bullar bone was removed to facilitate visualization of each cochlea and then the middle ear bones were also detached. Specimens were fixed in 4% paraformaldehyde (PFA) for 1–2 hours. Following fixation, cochleae were decalcified in a solution that was two-thirds formic acid and one-third 7% sucrose overnight. Prior to freezing, specimens were placed in foil molds and immersed in a 30% sucrose solution. Freezing was performed by placing the bottom of the container in contact with liquid nitrogen cooled 2-methyl-butane. Specimens were wrapped in parafilm and stored at −80°C until sectioning. 2.5 Cryostat Sectioning Samples were cut into 14 µm sections. For stereological samples, the cochleae were sectioned up to a depth of approximately 4000 µm. Every 6th section was collected. A random number generator was used to select a number between 1 and 6. The number generated identified the first slide for analysis in each cochlea. Thereafter, every 6th slide was evaluated such that the slides with numerical markings of 1, 7, 13, etc. were exhaustively assessed. A total of 61 slides were generated for each animal and 10 slides from each animal were assessed to ascertain particle number and distribution. For immunohistochemistry, up to 4 midmodiolar sections were taken from the MP infused cochlea of each animal. These sections were stained with CD45, a leukocyte antigen, to denote immune cell activity, and propidium iodide (PI) to indicate the presence of general cell structures such as nuclei. Cryosections of guinea pig liver were also made for use as positive and negative (in the absence of primary antibody) controls. The number of respective CD45+ and PI+ cells in cochlear cross sections were counted and the ratio CD45+ cells to total cells (CD45+ and PI+) was calculated to determine the percentage of CD45+ present within treated and untreated cochleae. 2.6 Infused particle number and persistence A sample of the particle solution was counted before the infusions using a hemacytometer. Persistence and distribution assessments were conducted using cryosections from cochleae that had particle infusions 7 days prior to harvest of the cochlea (n=3). Sections from various depths of the cochlea were sampled for particle number. The guinea pig cochlea consists of four turns with two perilymphatic compartments, the scala tympani and scala vestibule. During assessment, the location and distribution of particles within intact cochlea and cochlear cross-sections was determined. 2.7 Stereological Analysis Stereological analysis was used to determine particle distribution in animals infused with unloaded PLGA/dex particles (n=3) and to estimate the number of particles entering the cochlea at the time of infusion. In particular, the optical dissector method was used to systematically create image slices that contained particles in tissue at multiple planes within the cochlea. This technique is an unbiased method whereby an object is counted the first time it appears in an image18. A confocal microscope with a 405 nm laser was used to acquire z-stack image series of particles within cochlear cross-sections. The z-step size was 3.6 µm. Every 6th slide from each sample was evaluated using an unbiased counting grid composed of green inclusion lines and red exclusion lines. The grid was superimposed on top of specimen images and a particle was counted if it was inside of one of the squares in the counting grid or in contact with a green inclusion line. The total particle number within an infused cochlear sample could be estimated by multiplying the number of particles counted within the assessed samples by a factor of 6. 2.8 Auditory Brainstem Response (ABR) Animals were anesthetized with xylazine (10 mg/kg intramuscularly) and ketamine (40 mg/kg intramuscularly). Needle electrodes (active, reference, and ground) were inserted subcutaneously at the vertex and below each pinna and used to record the neurologic response. Up to 1024 responses were averaged for each stimulus level, with the stimulus consisting of a 15-msec tone burst, provided at 10/sec. Pure tones were delivered via a transducer coupled to the external auditory canal at 4, 8, and 20 kHz. Initial sound levels were set at 80 dB for pre- and post- infusion tests. Threshold determination was made by non-blinded evaluators based on the visual detection of maximum peak–peak amplitude of the resulting waveforms. ABRs were performed prior to infusion to enable exclusion of animals with abnormal hearing, and to enable detection, if present, of threshold shifts post-infusion. 2.9 Hair cell counts To prepare specimens for hair cell analysis, ears were harvested in the same manner as those used for cryosection preparation with a few notable differences. Following removal of the middle ear bones, the apex was visualized under stereoscopic magnification and slightly perforated with a 28G needle to create a small hole. Then the round window was opened and approximately 300 µL of 4% PFA was infused directly into the cochlea via the hole in the apex. Specimens were postfixed by immersion in 4% PFA overnight. The following day, cochleae were rinsed and the otic capsule, lateral wall, and tectorial membrane were carefully removed. Phalloidin was used to stain the modiolous and the attached organ of Corti. Following rinsing to remove excess stain, the organ of Corti was dissected from the modiolous and each turn was mounted onto a microscope slide and coverslipped. Phalloidin staining of the organ of Corti enabled visualization and counting of both inner and outer hair cells (as indicated by the presence of nuclei and/or stereocillia). The blinded counts were performed as described previously21. Counts were then plotted using a cytocochleogram program developed in house20, and depicted as percentage hair cell loss at a particular distance from the apex as compared to a database of normal guinea pigs (those not exposed to any external stimuli or agents that could induce hearing loss). Tracking distance along the cochlear spiral also facilitated the correlation of areas of loss with known frequency maps of the guinea pig cochlea. This provided insight on areas that may be functionally affected by the treatment. 2.10 Immunohistochemistry Specimens were prepared similarly to those used for cytocochleograms, except post fixation was performed for 2 hours rather than 12. Then the same protocol was followed as outlined in 2.4. Midmodiolar sections, 12 in total, were selected from 3 of the animals used for the collection of Piribedil exposed perilymph in vivo. These sections were assessed with CD45, a leukocyte antigen, to determine the extent to which particle infusion induced a local immune response. Sections from contralateral control ears and the liver from one of the guinea pigs (in the absence of primary antibody incubation) served as negative controls. In the presence of CD45 primary antibody, cryosections of the liver were used as positive controls. The specimens were pre-treated for 15 minutes with 0.3% Triton X-100 in PBS, followed by PBS rinsing (3×5 minutes). Each section was blocked for 1 hour with 5% goat serum, permeabilized for 30 minutes with 0.3% Triton in 5% goat serum, and incubated with a primary antibody for CD45 (mouse anti-guinea pig), and PBS (1:50). The diluted primary solution was incubated with plates overnight at 4°C in a sealed chamber. 2.11 Statistical analysis An analysis of variance (ANOVA) was performed on the hair cell counts of MP infused cochleae to determine whether significant quantitative differences existed between sensory cell viability in treated and untreated ears. Using SPSS software, a two factor model, with factor 1 = ear and factor 2 = turn was employed to assess whether any observed differences in hair cell loss were significantly different between treated and untreated ears and between turns. Further bonferroni post-hoc correction was used to account for the multiple comparisons made. The p value for significance was .05. 2.12 In vivo Piribedil release Liquid chromatography-mass spectrometry (LC-MS) analysis was performed to determine the Piribedil concentration in cochlear fluid samples. Use of LC allows separation of the drug from a biologically relevant background matrix (native or artificial perilymph) and mass spectrometry enables detection at very low concentrations. The extraction solvent contained 997.5 µL acetonitrile + 2.5 µL Acar mix. Six standards (0–3000 nM), were prepared by spiking native or artificial perilymph with 3 µM stock Piribedil solutions that had been diluted with the extraction solvent. Experimental samples obtained from animals exposed to Piribedil-loaded microparticles were mixed with extraction solvent and vortexed, followed by 5-minute incubation at 4°C. Two vortex/cooling cycles were completed prior to centrifugation of the samples at 15,000 rpm and 4°C for 5 minutes. Sample supernatants were transferred to autosampler vials and subjected to LC-MS analysis utilizing an Agilent 1200 RRLC coupled to an Agilent 6410 Triple Quad LC/MS. Analyte peaks were resolved with the use of a Waters Xbridge C18 (50 mm × 2.1 mm, particle size 2.5 µm) column. The liquid chromatographic conditions were as follows: mobile phases: A = 5mM ammonium acetate, adjusted to pH 9.9 with ammonium hydroxide; B = acetonitrile; flow rate: 0.25 mL/min; and injection volume: 2 µL. The gradient began at 25%B and followed a linear regime from 25% to 75%B over 5 minutes before increasing to and holding at 100% for 3 minutes. Then it was returned to 50%B and re-equilibrated for 4 minutes, thereby resulting in a total run time of 12 minutes/injection. The mass spectrometry source conditions were: gas temp: 325°C, gas flow: 10 L/minute, nebulizing pressure: 40 psi, and capillary voltage: 4000 V. All mass spectrometry readings were collected in positive ion mode. Though the total volume of perilymph in the cochlea is 10 µL, the volume of perilymph in the scala tympani, the chamber into which the particle solution is infused, is only 4.7 µL5. The remaining fluid consists of fluid from the scala vestibule or cerebrospinal fluid therefore a simple correction factor is attained by dividing total perilymph volume by collected perilymph volume. 3. Results 3.1 Particle Fabrication The PLGA/dex particles were fabricated as described previously in Rahmani, et al16,17. Average particle diameter was 7.7 ± 0.12 µm and was determined from ImageJ analysis of scanning electron microscope (SEM) images of the microparticles. Further characterization of the segmented microparticles, such as zeta potential measurements, morphology, and size distribution can be found in Supplemental information and Rahmani, et al16, 17. 3.2 Infused particle number and persistence The number of particles released into the cochlea at the time of infusion was examined. On average, 350,000 particles were infused into each cochlea as counted by a hemacytometer. The persistence of the aforementioned infused particles within the cochlea was also examined. On day 7 following microparticle infusions, untargeted unloaded particles were distributed in the cochlea (Figure 1). The vast majority of particles were located in the first turn of the cochlea (94±17%), followed by the second turn (6±2%) and the third and fourth turns (0%). The third and fourth turns were combined during analysis due to the relatively few numbers of particles found in each turn alone. 3.3 Stereological Analysis In order to facilitate an estimation of the remaining particle number, it was necessary to use stereology, an analysis method that utilizes random, systematic sampling to count objects, in this case, fluorescent particles within cochlear cross-sections. Though methodological restraints prevented retroactive analysis of the samples discussed in section 3.2, particle persistence and distribution of one timepoint, 7 days, was used for stereological analysis. In the new group of animals (n=3), the distribution profile was similar to the previous samples and the average number of particles persisting after 7 days was 20,000±4000. No particles were found in the contralateral ear and a comparison of the number of persisting particles to the number of infused particles demonstrated that approximately 5.7% of untargeted particles were retained in the cochlea following infusion. 3.4 Auditory brainstem responses (ABRs) The hearing thresholds of the animals used in this study were assessed both pre- and post-infusion. ABRs were administered before the start of the experiment to promote the inclusion of only those animals with behaviorally normal hearing (particularly in the lower two-thirds of the cochlea; the upper third was beyond the frequency threshold that can be measured non-invasively). Baseline hearing assessment also provided a basis for comparison to ABR results following infusion to assess particle impact on hearing. Post-infusion testing was conducted at the same three frequencies, 4, 8, and 20 kHz as the pre-test. If the intensity level required for the subject to detect the stimulatory tone at any of the frequencies changed, then a threshold shift occurred. Hearing was considered to have worsened if the sound intensity needed for the subject to detect the stimulatory tone was more than 10 decibels higher than in the pre-test. This value was selected because it has been shown to be outside the bounds of normal experimental variation and therefore can be indicative of an actual functional change in hearing capacity22. A total of eight animals underwent ABR assessment, five that received microparticle infusions and three that were infused with a vehicle solution consisting of artificial perilymph and guinea pig serum albumin. As seen in Figure 2, following microparticle infusion, three out of five animals had hearing within the normal range as compared to their pre-infusion values that is indicated by the threshold shift. One of the MP infused animals, GP 1, demonstrated a threshold shift (>10 dB) at one frequency, 20 kHz. Another MP infused animal, GP 3, demonstrated threshold shifts at all frequencies. Following vehicle infusion, only one animal, GPV 2, demonstrated a large threshold shift at 4 kHz (Figure 2). Therefore, it was imperative to evaluate these results in conjunction with hair cell viability, a morphological indicator of cochlear health, to enable the most complete interpretation of experimental outcomes. 3.5 Cytocochleograms In addition to functional performance, cochlear cell/tissue appearance, particularly the presence or absence of hair cells and immune cells, was evaluated following microparticle and vehicle infusion. Cytocochleograms provided a visual representation of areas of missing hair cells and the location of these cells relative to the apex and base of the cochlea. All infused cochleae were assessed via cytocochleogram and contralateral cochleae were evaluated as needed. Based on previous studies, an area was considered normal if the outer hair cell (OHC) loss occurred intermittently and was 20% or less23. Further, there should be almost negligible loss of inner hair cells (IHCs) observed. It should be noted that in the Hartley strain of guinea pig used in these studies, it is not uncommon to have untreated animals that have large amounts of hair cell loss in the apical third of the cochlea that cannot be detected by ABR during prescreening [unpublished observation] because of the technical limitations of the testing equipment. The distance from the aforementioned area is sufficiently far however, from the lower sixth of the cochlea where the particles were infused to enable the determination of the impact of particle delivery on hair cells/hair cell viability. Upon assessment, two of the microparticle infused animals used for cytocochleograms (GP 1 and GP 4) were found to have the aforementioned apical hair cell loss. However, in all but one case, microparticle infused animals had hair cell losses in all rows of less than 20% in the areas adjacent to the site of infusion (Figure 3). Further, the cytocochleogram of GP 3 demonstrated minimal hair cell loss; therefore, functional deficits identified at the lower frequencies during ABR were likely the result of conductive hearing losses due to the presence of middle ear fluid that was detected during gross dissection. Therefore, microparticle infusion was well tolerated in four out of five animals with only one animal, GP 1, displaying functional and histological losses. For vehicle infused animals, two of three animals, GPV 1 and GPV 3, did not exceed the functional loss criteria as determined by ABR (Figure 2). Further, for GPV 2, the decrement seen at 20 kHz could be explained by the presence of scar tissue, particularly when evaluated in conjunction with the hair cell counts obtained for that animal. Upon histological evaluation by cytocochleogram, all of the vehicle infused animals had minimal to limited hair cell losses near the cochleostomy, including GPV 2, which had hair cell losses in all rows of less than 20% in the areas adjacent to the site of infusion (Figure 4). A moderate apical hair cell loss was also demonstrated in GPV 3, one of the vehicle infused animals (Figure 4). In addition, within MP infused animals, a 2-factor analysis of variance (ANOVA) on OHC loss determined that both ear and turn losses were significant with a p-value below the α=.05 level indicating that treated ears and locations in the higher turns demonstrated greater losses. Subsequent analyses compared the OHC loss of treated and untreated ears at each individual turn (Figure 5). OHC loss in turn 1, near the infusion site, was not significantly different between treated and untreated ears. In turn 2, the difference was statistically significant at .02, however, as average losses for both the non-treated and treated ears were well under 10%, the difference may not be as functionally meaningful. Interestingly, turns 3 and 4 were significantly different when treated (left ears) are compared to non-treated (right ears). It is likely however, that differences observed are related to factors other than the physical presence of the particles because distribution studies demonstrated that particles were primarily located in the first and second turns. 3.6 Immunohistochemistry White blood cells were present in both treated and non-treated ears. The cells were primarily located near vasculature and within the spiral ligament. This finding correlates well with other work reporting the presence of macrophages within the native cochlea24. Further, the number of white blood cells present in treated ears (7.21% ± 4.62) was comparable to that seen in the untreated ears (6.57% ± 3.38) indicating that at 7 days post-infusion, the immune response to the particles was negligible (Figure 6). 3.7 In vivo Piribedil release Piribedil-loaded microparticles contained 2 w/w% Piribedil and a 7-day post-infusion timepoint was selected to perform perilymph sampling to ascertain in vivo concentration of Piribedil release from the particles. The total volume of fluid collected ranged from approximately 6.9–10.2 µL and Piribedil was detected in all analyzed samples (Table 2). Although Piribedil was present in all samples, the apparent concentrations of the drug at 7 days post-infusion were below therapeutic level10. 4. Discussion Historically, much of the focus with regard to cochlear drug delivery has been via systemic administration or acute injection of free drug. Carriers, particularly microparticles made from biodegradable, tunable polymers have the potential to facilitate sustained local drug release to the cochlea. In this study, the polymeric microparticles used were observed to accumulate primarily within the first turn. This finding was not unexpected as this location was the site of infusion and existence of fluid flow in the cochlea is negligible. During infusion, the dispersion of particles to other turns relies primarily upon diffusion and the small flow induced by the micropump. The infused particles were found in comparable numbers and distributions 1 and 7 days post-infusion thereby indicating that particles capable of surviving acute clearance (washout via eustachian tube, immune response, etc.) were able to remain in the cochlea for an extended period. Further, no particles were detected in the contralateral ear, indicating that the delivery platform does not diffuse to the contralateral cochlea via cerebrospinal fluid, nor were they seen in the liver, as can occur with nanoparticles25, thereby eliminating concerns about inadvertent effects. Though the percentage of remaining particles was small (5.7% ± 1.1), the utility of delivery from particles following infusion would be dependent on the dose of drug needed to achieve therapeutic level and the extent to which the drug was able to be incorporated (weight %) into the particles. Further, the percentage observed could be a consequence of the sample processing required to generate the particle cross-sections used in the analysis. During incubation in the decalcifying solution and/or the cryoprotection solution, particles located in the fluid spaces could be displaced, meaning that only particles closely associated with the lining of the cochlear chambers would remain for analysis. Therefore, it is possible that non-adherent particles were simply “washed out” of the system before they could be assessed. Potential modifications to the surface of the microparticles to include ligands targeting cochlear structures or implant materials may also help to increase the number of attached particles. The majority of infused animals had comparable or only slightly elevated post-infusion hearing thresholds and minimal hair cell loss following infusion. Hair cells act as mechanotransducers and enable the conversion of fluid movement into electrical signals within the cochlea that can be then be interpreted by the brain. These cells are sensitive to damage from both chemical and physical means and their absence impairs auditory function. One of the microparticle infused animals with an apparent post-infusion threshold shift was GP 3. This animal had an observable threshold increase at all frequencies, however the greatest shifts occurred at 4 and 8 kHz; this physiology correlated well with its gross anatomy in that fluid was present in its middle ear and excess fluid makes it more difficult for sound to be transmitted, particularly at lower frequencies. Further, GP 3 had similar numbers of viable hair cells as non-shifted animals, therefore conductive losses rather than particle presence caused the detected functional deficits. The almost negligible loss of inner hair cells and normal losses of outer hair cells near the site of infusion indicate that locally delivery of segmented microparticles was well tolerated in four out of five microparticle infused animals. One animal, GP 1, however, did experience functional and histological detriment following MP infusion. Within the three vehicle animals that received an infusion of the vehicle solution consisting of artificial perilymph and guinea pig serum albumin, one had an apparent threshold shift. This animal, GPV 2, had a large (greater than 40 dB) threshold shift at 4kHz. Examination and analysis following harvesting of the cochlea GPV 2 revealed a normal cytocochleogram and the presence of scar tissue on the treated cochlea although the tympanic membrane and middle ear were clear. This finding indicates that the functional deficit observed is also likely the result of conductive losses. The two remaining vehicle infused animals tolerated the procedure well and did not have extensive amounts of hair cell loss near the site of infusion. Upon further examination of the ANOVA analysis, the determination that both ear and turn were significant factors in predicting hair cell loss was found to be primarily based on differences between OHC presence in turns 3 and 4 of treated ears. The statistically significant difference observed in treated versus untreated ears in turns 3 and 4 seems unusual because distribution studies determined that the majority of particles were in the first and second turns. Further, based on the site of infusion, any potential hair cell loss would be expected to occur at the base of the cochlea. It is possible that the differences seen are related to an increase in intracochlear pressure induced by surgery or the viscosity of the delivery solution formed from artificial perilymph and guinea pig serum albumin as this increased trauma was also present when a vehicle solution was used. To alleviate potential pressure changes caused by displacement of native perilymph during infusion, an outlet hole could be drilled to provide an alternative exit to the cochlear aqueduct, thereby enabling removal of excess fluid from the perilymphatic space26. Further, future work could include assessment of the functional and pathophysiological impact of the viscosity of the delivery solution and incorporate this knowledge into future iterations of the particle delivery protocol. In addition, surface modification could enable particle attachment to implants and eliminate need for a delivery solution. The apparent biocompatibility of PLGA/dextran microparticles with the cochlea was not unexpected due to the composition of the constituent materials. PLGA is a FDA approved biodegradable polymer and it, along with its degradation products, is relatively biologically inert14. In addition, Dextran is a natural polysaccharide that is also well tolerated in the body14. Further, the detection of Piribedil, a proof of principle medication in this study, demonstrated that drug release from the particles is measurable albeit currently sub-therapeutic within cochlear fluids at 7 days post-infusion; Piribedil has poor solubility that resulted in low percentage incorporation into the polymer matrix of the microparticle. An alternative non-competitive glutamate agonist with better solubility could be utilized in future work to address the efficacy of intervention, thereby increasing the likelihood that therapeutic level could be attained in vivo. 5. Conclusions This work has demonstrated the feasibility of delivering microparticles with multiple distinct compartments to the cochlea in vivo. Specifically it has characterized the in vivo persistence, distribution, and drug release from the chosen particle system. Particles persisted within the cochlea for at least seven days. Following particle infusion, there was only a slight impact on cochlear functionality in the majority of animals and a robust immune response was not induced. Cell survival and morphology were also well maintained in areas of the cochlea with high numbers of particles. This technology represents a tunable platform with potential for intracochlear drug delivery as targeted free particle suspensions or particles attached to an auditory implant. Future work should consider the impact of the viscosity of the particle delivery solution on cochlear functionality and assess efficacy of drug delivery from segmented particles in a trauma model. Supplementary Material Supp Fig 01a Supp Fig 01b Supp Info The authors acknowledge funding from the National Institute of Deafness and Communication Disorders (NIH-NIDCD 5R01 DC011294-01), the Multidisciplinary University Research Initiative of the Department of Defense and the Army Research Office (W911NF-10-1-0518), the DOD through an idea award (W81XWH-11-1-0111), and the Tissue Engineering and Regenerative Medicine Training Grant (DE00007057-36). We also thank Noel Wys and Catherine Martin for their technical assistance. Figure 1 Particle distribution of unloaded microparticles observed 7 days after in vivo infusion. A) Percentage distribution of particles within the first four turns of the guinea pig cochlea where turn 1 (B–A) is the closest to the cochleostomy/infusion site and turn 2 (B–B) is the next adjacent turn. Figure 2 Cochlear function 7 days post MP or vehicle infusion as represented by auditory brainstem responses. For all animals, threshold shift (difference between pre- and post-infusion thresholds) at 4, 8, and 20 kilohertz are shown. A threshold shift greater than 10 dB (*) is indicative of a change in an animal’s hearing. At the majority of frequencies across animals, no shift is observed, though at 20 kHz GP 1 and 3 both demonstrate shifts and GP 3 also demonstrates shifts at 4 and 8 kHz. Among vehicle infused animals, GPV 2 demonstrates a notable shift at 4 kHz. Based on anatomical observations, the threshold shifts observed at lower frequencies are likely the result of conductive losses rather than the physical presence of the microparticles. Threshold shifts for GP and GPV animals are denoted by patterned or solid columns, respectively. GP= microparticle infused animals. GPV= vehicle infused animals. Figure 3 Cochlear hair cell death as represented by cytocochleograms (graphical representation of hair cell loss) at day 7 post-infusion for microparticle infused guinea pigs (n=5) receiving 5µL of microparticles. The regions of auditory brainstem response (ABR) functional testing as well as the cochleostomy site have been indicated along the x-axis. In all animals, hair cell loss near the cochleostomy site in the infused ear is minimal. Cytocochleograms for all left/treated ears are shown and a representative cytocochleogram (GP 1) of a right/untreated ear is provided in the upper right panel of the figure. Figure 4 Cochlear hair cell death as represented by cytocochleograms (graphical representation of hair cell loss) at day 7 post-infusion for vehicle infused guinea pigs (n=3) receiving 5µL of vehicle solution. The regions of auditory brainstem response (ABR) functional testing as well as the cochleostomy site have been indicated along the x-axis. In all animals, hair cell loss near the cochleostomy site in the infused ear is limited. Cytocochleograms for all left/treated and right/untreated ears are shown. Figure 5 Cochlear outer hair cell loss comparison. Graphical depiction of the OHC loss as a function of ear (treated/left vs. non-treated/right) and turn at 7 days post microparticle infusion. A two-factor analysis of variance (ANOVA) with α=.05 found that turn (p=0.00) and ear (p=0.00) were both significant. The differences between ears were primarily driven by OHC losses in turns 3 and 4. Differences between ears in turn 2 were functionally negligible and did not exist in turn 1. OHC= outer hair cell. Black bars are standard deviation. Figure 6 CD45 presence in treated and untreated cochleae. Resident white blood cells may be in the normal cochlea and no differences are seen in the number of these cells present at 7 days following MP infusion. A) A representative cochlear cross-section wherein leukocytes are identified as nuclei (red) whose surface/cell membrane is positive for CD45 (green). B) Quantification of the percentage of CD45+ cells present within treated and untreated cochlear cross-sections. Black bars are standard deviation. Table 1 Potential Strategies for Local Cochlear Drug Delivery Contains a synopsis of various platforms investigated for local cochlear drug delivery and outlines the advantages, limitations, and distribution profile of each type of carrier/strategy. Potential Inner Ear Carriers Advantages Limitations Cochlear Distribution (turn) Hydrogel/Sponge Ease of insertion/non-invasive; biodegradable Extracochlear/Intratympanic application; drug incorporation by diffusion; relies on RWM permeability Basal Microfluidics/Mini-Osmotic Pumps Continuous delivery; real time modulation of infusion rate Must be refilled/size; biofilm formation; finite power supply Basal to apex Viral Vector Induce regeneration Induce immune response; toxicity Variable Biodegradable Micro/nanoparticles Biodegradable; size; multifunctional Invasive; release rate predetermined Variable Table 2 Intracochlear Concentration of Piribedil at 7 Days Post-infusion of Piribedil-loaded Microparticles Contains the volumes, apparent concentrations, and volume corrected concentrations from each sample. The corrected concentrations were obtained by multiplying the apparent concentration by a correction factor derived from dividing total perilymph volume by collected perilymph volume. Specimen Volume (µL) Apparent Concentration (nM) Volume Corrected Concentration (nM) GP 11 10.20 339.23 736.20 GP 13 9.00 106.04 207.57 GP 22 8.40 207.41 370.68 GP 25 7.22 205.73 316.04 GP 15 6.88 223.43 327.07 GP 26 7.10 192.01 290.06 Average 8.13 212.31 374.60 Standard Deviation 1.31 74.84 185.20 3* Conflict of Interest: No benefit of any kind will be received either directly or indirectly by the author(s). References 1 Eshraghi AA van de Water TR Cochlear implantation trauma and noise-induced hearing loss: apoptosis and therapeutic strategies Anat Rec A Discov Mol Cell Evol Biol 2006 288A 473 481 2 Chien WW Monzack EL McDougald DS Cunningham LL Gene therapy for sensorineural hearing loss Ear Hearing 2015 36 1 1 7 25166629 3 Hendricks JL Chikar JA Crumling MA Raphael Y Martin DC Localized cell and drug delivery for auditory prostheses Hearing Res 2008 242 1–2 117 131 4 Pararas EEL Borkholder DA Borenstein JT Microsystems technologies for drug delivery to the inner ear Adv Drug Deliver Rev 2012 64 14 1650 1660 5 Salt AN Plontke SK Principles of local drug delivery to the inner ear Audiol Neurotol 2009 14 6 350 360 6 Anderson M Johnston AH Newman TA Dalton PD Rask-Andersen H Internalization of nanoparticles into spiral ganglion cells J of Nanoneurosci 2009 75 84 7 Staecker H Praetorius M Brough DE Development of gene therapy for inner ear disease: Using bilateral vestibular hypofunction as a vehicle for translational research Hearing Res 2011 276 44 51 8 Jager W Goiny M Herrera-Marschitz M Brundin L Fransson A Canlon B Noise-induced aspartate and glutamate efflux in the guinea pig cochlea and hearing loss Exp Brain Res 2000 134 4 426 434 11081824 9 Lü J-M Wang X Marin-Muller C Wang H Lin PH Yao Q Chen C Current advances in research and clinical applications of PLGA-based nanotechnology Expert Rev Mol Diagn 2009 9 4 325 341 19435455 10 d'Aldin C Puel J-L Leducq R Crambes O Eybalin M Pujol R Effects of a dopaminergic agonist in the guinea pig cochlea Hearing Res 1995 90 202 211 11 Champion JA Mitragotri S Shape Induced Inhibition of Phagocytosis of Polymer Particles Pharmaceutical Res 2009 26 1 244 249 12 Dunne M Corrigan OI Ramtoola Z Influence of particle size and dissolution conditions on the degradation properties of polylactide-co-glycolide particles Biomaterials 2000 21 16 1659 1668 10905407 13 Fu K Pack DW Klibanov AM Langer R Visual evidence of acidic environment within degrading poly(lactic-co-glycolic acid) (PLGA) microspheres Pharmaceutical Res 2000 17 1 100 106 14 Bachelder EM Beaudette TT Broaders KE Dashe J Fréchet JMJ Acetal-derivatized dextran: An acid-responsive biodegradable material for therapeutic applications J Am Chem Soc 2008 130 32 10494 10495 18630909 15 Lahann J Recent progress in nano-biotechnology: compartmentalized micro- and nanoparticles via electrohydrodynamic co-jetting Small 2011 7 9 1149 1156 21480519 16 Rahmani S Park T-H Dishman AF Lahann J Multimodal delivery of irinotecan from microparticles with two distinct compartments J Controlled Release 2013 172 1 239 245 17 Rahmani S Ross AM Park T-H Durmaz H Dishman AF Prieskorn D Jones N Dual release carriers for cochlear delivery Adv Healthc Mater 2015 Advance online publication 18 West MJ Design-based stereological methods for counting neurons Prog Brain Res 2002 135 43 51 Review 12143362 19 Prieskorn DM Miller JM Technical report: chronic and acute intracochlear infusion in rodents Hearing Res 2000 140 1–2 212 215 20 Piu F Wang X Fernandez R Dellamary L Harrop A Oinag Y Sweet J Tapp R Dolan DF Altschuler RA Lichter J LeBel C OTO-104: A sustained release dexamethasone hydrogel for the treatment of otic disorder Otol Neurotol 2011 32 171 179 21099726 21 Schacht J Altschuler R Burke DT Chen S Dolan D Galecki AT Alleles that modulate late life hearing in genetically heterogeneous mice Neurobiol of Aging 2012 33 8 15 29 22 Tanaka C Nguyen-Huynh A Loera K Stark G Reiss L Factors associated with hearing loss in a normal-hearing guinea pig model of hybrid cochlear implants Hear Res 2014 316 82 93 25128626 23 Ou HC Bohne BA Harding GW Noise damage in the C57BL/CBA mouse cochlea Hearing Res 2000 145 111 122 24 Shi X Resident macrophages in the cochlear blood-labyrinth barrier and their renewal via migration of bone-marrow-derived cells Cell Tissue Res 2010 342 1 21 30 20838812 25 Praetorius M Brunner C Lehnert B Klingmann C Schmidt H Staecker H Schick B Transsynaptic delivery of nanoparticles to the central auditory nervous system Acta OtoLaryngol 2007 127 5 486 490 17453474 26 Borkholder DA Zhu X Frisina RD Round window membrane intracochlear drug delivery enhanced by induced advection J. Controlled Release 2014 174 171 176

PMC005xxxxxx/PMC5111634.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7702118 4161 Immunol Rev Immunol. Rev. Immunological reviews 0105-2896 1600-065X 27782333 5111634 10.1111/imr.12499 NIHMS817361 Article The Immune System’s Role in Sepsis Progression, Resolution and Long-Term Outcome Delano Matthew J. 1 Ward Peter A. 2* 1 Department of Surgery, Division of Acute Care Surgery, University of Michigan 2 Department of Pathology, University of Michigan Medical School * Correspondence should be directed to: Peter A. Ward, M.D., Godfrey D. Stobbe Professor of Pathology, University of Michigan Medical School, Department of Pathology, 1301 Catherine Street, 7520 MSRB I, Ann Arbor, MI 48109-5602, Tel. 734-647-2921, Fax 734-764-4308, pward@umich.edu 22 9 2016 11 2016 01 11 2017 274 1 330353 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. SUMMARY Sepsis occurs when an infection exceeds local tissue containment and induces a series of dysregulated physiologic responses that result in organ dysfunction. A subset of patients with sepsis progress to septic shock, defined by profound circulatory, cellular, and metabolic abnormalities, and associated with a greater mortality. Historically, sepsis-induced organ dysfunction and lethality were attributed to the complex interplay between the initial inflammatory and later anti-inflammatory responses. With advances in intensive care medicine and goal-directed interventions, early 30-day sepsis mortality has diminished, only to steadily escalate long after “recovery” from acute events. Since so many sepsis survivors succumb later to persistent, recurrent, nosocomial and secondary infections, many investigators have turned their attention to the long-term sepsis-induced alterations in cellular immune function. Sepsis clearly alters the innate and adaptive immune responses for sustained periods of time after clinical recovery, with immune suppression, chronic inflammation, and persistence of bacterial representing such alterations. Understanding that sepsis-associated immune cell defects correlate with long-term mortality, more investigations have centered on the potential for immune modulatory therapy to improve long term patient outcomes. These efforts are focused on more clearly defining and effectively reversing the persistent immune cell dysfunction associated with long-term sepsis mortality. inflammation sepsis immune suppression sepsis innate immune dysfunction adaptive immune dysfunction INTRODUCTION Until recently, sepsis was defined as the constellation of symptoms occurring when a bacterial, viral or fungal infection leads to a systemic inflammatory response syndrome (SIRS), including fever, leukocytosis or leukopenia, and decreased vascular resistance frequently leading to hypotension (septic shock), organ failure (severe sepsis) and death(1, 2) (Fig.1). However, vagueness in definitions and ineffective clinical strategies have led to discrepancies in the incidence of sepsis and the observed mortality(3). In response, the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) were developed and address the limitations of previous definitions that were over focused on SIRS and inflammation(4). In addition, the Consensus also dispelled the longstanding notion that SIRS criteria possess adequate specificity and sensitivity to define and diagnose sepsis. Lastly, the report debunked the misleading model that sepsis always follows a linear continuum from the SIRS through severe sepsis and septic shock, and declared the term “severe sepsis” redundant and unnecessary. Instead, the Consensus report recommends that sepsis be defined as a life-threatening organ dysfunction caused by a dysregulated host response to an infection (Fig. 2). Organ dysfunction is now defined by an increase in the Sequential Organ Failure Assessment (SOFA) score of 2 points or more, which is associated with an in-hospital mortality greater than 10% (Fig. 3). Furthermore, septic shock is now defined as a subset of sepsis in which profound circulatory, cellular, and metabolic abnormalities are associated with a greater risk of mortality than with sepsis alone. Clinically, patients with septic shock can be identified by a vasopressor requirement to maintain a mean arterial pressure of 65 mmHg or greater and serum lactate level greater (>18mg/dL) in the absence of hypovolemia with inhospital mortality rates greater than 40% (Fig. 4). In order to identify patients with the highest probability of poor outcome associated with sepsis, a new bedside clinical score named the quickSOFA (qSOFA) was created which consist of at least 2 of the following clinical criteria including, respiratory rate of 22/min or greater, altered mentation, or systolic blood pressure of 100mmHg or less(4) (Fig. 5). Even though substantial advances in our clinical understanding, disease definition, and immune pathophysiology have improved overall sepsis survival, long-term sepsis mortality is abysmal, at 40–80%(5). Despite progress in antibiotic therapy, ventilator management, resuscitative strategies and blood glucose maintenance, sepsis remains the leading cause of death in the intensive care units (ICUs)(6). Even more alarming is the escalating cost of sepsis-associated medical care, which is estimated at $17 billion annually in the United States(7). Considering the rapidly expanding elderly population with large comorbidity burdens, physiologic frailty and immune senescence(8), sepsis mortality is expected to rise at an alarming rate over the next two decades(9). Despite over 100 therapeutic clinical trials in sepsis, no FDA-approved treatment options currently exist that improve sepsis survival(10). Although clinicians and investigators have opined in a sundry of editorials, reviews and commentaries incriminating a multitude of possible explanations such as, impaired cellular metabolism, tissue oxygenation, and myocardial dysfunction(11), the most accepted postulate describes a persistent and complex, immune/inflammatory interplay that is yet ill-defined(11, 12). Traditionally, sepsis investigations attempted to improve thirty day survival by dampening the inflammatory response by way of IL-1 and TNF blockade(13) with little success in reducing mortality. However, continued improvements in clinical treatment strategies over the past two decades(14) have resulted in more patients surviving life-threatening sepsis and organ dysfunction, only to manifest prolonged states of immune dysfunction, immune suppression(15), persistent inflammation and metabolic catabolism(16). These varying states of immune paralysis are characterized by impaired immune surveillance and the development of persistent, recurrent, secondary, and nosocomial infections which facilitate protracted events that often lead to death(17). Historically the sepsis death distribution has been biphasic, with an initial early peak at several days due to inadequate fluid resuscitation, resulting in cardiac and pulmonary failure, and a late peak at several weeks due to persistent organ injury or failure(18). Considering the recent recognition in mounting long-term sepsis mortality, a trimodal pattern is more indicative of the current death distribution(17–19). The early peaks in mortality exist, albeit of much less magnitude, and the third upswing occurs after 60–90 days and continues to soar over the ensuing three years(17, 19, 20) (Fig. 6A, B). These deaths are speculated to be the consequence of more sophisticated ICU care that keeps elderly and co-morbidly challenged(9) patients alive longer in spite of ongoing immune, physiologic, biochemical, and metabolic aberrations(21). Although the specific etiologies of long-term sepsis mortality are currently unclear, several reports suggest that advanced age, comorbidities, and persistent organ injury synergize to generate a damaging state of chronic and critically ill disease characterized by ongoing immune dysfunction, immune suppression, and catabolism and inflammation(15, 16, 22). Moreover, persistent inflammation combined with chronic immobility, catabolic drugs, and extended paralytic drugs all culminate to produce a state of immune dysregulation that facilitates infectious complications and terminates in chronic deterioration and death(23). Thus, investigators have been forced to refocus their efforts on the underlying innate and adaptive immune system derangements that facilitate the development of infectious complications, impair sepsis recovery and increase long-term mortality(24, 25). In this review we will outline the immune system’s role in sepsis progression, resolution and long term outcome and focus our attention on the clinical implications, and potential therapeutic interventions available to improve long-term survival. Immune Dysfunction in Sepsis Sepsis impacts the immune system by directly altering the life span, production and function of effector cells responsible for homeostasis(26, 27). The hematopoietic compartment constituently replenishes terminally differentiated innate and adaptive cells which are intrinsically responsible for immune surveillance against offending pathogens, and concomitantly prerequisite for successful tissue regeneration and wound healing. Over the last two decades a debate has persisted as to whether innate and adaptive immune dysfunction or inflammatory and anti-inflammatory processes are more detrimental to sepsis survival(28). Formerly, the inflammatory response was thought to drive early mortality in the first several days of sepsis, and the compensatory anti-inflammatory response was thought to induce organ failure, immune suppression and mortality weeks later (29). However new insights gathered using genomic analysis of septic patient tissue samples(15) and severely injured trauma patients, have identified an enduring and simultaneous inflammatory and anti-inflammatory state driven by dysfunctional innate and suppressed adaptive immunity that together culminate in persistent organ injury(30) and patient death(31, 32) (Fig. 7). Although shortcomings in each of these studies exist, when the collective results are compared with patient outcomes, it is clear that a paradigm shift is necessary to explain the long-term mortality surge after sepsis. It is evident that inflammatory and anti-inflammatory responses and innate and adaptive immune systems are each equally important, and likely present targets for future immune therapies to improve long-term sepsis outcomes(25–27, 33). The following discussion provides an overview of the sepsis-induced alterations in inflammation, innate and adaptive immune cell function, and the most promising immune response modifiers being considered for future human sepsis therapy. Hyperinflammation Once the host loses local-regional containment of an infection, the body is systemically exposed to microbes, microbial components and products of damaged tissue. This induces an inflammatory response and initiates sepsis-like responses through the recognition of pathogens and damaged tissue by way of pattern recognition receptors (PRRs) which are ubiquitous on immune cell surfaces. PRRs are expressed primarily on immune and phagocytic cells and on many types of somatic tissues. Microbial infections are recognized by pathogen-associated molecular patterns (PAMPs) that are expressed by both pathogenic and harmless microbes. PAMPs are recognized by PRRs such as Toll-like receptors (TLR), C-type and mannan binding lectin receptors, NOD-like receptors, and RIG-I-like receptors. Proteins and cellular products released by tissue damage are similarly recognized as damage-associated molecular patterns (DAMPs)(34). During sepsis, systemic activation of the innate immune system by PAMPs and DAMPs results in a severe and persistent inflammatory response characterized by an excessive release of inflammatory cytokines such as IL-1, TNF, and IL-17, collectively known as the “cytokine storm”(30). The exorbitant release of inflammatory cytokines occurs over a relatively short period of time (several days). In addition, intense complement activation and innate immune stimulation potentiate what should be a normal physiological response to infection, instead into an excessive inflammatory response resulting in tissue damage, cellular compromise, and molecular dysregulation that initiate organ dysfunction and even multi-organ failure(30). Although some patients recover from this inflammatory state, for unknown reasons elderly patients with heavy comorbidity burdens fail to resolve this initial condition and progress to a state of persistent ongoing inflammation, immune cell dysfunction, and catabolic metabolism, all of which degrades the immune system’s ability to clear infections and heal injured tissues(35). Recently, investigators have suggested that therapeutic interventions that curb hyperinflammation, shift catabolism toward anabolism, and bolster immune function maybe beneficial in combination, once the initial episode of sepsis has subsided(24, 36, 37). Although in other disease states such as severe burns(38), advanced cancers(24, 25, 39, 40), and autoimmune diseases(41), combination therapies that reduce inflammation, optimize metabolism, and decrease infections are common-place, there is as of yet no clear plan for the routine use of these or similar strategies to improve long-term outcome in sepsis(19). Until we embrace and adapt strategies for sepsis therapy that have been demonstrated to improve outcome in other inflammatory disease states, we may not be able to meaningfully improve sepsis survival. Immune Resolution Resolution of the hyperinflammatory cytokine cascade was thought to begin days after the initial sepsis episode had passed. However recently, it has been discovered that compensatory anti-inflammatory pathways are activated shortly after sepsis initiation(28). The hallmark cytokine is IL-10, which is produced by a variety of leukocytes, suppressing the production of IL-6 and interferon-γ (IFNγ), and stimulating the production of soluble TNF receptor and IL-1 receptor antagonist. These products neutralize proinflammatory TNFα and IL-1 signaling(42). At the subcellular level, autophagy provides a way to eliminate DAMPs and PAMPs by packaging pathogen components, damaged organelles and cellular proteins into vesicles targeted for lysosomal degradation, resulting in reduced inflammation and cellular activation(43). Resolution of inflammation after severe infection is not simply a passive process of curbing cytokine production and easing inflammatory pathways. Instead, dampening of inflammation involves an interdigitating, complex and coordinated array of cellular processes and recently recognized molecular signals. Soon after pathologic bacteria are eliminated from the host, damaged tissues, cells and leukocytes must be removed from the infection site. Under favorable circumstances the defunctionalized tissue cells and leukocytes undergo apoptosis, becoming engulfed by macrophages and removed from the inflamed field, triggering the production of anti-inflammatory IL-10 and transforming growth factor β. Furthermore, recently discovered bioactive lipids termed lipoxins, resolvins, protectins, and maresins have been shown to reduce ROS, endothelial permeability, and leukocyte recruitment, and further enhance macrophage phagocytosis(44, 45). In addition to anti-inflammatory cytokines, inflammation resolution is also governed by multiple subsets of regulatory immune cells such regulatory T cells (Tregs)(46, 47) and myeloid derived suppressor cells (MDSCs) that orchestrate inhibition over cytoxic effectors and curb inflammatory cytokine production(48). Historically, therapeutic interventions aimed at inhibiting the acute inflammatory response to infection and subsequent sepsis using glucocorticoids, nonsteroidal anti-inflammatory agents, and anti–TNFα antibodies have repeatedly failed to improve outcomes in patients with sepsis and septic shock(10). Conversely, strategies that stimulate endogenous mediators that actively resolve inflammation have not been thoroughly explored because the critical signaling factors that regulate these native processes have remained elusive. Serhan and colleagues, demonstrated that eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) derived lipids, [coined pro-resolving mediators (SPMs), including resolvins, protectins, and maresins], increase early after states of infection, regulate inflammation resolution, and enhance bacterial clearance(44). Recently, Dalli and coauthors reported a novel cohort of host-protective lipids termed 13-series resolvins (RvTs)(49). These host-protective lipids are formed early during inflammation, promote bacterial phagocytosis, increase production of reactive oxygen species, and augment host recovery from systemic infection by accelerating the resolution of the acute inflammation(49). Furthermore, the systemic administration of atorvastatin can increase RvT biosynthesis through activation of COX-2 and accelerate infection resolution which can be reversed by COX-2 inhibition using celecoxib(49). Together these findings shed light on a new group of endogenous mediators that resolve host inflammation without interfering with phagocytosis and identify a new mechanism for the pleiotropic effects of statins that mitigate inflammation and promote endothelial function(50). Although statin therapy in sepsis has yielded little benefit in clinical trials(51), the discovery of a new class of bioactive lipids that resolve inflammation could provide a biomarker to identify subgroups with impaired RvT biosynthesis who may benefit from statin therapy. Immune Suppression Generally, therapeutic strategies to treat sepsis have focused on inhibiting the early hyperinflammatory phase. However, it is evident that a state of immune suppression exists concomitantly with persistent inflammation and enables the development of persistent, recurrent, secondary, and nosocomial infections which lead to poorer outcomes and increased long-term mortality(25). Sepsis-induced immune suppression impacts both cellular effectors of the innate and adaptive immune systems. Neutrophils, essential for bacterial eradication, display defects in chemotaxis and recruitment to sites of infection, in the setting of sepsis(52, 53). Furthermore, the production and release of essential effector molecules, such as reactive oxygen species (ROS) and cytokines, is significantly impaired(53–55), leading to bacterial persistence and the development of infectious complications. Even though myeloid cell production of pro-inflammatory cytokines is reduced, overall myeloid cell production and release of anti-inflammatory cytokines (e.g., IL-10) is elevated(54). Defects in antigen presenting cell (APC) function, including reduced HLA-DR expression, endotoxin tolerance and impaired cytokine production all reduce the capabilities of APCs to stimulate lymphocyte driven immune functions following sepsis(56–59). Apoptosis of lymphocytes and APCs (dendritic cells, T cells and B cells) is considered a hallmark of septic immune suppression(60, 61). In addition to diminished innate function, adaptive immunity is similarly impaired. Splenocytes harvested from deceased sepsis patients demonstrate reduced numbers of CD4+ and CD8+ lymphocytes that emanate from substantial apoptosis(15). Moreover, CD4+ cell loss is associated with a reduced ability to mount appropriate immune responses to viral infections including cytomegalovirus, Epstein-Barr virus, herpes simplex virus, and human herpes virus reactivate after septic insults(62). However, reduced lymphocyte numbers are not just reflective of the risk for viral reactivation following sepsis. Lymphopenia four days after the onset of sepsis is associated with the development of secondary infection and is predictive of long-term mortality at one year after sepsis(63). Overall these specific cellular alterations coalesce into a chronic state of immune suppression, characterized by persistent, recurrent, secondary, and nosocomial infectious complications(64) often resulting in hospital readmissions(65–67), and poor long-term survival(68). Over the last 5 years it has become abundantly clear that in the weeks and months following hospital discharge, sepsis survivors have increased readmission rates(69) due to infectious complications(70), requiring more antibiotics, ICU days, and hospital resources, when compared to patients not experiencing sepsis(71). With the ever increasing, comorbidly-challenged elderly population, demonstrating persistent inflammation, immune suppression and immune senescence, the number of sepsis survivors that develop subsequent infections is predicted to rise substantially in the next decades(68, 72). It is evident that sepsis induces a pathologic state of immune suppression that prompts the development of secondary infections while still in the ICU setting(73). Several reports over the past five years demonstrate that sepsis survivors experience dramatically higher rates of subsequent infections long after the initial episode of sepsis has resolved(65, 69, 70, 74–76). The increased hospital readmission rates due to infectious complications among sepsis survivors is a sign of ongoing immune suppression and dysregulation that if not corrected, diminishes life quality and durable survival. Molecular Alterations in Sepsis Current sepsis observations suggest that multiple organ failure occurs even in the context of preserved cell morphology and in the absence of significant cell injury. In addition, organ dysfunction is often reversible, even in organs that regenerate poorly (heart, lung, central nervous system, kidneys). Therefore, it is apparent that sepsis-induced organ dysfunction occurs primarily though cellular and molecular dysregulation, as opposed to gross tissue damage. By this principle, immune dysfunction in sepsis is also associated with molecular alterations that alter cellular phenotype and function. Below, we outline several important pathways of cellular dysfunction that impact immune function during and after episodes of sepsis. Complement Activation It has long been known in human and rodent sepsis models that a robust and consumptive depletion of complement occurs, resulting in a sharp drop in hemolytic plasma complement and in the appearance in plasma of complement activation products(77). There is also evidence that sepsis in humans causes shedding of the C5a receptor into plasma, likely due to release of microparticles from neutrophils(78). Similar events occur in rats and mice with experimental polymicrobial sepsis(79). In addition to complement activation, there is also a robust stimulation of both the fibrinolytic and clotting systems(80). There is also evidence for activation of several clotting factors, some of which have C3 and C5 convertase activities, generating C3a as well as activation of thrombin which has C5 convertase activity and generates C5a and the terminal membrane attack complex (MAC)(81). There is well established evidence that activation of the complement system is often linked to activation of both the clotting and the fibrinolytic systems. Development of neutralizing C5a antibodies in murine models dramatically attenuated the intensity of sepsis, including greatly improved 7-day survival, reduced levels of plasma cytokines, and decreased multiple organ failure(79, 80, 82). The steady progress achieved in understanding complement and how its production and deposition increase systemic inflammation, organ failure and mortality, have resulted in the development and randomized phase 2 trial of a C5a inhibitor, CaCP29 (EudraCT Number: 2013-001037-40) which has shown great promise despite a historically large field of other failed antibody inhibitors(83). Redox Imbalance Clinical and experimental evidence has been well published that demonstrates septic patients exhibit overwhelming oxidative stress(84–86) which results from uncontrolled production of ROS and reactive nitrogen species (RNS). This severe state of oxidative stress is produced by activated immune and epithelial cells that overexpress oxide synthases including inducible nitric oxide synthase (iNOS), or via mitochondrial production of ROS(86–88). ROS and RNS play crucial roles in sepsis progression by interfering with nitric oxide signaling cascades and oxidation/nitrosylation of proteins or nucleic acid substrates that results in harmful molecular function(87, 89). Although anti-oxidant therapies have shown benefit during animal models of experimental sepsis, the same advantage has not translated into successful human clinical trials(90). Ca2+ Homeostasis Hypocalcemia is common in sepsis and correlates with disease specific scores during critical illness(91–94). Altered intracellular calcium handling is hypothesized to be responsible for the observed systemic hypocalcemia(94). Although systemic Ca2+ levels are reduced during sepsis, there are increased Ca2+ cytosolic levels which may stem from increased uptake. The same intracellular calcium fluctuations are elevated in a variety of tissues and cell types during sepsis(95–97) although the specific mechanisms underlying altered Ca2+ handling remain unclear. Heightened intracellular calcium leads to elevated inflammatory responses, cellular dysfunction, and can even be cytotoxic. In addition, the accumulation of Ca2+ in organs during sepsis is also associated with significant organ dysfunction(98). PARP1 and PARP2 Activation Poly(ADP-Ribose) Polymerase 1 (PARP1) and PARP2 are enzymes that catalyze poly(ADP-robosyl)ation of proteins. Catalytic activity of these enzymes is stimulated by DNA strand breaks. PARP activity is viewed as a sensor of DNA damage. PARP1 activation and initiation of the inflammatory response occur simultaneously(99). PARP1 activity upregulates proinflammatory gene expression(100), which is attributed to PARP1-induced alterations in chromatin structure and in transcriptional regulation(99, 101). Because PARP1 also directly contributes to cell death in affected tissues(99) it is hypothesized that PARP1 also plays a role in sepsis associated immune cell death. PARP1 genetic deficiency is protective during murine models of experimental sepsis, and is associated with significantly lowered plasma cytokine levels and reduced tissue/organ dysfunction(102). Inhibitors of PARP1 have been studied in clinical trials as potential cancer therapeutics, but trials for sepsis have not been initiated. Therefore, it is not clear whether inhibitors of PARP1 will be beneficial during the treatment of human sepsis. Mitochondrial Dysfunction Mitochondria are critical for maintaining an adequate supply of ATP for cellular processes. In addition, damaged mitochondria can trigger cell death pathways through the release of mitochondrial cytochrome c(103). Mitochondria are affected in several ways during sepsis. The generation of excessive amounts of ROS and RNS can directly inhibit respiration and damage respiratory chain components in mitochondria(104–106). In addition, impaired tissue perfusion (due to fluid loss, both intrinsic and extrinsic, as well as reduced vascular tone) leads to tissue hypoxia. Loss of tissue oxygenation significantly impairs oxidative phosphorylation and may trigger cell death pathways(107). Mitochondrial dysfunction, or direct damage of mitochondria, can directly affect the generation of ATP. Not only will the drop in ATP negatively affect cellular processes, but severe lack of ATP can trigger cellular anergy, whereby the cell will not necessarily die, but instead acquire a hibernation-like state resulting in tissue dysfunction and organ failure(108). The importance of mitochondrial dysfunction during sepsis is highlighted by the observations that cellular ATP levels are correlated with survival in both human and animal sepsis models(106, 109). Cellular Defects The following discussion provides a summary of the sepsis-induced immune alterations in the majority of innate and adaptive cell types, along with the most promising candidates under consideration to be employed as immune response modifiers for future human sepsis therapy. Although we have attempted to focus this discussion on human sepsis, many of the current insights have been gleaned from animal sepsis model recapitulating aspects of human sepsis. The authors recognize the recent and ongoing debate about the efficacy of murine and other animal models to accurately reflect human disease processes(110). The authors’ opinion is that both human and animal models are necessary if continued scientific progress and improved patient outcome is to continue. Even though we clearly realize that animals do not immunologically, metabolically, or genomically equal the state of human responses, it is not sustainable to merely test all of the proposed hypotheses in human disease systems without the benefit of animal correlation to explore scientific avenues not otherwise ethically amenable to the human condition. Innate Immunity Endothelium Endothelial cells (ECs) form a single cell layer called the endothelium, which line all of the vasculature and lymphatic systems in the body and comprise a semi-permeable barrier between blood and lymph within vessels and the surrounding tissue. ECs are a heterogeneous population of cells that fulfill many physiological processes and participate in innate and adaptive immune responses. ECs function as danger signal sensors, therefore are one of the first cell types to detect invading microbes via endogenous metabolite-related danger signals in the bloodstream(111). LPS exposure activates ECs, causing the production of pro-inflammatory cytokines and chemokines, which amplify the immune response by recruiting immune cells(112). Therefore, ECs function as innate force multipliers, cell mobilizers, and immune regulators by modulating cellular function(113). In special circumstances, ECs can serve as antigen presenting cells expressing both MHC I and II molecules and present endothelial antigens to T cells. ECs also express TLR-2 and TLR-4 enabling them to respond to LPS in states of bacterial infection(112). During sepsis a significant amount of EC dysfunction is present and manifests as several pathological processes including capillary leak, altered vasomotor tone and microvascular thrombosis(114). A further consequence of damage to the endothelium is the release of pathological quantities of von Willebrand factor, which promotes platelet aggregation and adhesion to the subendothelial layer and the formation of pathologic thrombi. In sepsis there may be direct destruction of the endothelial barrier, and an increased number of circulating endothelial cells that have been observed in patients with septic shock. Angiopoietin-2 (Ang-2) mediates endothelial microvascular leak and is an independent predictor of death and organ dysfunction during sepsis(115). Since the early outcome of sepsis is mainly determined by the degree of organ failure, macro- and microvascular endothelial dysfunction has been proposed as an early biomarker of immune function and outcome. In patients with severe sepsis, in vivo measured endothelial dysfunction coincides with lower numbers and reduced function of circulating progenitor cells implicated in endothelial repair(116). The results suggest that cellular markers of endothelial repair might be valuable in the assessment and evolution of organ dysfunction and even outcome following episodes of sepsis. ECs also release microparticles which are protective against vasomotor hyporeactivity, which accounts for hypotension in patients with septic shock(117). Taken together, there is a large amount of data to suggest that ECs are key regulators of the physiologic and immunologic dysfunction during and after sepsis, and that EC modulation is possibly beneficial to improve long-term human sepsis survival. Neutrophils Neutrophils are the most prevalent and integral cell type of innate function, essential for microbial containment and eradication, and prerequisite for long-term sepsis survival(118). They comprise the majority of the cellularity in the bone marrow (BM) and are the very first responders to sites of microbial infections (119). One of the most pronounced innate immune alterations in sepsis is a delayed state of neutrophil apoptosis(120), leading to ongoing neutrophil dysfunction. This delayed state of neutrophil apoptosis is further compounded by the release of immature band-like neutrophils from the BM that demonstrate clear deficits in oxidative burst(121), cellular migration patterns(122, 123), complement activation ability and bacterial eradication(54), which all combine and contribute to persistent immune dysfunction and inflammation. These findings combined with TLR signaling deficits (124), chemokine-induced chemotaxis reductions (125), altered apoptosis signaling pathways, and neutrophil immune senescence(126), result in a sundry of functional deficiencies that endure long after sepsis symptoms have subsided. A host of mounting scientific evidence also suggests that neutrophils may function as APCs in a broad array of pathological infections and act as crosstalkers between innate and adaptive responses through activation of CD4+ and CD8+ effector cells(127, 128). More importantly, multiple human studies have implicated the complex array of persistent neutrophil dysfunction in the development of hospital acquired infections(129). Furthermore, patients with the most pronounced derangements in neutrophil function following sepsis are the most susceptible to develop intensive care unit complications such as ventilator-associated pneumonia and other nosocomial infections(130). The vast majority of patients who die from sepsis have ongoing infections(15), suggesting that defects in innate immunity in general and neutrophil-mediated bacterial clearance in particular, could serve as potential therapeutic targets to regulate neutrophil apoptosis, production, maturation and function. In addition to the ability to eliminate pathogens by phagocytosis, oxidative burst and/or degranulation, it has recently been shown that neutrophils can eradicate a wide range of microorganisms by forming neutrophil extracellular traps (NETs)(131). This novel mechanism entails the release of antimicrobial proteins anchored to a chromatin network of activated neutrophils. A rapidly expanding body of evidence demonstrates that NET release is integral in the pathogenesis of diseases such as sepsis, atherosclerosis, and autoimmunity(132, 133). DNA is the major structural component of NETs. In addition, granule and cytoplasmic proteins, including neutrophil elastase (NE), myeloperoxidase (MPO), cathepsin G, proteinase 3 (PR3), gelatinase, LL-37, lactoferrin, and calprotectin as well as histones H1, H2A, H2B, H3, and H4, are all embedded in the DNA backbone comprising the NET. The DNA fibers promote physical containment of the microbes, whereas the histones and granular proteins confer an antimicrobial function to the NETs. Neutrophil NET production requires platelet-neutrophil interactions and can be inhibited by platelet depletion or disruption of integrin-mediated platelet-neutrophil binding. Released into the vasculature to ensnare bacteria from the bloodstream, NETs prevent further bacterial dissemination(134). During sepsis, NET release increases bacterial trapping by 4-fold in liver sinusoids. Furthermore, the correlation between the presence of NETs in peripheral blood and organ dysfunction was evaluated in 31 septic patients. Elevated NET concentrations were observed in septic patients in contrast with the healthy controls, and increased NET levels were associated with sepsis severity and organ dysfunction(135). Collectively these insights provide a foundation from which to explore the potential of immune modulatory therapy to exploit the benefits of NET formation in states of bacterial infection while minimizing the hazards of worsening organ dysfunction. Monocytes and Macrophages The impact of an episode of sepsis on human monocyte subpopulations has long been the subject of intense investigation over past half century. For decades it has been apparent that reduced mononuclear cell HLA-DR expression clearly correlates with human sepsis mortality (136). Moreover, the reduced capacity of blood monocytes from septic patients to release pro-inflammatory cytokines after endotoxin (LPS) challenge has been described as “endotoxin tolerance”, which has been suggested to facilitate poor short and long-term sepsis outcomes(137, 138). Although a sundry of complex mononuclear cell signaling pathways are altered and contribute to the establishment of endotoxin tolerance, the major implication on monocytes, and to a lesser extent macrophages, is reduced antigen presentation related to diminished HLA-DR cell surface expression(139). In addition to the clear and persistent reductions in HLA-DR cell surface expression, monocytes from septic patients also demonstrate a reduced ability to secrete the pro-inflammatory cytokines TNF, IL-1, IL-6, and IL-12 after LPS challenge. The reduced monocyte capacity to secrete pro-inflammatory cytokines suggest that intracellular signaling has shifted toward the production of anti-inflammatory mediators which are associated with hospital acquired, ongoing, and secondary infections which ultimately increase sepsis-associated mortality. Currently many investigators agree that reduced monocyte HLA-DR expression is a surrogate marker of monocyte “anergy”, development of ongoing infections, and patient death(140–142). In addition to diminished pro-inflammatory cytokine secretion, several reports associate low monocyte HLA-DR expression with reduced antigen-specific lymphocyte proliferation(143, 144). These findings suggest that sepsis induced monocyte anergy and immune suppression separately contribute to the increased risk of complications and adverse outcomes in sepsis. Although the mechanisms accounting for monocyte LPS tolerance are not clear at this moment, sepsis-induced monocyte epigenetic reprogramming may play a pivotal role the establishment of LPS tolerance, myeloid anergy and the overall immune-suppressive monocyte phenotype(145). Analysis of human monocyte mRNA clearly shows increased levels of inhibitory cytokine genes and reduced levels of pro-inflammatory chemokine genes(146). Recent reports on human monocytes make a substantial and convincing argument for the important impact of epigenetic reprograming on the establishment of monocyte anergy, however the true functional impact of these epigenetic alterations on long-term outcomes in human sepsis is still unknown(147). Natural Killer Cells (NK) Historically NK cells were thought of as undeveloped assassins of host cells that either lacked self-identification or were infected by viruses. However, thankfully the field on innate immunity has moved forward and we now clearly know that NK cells act as regulators immune complex immune functions. NK cells are divided into various different subgroups based on CD16 and CD56 cell surface expression(148). Human sepsis evidence indicates that both CD56hi and CD56low NK cell subgroups are significantly altered during episodes of sepsis. These observed alterations have recently been associated in several reports with increased lethality in human sepsis (149–151). Additionally, NK cell cytotoxic function in human sepsis is greatly decreased(152). Similarly, to LPS tolerance in monocytes, NK cell ex vivo production of IFNγ in response to TLR agonists is also gravely diminished. This evidence indicates that NK cell tolerance may be responsible for the reactivation of latent viruses such as CMV, which is frequently encountered in intensive care unit populations and more importantly may serve as a future target for therapeutic intervention(153). Despite the lack of published data, it is apparent that NK cells are major effectors of the final outcome in sepsis through modulation of INF-γ production(154). Contradictory results in clinical studies may be explained by sampling time variability and apparent patient heterogeneity. It is evident that NK cell activation provides protection from pneumonia through excess production of INF-γ which inhibits bacterial growth and has a major impact on sepsis outcome(155). Dendritic Cells (DCs) DCs are traditionally characterized as either conventional DCs (cDCs) or plasmacytoid DCs (pDCs). cDCs are similar to monocytes and secrete IL-12, while pDCs the are similar to plasma cells and secrete large amounts of IFNα. cDCs and pDCs are of particular interest due to their enhanced apoptosis during sepsis(60) and in patients who developed nosocomial infections(156). Although DCs have varying immune functions compared with monocytes, like monocytes, DCs also exhibit reduced HLA-DR expression and produce increased amounts of immune suppressive IL-10(157). Furthermore, co-culture of DCs with T effectors induces T cell anergy and Treg proliferation, which both correlate with sepsis-induced immune dysfunction. A couple of recent investigations have demonstrated that prevention of sepsis-induced DC apoptosis or augmentation of DC function enhances sepsis long-term survival(158, 159). Several reports demonstrate that immune suppression can be ameliorated by DC treatment with growth factor FMS-like tyrosine kinase 3 ligand (FLT3L). FLT3L therapy models of burn-wound sepsis enhances DC cytokine secretion (IL-12, IL-15 and IFNγ), and augments CD4+ T cell, NK, and neutrophil function(160). Still further investigations indicate that the DC-associated gains in sepsis survival occur through TLR signaling pathways, increased MHC class II antigen and costimulatory molecules CD80 and CD86 expression (57). These observations have led researchers and clinicians together to surmise that improvements in DC number and function may be high yield targets for future therapeutic interventions in sepsis(158, 159). Myeloid-Derived Suppressor Cells (MDSCs) and Myelopoiesis MDSCs are a heterogeneous population of immature myeloid cells that expand dramatically in sepsis, impede immune responses, and signal through TLR-mediated pathways(127, 161). MDSCs associated with sepsis are phenotypically similar to the MDSCs described in states of advanced cancer(127, 162). Although MDSCs have been demonstrated to inhibit CD8+ cell function, the actual and functional impact of MDSCs in human sepsis is still uncertain. A summation of the current literature implies a beneficial role centered on replenishing innate cell function and immune surveillance through emergency granulopoiesis (123). Prior to MDSC expansion, we have identified a window of susceptibility to secondary infections and subsequent mortality associated with reduced BM cells, and reduced blood and tissue neutrophil numbers and function(121). We have also demonstrated that robust MDSC expansion through enhanced granulopoiesis imparts lasting immunity to secondary and nosocomial infections in sepsis(163). Due to the inherent difficulty in immune phenotyping immature myeloid cells between mice and humans very limited investigations and clinical studies have closely examined the roles of MDSCs in human sepsis(164). Nonetheless, there is mounting interest being paid to myelopoiesis, MDSC expansion, emergency granulopoiesis and hematopoietic stem cell production and function(121, 127, 163, 165, 166). Due to the importance of efficiently regenerating functioning neutrophils, monocytes and DCs, it is no surprise that MDSCs expand to meet the continual need for functional innate immune cells. Considering that five to seven days are required for the BM to adequately produce a functioning neutrophil, an expansive immature pool (~18 x 1011) of myeloid lineage precursors in the BM and secondary lymphoid organs(167) is required to maintain a functioning cohort of innate immune cells. In humans, approximately 16 x 1010 neutrophils are produced daily which can be rapidly increased 5- to 10-fold in response to foreign microbial invaders and pathologic stated of infection. Our prior work has clearly demonstrated that myeloid expansion involving hematopoietic stem cells (HSCs) occurs through c-KIT-, type-I IFN- (IFN-I), and CXCL10-dependent mechanisms that involve IFN-I-secreting B cells(165, 166). Moreover, impaired HSC proliferation, development and function in human BM transplant models is clearly associated with increased mortality to secondary, chronic and nosocomial infections(168). Humans with reduced granulopoiesis ability undoubtedly experience more frequent, severe and anomalous infections, demonstrating the essential requirement for effective neutrophil production. Recently, patients with sepsis have been shown to have MDSCs persistently increased, functionally immune suppressive, and associated with adverse outcomes including increased nosocomial infections, prolonged intensive care unit stays, and poor functional status at discharge(169). Conversely, overzealous MDSC proliferation may facilitate a physiologic syndrome of persistent inflammation, such as in adult respiratory distress syndrome (ARDS) or persistent inflammation immunosuppression, and catabolism syndrome (PICS), causing patients with sepsis to experience a poor outcome(16). Recent work by Terashima and coauthors demonstrated that acute inflammation causes the reduction of peripheral lymphocytes and common lymphoid progenitors (CLPs) in the bone marrow, which is also associated with a dramatic decrease in the osteoblast number. Moreover, osteoblast-specific IL-7 production during myelopoiesis was shown to be pivotal in the regulation of lymphopoiesis during systemic inflammation and may serve as a target for improved lymphocyte production(170). The specific contributions of lymphopoiesis, myelopoiesis and MDSCs to sepsis recovery versus persistent inflammation and catabolism remain poorly understood. However, new insights into these processes and their roles in sepsis resolution and recovery will hopefully present new targets for immune modulatory therapy to improve sepsis outcomes. Adaptive Immunity Lymphoid Apoptosis and Immune Suppression In the majority of septic patients, circulating lymphocytes and gastrointestinal epithelial cells undergo significant apoptosis, while apoptosis/necrosis in the heart, kidneys, and lungs is not apparent(171). Lymphocyte apoptosis has now been accepted as an important step in the pathogenesis of sepsis and contributes to septic immunosuppression(172). Lymphocyte apoptosis has been described in both septic patients and animal sepsis models(173, 174). Importantly, lymphocyte apoptosis has been shown to occur through both the intrinsic (Fas, FasL) and extrinsic (TNF) pathways(174). Although the defining factors for this phenomenon are still not clear, recent evidence suggest that the release of extracellular histones during sepsis may drive lymphocyte apoptosis(175, 176). Therapeutic blockade of lymphocyte apoptosis and/or restoring lymphocyte function have generated promising preclinical data that may lead to new treatments after episodes of sepsis in humans(171, 177, 178). Gamma delta T cells (γδ T cells) γδ T cells are a diminutive subset of T cells that possess a rather distinct T cell receptor (TCR) on their cell surface. The majority of T cells have a TCR composed of two α and β glycoprotein chains, while γδ T cells have a TCR that is made up of one γ chain and one δ chain. This uniquely distinct group of T cells exists mainly in and around the gut mucosa within a population of intraepithelial lymphocytes(179). Despite the fact that the antigens to which γδ T cells respond are still unknown, it is suspected that these cells recognize lipid antigens from pathogens present on mucosal surfaces within the intestine(180). What is clear is that upon activation, γδ T cells release IFNγ, IL-17 and other inflammatory chemokines. In humans with sepsis the number of circulating γδ T cells is significantly diminished, and these reductions correlate clearly with the highest rates of sepsis lethality (181). The observed reduction in γδ T cells in the gut mucosa may serve to potentiate typically non-invasive intestinal bacteria types to become more invasive and translocate into the host systemic circulation, causing pathological infections following episodes of sepsis(182). In patients with acute sepsis, circulating neutrophils display a similar APC-like phenotype and readily process soluble proteins for cross-presentation of antigenic peptides to CD8+ T cells, at a time when peripheral γδ T cells are highly activated(128). From their findings the authors conclude that unconventional T cells represent key controllers of neutrophil-driven innate and adaptive responses to a broad range of pathogens and may serves as targets for additional immune enhancement. In addition, others have also postulated that tremendous potential exist for the therapeutic potential of NKT cellular targeting and immune enhancement in sepsis(183). TH cell subpopulations T helper cells (Th cells) assist other cell types, including B cell differentiation, cytotoxic T cell activation, and monocyte stimulation with immunological processes. When confronted with peptide antigens by MHC class II molecules expressed on APCs, CD4+ cells are quickly activated, rapidly proliferate, and efficiently secrete cytokines that regulate adaptive and innate responses. Upon activation, CD4+ cells have the capability to differentiate into one of several T cell subsets including Th1, Th2, Th3, Th17, Th22, Th9, or T follicular helper (TFH), which facilitate various immune responses through differing cytokine generation and secretion (184). Although numerous reports relate the effects of sepsis on circulating and peripheral CD4+ T cell subsets(46), the following section will highlight only the most relevant investigations to convey the important themes and potential areas of therapeutic interest. Similar to the phenomenon observed in neutrophils and monocytes, one of the most deleterious T cell defects induced by sepsis is development of apoptosis that decimates CD4+ populations(15, 185). In humans that die from sepsis, there was a much greater magnitude of lymphocyte (specifically CD4+) apoptosis than in T cell from sepsis survivors(15). Of the CD4+ cells that manage to persist, multiple investigations demonstrate that both Th1- and Th2-associated cytokine production is reduced during and long after sepsis subsides (186). Marked reductions in the transcription factors T-bet and GATA3, which modulate the Th1 and Th2 response, respectively, support the notion that CD4+ subsets are suppressed during sepsis(187). There are a sundry of factors and influences that regulate CD4+ Th subpopulation differentiation, including histone methylation and chromatin remodeling, which together are postulated to suppress Th1 and Th2 CD4+ T cell functions(188). However, the sepsis-induced immune impact is not only relegated to Th1 and Th2 CD4+ T cells but also to Th17 subsets and the other Th subsets as well. Th17 cytokine production is reduced in sepsis and probably negatively impacts long-term mortality(189). Given the fundamental role of Th17 in eradication of pathologic fungal infections, reduced Th17 cytokine production in sepsis is probably responsible for the increased susceptibility to fungal infections frequently observed in critically ill populations(190). Circulating CD4+ and CD8+ cells from patients with Candidemia display an immune phenotype consistent with immune suppression, T cell exhaustion and downregulation of positive co-stimulatory molecules(191). Moreover, IL-7 treatment has been demonstrated to increase Th17 cell responsiveness and reduce mortality from secondary fungal infections, making IL-17 a potential therapeutic agent(177). These findings may help explain why patients with fungal sepsis have a high mortality despite appropriate antifungal therapy. Regulatory T cells (Tregs) Tregs are a master regulators of adaptive immunity that suppresses responses of other effector T cells subsets, helping to maintain self-tolerance and suppress autoimmune disease. In states of sepsis, critical illness and states of inflammation, Tregs potentiate deleterious effector T cell (Teff) suppression that prolongs recovery and may dispose to increased complications. Increased Treg ratios are present early after episodes of sepsis and remain elevated in those patients who died from sepsis while hospitalized, placing a high level of attention on Treg function. Other reports relate that Treg number increases are due to effector Th cell loss from apoptosis rather than an absolute increase in Treg numbers(47). This observation suggests to many that Tregs are resistant to sepsis-induced apoptosis, thereby preventing the recovering immune system from mounting excessive autoimmune responses during the heightened initial inflammatory phase. Moreover, heat shock proteins and histones that induce mononuclear cell epigenetic changes also play a role as inducers of Tregs in sepsis(192). Recent murine reports demonstrate that Tregs are detrimental to Teff proliferation and immune function(193). This effect is ameliorated by the administration of therapeutic siRNAs that inhibit Treg differentiation(47). In other investigations, glucocorticoid-induced TNF-receptor-related protein (GITR) inhibitory antibodies were used to block Treg function, resulting in improved immune function and microbial killing(194). Treg-associated immune dysfunction in sepsis has also been linked to more rapid cancer and solid tumor cell growth, which probably stems from a reduced overall cytotoxic T cell (CTL) and mononuclear cell immune surveillance functions (195). In conclusion a strong notion exist that Tregs are resistant to apoptosis in sepsis, augment ongoing Teff cell dysregulation, contribute to infection development and potentially serve as targets for immune modulation. B cells B cells are a very diverse immune cell population with varying functional and phenotypical attributes. Historically B cell function was understood to only encompass producing antibodies and plasma cells for long-term antibody responses (196). Conversely, a rapidly growing body of knowledge and collection of recent reports demonstrate that B cells play a much more pivotal role in sepsis immune biology than previously suspected. Clearly humans with septic shock have overall reductions in B cell numbers, however the most significant deficit in B cell number is in CD5+ B1a-type cells, which correlate and are predictive of survivors and non-survivors following episodes of sepsis (197). In mouse models of human sepsis, B cells are necessary to improve cytokine production, reduce bacterial load and improve survival through type I interferon signaling (198). A recent investigation identified an innate response activator (IRA) B cell population, which is phenotypically and functionally distinct from B1a cells and depends on PRRs, which produces granulocyte-macrophage-CSF (GM-CSF). Inhibition of IRA B cells impairs bacterial eradication, enhances a cytokine storm, and perpetuates the symptoms of septic shock. These recent clarifications position IRA B cells as immunological gatekeepers of bacterial infection elimination and simultaneously recognize IRA B cells as a new therapeutic target to improve survival in human sepsis (199). Lastly, IRA B cell generation of IL-3 has been revealed to greatly enhance sepsis associated inflammation, induce myeloid production of Ly-6Chi mononuclear cells and potentiate cytokine production, while elevated plasma IL-3 levels correlate with increased human sepsis mortality(200). B cells stimulated ex vivo in both aging and sepsis patients demonstrate significant reductions in supernatant IgM production(76). This finding is clinically relevant and interesting and may explain why elderly patients with decreased IgM production are more susceptible to gram-negative bacteria and fungal infection. Reduced immunocompetent B cells may be related to increased secondary infection after sepsis, especially in the elderly. All things considered, the recent and vast advancement in our understanding of B cell biology has provided a great insight into the B cell role in immune modulation, emergency myelopoiesis, and IL-3 production which is also new potential therapeutic focus in sepsis(201). Immune Modulatory Therapies in Sepsis G-CSF and GM-CSF Granulocyte-colony stimulating factor (G-CSF) is a glycoprotein that stimulates the production of stem cells, progenitors and granulocytes of all maturity ranges(202). G-CSF is very efficacious in reducing the incidence of sepsis in patients with low absolute neutrophil counts, such as those undergoing BM transplant, cancer chemotherapy or autoimmune radiation(203). Two randomized controlled human trials with recombinant G-CSF have been conducted in a goal directed effort to bolster neutrophil production, maturity and overall function. Clinicians and research investigators alike hypothesized G-CSF that administration would improve granulocyte function and microbial elimination in such circumstances. Although an increase in blood leukocyte counts was realized, there was no improvement in 28-day patient mortality(204, 205). Although the initial conclusions of these two clinical studies were disheartening, the fact remains as the premise of this report that most of the sepsis-induced mortality occurs in a protracted process beyond 90 days(17). This makes one wonder if a longer study therapy or observation time would have changes the investigation outcomes. However as of this time the impact of G-CSF on mortality beyond 28 days is unknown. Given the ongoing and continuous alterations observed in granulocyte production, myelopoiesis and neutrophil function especially in comorbidly-challenged patients such as those with diabetes and physiologic frailty, prolonged G-CSF administration may be efficacious for improved immune surveillance, infection eradication, tissue regeneration and sepsis survival in future clinical trials. GM-CSF is a another cytokine that enhances stem cells and progenitors to produce neutrophils, monocytes and macrophages(206). In the immune suppressive phase of sepsis, patients that were ventilator-dependent and treated with recombinant GM-CSF and had fewer ventilator and ICU days(207, 208). Furthermore, recombinant GM-CSF treatment in immune suppressed pediatric populations with severe sepsis restored TNF production in lymphocytes and significantly reduced hospital associated infections(209). Even further evidence for GM-CSF therapy is gleaned from a meta-analysis of over 12 clinical studies involving either G-CSF or GM-CSF that demonstrated that either therapy significantly reduced the rate of infectious complications (210). In lite of the fact that that 70–80% patients who succumb to sepsis harbor persistent, chronic, ongoing, or secondary infections(15), GM-CSF and/or G-CSF in combination with other immune modulatory agents may prove invaluable for enhancing infection eradication during and after sepsis and improved long-term survival after sepsis(204, 211). Interferon gamma (IFNγ) IFNγ is unique and the sole member of the type II interferon family. Adequate IFNγ production and signaling is critical for appropriate immune function against viral, bacterial and protozoal invaders. Moreover, IFNγ is a central inducer of macrophage activation, inducing stimulating class I MHC expression(139). When treated with recombinant IFNγ, patients with severe sepsis and decreased monocyte HLA-DR levels demonstrate reversal of sepsis induced monocyte dysfunction and overall improved sepsis survival(212). Although the majority of the interventional therapeutic trials with IFNγ were done in burn and mixed severely injured trauma cohorts, the largest of these reports clearly demonstrated a decrease in infection-related mortality among patients treated with IFNγ(213). Moreover, a new study of severely injured trauma patients revealed that 42 of 63 genes identified as being differentially expressed in patients with uncomplicated vs complicated outcomes, were specifically associated with interferon signaling. The investigators discovered that IFN-associated genes were suppressed in trauma patients with complicated outcomes(31, 32), implying that this set of genes may be useful for identifying patients at risk for complications after trauma. In addition these patients prone to develop complications may preferentially respond to therapies utilizing IL-7, IL-15, IFNγ, and GITR agonists. IFNγ is very promising if it is targeted to the patient populations that may benefit the greatest such as those who demonstrate or are at risk for immune suppression, decreased monocyte HLA-DR expression, adaptive immune dysfunction or chronic inflammation in prolonged hospital stays. In a recent report, recombinant IFNγ treatment was able to partially restore immune metabolic defects associated with immune paralysis in humans after sepsis further suggesting that IFNγ therapy after sepsis may benefit a multitude of cellular immune functions(214). Though IFNγ offers promise as a potential immune modulatory therapy given its ability to rejuvenate monocyte/macrophage function and adaptive immunity, IFNγ may be more efficacious if administered in a time-phased approach coupled with the likes of GM-CSF and G-CSF or even IL-7 and/or PD-1 inhibitors to bolster specific immune function, reduce secondary and nosocomial infections, and improve long-term survival as sepsis recovery evolves. Programmed Cell Death Protein-1 and Ligand (PD-1 and PD-L1) Steady progress in cancer biology has generated a new class of drugs that inhibit programmed cell death. Programmed Cell Death Protein-1 (PD-1) is a protein that under homeostasis generates an inhibitory signal that reduces CD8+ T cell proliferation and accumulation in secondary lymphoid organs such as lymph nodes. PD-1 is expressed on most T- and B-lymphocytes and myeloid cells. The ligand for PD-1, (PD-L1) is ubiquitously expressed on epithelial and endothelial cells, monocytes, macrophages, and DCs(215). Considering that PD-1 is upregulated on both CD4+ and CD8+ cells during viral infections and cancer states, it is often associated with the phenomenon of “T cell exhaustion”, which is thought to stem from prolonged periods of exposure to self-antigens (216). Both anti-PD-1 and anti-PD-L1 therapies have demonstrated promising results in human trials involving cancer and viral infection. Therefore, sepsis biologist have postulated that anti-PD-1 and anti-PD-L1 therapy could also have similar beneficial results with reducing sepsis-induced immune dysfunction that drives ongoing infectious complications (217). Patients with septic shock demonstrate increased levels of PD-1 and PD-L1 on their monocyte and T-lymphocyte cell types (218). Recent studies demonstrate granulocyte PD-L1 upregulation results in potentiation of lymphocyte apoptosis through contact inhibition, which correlates with outcome(219). Moreover, in clinically relevant animal models of experimental sepsis, inhibition of PD-1 and PD-L1 signaling pathways improved survival and reduced the number of fungal infections(220). Considering the beneficial impact on adaptive immunity and tumor eradication strategies, it makes sense that PD-1 and PD-1L could concomitantly serve as biomarkers of sepsis-initiated immune suppression as well as prospective therapeutic targets to reverse adaptive immune dysfunction and improve long-term survival. Recombinant Human IL-3, IL-7, IL-15 Considering the well documented and significant loss of lymphocytes in sepsis, IL-7 administration has been suggested as possible treatment strategy. IL-7 is a hematopoietic cytokine produced by BM stromal cells and is prerequisite for B and T cell production, maturation, homeostasis, and maintenance(221). IL-7 is an attractive therapy due to its ability to upregulate the anti-apoptotic protein BCL-2, which intern causes increased numbers of peripheral blood CD4+ and CD8+ cells. IL-7 also augments TCR diversity, which is lost in septic patients and is associated with increased development of nosocomial infections and hospital complications (177, 222). Low doses of recombinant human IL-7 preferentially activate Teffs from patients with sepsis(223). Although recent evidence exists that IL-7 can improve lymphocyte PD-1 expression, cytokine-driven peripheral T cell expansion and survival remain intact(224). However, when IL-7 was administered to patients with HIV, lymphocyte expression of PD-1 was decreased(225). Although IL-7 has the potential to work synergistically with other therapies targeting PD-1 or PD-L1 to improve aspects of T cell function lost in sepsis, the capability of PD-1 or PD-L1 to reduce hospital acquired infections or improve survival in human sepsis is currently unknown. Emerging evidence suggest that sepsis reduces bone osteoblast number, which induces lymphopenia through IL-7 downregulation, demonstrating a reciprocal interaction between the immune and bone systems and identifying bone cells as potential therapeutic targets in sepsis(170). Although clinical studies do support IL-7 therapy in HIV-infected patients receiving antiretroviral therapy, no human sepsis trials currently exist. Given the promise that IL-7 has demonstrated in other states inflammatory disease, sepsis clinical trials with IL-7 either alone or in combination with other immune modulators such as anti-PD-1 should be considered. Although still in preclinical scientific and experimental phases, it is worth mentioning IL-3 and IL-15 as potential sepsis therapies that have gained considerable attention. Considering the central role of IL-15 in the development and activation of effector and memory T, NK and NKT cells and neutrophils, it has become a hopeful therapeutic candidate for future sepsis trials (226). In murine models of experimental sepsis, IL-15 treatment diminished overall immune dysfunction and reduced mortality(227). The synergistic impact that IL-3 plays in stem cell and progenitor maturation and development along with IL-7 makes IL-3 an appealing therapy to augment the potential impact of IL-7therapy(200, 201). However, there is very little evidence relating to the impact of either IL-3 or IL-15 in states of human sepsis and further discourse is purely speculative. Vagal Nerve Modulation A growing body of evidence over the last two decades has revealed that systemic cytokine production and release is controlled by the 10th cranial nerve (vagus nerve). The scientific thrust of this discovery was done by the Tracey laboratory which elucidated the efferent arm of the inflammatory reflex(228, 229) As the cholinergic anti-inflammatory pathway(230). Accumulating experimental evidence establishes that vagus nerve activation operated via cholinergic anti-inflammatory signaling via [alpha]7 nicotinic acetylcholine receptors (7nAChR) expressed on nonneuronal cytokine-producing cells. Therefore, 7nAChR receptor agonists inhibit sepsis associated cytokine release and protect animals in experimental models of sepsis and in a variety of other experimental inflammatory models(231). Vagus nerve stimulation has an anti-inflammatory effect in sepsis and downregulates proinflammatory cytokine production in sepsis, decreasing the plasma protein levels of high mobility group box 1 (HMGB1) and improving survival in septic peritonitis models(232). Surgical vagotomy increases the plasma levels of pro-inflammatory cytokines, tissue damage and mortality in sepsis(233). Conversely, in rat models of sepsis, vagus nerve electrical stimulation attenuates and prevents hypotension(234), modulates coagulation, and fibrinolysis activation which all decrease organ dysfunction(235). Hence, the therapeutic potential of vagal efferent fiber modulation to treat disorders characterized by cytokine dysregulation is of great interest(236). However, until very recently, a paucity of real human data existed indicating that vagus nerve modulation was achievable let alone beneficial to outcomes in diseases of inflammation(237). Recently, clinical trial data demonstrates that stimulating the vagus nerve with a tiny implantable bioelectronic device significantly improves measures of disease activity in patients with rheumatoid arthritis, specifically reducing TNF production(238). Given the positive results of vagus nerve modulation in experimental models of sepsis, this latest report offers the real potential for systemic cytokine modulation over the long-term. In addition, the potential also exists for cellular modulation at the immune phenotype level as demonstrated by a recent report from Mucida and coauthors(239). Incorporating technologically advanced imaging and transcriptional profiling techniques, the authors demonstrate that lamina propria zone macrophages, residing close to fecal contents, are primarily proinflammatory, poised to mount robust inflammatory responses should the epithelial barrier be breeched. Conversely, muscularis macrophages, residing deeper in the gut wall, are primarily anti-inflammatory, expressing a tissue protective phenotype. Their search for the mechanism of this phenotypic switch revealed a breakthrough discovery that adrenergic neural signals regulate the tissue protective switch by norepinephrine signaling through β2 AR(239). This is the first direct evidence of a closed loop neural reflex that regulates gut immunity. Moreover, considering the newest insights into gut barrier function, host pathogen interaction, and microbiota contribution to gastrointestinal dysfunction in sepsis, the novel finding that neuronal reflexes control macrophage phenotypic regulation adds importantly to the growing list of neural reflexes implicated in modulating innate and adaptive cellular immune function(240). Metabolic Regulation of Immunity The immune system protects against foreign invaders, maintains optimal tissue homeostasis, and facilitates wound healing throughout the life of the organism. These diverse and integral functions require precise control of cellular, metabolic and bioenergetics pathways. While these initial observations were described at the turn of the century(241), more recent investigations have better defined the molecular basis for how extracellular signals control the uptake, anabolism and catabolism of nutrients in quiescent, activated, and inflammatory immune cells. These reports collectively reveal that oxidative metabolism, glycolysis, and glutaminolysis are preferentially utilized by immune cells and decide cell fate and effector functions(242–244). For more than a century, we have known that propagation of a successful innate effector response is dependent on glucose metabolism(245). In addition, we have also long understood that mitogen-driven proliferation of adaptive immune cells is predicated on the utilization of extracellular glutamine(246, 247). Under homeostatic conditions immune cells rely on oxidative phosphorylation and β-oxidation as energy sources for ATP production to maintain cellular equipoise(248). However, after stimulation, leukocytes shift their metabolism toward aerobic glycolysis in a process known as the Warburg effect(249). In this metabolic shift, cellular energy is predominantly manufactured by an increase in glycolysis followed by lactic acid fermentation (lactate production) in the cytosol, rather than a low rate of glycolysis followed by oxidation of pyruvate in mitochondria(250). Hypoxia-inducible factor–1a(HIF-1a) and the mammalian target of rapamycin (mTOR) are major drivers of this metabolic switch and hence cellular fate(251, 252) (Fig. 8). When innate immune cells are deficient in myeloid specific HIF1a, mice are not protected against S. aureus sepsis, indicating that this HIF1a pathway and glycolytic flux is integral for septic immune responses(253). However, the HIF-1a is not the only metabolic intermediate that drives immune responses in sepsis. Upon exposure to LPS macrophages demonstrate a shift from oxidative phosphorylation to glycolysis and succinate and induce IL-1β production(254, 255). The Glucose transporter 1 (Glut1) mediates an increase in glycolysis that facilitates a macrophage proinflammatory phenotype(256). In a recent report, human leukocytes rendered tolerant by exposure to LPS after isolation from patients with sepsis and immune paralysis demonstrated a generalized metabolic defect at the level of glycolysis and oxidative metabolism, which was restored after patient recovery(214). Compared with baseline, blood gene expression in patients who developed ICU-acquired infections post-sepsis revealed reduced expression of genes involved in gluconeogenesis and glycolysis(73). Furthermore, the genomic response of patients with sepsis was consistent with immune suppression at the onset of secondary infection in the ICU setting. Another report on sepsis patients, identified regulatory genetic variants involving key mediators of gene networks implicated in the hypoxic response and the switch to glycolysis that occurs in sepsis, including HIF1α and mTOR, and mediators of endotoxin tolerance, T-cell activation, and viral defense(257). A clearer understanding of the metabolic checkpoints that control immune cell function, transition, and maturation will provide new insights for modulating systemic inflammation, cellular immunity, and sepsis recovery. The realization that oncogenesis and immune responses have a common mechanism for metabolic switching after stimulation indicates a fundamental cellular processes that can serve as a potential therapeutic targets. Propranolol, Oxandrolone, Dronabinol The adrenergic system is a powerful modulator of the immune system(258). Hematopoietic and lymphopoietic tissues such as the spleen, thymus, lymph nodes, and bone marrow are all predominantly innervated by the sympathetic neurons. Except for CD4+ Th2 cells, the majority of lymphoid cells express beta-adrenergic receptors on their cell surface. Bone marrow production and differentiation of monocytes is influenced by the adrenergic system and immune cell apoptosis is at least partly mediated by catecholamines, via alpha-adrenergic and beta-adrenergic pathways(259, 260). The adrenergic system also modulates cell death, mitochondrial function, and inflammatory signaling(261). Although a great deal of focus has centered on the cardiovascular benefits of beta blockade in sepsis(262), the ubiquitous nature of the adrenergic system begs the question whether there are additional mechanisms whereby beta blockers may exert beneficial influences on immune process. It has been recognized for over fifty years that epinephrine enhances bacterial infections and reduces the absolute number of bacteria necessary for a lethal dose in both Clostridia species and pathogenic aerobic organisms(263). Catecholamines also enhance biofilm formation and stimulate bacterial growth in Staphylococcus epidermidis infections(264). Escherichia coli O157:H7, Salmonella enterica, and Yersinia enterocolitica proliferation are all greatly enhanced by dopamine and norepinephrine through the catechol moiety and its subsequent acquisition by the bacteria(265, 266). Although there is a clear connection between the adrenergic and immune systems, further investigation is still required to elucidate the fundamental mechanisms. For instance, beta blockade reduces proinflammatory cytokines in heart failure(267), critically ill trauma patients(268), and appears to have a beneficial effect on Th1 to Th2 helper T cell ratio(269). In a clinical trial including 55 severely injured patients at increased risk for heart disease, administration of metoprolol or esmolol decreased serum interleukin- (IL-) 6 levels(270). Christensen and coinvestigators conducted a retrospective study on 8087 ICU patients over 6 years. In this case-matched study of 3112 patients the 30-day mortality was 25.7% among beta blocker users and 31.4% among nonusers (OR 0.74 (95% CI: 0.63 to 0.87))(271). Herndon and colleagues have successfully shown that propranolol reduces heart rate by 20% in burned septic children, decreases systemic hypermetabolism and diminishes of muscle-protein catabolism over the ensuing 12 months(272). A retrospective study in trauma patients suggests that pretreatment with beta blockade is associated with a significant decrease in fatal outcome and healing time(273). Given the multiple studies demonstrating a benefit from beta blockade in general and propranolol in particular, the administration of propranolol in burn wounds greater than 20% is considered the standard of care. However, the propranolol induced benefit of reducing hypermetabolism and muscle-protein catabolism in the recovery and long term survival of sepsis patients requires further investigation to prove efficacy. Despite the fact that oxandrolone (testosterone derivative) has not been shown to impact immune function in patients with severe burns or sepsis per se, it has become a mainstay of therapy in severely burned populations to improve weight loss, protect muscle mass, and minimize long term cachectic metabolism(274). In addition, oxandrolone is also associated with improved donor site wound healing(275). In one prospective study, treatment with 10mg of oxandrolone every 12 hours decreased hospital length of stay(276). Lastly in a prospective, double blind, randomized single center trial, oxandrolone administered at a dose of 0.1mg/kg every 12 hours decreased hospital length of stay, preserved lean body mass, improved overall body composition, and enhanced liver protein synthesis(38). Taken together, there is ample evidence that oxandrolone therapy ameliorates the protein wasting, hypermetabolic state, and long term cachectic milieu facilitated by severe burn injury. Although direct evidence establishing oxandrolone as a modulator of immune cell function is lacking, insurmountable data exist indicating that protein wasting, cachexia, weight loss and hypermetabolism are all associated with poor immune function and abysmal outcome in severely injured trauma and critically ill sepsis cohorts(277). This observation has led some to garner optimism for the use of oxandrolone to counteract the well documented catabolism syndrome observed after sepsis(16). Dronabinol is the International Nonproprietary Name for the pure isomer of THC, (–)-trans-Δ9-tetrahydrocannabinol, which is the main isomer found in cannabis. It is used to treat anorexia in people with HIV/AIDS as well as for refractory nausea and vomiting in people undergoing chemotherapy(278). Antagonists at the cannabinoid receptors 1 or 2, (CB1 or CB2) prevents the delay of GI transit and thus may be powerful tools in the future treatment of septic ileus(279). Dexanabinol a synthetic cannabinoid devoid of psychotropic effects, improves neurological outcome in models of brain trauma, ischemia and meningitis. In addition, when given 2 to 3 min before LPS induced rat endotoxemia, completely abolishes the typical hypotensive response. Furthermore, the drug also markedly suppressed in vitro TNFα production and nitric oxide generation in murine peritoneal macrophages and rat alveolar macrophage cell line exposed to LPS. Dexanabinol may, therefore, have therapeutic implications in the treatment of TNFα-mediated pathologies such as sepsis(280). Biomarkers Over 180 biomarkers have been unsuccessfully evaluated for use in sepsis over the past 5 decades(281). Most of the biomarkers had been tested clinically, primarily as prognostic markers in sepsis; relatively few have been used for sepsis diagnosis and prognosis. None of the markers have demonstrated sufficient specificity or sensitivity for reasonable utility in clinical practice. In the past, procalcitonin and C-reactive protein have been most widely used but are limited in their ability to distinguish sepsis from other inflammatory conditions or to predict outcome. More recently, the use of serum lactate levels greater than 2 mmol/L (>18 mg/dL) in the absence of hypovolemia is now a biomarker used in the diagnosis of septic shock(4). The identification of precise biomarkers to detect and quantitate immune suppression in septic patients will be key to successful interdiction with immune modulatory therapies in sepsis. At the present time, reduced HLA–DR expression on monocytes is the most reliable biomarker to assess the immune status of critically ill patients. Reduced monocyte HLA-DR expression and failure of HLA–DR expression restoration are indicative of immune paralysis, the susceptibility to opportunistic infections and overall sepsis outcome(282). In addition the decreased expression of monocyte HLA-DR, the most robust biomarker of sepsis-induced immune suppression, is easily measured by flow cytometry analysis which is quick, easy and affordable(283). In a recent report, investigators were able to demonstrate that monocyte PD-L1 expression after 3–4 days of sepsis was associated with risk stratification and mortality. Moreover, monocyte PD-L1 expression was a promising independent prognostic marker for septic shock patients(284). As our knowledge of cellular phenotype and prognosis grows we will hopefully be able to capitalize on our newfound knowledge of immune suppression and intervene prior to adverse outcomes. Considering our discussion of cell specific immune dysregulation and targeted approaches to restore or enhance cellular function also implies that sensitive and specific tests exist for measuring cellular function and dysfunction that easily identify sepsis survivors who may potentially benefit from immune therapy. For instance, the ability to determine that neutrophil phagocytosis and oxidative burst capability is reduced, CD4+ cell immune exhaustion is present and that Th17 cell cytokine production is diminished are all valuable functional readouts that may serve as new markers of immune dysfunction that can specifically be augmented by targeted therapy. Lastly, genomic prediction modeling can be implemented to identify the inter-individual variation between transcriptomes of patients with sepsis and used to predict long-term outcome(257). As we more clearly begin to understand which aspects of immune dysfunction and suppression are the most crucial for successful outcome, we will be able to develop comprehensive panels of biomarkers comprised of immune phenotypes, cellular functions, genomic alterations, and signaling intermediates to more precisely guide specific immune therapy interventions. Immune Modulatory Intervention The sepsis landscape is littered with failed therapeutic interventions to block particular pathways or processes in humans(10). Although there are as many explanations as failures, continued attempts to augment immune processes at early time points with the expectation that sepsis mortality will be halved at 2 years are doomed to fail(26). However, if the sepsis-induced immune derangements are analyzed alongside other immune-mediated disease processes such as cancer, autoimmune disease, or HIV, where immune modulatory therapy has improved patient survival, it is obvious that single-agent therapy is ineffective. Rather, it is preferable to use multiple agents in combination that are introduced at the correct moment and synergistically altered over time, according to disease-specific progression, patient immune responses, and defined host/pathogen genomic interactions(285). Cancer chemotherapy strategies already utilize a personalized combination of therapies to induce, maintain and prolong cancer disease-free survival based on patient disease progression and tumor-specific genetic patterns. We propose a similar strategy be applied in sepsis therapy with a high probability of success if the interventions are tailored to specific host/pathogen genomic patterns and immune perturbations that occur in the elderly and in patients with co-morbidities as post-sepsis recovery evolves(22). Considering the ease of modern genomic determination and patient screening for specific genetic variations, therapies that enhance microbial eradication could be used to develop a personalized treatment plan. In addition, the widespread and routine use of flow cytometry technology for immune cell phenotyping in blood borne and other cancers types already provides a solid platform for immune cell profiling during cancer, sepsis and other inflammatory states. Although no FDA approved sepsis biomarkers exist at present, using a thoughtful approach to plan patient follow-up, evaluate nutritional status, determine immune cell function, and assess inflammation status, will serve as a foundation to gain the necessary insights to identify the appropriate times and immune therapies to provide to the patient. The collective willingness of sepsis investigators and patient care providers to replace their favorite molecules and interesting mechanistic pathways with a transparent and unified approach to streamline research and patient outcomes to learn meaningful lessons from prior negative intervention trials is paramount. The adoption of the adaptive trail design that has been successfully employed in other disease states should be employed to quickly adjust interventional sepsis trials based on interim analysis and quality reassessment instead of the classic and dogmatic trail design that is methodical and slow natured(286). Finally, the recognition that an episode of sepsis is associated with long-term debilitation, deterioration, and demise in very definable patient populations begs the question why we have such poor follow up and care after hospital discharge? Patients with cancer follow up with their oncologist and receive chemotherapy for years. Patients with HIV and hepatitis C routinely follow-up with their infectious disease specialist for ongoing disease evaluation and microbial therapy alterations over the long term to achieve remission. Until the greater body of inflammation investigators and sepsis clinicians alike coalesce to provide the same long term care and follow up, as done in other specialties in other disease states, the long-term sepsis mortality will continue to climb with time. Although the specific combinations of immune modulation therapy are numerous and all associated with select pros and cons, the following combinations are to serve as an example, to illustrate how long-term sepsis treatment strategies could be employed. These are by no means the only options or even the most correct for all or any patients, but hopefully will convey to the reader how a successful strategy to curb long-term mortality based on the observed immune defects could be employed. For example, GM-CSF treatment with IL-3 and may enhance monocyte and neutrophil production and function early after sepsis when the mature pool of these cells is depleted. Next, a combination of anti-PD-1 or anti-PD-L1 coupled with IFNγ may prove beneficial for lymphocyte activation and augmentation of innate immune surveillance to prevent secondary and nosocomial infections (Table 1). Lastly, a low-dose combination of oxandrolone, propranolol, and even dronabinol may ameliorate protein catabolism and persistent inflammation and promote anabolism, which has already been implemented in promoting recovery of severely burned patients (Table 2). Moreover, with improvements in biomarkers, cellular function determination and host/pathogen genomic prediction models, strategically engineered combinations of immune modulators could be employed in a goal-directed manner based on patient comorbidity profiles, immune function, and recovery course. For example, poorly maintained type 2 diabetic patients recovering from sepsis are predisposed to develop secondary and nosocomial infections associated with poor neutrophil and lymphocyte function. This group of patients may benefit from the concomitant administration of G-CSF, GM-CSF and anti-PD-1 or anti-PD-1L early in the sepsis recovery phase to prevent ongoing infection and subsequent mortality, followed by IFNγ therapy to facilitate an infection-free period and promote more durable recovery. Employing this strategy in sepsis would allow for tailored and monitored interventions that vary over time with specific patient populations, minimizing the human physiologic heterogeneity that has plagued prior sepsis clinical trials. Sepsis Trial Designs In clinical studies, the enrollment criteria are typically very broad, the agent is administered on the basis of a standard formula for only a short period (a few days). There is little information on how an agent changes the host response and host-pathogen interactions, and the primary end point is death from any cause. Such a research strategy is probably overly simplistic in that it does not select patients who are most likely to benefit, cannot adjust therapy on the basis of the evolving host response and clinical course, and does not capture potentially important effects except for effects on 28-day mortality. A more advantageous strategy would be to implement an adaptive clinical trial design that evaluates a sepsis treatment by observing clinical outcomes and side-effects on a routine schedule(287). This information would then be used to modify the study protocols in accord with those newly acquired observations. The “adaptation process” would continue throughout the trial as described in the trial protocol. Modifications may include drug dosage, sample size, patient selection criteria and drug cocktail administration. In some instances, trials have become an ongoing process that routinely adds and drops therapies and patient cohorts as more insight is gained(288). Most importantly, the aim of an adaptive trial is to more quickly identify therapies that have a beneficial effect and the patient populations that are appropriate for such treatment(289). CONCLUSIONS Sepsis induces a multitude of defects in immunity that cause protracted inflammation, immune suppression, susceptibility to infections and insurmountable death. Although there are new cell-based methodologies available to identify patients with post-sepsis immune dysregulation, it is still unclear which interventions and at what time points targeting cell-specific deficits will be most beneficial for sepsis survival. Considering the overlapping, inter-related and interdigitating complexity of immune cell derangements, as well as the protracted and convoluted road to mortality, we believe that single-agent immune modulatory intervention as attempted in past sepsis trials will fail. Conversely, the notion of more thorough and rigorous patient stratification and selection, coupled with strategic and thoughtful long-term monitoring of immune function, combined with goal-directed immune modulatory therapy will, over time, provide optimal clinical benefit to those surviving initial sepsis. MJD would like to acknowledge the Research and Education Foundation Scholarship that supported this work and was generously provided by the 2015 American Association for the Surgery of Trauma and the 2016 Shock Society Faculty Research Award for Early Career Faculty Investigators. Additional support was provided by National Institutes of Health grants, GM-29507 and GM-61656, and from the Godfrey D. Stobbe Endowment (PAW). Figure 1 Earlier Conceptual View and Definition of Systemic Inflammatory Response Syndrome (SIRS), Sepsis, Severe Sepsis, and Septic Shock (A.) The concept of an infection exceeding local regional control and inducing an inflammatory SIRS response has been the fundamental premise conceptualizing sepsis for over two decades. (B.) Until recently, sepsis was defined as the constellation of symptoms occurring when a bacterial, viral or fungal infection leads to a systemic inflammatory response syndrome, including fever, leukocytosis or leukopenia, and decreased vascular resistance frequently leading to hypotension (septic shock), organ failure (severe sepsis) and death. However, confounding the prior definition of sepsis is that other states of inflammation such as pancreatitis, trauma and burns can also produce a SIRS response making the definition overly nebulous and misapplied in many instances. (C.) In addition to the conceptual vagueness, the prior sepsis definition also implied that SIRS criteria possess adequate specificity and sensitivity to define and diagnose sepsis which is not always the case. Moreover, the prior sepsis model inferred that sepsis always follows a linear trajectory from SIRS through severe sepsis and septic shock, which is offend times does not occur. Adapted from Bone RC et al: Chest. 1992,101:1644–55. Figure 2 The Third International Consensus Definitions for Sepsis and Septic Shock The current definitions for sepsis and septic shock were developed to address the limitations of previous definitions that were over focused on SIRS and inflammation. In addition, the Sepsis-3 also dispelled the longstanding notion that SIRS criteria possess adequate specificity and sensitivity to define and diagnose sepsis. Lastly, the report debunked the misleading model that sepsis always follows a linear continuum from the SIRS through severe sepsis and septic shock, and declared the term “severe sepsis” redundant and unnecessary. Instead, the Consensus report recommends that sepsis be defined as a life-threatening organ dysfunction caused by a dysregulated host response to an infection. Figure 3 Organ Dysfunction in Sepsis and Associated Mortality The Sepsis-3 consensus report defined organ dysfunction by an increase in the Sequential Organ Failure Assessment (SOFA) score of 2 points or more, which is associated with an inhospital mortality greater than 10%. Figure 4 Definition of Septic Shock and Associated Mortality Septic shock is now defined as a subset of sepsis in which profound circulatory, cellular, and metabolic abnormalities are associated with a greater risk of mortality than with sepsis alone. Clinically, patients with septic shock can be identified by a vasopressor requirement to maintain a mean arterial pressure of 65 mmHg or greater and serum lactate level greater than 2 mmol/L (>18mg/dL) in the absence of hypovolemia with in-hospital mortality rates greater than 40%. Figure 5 Bedside Criteria Defined to Identify To identify patients with the highest probability of poor outcome associated with sepsis, a new bedside clinical score named the quickSOFA (qSOFA) was created which consist of at least 2 of the following clinical criteria including, respiratory rate of 22/min or greater, altered mentation (GCS 14), or systolic blood pressure of 100mmHg or less. Figure 6 Past and Present Mortality Distribution Sepsis (A.) Classically, the mortality distribution from sepsis occurred in a biphasic pattern, with an initial peak due to inadequate resuscitation resulting in cardiac and pulmonary failure and a second peak at several weeks from persistent organ dysfunction. Considering the recent trends in physiologic frailty, the growing elderly population, and mounting long-term mortality, a trimodal distribution is more indicative of the current sepsis-associated mortality. (B.) The two early peaks in mortality still do exist but with much less magnitude than in the past. The third and largest upswing occurs beginning at 2–3 months after sepsis and continues to steeply climb as time progresses. This delay in sepsis mortality is attributed to significant advances in ICU care that keeps the elderly and co-morbidly challenged patients alive longer despite ongoing immune, physiologic, metabolomics and biochemical aberrations. Figure 7 Inflammatory vs Anti-Inflammatory Responses An ongoing debate persist as to whether innate and adaptive immune dysfunction or inflammatory and anti-inflammatory processes are more detrimental to overall sepsis survival. In the past, the inflammatory response was thought to drive early mortality in the initial days of sepsis, and the compensatory anti-inflammatory response was thought to induce mortality weeks later through immune suppression and organ failure. However new insights gathered from septic patient tissue samples and severely injured trauma patients, have identified an enduring and simultaneous inflammatory and anti-inflammatory state of affairs driven by dysfunctional innate and suppressed adaptive immunity that together culminate in persistent organ injury, infectious complications requiring hospital readmission, and ultimately patient death. It is evident that the inflammatory and anti-inflammatory responses and innate and adaptive immune systems are each equally important, continually in a state of fluctuation, and ever at odds with one another, as sepsis recovery progresses. This perpetual state of immunologic yin and yang is thought to drive ongoing inflammation, facilitate organ injury, and enable infectious complications that all preclude durable sepsis survival. Figure 8 Alterations in Metabolic Function Determine Immune Phenotype Immune cells rely on oxidative phosphorylation and β-oxidation as energy sources for ATP production to maintain cellular equipoise at homeostasis. However, after stimulation, leukocytes shift their metabolism toward aerobic glycolysis in a process known as the Warburg effect. In this metabolic shift, cellular energy is predominantly manufactured by an increase in glycolysis followed by lactic acid fermentation (lactate production) in the cytosol, rather than a low rate of glycolysis followed by oxidation of pyruvate in mitochondria. Hypoxia-inducible factor–1a(HIF-1a) and the mammalian target of rapamycin (mTOR) are major drivers of this metabolic switch and hence determines cellular fate. These metabolic shifts have been incriminated in immune suppression and secondary infection progression in humans. A clearer understanding of the metabolic checkpoints that control immune cell function, transition, and maturation will provide new insights for modulating systemic inflammation, cellular immunity, and sepsis recovery. Table 1 Immune Modulator G-CSF GM-CSF IFNγ PD-1 and PD-L1 IL-3 IL-7 IL-15 Cellular Benefit Improve neutrophil and monocyte production and release Improve neutrophil and monocyte production and function Improve monocyte HLA-DR expression and function Biomarker to identify candidates for immune modulatory therapy Promote stem cell and progenitor development Increase T cell proliferation and recruitment Decrease NK, T cell, and NKT cell apoptosis Improve meylopoiesis and granulopoiesis Enhance monocyte and lymphocyte cytotoxicity Reduce infection and related complications Reverse T cell exhaustion Enhance lymphopoiesis in combination with IL-7 Decrease lymphocyte apoptosis Increase NK, T cell, and NKT cell proliferation and activation Augment T cell responses Improve immunity against fungal infections Promote lymphocyte proliferation Increase T cell IFNγ secretion Reduce nosocomial infection acquisition Augment neutrophil and monocyte cytotoxicity Improve T cell homing and pathogen clearance Reduce ventilator days Reduce opportunistic infections Increases T cell receptor diversity Table 2 Immune Modulator Propranolol Oxandrolone Dronabinol Benefit Reduce inflammatory cytokine production Improve wieght loss Improve gut transit, Increase appetite Diminish muscle protien catabolism Protect muscle mass, Reduce length of stay Reduce TNFα production Improve 30 day survival Minimize cachectic metabolism Reduce nitric oxide generation The authors declare there are no commercial or financial conflicts of interest related to the studies. 1 American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis Crit Care Med 1992 20 864 874 1597042 2 Levy MM 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference Crit Care Med 2003 31 1250 1256 12682500 3 Kaukonen KM Bailey M Pilcher D Cooper DJ Bellomo R Systemic inflammatory response syndrome criteria in defining severe sepsis N Engl J Med 2015 372 1629 1638 25776936 4 Singer M The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) JAMA 2016 315 801 810 26903338 5 Gaieski DF Edwards JM Kallan MJ Carr BG Benchmarking the incidence and mortality of severe sepsis in the United States Crit Care Med 2013 41 1167 1174 23442987 6 Mayr FB Yende S Angus DC Epidemiology of severe sepsis Virulence 2014 5 4 11 24335434 7 Coopersmith CM A comparison of critical care research funding and the financial burden of critical illness in the United States Crit Care Med 2012 40 1072 1079 22202712 8 Martin GS Mannino DM Moss M The effect of age on the development and outcome of adult sepsis Crit Care Med 2006 34 15 21 16374151 9 Kahn JM The epidemiology of chronic critical illness in the United States* Critical care medicine 2015 43 282 287 25377018 10 Marshall JC Why have clinical trials in sepsis failed? Trends Mol Med 2014 20 195 203 24581450 11 Deutschman CS Tracey KJ Sepsis: current dogma and new perspectives Immunity 2014 40 463 475 24745331 12 Ward PA Bosmann M A historical perspective on sepsis Am J Pathol 2012 181 2 7 22642906 13 Tracey KJ Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia Nature 1987 330 662 664 3317066 14 Schorr CA Dellinger RP The Surviving Sepsis Campaign: past, present and future Trends Mol Med 2014 20 192 194 24698888 15 Boomer JS Immunosuppression in patients who die of sepsis and multiple organ failure JAMA 2011 306 2594 2605 22187279 16 Gentile LF Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care J Trauma Acute Care Surg 2012 72 1491 1501 22695412 17 Winters BD Eberlein M Leung J Needham DM Pronovost PJ Sevransky JE Long-term mortality and quality of life in sepsis: a systematic review Crit Care Med 2010 38 1276 1283 20308885 18 Moore FA Moore EE Evolving concepts in the pathogenesis of postinjury multiple organ failure Surg Clin North Am 1995 75 257 277 7899997 19 Delano MJ Ward PA Sepsis-induced immune dysfunction: can immune therapies reduce mortality? J Clin Invest 2016 126 23 31 26727230 20 Needham DM Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders' conference Crit Care Med 2012 40 502 509 21946660 21 Yende S Inflammatory markers at hospital discharge predict subsequent mortality after pneumonia and sepsis Am J Respir Crit Care Med 2008 177 1242 1247 18369199 22 Elliott D Exploring the scope of post-intensive care syndrome therapy and care: engagement of non-critical care providers and survivors in a second stakeholders meeting Crit Care Med 2014 42 2518 2526 25083984 23 Annane D What Is the Evidence for Harm of Neuromuscular Blockade and Corticosteroid Use in the Intensive Care Unit? Semin Respir Crit Care Med 2016 37 51 56 26820274 24 Hotchkiss RS Moldawer LL Parallels between cancer and infectious disease N Engl J Med 2014 371 380 383 25054723 25 Hotchkiss RS Monneret G Payen D Sepsis-induced immunosuppression: from cellular dysfunctions to immunotherapy Nat Rev Immunol 2013 13 862 874 24232462 26 Delano MJ Moldawer LL Magic bullets and surrogate biomarkers circa 2009 Crit Care Med 2009 37 1796 1798 19373047 27 Bosmann M Ward PA The inflammatory response in sepsis Trends Immunol 2013 34 129 136 23036432 28 Hotchkiss RS Coopersmith CM McDunn JE Ferguson TA The sepsis seesaw: tilting toward immunosuppression Nat Med 2009 15 496 497 19424209 29 Bone RC Grodzin CJ Balk RA Sepsis: a new hypothesis for pathogenesis of the disease process Chest 1997 112 235 243 9228382 30 Rittirsch D Flierl MA Ward PA Harmful molecular mechanisms in sepsis Nat Rev Immunol 2008 8 776 787 18802444 31 Xiao W A genomic storm in critically injured humans J Exp Med 2011 208 2581 2590 22110166 32 Cuenca AG Development of a genomic metric that can be rapidly used to predict clinical outcome in severely injured trauma patients Crit Care Med 2013 41 1175 1185 23388514 33 Hutchins NA Unsinger J Hotchkiss RS Ayala A The new normal: immunomodulatory agents against sepsis immune suppression Trends Mol Med 2014 20 224 233 24485901 34 Medzhitov R Janeway C Jr Innate immunity N Engl J Med 2000 343 338 344 10922424 35 Goodwin AJ Rice DA Simpson KN Ford DW Frequency, cost, and risk factors of readmissions among severe sepsis survivors Crit Care Med 2015 43 738 746 25746745 36 Gauglitz GG Williams FN Herndon DN Jeschke MG Burns: where are we standing with propranolol, oxandrolone, recombinant human growth hormone, and the new incretin analogs? Curr Opin Clin Nutr Metab Care 2011 14 176 181 21157309 37 Norbury WB Jeschke MG Herndon DN Metabolism modulators in sepsis: propranolol Crit Care Med 2007 35 S616 620 17713418 38 Jeschke MG Finnerty CC Suman OE Kulp G Mlcak RP Herndon DN The effect of oxandrolone on the endocrinologic, inflammatory, and hypermetabolic responses during the acute phase postburn Ann Surg 2007 246 351 360 discussion 360–352 17717439 39 Goldszmid RS Dzutsev A Trinchieri G Host immune response to infection and cancer: unexpected commonalities Cell Host Microbe 2014 15 295 305 24629336 40 Bohlius J Herbst C Reiser M Schwarzer G Engert A Granulopoiesis-stimulating factors to prevent adverse effects in the treatment of malignant lymphoma Cochrane Database Syst Rev 2008 CD003189 18843642 41 Smilek DE Ehlers MR Nepom GT Restoring the balance: immunotherapeutic combinations for autoimmune disease Dis Model Mech 2014 7 503 513 24795433 42 Gilroy DW Vallance P Resolution for sepsis? Circulation 2005 111 2 4 15630034 43 Fullerton JN Gilroy DW Resolution of inflammation: a new therapeutic frontier Nat Rev Drug Discov 2016 44 Serhan CN Pro-resolving lipid mediators are leads for resolution physiology Nature 2014 510 92 101 24899309 45 Buckley CD Gilroy DW Serhan CN Proresolving lipid mediators and mechanisms in the resolution of acute inflammation Immunity 2014 40 315 327 24656045 46 Venet F Regulatory T cell populations in sepsis and trauma J Leukoc Biol 2008 83 523 535 17913974 47 Venet F Increased circulating regulatory T cells (CD4(+)CD25 (+)CD127 (-)) contribute to lymphocyte anergy in septic shock patients Intensive Care Med 2009 35 678 686 18946659 48 Brudecki L Ferguson DA McCall CE El Gazzar M Myeloid-derived suppressor cells evolve during sepsis and can enhance or attenuate the systemic inflammatory response Infect Immun 2012 80 2026 2034 22451518 49 Dalli J Chiang N Serhan CN Elucidation of novel 13-series resolvins that increase with atorvastatin and clear infections Nat Med 2015 21 1071 1075 26236990 50 Lee CR Zeldin DC Resolvin Infectious Inflammation by Targeting the Host Response N Engl J Med 2015 373 2183 2185 26605933 51 Deshpande A Pasupuleti V Rothberg MB Statin therapy and mortality from sepsis: a meta-analysis of randomized trials Am J Med 2015 128 410 417 e411 25526798 52 Cummings CJ Expression and function of the chemokine receptors CXCR1 and CXCR2 in sepsis J Immunol 1999 162 2341 2346 9973513 53 Kovach MA Standiford TJ The function of neutrophils in sepsis Curr Opin Infect Dis 2012 25 321 327 22421753 54 Grailer JJ Kalbitz M Zetoune FS Ward PA Persistent neutrophil dysfunction and suppression of acute lung injury in mice following cecal ligation and puncture sepsis J Innate Immun 2014 6 695 705 24861731 55 Huber-Lang M Role of C5a in multiorgan failure during sepsis J Immunol 2001 166 1193 1199 11145701 56 Cavaillon JM Adib-Conquy M Monocytes/macrophages and sepsis Crit Care Med 2005 33 S506 509 16340435 57 Benjamim CF Lundy SK Lukacs NW Hogaboam CM Kunkel SL Reversal of long-term sepsis-induced immunosuppression by dendritic cells Blood 2005 105 3588 3595 15604223 58 Pene F Dendritic cells modulate lung response to Pseudomonas aeruginosa in a murine model of sepsis-induced immune dysfunction J Immunol 2008 181 8513 8520 19050269 59 Grimaldi D Llitjos JF Pene F Post-infectious immune suppression: a new paradigm of severe infections Med Mal Infect 2014 44 455 463 25169939 60 Hotchkiss RS Depletion of dendritic cells, but not macrophages, in patients with sepsis J Immunol 2002 168 2493 2500 11859143 61 Hotchkiss RS Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction Crit Care Med 1999 27 1230 1251 10446814 62 Walton AH Reactivation of multiple viruses in patients with sepsis PLoS One 2014 9 e98819 24919177 63 Drewry AM Samra N Skrupky LP Fuller BM Compton SM Hotchkiss RS Persistent lymphopenia after diagnosis of sepsis predicts mortality Shock 2014 42 383 391 25051284 64 Otto GP The late phase of sepsis is characterized by an increased microbiological burden and death rate Crit Care 2011 15 R183 21798063 65 Prescott HC Langa KM Iwashyna TJ Readmission diagnoses after hospitalization for severe sepsis and other acute medical conditions JAMA 2015 313 1055 1057 25756444 66 Jones TK Post-Acute Care Use and Hospital Readmission after Sepsis Ann Am Thorac Soc 2015 12 904 913 25751120 67 Chang DW Tseng CH Shapiro MF Rehospitalizations Following Sepsis: Common and Costly Crit Care Med 2015 43 2085 2093 26131597 68 Jensen US Knudsen JD Wehberg S Gregson DB Laupland KB Risk factors for recurrence and death after bacteraemia: a population-based study Clin Microbiol Infect 2011 17 1148 1154 21714830 69 Prescott HC Dickson RP Rogers MA Langa KM Iwashyna TJ Hospitalization Type and Subsequent Severe Sepsis Am J Respir Crit Care Med 2015 192 581 588 26016947 70 Prescott HC Toward a Nuanced Understanding of the Role of Infection in Readmissions After Sepsis Crit Care Med 2016 44 634 635 26901548 71 Sun A Association Between Index Hospitalization and Hospital Readmission in Sepsis Survivors Crit Care Med 2016 44 478 487 26571185 72 Leibovici L Long-term consequences of severe infections Clin Microbiol Infect 2013 19 510 512 23397980 73 van Vught LA Incidence, Risk Factors, and Attributable Mortality of Secondary Infections in the Intensive Care Unit After Admission for Sepsis JAMA 2016 315 1469 1479 26975785 74 Wang T Derhovanessian A De Cruz S Belperio JA Deng JC Hoo GS Subsequent infections in survivors of sepsis: epidemiology and outcomes J Intensive Care Med 2014 29 87 95 23753224 75 Sogaard M Thomsen RW Bang RB Schonheyder HC Norgaard M Trends in length of stay, mortality and readmission among patients with community-acquired bacteraemia Clin Microbiol Infect 2015 21 789 e781 787 26003278 76 Suzuki K Reduced Immunocompetent B Cells and Increased Secondary Infection in Elderly Patients with Severe Sepsis Shock 2016 77 Martin GS Sepsis, severe sepsis and septic shock: changes in incidence, pathogens and outcomes Expert Rev Anti Infect Ther 2012 10 701 706 22734959 78 Unnewehr H Changes and regulation of the C5a receptor on neutrophils during septic shock in humans J Immunol 2013 190 4215 4225 23479227 79 Czermak BJ Protective effects of C5a blockade in sepsis Nat Med 1999 5 788 792 10395324 80 Huber-Lang MS Protective effects of anti-C5a peptide antibodies in experimental sepsis FASEB J 2001 15 568 570 11259369 81 Huber-Lang M Generation of C5a in the absence of C3: a new complement activation pathway Nat Med 2006 12 682 687 16715088 82 Ward PA The harmful role of c5a on innate immunity in sepsis J Innate Immun 2010 2 439 445 20588003 83 Melis JP Strumane K Ruuls SR Beurskens FJ Schuurman J Parren PW Complement in therapy and disease: Regulating the complement system with antibody-based therapeutics Mol Immunol 2015 67 117 130 25697848 84 Ogilvie AC Groeneveld AB Straub JP Thijs LG Plasma lipid peroxides and antioxidants in human septic shock Intensive Care Med 1991 17 40 44 1645378 85 Cowley HC Bacon PJ Goode HF Webster NR Jones JG Menon DK Plasma antioxidant potential in severe sepsis: a comparison of survivors and nonsurvivors Crit Care Med 1996 24 1179 1183 8674332 86 Santos SS Brunialti MK Rigato O Machado FR Silva E Salomao R Generation of nitric oxide and reactive oxygen species by neutrophils and monocytes from septic patients and association with outcomes Shock 2012 38 18 23 22575991 87 Andrades ME Morina A Spasic S Spasojevic I Bench-to-bedside review: sepsis - from the redox point of view Crit Care 2011 15 230 21996422 88 Ojeda Ojeda M Temporal trends of circulating nitric oxide and pro-inflammatory cytokine responses ex vivo in intra-abdominal sepsis: results from a cohort study Inflamm Res 2011 60 289 297 20976525 89 Dare AJ A systematic review of experimental treatments for mitochondrial dysfunction in sepsis and multiple organ dysfunction syndrome Free Radic Biol Med 2009 47 1517 1525 19715753 90 Mishra V Oxidative stress and role of antioxidant supplementation in critical illness Clin Lab 2007 53 199 209 17447658 91 Muller B Disordered calcium homeostasis of sepsis: association with calcitonin precursors Eur J Clin Invest 2000 30 823 831 10998084 92 Zivin JR Gooley T Zager RA Ryan MJ Hypocalcemia: a pervasive metabolic abnormality in the critically ill Am J Kidney Dis 2001 37 689 698 11273867 93 Desai TK Carlson RW Geheb MA Prevalence and clinical implications of hypocalcemia in acutely ill patients in a medical intensive care setting Am J Med 1988 84 209 214 3407650 94 Sayeed MM Signaling mechanisms of altered cellular responses in trauma, burn, and sepsis: role of Ca2+ Arch Surg 2000 135 1432 1442 11115349 95 Zaloga GP Washburn D Black KW Prielipp R Human sepsis increases lymphocyte intracellular calcium Crit Care Med 1993 21 196 202 8428469 96 Thompson M Increased cardiomyocyte intracellular calcium during endotoxin-induced cardiac dysfunction in guinea pigs Pediatr Res 2000 47 669 676 10813595 97 Song SK Karl IE Ackerman JJ Hotchkiss RS Increased intracellular Ca2+: a critical link in the pathophysiology of sepsis? Proc Natl Acad Sci U S A 1993 90 3933 3937 8483913 98 Collage RD Calcium supplementation during sepsis exacerbates organ failure and mortality via calcium/calmodulin-dependent protein kinase kinase signaling Crit Care Med 2013 41 e352 360 23887235 99 Ba X Garg NJ Signaling mechanism of poly(ADP-ribose) polymerase-1 (PARP-1) in inflammatory diseases Am J Pathol 2011 178 946 955 21356345 100 Liaudet L Oddo M Role of poly(adenosine diphosphate-ribose) polymerase 1 in septic peritonitis Curr Opin Crit Care 2003 9 152 158 12657979 101 Wongsrichanalai C Namsiripongpun V Pornsilapatip J Kyle DE Wilde H Sensitivity of QBC malaria test Lancet 1992 340 792 793 102 Soriano FG Resistance to acute septic peritonitis in poly(ADP-ribose) polymerase-1-deficient mice Shock 2002 17 286 292 11954828 103 Hotchkiss RS Strasser A McDunn JE Swanson PE Cell death N Engl J Med 2009 361 1570 1583 19828534 104 Larsen FJ Schiffer TA Weitzberg E Lundberg JO Regulation of mitochondrial function and energetics by reactive nitrogen oxides Free Radic Biol Med 2012 53 1919 1928 22989554 105 Szabo C Modis K Pathophysiological roles of peroxynitrite in circulatory shock Shock 2010 34 Suppl 1 4 14 106 Brealey D Association between mitochondrial dysfunction and severity and outcome of septic shock Lancet 2002 360 219 223 12133657 107 Srinivasan S Avadhani NG Cytochrome c oxidase dysfunction in oxidative stress Free Radic Biol Med 2012 53 1252 1263 22841758 108 Singer M The role of mitochondrial dysfunction in sepsis-induced multi-organ failure Virulence 2014 5 66 72 24185508 109 Brealey D Mitochondrial dysfunction in a long-term rodent model of sepsis and organ failure Am J Physiol Regul Integr Comp Physiol 2004 286 R491 497 14604843 110 Gentile LF A better understanding of why murine models of trauma do not recapitulate the human syndrome Crit Care Med 2014 42 1406 1413 24413577 111 Opitz B Eitel J Meixenberger K Suttorp N Role of Toll-like receptors, NOD-like receptors and RIG-I-like receptors in endothelial cells and systemic infections Thromb Haemost 2009 102 1103 1109 19967140 112 Faure E Bacterial lipopolysaccharide activates NF-kappaB through toll-like receptor 4 (TLR-4) in cultured human dermal endothelial cells. Differential expression of TLR-4 and TLR-2 in endothelial cells J Biol Chem 2000 275 11058 11063 10753909 113 Ince C The Endothelium in Sepsis Shock 2016 45 259 270 26871664 114 Colbert JF Schmidt EP Endothelial and Microcirculatory Function and Dysfunction in Sepsis Clin Chest Med 2016 37 263 275 27229643 115 Mikacenic C Biomarkers of Endothelial Activation Are Associated with Poor Outcome in Critical Illness PLoS One 2015 10 e0141251 26492036 116 van Ierssel SH Van Craenenbroeck EM Hoymans VY Vrints CJ Conraads VM Jorens PG Endothelium dependent vasomotion and in vitro markers of endothelial repair in patients with severe sepsis: an observational study PLoS One 2013 8 e69499 23936333 117 Mostefai HA Circulating microparticles from patients with septic shock exert protective role in vascular function Am J Respir Crit Care Med 2008 178 1148 1155 18723433 118 Nathan C Neutrophils and immunity: challenges and opportunities Nat Rev Immunol 2006 6 173 182 16498448 119 Link DC Neutrophil homeostasis: a new role for stromal cell-derived factor-1 Immunol Res 2005 32 169 178 16106067 120 Hotchkiss RS Nicholson DW Apoptosis and caspases regulate death and inflammation in sepsis Nat Rev Immunol 2006 6 813 822 17039247 121 Delano MJ Sepsis induces early alterations in innate immunity that impact mortality to secondary infection J Immunol 2011 186 195 202 21106855 122 Eash KJ Greenbaum AM Gopalan PK Link DC CXCR2 and CXCR4 antagonistically regulate neutrophil trafficking from murine bone marrow J Clin Invest 2010 120 2423 2431 20516641 123 Delano MJ Neutrophil mobilization from the bone marrow during polymicrobial sepsis is dependent on CXCL12 signaling J Immunol 2011 187 911 918 21690321 124 Lerman YV Sepsis lethality via exacerbated tissue infiltration and TLR-induced cytokine production by neutrophils is integrin alpha3beta1-dependent Blood 2014 124 3515 3523 25278585 125 Alves-Filho JC Regulation of chemokine receptor by Toll-like receptor 2 is critical to neutrophil migration and resistance to polymicrobial sepsis Proc Natl Acad Sci U S A 2009 106 4018 4023 19234125 126 Nacionales DC Aged mice are unable to mount an effective myeloid response to sepsis J Immunol 2014 192 612 622 24337739 127 Delano MJ MyD88-dependent expansion of an immature GR-1(+)CD11b(+) population induces T cell suppression and Th2 polarization in sepsis J Exp Med 2007 204 1463 1474 17548519 128 Davey MS Microbe-specific unconventional T cells induce human neutrophil differentiation into antigen cross-presenting cells J Immunol 2014 193 3704 3716 25165152 129 Morris AC C5a-mediated neutrophil dysfunction is RhoA-dependent and predicts infection in critically ill patients Blood 2011 117 5178 5188 21292772 130 Wilkinson TS Ventilator-associated pneumonia is characterized by excessive release of neutrophil proteases in the lung Chest 2012 142 1425 1432 22911225 131 Brinkmann V Neutrophil extracellular traps kill bacteria Science 2004 303 1532 1535 15001782 132 Warnatsch A Ioannou M Wang Q Papayannopoulos V Inflammation. Neutrophil extracellular traps license macrophages for cytokine production in atherosclerosis Science 2015 349 316 320 26185250 133 Sorensen OE Borregaard N Neutrophil extracellular traps - the dark side of neutrophils J Clin Invest 2016 126 1612 1620 27135878 134 McDonald B Urrutia R Yipp BG Jenne CN Kubes P Intravascular neutrophil extracellular traps capture bacteria from the bloodstream during sepsis Cell Host Microbe 2012 12 324 333 22980329 135 Czaikoski PG Neutrophil Extracellular Traps Induce Organ Damage during Experimental and Clinical Sepsis PLoS One 2016 11 e0148142 26849138 136 Cheadle WG Wilson M Hershman MJ Bergamini D Richardson JD Polk HC Jr Comparison of trauma assessment scores and their use in prediction of infection and death Ann Surg 1989 209 541 545 discussion 545–546 2705819 137 Livingston DH Appel SH Wellhausen SR Sonnenfeld G Polk HC Jr Depressed interferon gamma production and monocyte HLA-DR expression after severe injury Arch Surg 1988 123 1309 1312 3140765 138 Munoz C Carlet J Fitting C Misset B Bleriot JP Cavaillon JM Dysregulation of in vitro cytokine production by monocytes during sepsis J Clin Invest 1991 88 1747 1754 1939659 139 Docke WD Monocyte deactivation in septic patients: restoration by IFN-gamma treatment Nat Med 1997 3 678 681 9176497 140 Cazalis MA Decreased HLA-DR antigen-associated invariant chain (CD74) mRNA expression predicts mortality after septic shock Crit Care 2013 17 R287 24321376 141 Cheron A Lack of recovery in monocyte human leukocyte antigen-DR expression is independently associated with the development of sepsis after major trauma Crit Care 2010 14 R208 21092108 142 Landelle C Low monocyte human leukocyte antigen-DR is independently associated with nosocomial infections after septic shock Intensive Care Med 2010 36 1859 1866 20652682 143 Venet F Human CD4+CD25+ regulatory T lymphocytes inhibit lipopolysaccharide-induced monocyte survival through a Fas/Fas ligand-dependent mechanism J Immunol 2006 177 6540 6547 17056586 144 Monneret G Venet F Pachot A Lepape A Monitoring immune dysfunctions in the septic patient: a new skin for the old ceremony Mol Med 2008 14 64 78 18026569 145 Carson WF Cavassani KA Dou Y Kunkel SL Epigenetic regulation of immune cell functions during post-septic immunosuppression Epigenetics 2011 6 273 283 21048427 146 Ishii M Epigenetic regulation of the alternatively activated macrophage phenotype Blood 2009 114 3244 3254 19567879 147 Saeed S Epigenetic programming of monocyte-to-macrophage differentiation and trained innate immunity Science 2014 345 1251086 25258085 148 Poli A Michel T Theresine M Andres E Hentges F Zimmer J CD56bright natural killer (NK) cells: an important NK cell subset Immunology 2009 126 458 465 19278419 149 Halstead ES Carcillo JA Schilling B Greiner RJ Whiteside TL Reduced frequency of CD56 dim CD16 pos natural killer cells in pediatric systemic inflammatory response syndrome/sepsis patients Pediatr Res 2013 74 427 432 23857294 150 Giamarellos-Bourboulis EJ Early changes of CD4-positive lymphocytes and NK cells in patients with severe Gram-negative sepsis Crit Care 2006 10 R166 17129388 151 Souza-Fonseca-Guimaraes F Toll-like receptors expression and interferon-gamma production by NK cells in human sepsis Crit Care 2012 16 R206 23098236 152 Chiche L Interferon-gamma production by natural killer cells and cytomegalovirus in critically ill patients Crit Care Med 2012 40 3162 3169 22971588 153 Limaye AP Cytomegalovirus reactivation in critically ill immunocompetent patients JAMA 2008 300 413 422 18647984 154 Giamarellos-Bourboulis EJ Natural killer cells in sepsis: detrimental role for final outcome Crit Care Med 2014 42 1579 1580 24836809 155 Broquet A Roquilly A Jacqueline C Potel G Caillon J Asehnoune K Depletion of natural killer cells increases mice susceptibility in a Pseudomonas aeruginosa pneumonia model Crit Care Med 2014 42 e441 450 24732238 156 Grimaldi D Profound and persistent decrease of circulating dendritic cells is associated with ICU-acquired infection in patients with septic shock Intensive Care Med 2011 37 1438 1446 21805160 157 Pastille E Modulation of dendritic cell differentiation in the bone marrow mediates sustained immunosuppression after polymicrobial sepsis J Immunol 2011 186 977 986 21160046 158 Efron PA Characterization of the systemic loss of dendritic cells in murine lymph nodes during polymicrobial sepsis J Immunol 2004 173 3035 3043 15322163 159 Scumpia PO CD11c+ dendritic cells are required for survival in murine polymicrobial sepsis J Immunol 2005 175 3282 3286 16116220 160 Toliver-Kinsky TE Lin CY Herndon DN Sherwood ER Stimulation of hematopoiesis by the Fms-like tyrosine kinase 3 ligand restores bacterial induction of Th1 cytokines in thermally injured mice Infect Immun 2003 71 3058 3067 12761083 161 Ochoa JB Arginase I expression and activity in human mononuclear cells after injury Ann Surg 2001 233 393 399 11224628 162 Gabrilovich DI Ostrand-Rosenberg S Bronte V Coordinated regulation of myeloid cells by tumours Nat Rev Immunol 2012 12 253 268 22437938 163 Cuenca AG A paradoxical role for myeloid-derived suppressor cells in sepsis and trauma Mol Med 2011 17 281 292 21085745 164 Makarenkova VP Bansal V Matta BM Perez LA Ochoa JB CD11b+/Gr-1+ myeloid suppressor cells cause T cell dysfunction after traumatic stress J Immunol 2006 176 2085 2094 16455964 165 Kelly-Scumpia KM Type I interferon signaling in hematopoietic cells is required for survival in mouse polymicrobial sepsis by regulating CXCL10 J Exp Med 2010 207 319 326 20071504 166 Scumpia PO Cutting edge: bacterial infection induces hematopoietic stem and progenitor cell expansion in the absence of TLR signaling J Immunol 2010 184 2247 2251 20130216 167 Cartwright GE Athens JW Wintrobe MM The Kinetics of Granulopoiesis in Normal Man Blood 1964 24 780 803 14235362 168 McCabe A Zhang Y Thai V Jones M Jordan MB MacNamara KC Macrophage-Lineage Cells Negatively Regulate the Hematopoietic Stem Cell Pool in Response to Interferon Gamma at Steady State and During Infection Stem Cells 2015 33 2294 2305 25880153 169 Mathias B Human Myeloid-derived Suppressor Cells are Associated With Chronic Immune Suppression After Severe Sepsis/Septic Shock Ann Surg 2016 170 Terashima A Okamoto K Nakashima T Akira S Ikuta K Takayanagi H Sepsis-Induced Osteoblast Ablation Causes Immunodeficiency Immunity 2016 44 1434 1443 27317262 171 Hotchkiss RS Coopersmith CM Karl IE Prevention of lymphocyte apoptosis--a potential treatment of sepsis? Clin Infect Dis 2005 41 Suppl 7 S465 469 16237649 172 Lang JD Matute-Bello G Lymphocytes, apoptosis and sepsis: making the jump from mice to humans Crit Care 2009 13 109 19216722 173 Hotchkiss RS Swanson PE Cobb JP Jacobson A Buchman TG Karl IE Apoptosis in lymphoid and parenchymal cells during sepsis: findings in normal and T- and B-cell-deficient mice Crit Care Med 1997 25 1298 1307 9267941 174 Hotchkiss RS Osmon SB Chang KC Wagner TH Coopersmith CM Karl IE Accelerated lymphocyte death in sepsis occurs by both the death receptor and mitochondrial pathways J Immunol 2005 174 5110 5118 15814742 175 Liu ZG Histones-mediated lymphocyte apoptosis during sepsis is dependent on p38 phosphorylation and mitochondrial permeability transition PLoS One 2013 8 e77131 24167561 176 Grailer JJ Fattahi F Dick RS Zetoune FS Ward PA Cutting edge: critical role for C5aRs in the development of septic lymphopenia in mice J Immunol 2015 194 868 872 25539817 177 Unsinger J IL-7 promotes T cell viability, trafficking, and functionality and improves survival in sepsis J Immunol 2010 184 3768 3779 20200277 178 Brahmamdam P Inoue S Unsinger J Chang KC McDunn JE Hotchkiss RS Delayed administration of anti-PD-1 antibody reverses immune dysfunction and improves survival during sepsis J Leukoc Biol 2010 88 233 240 20483923 179 Holtmeier W Kabelitz D gammadelta T cells link innate and adaptive immune responses Chem Immunol Allergy 2005 86 151 183 15976493 180 Andreu-Ballester JC Association of gammadelta T cells with disease severity and mortality in septic patients Clin Vaccine Immunol 2013 20 738 746 23515014 181 Grimaldi D Specific MAIT cell behaviour among innate-like T lymphocytes in critically ill patients with severe infections Intensive Care Med 2014 40 192 201 24322275 182 Tomasello E Bedoui S Intestinal innate immune cells in gut homeostasis and immunosurveillance Immunol Cell Biol 2013 91 201 203 23478396 183 Szabo PA Anantha RV Shaler CR McCormick JK Haeryfar SM CD1d- and MR1-Restricted T Cells in Sepsis Front Immunol 2015 6 401 26322041 184 Gutcher I Becher B APC-derived cytokines and T cell polarization in autoimmune inflammation J Clin Invest 2007 117 1119 1127 17476341 185 Hotchkiss RS Sepsis-induced apoptosis causes progressive profound depletion of B and CD4+ T lymphocytes in humans J Immunol 2001 166 6952 6963 11359857 186 O'Sullivan ST Lederer JA Horgan AF Chin DH Mannick JA Rodrick ML Major injury leads to predominance of the T helper-2 lymphocyte phenotype and diminished interleukin-12 production associated with decreased resistance to infection Ann Surg 1995 222 482 490 discussion 490–482 7574928 187 Pachot A Longitudinal study of cytokine and immune transcription factor mRNA expression in septic shock Clin Immunol 2005 114 61 69 15596410 188 Romani L Immunity to fungal infections Nat Rev Immunol 2011 11 275 288 21394104 189 Wu HP Chung K Lin CY Jiang BY Chuang DY Liu YC Associations of T helper 1, 2, 17 and regulatory T lymphocytes with mortality in severe sepsis Inflamm Res 2013 62 751 763 23670410 190 Smeekens SP Functional genomics identifies type I interferon pathway as central for host defense against Candida albicans Nat Commun 2013 4 1342 23299892 191 Spec A T cells from patients with Candida sepsis display a suppressive immunophenotype Crit Care 2016 20 15 26786705 192 Zanin-Zhorov A Cahalon L Tal G Margalit R Lider O Cohen IR Heat shock protein 60 enhances CD4+ CD25+ regulatory T cell function via innate TLR2 signaling J Clin Invest 2006 116 2022 2032 16767222 193 Scumpia PO Increased natural CD4+CD25+ regulatory T cells and their suppressor activity do not contribute to mortality in murine polymicrobial sepsis J Immunol 2006 177 7943 7949 17114466 194 Scumpia PO Treatment with GITR agonistic antibody corrects adaptive immune dysfunction in sepsis Blood 2007 110 3673 3681 17690255 195 Cavassani KA The post sepsis-induced expansion and enhanced function of regulatory T cells create an environment to potentiate tumor growth Blood 2010 115 4403 4411 20130237 196 Mauri C Bosma A Immune regulatory function of B cells Annu Rev Immunol 2012 30 221 241 22224776 197 Monserrat J Early alterations of B cells in patients with septic shock Crit Care 2013 17 R105 23721745 198 Kelly-Scumpia KM B cells enhance early innate immune responses during bacterial sepsis J Exp Med 2011 208 1673 1682 21746813 199 Rauch PJ Innate response activator B cells protect against microbial sepsis Science 2012 335 597 601 22245738 200 Weber GF Interleukin-3 amplifies acute inflammation and is a potential therapeutic target in sepsis Science 2015 347 1260 1265 25766237 201 Hotchkiss RS Sherwood ER Immunology. Getting sepsis therapy right Science 2015 347 1201 1202 25766219 202 Lieschke GJ Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization Blood 1994 84 1737 1746 7521686 203 Petit I G-CSF induces stem cell mobilization by decreasing bone marrow SDF-1 and up-regulating CXCR4 Nat Immunol 2002 3 687 694 12068293 204 Nelson S A randomized controlled trial of filgrastim as an adjunct to antibiotics for treatment of hospitalized patients with community-acquired pneumonia. CAP Study Group J Infect Dis 1998 178 1075 1080 9806037 205 Root RK Multicenter, double-blind, placebo-controlled study of the use of filgrastim in patients hospitalized with pneumonia and severe sepsis Crit Care Med 2003 31 367 373 12576938 206 Francisco-Cruz A Granulocyte-macrophage colony-stimulating factor: not just another haematopoietic growth factor Med Oncol 2014 31 774 24264600 207 Paine R 3rd A randomized trial of recombinant human granulocyte-macrophage colony stimulating factor for patients with acute lung injury Crit Care Med 2012 40 90 97 21926600 208 Meisel C Granulocyte-macrophage colony-stimulating factor to reverse sepsis-associated immunosuppression: a double-blind, randomized, placebo-controlled multicenter trial Am J Respir Crit Care Med 2009 180 640 648 19590022 209 Hall MW Immunoparalysis and nosocomial infection in children with multiple organ dysfunction syndrome Intensive Care Med 2011 37 525 532 21153402 210 Bo L Wang F Zhu J Li J Deng X Granulocyte-colony stimulating factor (G-CSF) and granulocyte-macrophage colony stimulating factor (GM-CSF) for sepsis: a meta-analysis Crit Care 2011 15 R58 21310070 211 Mathias B Szpila BE Moore FA Efron PA Moldawer LL A Review of GM-CSF Therapy in Sepsis Medicine (Baltimore) 2015 94 e2044 26683913 212 Nalos M Santner-Nanan B Parnell G Tang B McLean AS Nanan R Immune effects of interferon gamma in persistent staphylococcal sepsis Am J Respir Crit Care Med 2012 185 110 112 22210794 213 Dries DJ Effect of interferon gamma on infection-related death in patients with severe injuries. A randomized, double-blind, placebo-controlled trial Arch Surg 1994 129 1031 1041 discussion 1042 7944932 214 Cheng SC Broad defects in the energy metabolism of leukocytes underlie immunoparalysis in sepsis Nat Immunol 2016 17 406 413 26950237 215 Chen L Flies DB Molecular mechanisms of T cell co-stimulation and co-inhibition Nat Rev Immunol 2013 13 227 242 23470321 216 Day CL PD-1 expression on HIV-specific T cells is associated with T-cell exhaustion and disease progression Nature 2006 443 350 354 16921384 217 Topalian SL Safety, activity, and immune correlates of anti-PD-1 antibody in cancer N Engl J Med 2012 366 2443 2454 22658127 218 Zhang Y Upregulation of programmed death-1 on T cells and programmed death ligand-1 on monocytes in septic shock patients Crit Care 2011 15 R70 21349174 219 Wang JF Up-regulation of programmed cell death 1 ligand 1 on neutrophils may be involved in sepsis-induced immunosuppression: an animal study and a prospective case-control study Anesthesiology 2015 122 852 863 25437496 220 Chang KC Blockade of the negative co-stimulatory molecules PD-1 and CTLA-4 improves survival in primary and secondary fungal sepsis Crit Care 2013 17 R85 23663657 221 Mackall CL Fry TJ Gress RE Harnessing the biology of IL-7 for therapeutic application Nat Rev Immunol 2011 11 330 342 21508983 222 Perales MA Recombinant human interleukin-7 (CYT107) promotes T-cell recovery after allogeneic stem cell transplantation Blood 2012 120 4882 4891 23012326 223 Demaret J STAT5 phosphorylation in T cell subsets from septic patients in response to recombinant human interleukin-7: a pilot study J Leukoc Biol 2015 97 791 796 25691382 224 Kinter AL The common gamma-chain cytokines IL-2, IL-7, IL-15, and IL-21 induce the expression of programmed death-1 and its ligands J Immunol 2008 181 6738 6746 18981091 225 Levy Y Effects of recombinant human interleukin 7 on T-cell recovery and thymic output in HIV-infected patients receiving antiretroviral therapy: results of a phase I/IIa randomized, placebo-controlled, multicenter study Clin Infect Dis 2012 55 291 300 22550117 226 Pelletier M Ratthe C Girard D Mechanisms involved in interleukin-15-induced suppression of human neutrophil apoptosis: role of the anti-apoptotic Mcl-1 protein and several kinases including Janus kinase-2, p38 mitogen-activated protein kinase and extracellular signal-regulated kinases-1/2 FEBS Lett 2002 532 164 170 12459483 227 Inoue S IL-15 prevents apoptosis, reverses innate and adaptive immune dysfunction, and improves survival in sepsis J Immunol 2010 184 1401 1409 20026737 228 Tracey KJ The inflammatory reflex Nature 2002 420 853 859 12490958 229 Tracey KJ Physiology and immunology of the cholinergic antiinflammatory pathway J Clin Invest 2007 117 289 296 17273548 230 Borovikova LV Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin Nature 2000 405 458 462 10839541 231 Wang H Cholinergic agonists inhibit HMGB1 release and improve survival in experimental sepsis Nat Med 2004 10 1216 1221 15502843 232 Ulloa L The vagus nerve and the nicotinic anti-inflammatory pathway Nat Rev Drug Discov 2005 4 673 684 16056392 233 Tracey KJ Reflex control of immunity Nat Rev Immunol 2009 9 418 428 19461672 234 Song XM The protective effect of the cholinergic anti-inflammatory pathway against septic shock in rats Shock 2008 30 468 472 18391858 235 van Westerloo DJ The cholinergic anti-inflammatory pathway regulates the host response during septic peritonitis J Infect Dis 2005 191 2138 2148 15898001 236 Rosas-Ballina M Tracey KJ Cholinergic control of inflammation J Intern Med 2009 265 663 679 19493060 237 Kox M Pickkers P Modulation of the Innate Immune Response through the Vagus Nerve Nephron 2015 131 79 84 26491975 238 Koopman FA Vagus nerve stimulation inhibits cytokine production and attenuates disease severity in rheumatoid arthritis Proc Natl Acad Sci U S A 2016 239 Gabanyi I Muller PA Feighery L Oliveira TY Costa-Pinto FA Mucida D Neuro-immune Interactions Drive Tissue Programming in Intestinal Macrophages Cell 2016 164 378 391 26777404 240 Tracey KJ Reflexes in Immunity Cell 2016 164 343 344 26824649 241 Kempner W The Nature of Leukemic Blood Cells as Determined by Their Metabolism J Clin Invest 1939 18 291 300 16694664 242 Newsholme P Curi R Gordon S Newsholme EA Metabolism of glucose, glutamine, long-chain fatty acids and ketone bodies by murine macrophages Biochem J 1986 239 121 125 3800971 243 Newsholme P Newsholme EA Rates of utilization of glucose, glutamine and oleate and formation of end-products by mouse peritoneal macrophages in culture Biochem J 1989 261 211 218 2775207 244 Oren R Farnham AE Saito K Milofsky E Karnovsky ML Metabolic patterns in three types of phagocytizing cells J Cell Biol 1963 17 487 501 13940299 245 Levene PA Meyer GM On the action of leucocytes on glucose Journal of Biological Chemistry 1912 12 265 273 246 Ardawi MS Newsholme EA Maximum activities of some enzymes of glycolysis, the tricarboxylic acid cycle and ketone-body and glutamine utilization pathways in lymphocytes of the rat Biochem J 1982 208 743 748 7165729 247 Newsholme EA Crabtree B Ardawi MS Glutamine metabolism in lymphocytes: its biochemical, physiological and clinical importance Q J Exp Physiol 1985 70 473 489 3909197 248 Vachharajani V Liu T McCall CE Epigenetic coordination of acute systemic inflammation: potential therapeutic targets Expert Rev Clin Immunol 2014 10 1141 1150 25088223 249 Palsson-McDermott EM O'Neill LA The Warburg effect then and now: from cancer to inflammatory diseases Bioessays 2013 35 965 973 24115022 250 Warburg O On the origin of cancer cells Science 1956 123 309 314 13298683 251 Lu H Forbes RA Verma A Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis J Biol Chem 2002 277 23111 23115 11943784 252 Sun Q Mammalian target of rapamycin up-regulation of pyruvate kinase isoenzyme type M2 is critical for aerobic glycolysis and tumor growth Proc Natl Acad Sci U S A 2011 108 4129 4134 21325052 253 Cheng SC mTOR- and HIF-1alpha-mediated aerobic glycolysis as metabolic basis for trained immunity Science 2014 345 1250684 25258083 254 Rodriguez-Prados JC Substrate fate in activated macrophages: a comparison between innate, classic, and alternative activation J Immunol 2010 185 605 614 20498354 255 Tannahill GM Succinate is an inflammatory signal that induces IL-1beta through HIF-1alpha Nature 2013 496 238 242 23535595 256 Freemerman AJ Metabolic reprogramming of macrophages: glucose transporter 1 (GLUT1)-mediated glucose metabolism drives a proinflammatory phenotype J Biol Chem 2014 289 7884 7896 24492615 257 Davenport EE Genomic landscape of the individual host response and outcomes in sepsis: a prospective cohort study Lancet Respir Med 2016 4 259 271 26917434 258 Elenkov IJ Wilder RL Chrousos GP Vizi ES The sympathetic nerve--an integrative interface between two supersystems: the brain and the immune system Pharmacol Rev 2000 52 595 638 11121511 259 Muthu K Thermal injury and sepsis modulates beta-adrenergic receptors and cAMP responses in monocyte-committed bone marrow cells J Neuroimmunol 2005 165 129 138 15955567 260 Cohen MJ Shankar R Stevenson J Fernandez R Gamelli RL Jones SB Bone marrow norepinephrine mediates development of functionally different macrophages after thermal injury and sepsis Ann Surg 2004 240 132 141 15213629 261 Rudiger A Singer M Mechanisms of sepsis-induced cardiac dysfunction Crit Care Med 2007 35 1599 1608 17452940 262 Rudiger A Beta-block the septic heart Crit Care Med 2010 38 S608 612 21164404 263 Evans DG Miles AA Niven JS The enhancement of bacterial infections by adrenaline Br J Exp Pathol 1948 29 20 39 18865102 264 Lyte M Stimulation of Staphylococcus epidermidis growth and biofilm formation by catecholamine inotropes Lancet 2003 361 130 135 12531580 265 Freestone PP Haigh RD Lyte M Specificity of catecholamine-induced growth in Escherichia coli O157:H7, Salmonella enterica and Yersinia enterocolitica FEMS Microbiol Lett 2007 269 221 228 17229058 266 Freestone PP Lyte M Neal CP Maggs AF Haigh RD Williams PH The mammalian neuroendocrine hormone norepinephrine supplies iron for bacterial growth in the presence of transferrin or lactoferrin J Bacteriol 2000 182 6091 6098 11029429 267 Ohtsuka T Effect of beta-blockers on circulating levels of inflammatory and anti-inflammatory cytokines in patients with dilated cardiomyopathy J Am Coll Cardiol 2001 37 412 417 11216955 268 Friese RS Barber R McBride D Bender J Gentilello LM Could Beta blockade improve outcome after injury by modulating inflammatory profiles? J Trauma 2008 64 1061 1068 18404076 269 Tian X Effects of cAMP and beta-adrenergic receptor antagonists on the function of peripheral T helper lymphocytes in patients with heart failure Neuroimmunomodulation 2011 18 73 78 20938210 270 Bergmann M Attenuation of catecholamine-induced immunosuppression in whole blood from patients with sepsis Shock 1999 12 421 427 10588509 271 Christensen S Preadmission beta-blocker use and 30-day mortality among patients in intensive care: a cohort study Crit Care 2011 15 R87 21385356 272 Herndon DN Long-term propranolol use in severely burned pediatric patients: a randomized controlled study Ann Surg 2012 256 402 411 22895351 273 Arbabi S Beta-blocker use is associated with improved outcomes in adult trauma patients J Trauma 2007 62 56 61 discussion 61–52 17215733 274 Hart DW Anabolic effects of oxandrolone after severe burn Ann Surg 2001 233 556 564 11303139 275 Demling RH Orgill DP The anticatabolic and wound healing effects of the testosterone analog oxandrolone after severe burn injury J Crit Care 2000 15 12 17 10757193 276 Wolf SE Effects of oxandrolone on outcome measures in the severely burned: a multicenter prospective randomized double-blind trial J Burn Care Res 2006 27 131 139 discussion 140–131 16566555 277 Delano MJ Moldawer LL The origins of cachexia in acute and chronic inflammatory diseases Nutr Clin Pract 2006 21 68 81 16439772 278 Haney M Dronabinol and marijuana in HIV-positive marijuana smokers. Caloric intake, mood, and sleep J Acquir Immune Defic Syndr 2007 45 545 554 17589370 279 Li YY Involvement of cannabinoid-1 and cannabinoid-2 receptors in septic ileus Neurogastroenterol Motil 2010 22 350 e388 19840270 280 Gallily R Protection against septic shock and suppression of tumor necrosis factor alpha and nitric oxide production by dexanabinol (HU-211), a nonpsychotropic cannabinoid J Pharmacol Exp Ther 1997 283 918 924 9353414 281 Pierrakos C Vincent JL Sepsis biomarkers: a review Crit Care 2010 14 R15 20144219 282 Monneret G Venet F Monocyte HLA-DR in sepsis: shall we stop following the flow? Crit Care 2014 18 102 24393356 283 Monneret G Venet F Sepsis-induced immune alterations monitoring by flow cytometry as a promising tool for individualized therapy Cytometry B Clin Cytom 2015 284 Shao R Fang Y Yu H Zhao L Jiang Z Li CS Monocyte programmed death ligand-1 expression after 3–4 days of sepsis is associated with risk stratification and mortality in septic patients: a prospective cohort study Crit Care 2016 20 124 27156867 285 Sironi M Cagliani R Forni D Clerici M Evolutionary insights into host-pathogen interactions from mammalian sequence data Nat Rev Genet 2015 16 224 236 25783448 286 Opal SM Dellinger RP Vincent JL Masur H Angus DC The next generation of sepsis clinical trial designs: what is next after the demise of recombinant human activated protein C?* Crit Care Med 2014 42 1714 1721 24717456 287 Coffey CS Kairalla JA Adaptive clinical trials: progress and challenges Drugs in R&D 2008 9 229 242 18588354 288 Press WH Bandit solutions provide unified ethical models for randomized clinical trials and comparative effectiveness research Proceedings of the National Academy of Sciences of the United States of America 2009 106 22387 22392 20018711 289 Petroni GR Wages NA Paux G Dubois F Implementation of adaptive methods in early-phase clinical trials Statistics in medicine 2016

PMC005xxxxxx/PMC5111811.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101656080 43733 Cell Syst Cell Syst Cell systems 2405-4712 2405-4720 27720632 5111811 10.1016/j.cels.2016.09.003 NIHMS816984 Article Extended twilight among isogenic C. elegans causes a disproportionate scaling between lifespan and health Zhang William B. 12 Sinha Drew B. 123 Pittman William E. 123 Hvatum Erik 12 Stroustrup Nicholas 4 Pincus Zachary 12* 1 Department of Genetics, Washington University in St. Louis, St. Louis, MO 63110, USA 2 Department of Developmental Biology, Washington University in St. Louis, St. Louis, MO 63110, USA 3 Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, MO 63130, USA 4 Department of Systems Biology, Harvard Medical School, Boston, MA 02115, USA * Corresponding Author and Lead Contact: 660 South Euclid Avenue, Department of Genetics, CB 8232, St. Louis, MO 63110-1031, zpincus@wustl.edu 23 9 2016 6 10 2016 26 10 2016 26 10 2017 3 4 333345.e4 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Although many genetic factors and lifestyle interventions are known to affect the mean lifespan of animal populations, the physiological variation displayed by individuals across their lifespans remains largely uncharacterized. Here, we use a custom culture apparatus to continuously monitor five aspects of aging physiology across hundreds of isolated Caenorhabditis elegans individuals kept in a constant environment from hatching until death. Aggregating these measurements into an overall estimate of senescence, we find two chief differences between longer- and shorter-lived individuals. First, though long- and short-lived individuals are physiologically equivalent in early adulthood, longer-lived individuals experience a lower rate of physiological decline throughout life. Second, and counter-intuitively, long-lived individuals have a disproportionately extended “twilight” period of low physiological function. While longer-lived individuals experience more overall days of good health, their proportion of good to bad health, and thus their average quality of life, is systematically lower than that of shorter-lived individuals. We conclude that within a homogeneous population reared under constant conditions, the period of early-life good health is comparatively uniform and the most plastic period in the aging process is end-of-life senescence. eTOC Zhang et al. use a novel culture system to study inter-individual variation in aging in C. elegans. They uncover systematic, qualitative differences in the aging process of short- vs. long-lived individuals within a wild-type population. Introduction Pioneering work over the last quarter century has identified many molecular pathways involved in determining lifespan and illustrated that aging is a plastic process (Kenyon, 2010, 2005; Guarente and Kenyon, 2000). Many genes, small molecules, and environmental interventions have been found that alter a population’s mean lifespan. However, even in very homogeneous conditions, there is a large degree of variability in the lifespan of individuals around that population’s mean (Vaupel et al., 1998). Comparatively little is known, however, about the origins and consequences of inter-individual differences in the aging process. In general, the bulk of variability in lifespan is of not of genetic origin, and persists even in homogeneous environmental conditions. Studies estimate that only 15–25% of variation in human lifespan is attributable to genetic variation (Christensen et al., 2006; Gögele et al., 2011; Pettay et al., 2005; Herskind et al., 1996). Moreover, genetically identical populations of model organisms, reared in tightly controlled laboratory conditions have a similar degree of variability in lifespan as outbred humans (relative to the population mean) (Kirkwood et al., 2005; Vaupel et al., 1998). Even conditions and mutations that dramatically extend or shorten lifespan do not generally change the degree of inter-individual variation in longevity (Stroustrup et al., 2016). This raises a fundamental question: what distinguishes long-lived from short-lived individuals, when they are genetically identical and reared in the same cage, vial, or culture dish? Previous studies of single individuals have identified inter-individual differences in gene expression that are predictive of future lifespan in C. elegans, either early (Pincus et al., 2011; Golden et al., 2008) or late (Sánchez-Blanco and Kim, 2011; Golden et al., 2008) in adulthood. However, chronological lifespan merely measures the end of a complex process of aging and senescence. Pioneering studies of individual animals have therefore sought correlates of early and mid-life physiological health as well (Glenn et al., 2004; Chow et al., 2006; Golden et al., 2008; Eckley et al., 2013). However, due to technical limitations and the use of invasive measurements, many of these efforts were unable to follow an individual’s health longitudinally through time. Therefore, the relationship between the complex progression of aging and ultimate lifespan has been relatively unstudied. In this work, we focus on the physiological changes that characterize senescent decline in individual C. elegans (Klass, 1977; Hosono et al., 1980; Herndon et al., 2002; Huang et al., 2004). In particular, we determined how the physiological process of aging differs between long- and short-lived wild-type C. elegans. Do individuals with different lifespans senesce in a qualitatively different manner? Or does an identical aging process simply play out at a different rate for short- and long-lived individuals? In order to address these questions, we developed an experimental technique to isolate and image many individual C. elegans over each animal’s entire lifespan. This is not possible with existing automated vermiculture methods, which generally focus on enumeration of lifespans for screening purposes (Gill et al., 2003; Hertweck and Baumeister, 2005; Stroustrup et al., 2013), and are not designed for making detailed measurements of physiology. Here, using custom microscopy hardware and image-analysis software, we performed a battery of non-invasive physiological measurements of each individual, every three hours over their approximately two-week lifespans. This dataset of longitudinal measures of 734 animals, based on over 400,000 images obtained at subcellular resolution, allowed us to retrospectively understand how and when the aging process diverges between long- and short-lived individuals. Results Longitudinal Measurements of Physiology in Individual C. elegans In order to measure physiological changes throughout aging, we developed a specialized culture system to allow for observation of many freely moving, isolated individuals throughout their entire lives. This new system improves on our previous culture techniques (Pincus et al., 2011), allowing for denser culture (~100 animals per standard microscope slide) and fully automated image acquisition and processing. In brief, we produce a smooth hydrogel surface atop standard microscope slides by polymerizing polyethylene glycol (PEG) monomers in situ (see Methods and Resources). We then dispense a grid of ~2 mm diameter droplets of concentrated bacteria onto the gel, to provide a food source. After allowing each droplet to adsorb to the surface, we transfer one egg from the temperature-sensitive sterile strain spe-9(hc88) to each food pad. Finally, we pour a thin layer of polydimethylsiloxane (PDMS) over the gel. While polymerizing overnight at room temperature, the PDMS also cross-links with unreacted acrylate moieties in the PEG gel. This cross-linking proceeds everywhere except where the PEG is separated from the PDMS by the bacterial food pads. Thus while each individual C. elegans is free to move about the 2-dimensional surface of its food pad, it is constrained at all borders by strong covalent bonds within and between the PEG and PDMS polymers (Figure 1A). We placed slides constructed in this fashion on the stage of a computer-controlled microscope, housed in a temperature- and humidity-controlled enclosure. Every three hours, timepoint data for each individual was acquired autonomously, driven by custom microscope-control scripts (Figure 1B). We designed a comprehensive panel of phenotypic measurements encompassing five diverse aspects of aging physiology. For this, we drew from previously validated “biomarkers of aging” (Baker and Sprott, 1988; Pincus and Slack, 2010). These biomarkers measure aspects of physiology that differ prospectively between long- and short-lived individuals, and can thus be used to predict an individual’s future lifespan. First, we examined C. elegans neuromuscular function, as measured by locomotory ability. Second, we assessed age-associated tissue deterioration through quantitative measurements of textural order and disorder in brightfield images (Johnston et al., 2008; Pincus et al., 2011). Third, we measured age-related declines in homeostatic ability, as manifested in the accumulation of fluorescent non-hydrolyzable materials in intestinal endosomes (Klass, 1977; Clokey and Jacobson, 1986; Gerstbrein et al., 2005; Hermann et al., 2005; Pincus et al., 2016). Fourth, we evaluated nutritional history and somatic investment through cross-sectional body size (Pincus et al., 2011; Hulme et al., 2010). Fifth, we quantified reproductive output and investment through the number of oocytes laid (Huang et al., 2004; Pickett et al., 2013). These measurements were all performed via custom, fully automated image segmentation and analysis software (Figure 1C and D; see Methods and Resources). Taken together, this panel comprises a diverse set of individual measurements, allowing us to characterize longitudinal changes in distinct aspects of aging physiology. Distinct Physiology of Long- vs. Short-Lived Individuals In total, we observed 734 individuals from hatching to death, through the stages of larval development, reproductive maturity, and senescence. The population, maintained throughout life at 25°C, had a mean lifespan of 12.1 days with a stand ard deviation of 2.3 days from hatch, within the range of lifespans previously observed at this temperature (Byerly et al., 1976; Fabian and Johnson, 1994; Pincus et al., 2011). In order to rule out potential systematic biases in our data, we confirmed that overall lifespan distributions are consistent across experimental runs spanning roughly three months of time (Figure S1A), and are not influenced by the specific culture slide used within an experimental run, or by the spatial position of an individual’s food pad within the slide (Figure S1B–G). We find that virtually all variation in lifespan is due to differences in the period between reproductive maturity and death, rather than in the length of larval development (Figure S1H–J). Though larval development takes 2.1 days on average (17.3% of an average animal’s lifespan), the variability of time in development accounts for less than 0.1% of the variability in total lifespan. Therefore, we limited our analysis to the animals’ lifespans after reproductive maturity. The mean adult lifespan in our population is 10.0 days, ranging from 2.2–15.5 days. For illustration, we grouped the animals into seven cohorts by lifespan, with the first cohort containing individuals with adult lifespans of 2–4 days, the second containing individuals with adult lifespans of 4–6 days, and so forth (Figures 2A and B, S2), and plotted cohort averages of each of our physiological measurements over time (Figure 2C). For our measures of autofluorescence, body texture, and long-term movement, we observed a graded difference between shorter-lived and longer-lived cohorts (Figure 2D–F). At any given age, the shorter-lived cohorts had higher levels of autofluoresence, worse tissue maintenance, and were less mobile than their longer-lived peers. In contrast, for our measures of body size and reproductive output, we noticed a nonlinear trend among our lifespan cohorts (Figure 2G and H). For animals with adult lifespans from 2 to 8 days, larger body size and greater reproductive output were correlated with longer life, as in the existing literature (Huang et al., 2004; Hulme et al., 2010; Pincus et al., 2011). Among the longest-lived cohorts, however, this trend reverses itself: of individuals with lifespans between 10 and 16 days, longer-lived individuals tend to be smaller and produce fewer oocytes. We therefore examined this nonlinear relationship between longevity and certain aspects of physiology more closely. Overall, this effect is largely driven by a small subpopulation (13.7% of the total) of small, sickly-looking individuals that appear unhealthy throughout their lives but have very long lifespans (Figure S3A–G). We have also found evidence of this population in standard C. elegans culture conditions (Figure S3D). Overall, however, the existence of this sub-population suggests that there may be a component of functional health that can be de-coupled from lifespan. In particular, it appears that some longer-lived individuals may be qualitatively less healthy over the course of their lives compared to shorter-lived peers. Regardless, excluding this population does not substantially alter the results described below (Figure S4A–H). Differences in Aging Rate and Health at Death in Long- vs. Short-Lived Individuals To understand how aging differs in long- and short-lived individuals, we developed a physiological measure of overall senescence. Since the main hallmark of aging is increased mortality over time, more senescent (i.e. less healthy) individuals have shorter expected future lifespans. As each of the diverse physiological parameters we measured is a known biomarker of aging – that is, a predictor of future lifespan – we therefore aggregated these measures into a single, maximally informative estimate of future lifespan. We use this estimate, which we term prognosis, as a measure of an individual’s degree of senescence, or, equivalently, its state of health. To construct this prognosis, we first verified that in our dataset, each of our measures is a bona fide biomarker of aging and of mortality (Table S3). The trends over time of the raw measurements also indicate that the relationship between each parameter and future lifespan is generally nonlinear (Figure 2D–H). Therefore, we used support vector regression to define a nonlinear mapping from an individual’s measured physiological parameters at any given timepoint to an estimate of its future lifespan at that time. The overall r2 for this regression is 0.695 (10-fold cross-validated r2 = 0.669), suggesting that throughout life, these measures are able to explain the bulk of the total variability in future lifespan. While some of the measurements correlate with one another and thus redundantly measure certain aspects of physiology (Table S1), each contributes a measurable amount of independent, nonredundant information about future lifespan (Tables S2 and S3). This indicates the physiological parameters as a set, and thus our prognosis score, report on multiple distinct aspects of the aging process. We also verified that this prognosis score is neither driven by our choice of regression methodology, nor by any particular physiological measurement (these analyses are described in the STAR Methods section). We next use this prognosis of future lifespan to distinguish among several possibilities for how long- and short-lived individuals differ in physiological aging. First, it could be that short-lived individuals simply start their adulthoods in worse health (the “starting point hypothesis”; Figure 3A). Second, it is possible that senescence proceeds more rapidly in short-lived individuals (the “rate of aging hypothesis”; Figure 3B). Finally, it may be the case there is no difference in the trajectory of senescent decline between long- and short-lived individuals, and all individuals start out with similar prognoses and decline at the same rate. Here, shorter-lived animals will be those that die “prematurely”, in more healthy-appearing states, before the full process of senescence has played out (the “premature death hypothesis”; Figure 3C). In this latter case, all individuals would be, by our measurements, indistinguishable over the course of aging, and death would be stochastic and unpredictable. Short-lived individuals would differ from long-lived ones by simply happening to be the ones that died early. The trajectories of our health score look very similar to those predicted by the “rate of aging” and “premature death” hypotheses, even from a cursory visual inspection of Figure 3D. To test this more rigorously, we examined the relationships between lifespan and an individual’s prognosis at the first day of adulthood (i.e. its starting health), its rate of decline in prognosis (its rate of aging, or senescence), and its prognosis at the time of death (how “premature” its demise appears to have been, based on its health immediately prior to death). As shown in Figure 3E and F, there is no substantial relationship between starting health and lifespan, but a strong correlation between the rate of decline (as measured by simply subtracting an individual’s ending health from its starting health and dividing by its lifespan) and lifespan (Pearson r2 = 0.676; p < 10−180, F-test; Spearman ρ2 = 0.615; p < 10−153, F-test) and a modest one between prematurity of death and lifespan (Pearson r2 = 0.096; p < 10−30, F-test; Spearman ρ2 = 0.123; p < 10−16, F-test). (Quadratic fits were used to estimate r2 values due to the clear nonlinearity of these two relationships.) This latter trend indicates a modest but significant contribution either from stochastic death, or from differences in health not captured by our measurements. Overall, we conclude that long-lived individuals are not physiologically different from short-lived individuals at the start of reproductive maturity, but age more slowly throughout adulthood and experience the full scope of senescent decline. Uneven Rates of Aging Produce Lower Quality of Life in Longer-Lived Individuals Next, we asked whether long and short-lived individuals exhibit qualitatively different aging processes, in addition to having quantitatively different rates of aging. The black line in Figure 4A presents a “neutral rate of aging” for both long- and short-lived individuals: the level of health decreases uniformly throughout life. However, an individual might have a positive deviation (green line) from the neutral, straight-line decline, such that the individual appears to maintain a high level of function until a precipitous decline at the very end of its life – a phenomenon known as “morbidity compression” (Fries, 1980). Alternately, an individual may have a negative deviation (red line) from the neutral decline, and will thus senesce relatively early in its life, and persist in an extended “twilight” period of low physiological function for a larger fraction of its life. To visualize these qualitatively different classes of functional declines independently of their quantitative rate, we represent physiological change not as a function of chronological time, but as a function of the relative fraction of lifespan elapsed. If the process of aging is identical between long- and short-lived individuals except for differences in rate, then the trajectory of physiological decline of cohorts with different lifespans should fully overlap when rescaled to “relative time” (Figure 4C, left). Alternately, long-lived individuals may have systematically more positive deviations, and thus will remain healthy for a greater proportion of their lives (Figure 4C, center). This would lead long-lived animals to experience a higher average quality of life, as measured by the area under the health vs. relative-lifespan curve for long-lived vs. short-lived individuals. Finally, the opposite may be true: long-lived individuals may exhibit declines in health relatively earlier in life, leading to a greater proportion of life spent in poor health (Figure 4C, right). In this scenario, due to an extended period of poor physiological function, longer-lived individuals would have an overall lower average quality of life than shorter-lived individuals. Figure 4D shows that this latter hypothesis is most accurate, and that long-lived individuals experience a lower average quality of life compared to their short-lived counterparts. This is visible not only in the cohort averages of Figure 4D but in the clear negative relationship between each individual’s lifespan and the deviation from a neutral, straight-line decline from the population mean starting prognosis to the mean prognosis at death (Figure 4E; Pearson r2 = 0.207; p < 10−37, F-test; Spearman ρ2 = 0.183; p < 10−33, F-test). Longer-lived animals typically have negative deviations (i.e. they decline in health relatively early in life), while shorter-lived animals have more positive deviations and decline in health relatively late in life. To control for shorter-lived individuals dying in healthier states, we repeated the analysis with each animal’s health set to 1 at the onset of reproductive maturity and to 0 at the time of death (Figure 4F and G). Even when adjusted for each individual animal’s starting and ending health, the association between long lifespan and negative deviation remains (Pearson r2 = 0.123; p < 10−21, F-test; Spearman ρ2 = 0.106; p < 10−18, F-test). Taken together, these results demonstrate that longer-lived individuals have systematically worse qualitative physiological declines, despite the fact that longer-lived individuals typically have a better health prognosis on any chronological day of life (Figure 3D). The extended end-of-life period of low function in long-lived animals simply drags the overall average quality of life below that of short-lived animals. Longer-Lived Individuals Have Disproportionately Extended Gerospans The qualitatively worse senescent declines of long-lived individuals can be observed more starkly by partitioning each individual’s life into two segments: “healthspan”, the period of high physiological function, and “gerospan”, the period of low physiological function. Given a threshold prognosis score, we define healthspan as the period of time from the beginning of adulthood until an individual’s prognosis dips below that threshold for the first time. Gerospan is then the remainder of life thereafter (Figure 5A and B). In order to make a neutral comparison, we selected a threshold value that yields identical average healthspans and gerospans across the population: 5.0 days each. (Similar analysis with different thresholds yields essentially identical results to the below; see Figure S5A–P.) Figure 5A and C show that, as expected, longer-lived individuals enjoy longer healthspans in absolute chronological time. However, the differences in healthspans among long- and short-lived cohorts are small compared to the differences in gerospans. In contrast, we observe that despite this long chronological healthspan, longer-lived individuals nevertheless experience a systematically larger fraction of their total lives in senescent gerospan (Figure 5B and D). This is analogous to the results of our analysis of the full trajectories of senescence, above: long-lived individuals enjoy a better prognosis at any point in absolute chronological time, but experience declines in health relatively early in their lives. In order to validate this key result and ensure it is not an artifact of our culture system or analysis methods, we turned to data from the “Lifespan Machine”, which measures the movement individual C. elegans on standard culture plates in order to tally lifespans in an automated fashion (Stroustrup et al., 2013). The Lifespan Machine tracks individual animals once they are no longer able to move more than a few hundred microns in any direction; therefore this system inherently measures a “fast moving span” and a “slow moving” span for each individual. We manually validated these span annotations from an extant dataset with a similar genetic background and culture conditions, and computed the “slow moving span” (i.e. gerospan) as a fraction of total lifespan for each individual. As shown in Figure 5E, longer-lived individuals in these conditions experience a larger portion of life in senescent, slow-moving states. If we define the slow-moving and fast-moving spans similarly in our present dataset, we observe a very similar trend in (Figure 5F). This analysis provides an intuitive understanding of why longer-lived individuals spend more of their lives in gerospan. Figure 5G and H plot the relationship between total lifespan and healthspan or gerospan, respectively. Longer-lived individuals are both likelier to have longer healthspans and longer gerospans. However, while inter-individual differences in healthspan account for about 30% of the total variability in lifespan, differences in gerospan account for about 67% of variability in lifespan. As shown in Figure 5I, though we selected a threshold that yields equal mean healthspan and gerospan, there is simply more inter-individual variability in gerospan than healthspan. Indeed, this relationship holds for several choices of threshold (Figure S5Q and R) and our full deviation analysis, which is not subject to any arbitrary threshold. In sum, variability in lifespan is in large part driven by differences in individual gerospans, and much less so by differences in healthspans. Taken together, this indicates that gerospan may be inherently more flexible than healthspan. Discussion In this work, we examined the substantial variation in lifespan and aging physiology that exists even in isogenic individuals reared in identical environments. We developed a culture method that, to the best of our knowledge, allows for the first time high-resolution imaging of a large number of individual C. elegans throughout life. This enabled us to analyze many different aspects of each individual animal’s physiology in longitudinal fashion, from hatching until death. Motivated by similar efforts to define and characterize a “biological age” (Borkan and Norris, 1980; Baker and Sprott, 1988) or “frailty score” (Fried et al., 2001; Hubbard, 2015) in humans, we aggregated these measurements into a prognostic estimate of days of life remaining. Using this prognosis, we were able to determine how the progress of senescence differs between longer- and shorter-lived individuals. As one might expect for a genetically identical population reared in homogenous conditions, long-lived C. elegans do not begin adulthood with healthier physiology according to any of our measures. This is consistent with our previous work, which found that while physiological measurements made on the third and fourth day of adulthood can readily distinguish long- from short-lived individuals, physiological measures earlier in life cannot (Pincus et al., 2011). However, several fluorescent gene-expression reporters can be used to predict future lifespan even at the onset of adulthood: the level of the microRNA mir-71, which regulates insulin signaling (Pincus et al., 2011), and the expression, after a heat shock, of the heat-shock response protein hsp-16.2 (Rea et al., 2005). Thus, while long- and short-lived animals are not physiologically distinguishable at the start of adulthood, gene-regulatory and biochemical differences between individuals with different future fates are already established, and only later manifest themselves as phenotypic differences. Indeed, we find that the population does not remain physiologically identical for long. Systematic differences between the longest- and shortest-lived individuals are detectable less than one day after reproductive maturity. Overall, shorter-lived animals age more rapidly throughout adulthood. That is, short-lived individuals are generally in worse physiological health than their long-lived counterparts at any particular chronological age. In addition, short-lived individuals typically die before experiencing the most advanced stages of senescence. However, senescence typically occurs relatively late in life in shorter-lived individuals, which thus remain healthy-appearing for a large fraction of their lives. In contrast, longer-lived animals undergo the full range of senescent decline, from good to ill health. These declines typically begin relatively earlier in life, leading long-lived animals to die after an extended twilight period of low physiological function (Figure 6). Moreover, we identified a previously uncharacterized subpopulation of individuals that demonstrate these trends in the extreme. Small in size, with very poor oocyte production, and qualitatively and quantitatively ill appearing, these individuals are, however, extremely long-lived. While the mechanisms underlying this expansion in fractional gerospan in long-lived individuals are unclear, it is possible that they may result from diminished homeostatic capacity in older animals. One possibility is that phenotypically similar individuals entering gerospan may nevertheless have different degrees of homeostatic capacity (also known as “organ reserve” (Montgomery, 2000; Ghezzi and Ship, 2003) or “resilience” (McEwen, 2003; Stroustrup et al., 2016). Such latent, currently unobservable variability could allow some individuals to persist longer than others in highly senescent states. Alternately, a uniformly diminished homeostatic reserve across all individuals in gerospan might leave individuals susceptible to lethal stochastic insults (e.g. free-radical damage or protein translation errors) that would be survivable by a healthier individual. In this scenario, the variability in gerospan would be driven by the rate at which these insults occur. Therefore, determining the mechanistic underpinnings of these systematic trends in aging physiology is an important direction for future study. Our results highlight how the physiological determinants of lifespan and of health need not completely overlap. As shown in Figure 3, some individuals die in outward good health, while others spend unexpectedly long in a twilight of ill health before death. And though we observed distinct phenotypes of late-life senescence – gonadal hypertrophy, large collections of clear fluid, “wrinkling” of the cuticle due to shrinkage, and distended intestines packed with bacteria (Figure S3H–K) – we did not observe significant differences in lifespan among individuals with these different phenotypes (Figure S3L and M). Overall, individuals exhibiting ostensibly similar phenotypic health can have very different lifespans, while at the same time individuals with seemingly very different senescent pathologies can have very similar lifespans. These results provide additional context for recent work on physiologic determinants of lifespan in C. elegans (Stroustrup et al., 2016). Stroustrup and colleagues found that diverse interventions, including oxidative stress, changes in body temperature, and lifespan-extending mutations, do not change the overall shape of the population survival curve, but instead scale it uniformly in time. The data presented here show that the health declines of long- vs. short-lived individuals within a population do have qualitatively different shapes, however (Figure 4), and thus do not reflect a simple rescaling of time. Perhaps the decoupling of the health measures we observe from lifespan, and the similarity in lifespan of animals dying with different senescent pathologies, may be an important element in explaining how diverse interventions all produce similar effects on lifespan distributions. Our data also offer a new perspective on recent studies of functional declines in health in long-lived mutants. Bansal et al. measured the population average of several tests of stress resistance and physiological function in wild-type and long-lived mutant C. elegans, defining “healthspan” for any particular test as the period of time with greater than 50% of the maximal wild-type functional capacity (Bansal et al., 2015). These investigators found that while some long-lived mutants had longer healthspans in absolute time on certain functional tests, the relative fraction of life spent in good health was diminished in many long-lived mutant strains. (A notable exception is the insulin-signaling deficient mutant daf-2; see also (Hahm et al., 2015).) This overall result has been taken by some to suggest that these long-lived mutants are not good models for wild-type aging, and that the observed extension of relative time spent in poor life may have been a pleiotropic effect of the lifespan-extending genetic mutations. However, we show here that this is not necessarily the case. Our study, which incorporates comprehensive, lifelong measurements and an integrated definition of overall health (both suggested to be critical factors for such an analysis (Melov, 2016)), shows similar results in a genetically identical, wild-type population. Specifically, we observe a continuum in which the fraction of lifespan spent in good health systematically diminishes from short- to long-lived individuals. The results of Bansal et al. can thus be seen as placing many longevity mutants along this same continuum. As such, a relative expansion in gerospan may not be a “bug” specific to certain long-lived mutants, but rather a general property of the aging process itself. Together, these results demonstrate that extended lifespan, whether induced by stochastic events (in the case of inter-individual variability) or mutations (in the case of inter-strain differences), is often due to a disproportionate extension of a highly senescent “twilight period” of low physiological function. Put simply, it appears that an individual’s gerospan is inherently more plastic than its healthspan. Determining whether these results hold in more complex organisms is an important task. First, it is certainly true that in humans and other mammals, longer-lived individuals typically enjoy more total days of healthy life. Likewise, it is well established that individuals with higher physiological function and better physical fitness generally live longer. In particular, several remarkable studies in humans have shown that heritable factors that favor exceptional longevity also increase healthspan (Sebastiani et al., 2013; Ash et al., 2015). Further, longevity is more generally associated with a longer chronological span of healthy life, both within a relatively homogeneous Ashkenazi Jewish population (Ismail et al., 2016), and among the ethnically diverse general Chinese population (Gu et al., 2009). Our results in C. elegans are identical: longer-lived C. elegans enjoy longer absolute healthspans (Figure 4C), and better health at any point in time is correlated with extended future lifespan (Figure 4A). It is an open question, however, whether longer-lived individuals within mammalian populations spend a smaller fraction of their lives in good health, as we observe in C. elegans (Figure 4D). As above, several studies in humans have demonstrated in different contexts that extended lifespan is associated with extended chronological healthspan. Nevertheless, it remains a matter of some debate whether longer life brings with it a proportionate extension in healthspan (Rechel et al., 2013; Ash et al., 2015). As most of the current studies do not include data on ultimate lifespan, either due to cross-sectional study design or ongoing follow-up in a longitudinal study, these analyses cannot yet shed light on the proportion of life spent in good health in short- vs. long-lived individuals. Some recent work has been able to address this question by asking whether modern advances in medicine have been more successful in extending disability-free life (healthspan) or activity-limited life (gerospan). There is evidence that, especially for females, medical advances have prolonged longevity primarily by expanding the late-life period of morbidity, without much effect on the span of disability-free life (Freedman et al., 2016). Fundamentally, however, whether it is worthwhile to extend lifespan in this fashion is a value judgment that may vary between individuals, and is thus more in the realm of clinical decision analysis (Kassirer, 1976; Weinstein and Stason, 1977; Lee et al., 2009) than experimental biology. (The simple analysis in Figure S5S–U shows how the desirability of such extensions can depend on the threshold for acceptable quality of life.) It is, however, an experimental question whether there are conditions, interventions, or mutations that can extend fractional healthspan as easily as fractional gerospan. In mammals, exercise (Garcia-Valles et al., 2013) and caloric restriction (Mattison et al., 2012) may extend the ratio of healthspan to gerospan. In C. elegans, the data from Bansal et al. and Hahm et al. both show that the long-lived daf-2 mutant enjoys a proportionate extension of both healthspan and gerospan in relevant physiological assays. Thus the daf-2 mutant likely departs from the relationship between long life and extended fractional gerospan that we have characterized. Our longitudinal approach now allows for a more detailed understanding of how health integrates across physiological systems and evolves over time in different genetic backgrounds, filling a previously unmet experimental need (Melov, 2016). Future studies with these approaches promise to unravel the complex relationship between the plasticity of different phases of senescence and lifespan extension in general. Methods and Resources Contact for Reagent and Resource Sharing Zachary Pincus is the Lead Contact and may be contacted at 660 South Euclid Avenue; Department of Genetics, CB 8232; St. Louis, MO 63110-1031 and at zpincus@wustl.edu. Experimental Model and Subject Details Strains The C. elegans strain spe-9(hc88), a temperature-sensitive fertilization-defective mutant, was provided by the Caenorhabditis Genetics Center (CGC). The strain was maintained at 15°C and assays were conducted at a restrictive temperature of 25°C. spe-9(hc88) strains are widely used as an alternative to 5-fluoro-2′-deoxyuridine (FUDR) sterilization, and have been validated to have wild-type lifespans at the restrictive temperature of 25.5°C (Fabian and Johnson, 1994) and wild-type brood sizes at the permissive temperatures of 16°C and 20 °C (Singson et al., 1998). We used spe-9(hc88) to avoid the known confounding effects of FUDR (Anderson et al., 2016) and to eliminate issues due to the timing of administration (such as being administered at different phases of life for fast- and slow-developing individuals on the same slide). Method Details Single-Animal Vermiculture Standard 25 mm × 75 mm (1.2 mm thick) glass microscopy slides (obtained from VWR International, LLC; Radnor, PA, USA) were used as the base support for our culture device. Before use, the glass slides were bonded to custom-machined aluminum frames (outer dimensions 25 mm × 75 mm; inner dimensions 20 mm × 70 mm; thickness 2.36 mm) using 120 μL of polydimethylsiloxane (PDMS; Sylgard 184 Silicone Elastomer Kit obtained from Dow Corning Corporation; Midland, MI, USA). The PDMS was then cured in an oven at 100°C f or 3 hours. The device was then cleaned with distilled water and ethanol, sealed in aluminum foil, and dry heat sterilized at 160°C for 2 hours. Next, a modified version of standard nematode growth media (NGM) (Brenner, 1974) was made by mixing 97.5 mL of distilled water with 0.3 g of sodium chloride, 0.25 g of peptone, 0.1 mL of 1M magnesium sulfate, and 2.5 mL of 1M potassium phosphate buffer (pH 6.3, titrated with sodium hydroxide and hydrochloric acid). We omit the calcium chloride used in standard NGM, and modify the pH of the potassium phosphate buffer from 6.0 to 6.3 to allow for polymerization of the polyethylene glycol (PEG) gel. Instead of autoclaving, which can darken sugar-containing solutions, we filter-sterilized the modified NGM to maximize optical clarity. Then, 8 μL/mL of cholesterol stock (5 mg/mL in 95% ethanol) was added to the NGM, which was then used to separately dissolve an 8-armed PEG-thiol (Jenkem Technology; Beijing, P. R. China; Item Number: 8ARM(TP)-SH-10K) and PEG-diacrylate (Sigma-Aldrich; St. Louis, MO, USA; Catalog Number: 455008 Aldrich) at 140 mg/mL and 40 mg/mL, respectively. The NGM-PEG-thiol and NGM-PEG-diacrylate solutions were then mixed in a 1:1 ratio and vortexed vigorously. 1.7 mL of the mixture was then added to the center of the aluminum frame on the culture device. The device was tilted to break the surface tension of the PEG mixture and to achieve an even layer of fluid. The device was then placed in a standard 100 mm × 15 mm petri dish. A delicate task wipe was cut in half and added to the petri dish, and 700 μL of distilled water was pippetted onto each half wipe in order to maintain the humidity of the petri dish and prevent excessive evaporation of water from the PEG gel. Finally, an additional 25 mm × 75 mm glass slide was sterilized with ethanol, dried, and placed on top of the aluminum frame to further isolate the gel during polymerization. The petri dish containing the device is then left at room temperature for 90 minutes, during which the 8-armed PEG-thiol crosslinks to the PEG-diacrylate via Michael addition. Next, an array of 0.2 μL droplets of an OP50 E. coli (50% by mass) food source was deposited onto the gel. Single, individual pretzel-stage eggs were picked into each droplet using an eyelash pick. Approximately 100 eggs can be picked onto each slide in 20 rows of 5 eggs each. After the E. coli droplets were dried, 1.2 mL of PDMS was deposited over the PEG gel using a syringe. The PDMS polymerizes overnight at 25°C. During polymerizatio n, each individual is trapped on its food pad by hydrosilylation chemistry between unreacted moieties in the PDMS cure agent and unreacted acrylate groups in the PEG-diacrylate. This reaction results in a covalent bond between the PDMS and the PEG gel, formed at all locations where the gel is in direct contact with the PDMS (i.e. everywhere except the bacterial food pads). Immediately after deposition of the PDMS, the glass slide was placed on the stage of our computer-controlled microscope, and automated image acquisition was initiated. Image Acquisition All images were acquired at 5× magnification using custom-built hardware and software. The microscope itself was housed in a climate-controlled enclosure to provide consistent temperature and humidity for the animals studied. A thermoelectric cooler with continuous temperature monitoring (Torrey Pines Scientific, La Jolla, CA) was used to maintain the temperature of the enclosure at 25.0±0.1°C. Humidity was maintained at 85±10% relative humidity by including a tub of distilled water in the enclosure and using two aquarium air diffusion stones along with an air pump to aerate the water. Consistent air circulation was achieved with the use of several 80 mm DC fans attached to the thermoelectric cooler and the walls of the enclosure. Custom in-house control software was developed and used to move to and automatically focus (Firestone et al., 1991; Brenner et al., 1976) on each animal every three hours. At each time point, a series of six brightfield and one fluorescence image was acquired, taking approximately 17 seconds per animal. The fluorescence image and one bright-field image were taken first, taking less than 60 ms combined time to acquire. The fluorescence image taken with a 50 ms exposure using a TRITC filter (Semrock part DA/FI/TR-3X-A-000), with light from a Lumencor Spectra X. The center wavelengths of the excitation and emission bands were 556 nm and 613 nm, respectively. Next, a series of 5 bright-field images to assess short-term movement were taken over 4.5 seconds, with 0.5 seconds between the first and second images, 1.0 seconds between the second and third images, during which time the animals were stimulated with a 0.5 second pulse of cyan light (Edwards et al., 2008), 1.5 seconds between the third and fourth images, and 1.5 seconds between the fourth and fifth images. For the first 10 timepoints, the 0.5 second pulse of cyan light and the fluorescence image were omitted to avoid excessive stimulation of the animals during early larval development. The microscope was automatically re-calibrated at every three-hour timepoint for spatial and temporal variation in light-source intensity for both the bright-field lamp and the fluorescence light source. In addition, images were corrected for variation in sensitivity of the camera image sensor. Finally, bright-field exposure times and lamp intensities were adjusted at every time point to optimize the dynamic range of the acquired images. Overall, all images were corrected for background camera noise (dark current), spatial illumination inhomogeneity (flat field) and temporal variation in illumination. After correction, image intensity values were divided by the exposure time to render all images comparable. Image Segmentation The acquired image series were then manually annotated for time of hatching, reproductive maturity as indicated by first oocyte laid, and death as indicated by total cessation of coordinated movement. Eight developmentally defective animals that never reached reproductive maturity were censored from our analysis, while 734 were included in the final analysis. All animals were observed for an additional 30 hours post-mortem to confirm death. A suite of custom image analysis software was written to automatically identify the location of the animal in the over 400,000 bright-field images. First, the rough location of the animal was determined using background subtraction (Piccardi, 2004), which takes advantage of the fact that the animal moves around in the image, while other objects such as the bacterial lawn are stationary. The current image was combined with the previous nine images taken for the same individual by computing a pixel-wise median, which produces a model of the background. This background image was then subtracted from the current image, and thresholded for only pixels above the 97.5th percentile of brightness. The largest contiguous object in the thresholded image was then identified as the animal, and all internal holes in the object were filled. This location was then masked out of future background contexts to improve the robustness of the background model. Finally, the algorithm was run backwards in time to identify the position of the animal during the first nine time points. Next, the location of the individual was refined using the Canny edge detector (Canny, 1986). While the background subtraction segmentation is sensitive to changes in the bacterial lawn surrounding the animal due to illumination changes or physical churning by the animal’s movement, edge-detection defines sharper borders based on local spatial intensity gradients. The largest object in the field of view (excluding the edge of illumination from vignetting and the border of the bacterial lawn itself) was identified as the worm, and all internal holes in the object were filled. However, edge detection can occasionally locate the wrong object due to a large pattern of churned bacteria or the introduction of a piece of dust or air bubble into the field of view. Therefore, the background subtraction-derived location was used as a fallback for cases in which the location of the animal as defined by edge detection was more than 50 pixels from the location as identified by the more reliable background subtraction. Finally, a support vector classifier (SVC) (Cortes and Vapnik, 1995) with a radial basis function kernel (γ = 4.88 × 10−4, ν = 8.99 × 10−4) was trained on the intensities from 10,000 5 × 5 pixel patches from 1,617 manually drawn outlines of worms to distinguish between the animals and their surrounding bacteria. This classifier was then applied to patches centered on each individual pixel within the animal’s location as computed by the previous two methods. Using the SVC allowed us to improve our image segmentation significantly in cases in which the animal was in a curled posture and looped back on itself. While the simple hole-filling would naïvely identify patches of bacteria surround by the animal as part of the animal itself, the SVC is generally able to distinguish between the two. Overall, the automated segmentation agreed strongly with manual segmentation, with Pearson correlation coefficients between automated size and manual size of r2 = 0.81 and between automated position and manual position of r2 > 0.95 (Figure S1K–O). Physiological Panel Design Several “biomarkers of aging” (Baker and Sprott, 1988; Pincus and Slack, 2010) – properties that can predict an individual’s future lifespan better than age alone – have previously been identified in C. elegans. As each biomarker represents an aspect of physiology that differs between long- and short-lived individuals at some point in life, we selected a number of these markers to follow longitudinally across our population. First, we examined C. elegans neuromuscular function, as measured by locomotory ability. Previous work has shown that locomotion in age-matched animals diminishes over time, correlates with remaining lifespan (Hosono et al., 1980; Herndon et al., 2002; Huang et al., 2004; Hsu et al., 2009; Hulme et al., 2010; Pincus et al., 2011) and with the degree of sarcopenia (Herndon et al., 2002). Second, we assay the accumulation of non-hydrolyzable autofluorescent material in intestinal endosomes (Klass, 1977; Clokey and Jacobson, 1986; Gerstbrein et al., 2005; Hermann et al., 2005; Pincus et al., 2016). This is a marker for declining macromolecular homeostatic capacity over time, as individuals fail to clear the fluorescent material at a pace that balances its production. We chose to follow red-wavelength autofluorescence, which is most predictive of future lifespan (Pincus et al., 2016), instead of blue or green autofluorescence, which largely report only incipient death (Coburn et al., 2013; Pincus et al., 2016). Third, we also measure declining tissue organization through quantitative measurements of textural order and disorder in brightfield images, which has been shown to correlate with lifespan (Herndon et al., 2002; Johnston et al., 2008; Shamir et al., 2009; Pincus et al., 2011). Fourth, C. elegans reproductive output has also been identified as a biomarker of aging and has been shown to decrease with age (Pickett et al., 2013). While Huang et al. found no correlation between reproduction and lifespan was found in unmated, and hence sperm-limited, hermaphrodite C. elegans (Huang et al., 2004), we here examine the temperature-sensitive sterile strain spe-9, which lays unfertilized oocytes, and may thus not be sperm-limited in the same way. Finally, we and others have shown that overall body size, and rate growth and/or shrinkage correlate with future lifespan (Pincus et al., 2011; Hulme et al., 2010). Taken together, this panel comprises a diverse set of individual measurements, allowing us to characterize longitudinal changes in distinct aspects of aging physiology (Figure 1C–E). Image Measurements Autofluorescence measurements (in red wavelengths; excitation 556 nm, emission 613 nm (Pincus et al., 2016)) were made using pixel values within the defined animal location from automated image segmentation of the paired bright field image. Intensity values were extracted and summary statistics such as 80th percentile of intensity and integrated total body fluorescence were computed. Using a percentile score provides robustness against brightly autofluorescent clumps of debris or oocytes overlapping the individual (see Figures S6A for example illustrative images). Four measures of motion were made at each time point. Long-term movement over a three-hour time scale was assessed by measuring the displacement between the centroid of the animal’s current position and that of its previous position. Unstimulated movement and stimulated movement immediately before and after blue-light stimulus were assessed by measuring distances between centroids of animal positions from the sequence of five bright-field images (see Figures S6B for example illustrative images). An estimate of the number of oocytes laid was used as a proxy for reproductive investment. This measure was made by measuring the total area of objects detected within the bacterial lawn, excluding the animal itself and thresholding by size to remove small debris. This total area was then divided by the area of an average oocyte to obtain an estimate of the total number laid (see Figure S6C for example illustrative images). Cross-sectional size of the animal was measured by simply counting the number of pixels within the segmented animal region. The raw measured size was adjusted for a slight systematic bias towards overestimating the animal’s size by a linear regression between automatically measured size and size from 1,617 human-drawn outlines of the animal (see Figure S6D for example illustrative images). Finally, a quantitative measure of tissue degeneration similar to the score used in (Pincus et al., 2011) was made by analyzing pixel patches within animals’ outlines. First, representative texture patches (“textons”) for different stages of decrepitude were obtained by grouping images by the number of days remaining in the animal’s life at that point in time (bins of 0–3, 3–6, 6–9, etc. days left to live), sampling 200,000 17 × 17 pixel patches from images in each bin, and using k-means classification to obtain 60 representative patches for bin (300 total). Then, the texture pattern of an image was defined as the 300-element histogram containing the count of closest- matching 17 × 17 pixel patches for each texton, normalized by the total number of patches in that image. These histograms were then used to train a support vector regression procedure to produce a tissue degeneration score in terms of predicted days of remaining life (see Figure S6E for example illustrative images). Quantification and Statistical Analysis Smoothing Individual animals’ time traces for each measurement were smoothed using 3 iterations of the one-dimensional Savitzsky–Golay filter (Savitzky and Golay, 1964) with a 1-degree polynomial and a window length of 9. Smoothing parameters were optimized for signal-to-noise ratio by checking the physiological measurements’ cross-validated partial Pearson r2 with remaining lifespan, controlled for age. Pearson Correlations p-values for Pearson correlations were computed using an F-test for linear regression. The F-statistic was computed in the usual way as F = r2(df)/(1 − r2), where df = n − 2, n is the sample size, and r is the Pearson coefficient of correlation. The test was performed under the assumption of normality, so for computational efficiency an appropriately renormalized chi-squared distribution was used as an approximation for the exact Fisher–Snedecor distribution. Non-parametric Spearman Correlations To validate our results without the assumption of normality made in computing Pearson correlations and their associated p-values, we also performed non-parametric Spearman correlations on the ranks of the data for each statistical test. These were performed identically to the Pearson correlation p-values. Overfitting and Multiple Hypothesis Testing To guard against overfitting, we analyzed the cross-validated Pearson r2 for defining our prognosis measurement, finding that it was not substantially lower than the non-cross-validated version. All p-values are significant even after applying the conservative Bonferroni correction for the 6 hypothesis tests conducted. Alternate Health Regression Approaches To ensure that our results were not artifacts of our particular methodology for computing a prognosis and measuring health, we examined several alternate approaches, none of which altered our key findings. Table S4 demonstrates our key results are visible from each of the individual aspects of aging physiology we measured, without combining them to form the prognosis score. Figure S4I–P shows that these results also hold when using linear regression to create the prognosis, which also allows the relative contributions of each raw measurement used to be compared (Table S5). This indicates that no specific measurement, nor the regression methodology, drives the central findings of this study. Next, we used multivariate support vector regression to create alternates to our “prognosis” score. By regressing against age rather than remaining lifespan we created a “youthfulness” or “biological age” score (Borkan and Norris, 1980; Baker and Sprott, 1988). Similarly, regressing against predicted three-day survival created a “frailty” score (Fried et al., 2001; Hubbard, 2015) (Figure S4Y, inset). Both frailty and youthfulness scores produced similar findings to our approach above (Figure S4Q–X and Y–FF). Data and Software Availability All physiological data used in this study is included with this manuscript in Supplemental_Data.zip, which is related to Figures 1–6. The image acquisition software is freely available at https://github.com/zplab/rpc-scope, and the image processing and statistical analysis software is freely available at https://github.com/zplab/wormPhysiology. Additional Resources Additional galleries of randomly selected images are available on Mendeley Data at DOI: 10.17632/9xdthhmm75.1. Supplementary Material 1 2 We would like to thank Shin-ichiro Imai, Heidi Tissenbaum, Brian Kim, Amy Xu, and Seth Pincus for helpful discussions and thoughtful feedback. ZP began this work in the lab of Frank J. Slack, supported by NIH R01 AG033921. WZ, DS, WP, EH and ZP are supported by NIH R00 AG042487. WZ and ZP are supported by Longer Life Foundation grant 2015-008. WZ is additionally supported by NIH T32 GM07200. NS is supported by a Glenn Award from the Glenn Foundation for Medical Research. Some strains were provided by the CGC, which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440). Figure 1 Experimental workflow (A) A novel culture device allows for lifelong, longitudinal observation of a high-density array of isolated individual C. elegans. Individuals are free to move about on pads of bacterial food, but cannot depart the pads. (B) Images of each individual C. elegans are acquired every three hours throughout life, at 2.2-μm resolution. The spe-9(hc88) temperature-sensitive sterile strain is used to prevent reproduction. Unfertilized oocytes are visible as dark clumps on the bacterial food pad. (C) Custom software automatically annotates the region of the food pad and the position of the animal. (D) These annotations are used to make several physiological measurements (left to right): Movement between and within timepoints is scored as a measure of neuromuscular function. The brightfield image of the animal is used to automatically score tissue integrity. Red autofluorescence is used as a measure of macromolecular homeostasis. To characterize somatic maintenance, body size is measured as cross-sectional area. Last, the number of oocytes laid is counted as a measure of reproductive investment. (E) This battery of five longitudinal physiological measurements is aggregated together to produce an estimate of remaining lifespan at each timepoint. We use this “prognosis” as an operational definition of an individual’s overall degree of senescence. (F) Examples of prognosis scores. Left: a healthy, young adult with 10.3 estimated days of life remaining. Right: an unhealthy, late-life animal with 0.5 estimated days remaining. Figure 2 Variation in lifespan and physiology within a homogeneous population (A) The survival curve for all 734 individuals in our study population, as a function of age after reproductive maturity. (B) Histogram of the number of individuals in each of seven cohorts, grouped by lifespan into 2-day-wide bins. A kernel density estimate of the underlying distribution of ages at death is shown in black. (C) An illustration of how we calculate trends of a given measurement over time. At left, the distributions of values for some lifespan-predictive measurement are shown for a long- and short-lived cohort, at day three of adulthood. Dashed lines show the mean of each distribution. At right, the trend in these means over time is plotted for each cohort. (D) Trends of bulk movement between three-hour timepoints over time, for each lifespan cohort. Due to the limited size of the food pad, this measurement saturates around 160 μm/hour. (E) Trends in autofluorescence over time, measured as the 80th percentile of whole-body red-channel autofluorescence. (F) Trends in tissue integrity score over time. This score is produced via support vector regression that maps brightfield image texture into an estimate of remaining days of life. (G) Trends in cumulative oocytes laid. (H) Trends in body size. Figure 3 How does aging physiology differ between long- and short-lived individuals? (A) “Starting Point” hypothesis: long-lived individuals start their adulthood healthier than short-lived individuals. (B) “Rate of Aging” hypothesis: long-lived individuals age more slowly than do short-lived individuals. (C) “Premature Death” hypothesis: short- and long-lived individuals are indistinguishable over the course of their lives. In this case, differences in lifespan arise from stochastic, inherently unpredictable causes of death or from factors outside our prognostic criteria. (D) Trends in the decline of prognosis over time, for each of the seven lifespan cohorts in Figure 2. These qualitatively match the “rate of aging” and “premature death” hypotheses from panels b and c. (E) We observe little quantitative relationship between lifespan and starting health (measured by an individual’s prognosis score at reproductive maturity). (F) In contrast, there is a strong negative correlation between lifespan and the rate of decline of physiological health. (G) Last, there is a moderate negative correlation between health at death and lifespan, suggesting stochasticity in death or unmeasured differences in functional health. Figure 4 Systematic differences in the trajectories of physiological decline (A) The pattern of decline can differ even among individuals with identical lifespans, starting prognoses, and ending prognoses. Individuals may experience senescence evenly throughout life (black; “neutral decline”). Alternately, senescence can accelerate early (red; a “negative deviation”), which produces a relatively extended period of low function. Finally, senescence can be delayed (green; “positive deviation”), leading to a compressed period of low function. (B) We quantify an individual’s trajectory deviation as the area between the actual trajectory (green) and a linear decline (black). This value is positive for trajectories above neutral decline, and negative for those below. To compare individuals with different lifespans, we divide this total area by each individual’s lifespan to produce the “average deviation” throughout life. (C) How trajectories of senescent decline may differ between long- and short-lived individuals. Left: Senescence is identical between long- and short-lived individuals and is merely stretched in time, causing the trajectories to align when plotted in terms of the relative fraction of life elapsed. Middle: Longer-lived individuals might have more positive deviations from neutral declines. Right: Shorter-lived individuals may have more positive deviations. (D) Trajectories of senescence for different lifespan cohorts in relative time. As in the “negative hypothesis”, shorter-lived individuals are systematically healthier at any given fraction of adult lifespan. (E) Lifespan is plotted against “average deviation”, where neutral decline is defined as the straight line between the population mean prognoses at reproductive maturity and death. Longer-lived individuals have negative deviation, while short-lived individuals experience positive deviation. (F) After controlling for differences in starting and ending prognosis, the “negative hypothesis” still applies qualitatively and (G) quantitatively. Figure 5 Healthspan and gerospan analysis In both chronological (A) and relative (B) time, the trajectory of physiological aging can be thresholded into a span of high physiological function (“healthspan”; time spent above dotted line) and a span of low function (“gerospan”; time below dotted line). (C) In chronological time, longer-lived cohorts generally have longer period of good prognosis. However the differences in healthspan between these cohorts are small compared to the differences in gerospan. (D) In relative time, it is clear that longer-lived individuals are healthy for a smaller fraction of their total lifespan than shorter-lived individuals. (E) Data from a previous experiment using spe-9(hc88); fer-15(b26) individuals on standard C. elegans culture conditions (Stroustrup et al., 2013) confirm that individuals with longer lifespans spend a larger fraction of their life in poor physiological function, as measured by fraction of life spent moving very poorly or not at all (n = 55 in each group). (F) An equivalent analysis of movement data from our culture apparatus produces similar results (n = 146 in each group). Across our population, lifespan positively correlates with both (G) healthspan (as calculated in panel A), and (H) gerospan. Compared to healthspan, variability in gerospan explains almost twice as much of the variability in lifespan (r2 of 0.302 vs. 0.672). (I) This is because, despite having the same mean duration by construction, gerospan (red curve) is more variable across our study population than healthspan (green curve). The mean of both distributions is 5.0 days, with standard deviations of 1.3 and 1.9 days for healthspan and gerospan respectively. Figure 6 Summary of key findings We evaluated the process of senescence in long- vs. short-lived individual, based on a “prognosis” score that aggregates multiple measures of aging physiology. First, based on our physiological measurements, long- and short-lived individuals are indistinguishable at the beginning of adulthood. Second, soon after the onset of reproductive maturity, short- and long-lived individuals diverge rapidly. Overall, short-lived individuals experience faster physiological aging than long-lived individuals. In addition, short-lived individuals often die “prematurely”, while still healthy-appearing. Third, we discovered that long-lived individuals spend a disproportionate fraction of their lives in highly senescent, ill-appearing states. This extended “twilight period” of low physiological function has the paradoxical effect of reducing the overall average physiological health of long-lived individuals to below that of short-lived individuals. Highlights Fully automated C. elegans culture allows for study of inter-individual variation. Short and long-lived individuals start adulthood in equal physiological health. Short-lived individuals age more quickly but have a better average quality of life. The span of poor health is more variable among individuals than that of good health. Author Contributions WZ and ZP designed the experiments. WZ designed and implemented the image segmentation and analysis software and analyzed the data. WZ, DS, and WP conducted the experiments. WP, WZ, DS, and ZP developed the single-worm culture system. ZP designed and built the custom microscope control and incubation hardware, and ZP and EH wrote the custom microscope control software. NS provided data and analysis for validation of key results in standard culture conditions. WZ and ZP wrote the manuscript with feedback from DS and NS. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Anderson EN Corkins ME Li JC Singh K Parsons S Tucey TM Sorkaç A Huang H Dimitriadi M Sinclair DA 2016 C. elegans lifespan extension by osmotic stress requires FUDR, base excision repair, foxo, and sirtuins Mechanisms of ageing and development 154 30 42 26854551 Ash AS Kroll-Desrosiers AR Hoaglin DC Christensen K Fang H Perls TT 2015 Are members of long-lived families healthier than their equally long-lived peers? Evidence from the long life family study The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 70 971 976 Baker GT Sprott RL 1988 Biomarkers of aging Experimental gerontology 23 223 239 Bansal A Zhu LJ Yen K Tissenbaum HA 2015 Uncoupling lifespan and healthspan in Caenorhabditis elegans longevity mutants Proceedings of the National Academy of Sciences 112 E277 E286 Borkan GA Norris AH 1980 Assessment of biological age using a profile of physical parameters Journal of Gerontology 35 177 184 6967883 Brenner JF Dew BS Horton JB King T Neurath PW Selles WD 1976 An automated microscope for cytologic research a preliminary evaluation Journal of Histochemistry & Cytochemistry 24 100 111 1254907 Brenner S 1974 The genetics of Caenorhabditis elegans Genetics 77 71 94 4366476 Byerly L Cassada R Russell R 1976 The life cycle of the nematode Caenorhabditis elegans: I. wild-type growth and reproduction Developmental biology 51 23 33 988845 Canny J 1986 A computational approach to edge detection Pattern Analysis and Machine Intelligence, IEEE Transactions on 679 698 Chow DK Glenn CF Johnston JL Goldberg IG Wolkow CA 2006 Sarcopenia in the Caenorhabditis elegans pharynx correlates with muscle contraction rate over lifespan Experimental gerontology 41 252 260 16446070 Christensen K Johnson TE Vaupel JW 2006 The quest for genetic determinants of human longevity: challenges and insights Nature Reviews Genetics 7 436 448 Clokey GV Jacobson LA 1986 The autofluorescent “lipofuscin granules” in the intestinal cells of Caenorhabditis elegans are secondary lysosomes Mechanisms of ageing and development 35 79 94 3736133 Coburn C Allman E Mahanti P Benedetto A Cabreiro F Pincus Z Matthijssens F Araiz C Mandel A Vlachos M 2013 Anthranilate fluorescence marks a calcium-propagated necrotic wave that promotes organismal death in C. elegans PLoS Biol 11 e1001613 23935448 Cortes C Vapnik V 1995 Support-vector networks Machine learning 20 273 297 Eckley DM Rahimi S Mantilla S Orlov NV Coletta CE Wilson MA Iser WB Delaney JD Zhang Y Wood W III 2013 Molecular characterization of the transition to mid-life in Caenorhabditis elegans Age 35 689 703 22610697 Edwards SL Charlie NK Milfort MC Brown BS Gravlin CN Knecht JE Miller KG 2008 A novel molecular solution for ultraviolet light detection in Caenorhabditis elegans PLoS Biol 6 e198 18687026 Fabian TJ Johnson TE 1994 Production of age-synchronous mass cultures of Caenorhabditis elegans Journal of gerontology 49 B145 B156 7516947 Firestone L Cook K Culp K Talsania N Preston K 1991 Comparison of autofocus methods for automated microscopy Cytometry 12 195 206 2036914 Freedman VA Wolf DA Spillman BC 2016 Disability-free life expectancy over 30 years: A growing female disadvantage in the us population American Journal of Public Health e1 e7 Fried LP Tangen CM Walston J Newman AB Hirsch C Gottdiener J Seeman T Tracy R Kop WJ Burke G 2001 Frailty in older adults evidence for a phenotype The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 56 M146 M157 Fries JF 1980 Aging, natural death, and the compression of morbidity New England journal of medicine 303 130 135 7383070 Garcia-Valles R Gomez-Cabrera MC Rodriguez-Mañas L Garcia-Garcia FJ Diaz A Noguera I Olaso-Gonzalez G Viña J 2013 Life-long spontaneous exercise does not prolong lifespan but improves health span in mice Longevity & healthspan 2 1 24764515 Gerstbrein B Stamatas G Kollias N Driscoll M 2005 In vivo spectrofluorimetry reveals endogenous biomarkers that report healthspan and dietary restriction in Caenorhabditis elegans Aging cell 4 127 137 15924569 Ghezzi E Ship J 2003 Aging and secretory reserve capacity of major salivary glands Journal of dental research 82 844 848 14514768 Gill MS Olsen A Sampayo JN Lithgow GJ 2003 An automated high-throughput assay for survival of the nematode Caenorhabditis elegans Free Radical Biology and Medicine 35 558 565 12957648 Glenn CF Chow DK David L Cooke CA Gami MS Iser WB Hanselman KB Goldberg IG Wolkow CA 2004 Behavioral deficits during early stages of aging in Caenorhabditis elegans result from locomotory deficits possibly linked to muscle frailty The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 59 1251 1260 Gögele M Pattaro C Fuchsberger C Minelli C Pramstaller PP Wjst M 2011 Heritability analysis of life span in a semi-isolated population followed across four centuries reveals the presence of pleiotropy between life span and reproduction The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 66 26 37 Golden TR Hubbard A Dando C Herren MA Melov S 2008 Age-related behaviors have distinct transcriptional profiles in Caenorhabditis elegans Aging cell 7 850 865 18778409 Gu D Dupre ME Sautter J Zhu H Liu Y Yi Z 2009 Frailty and mortality among Chinese at advanced ages The Journals of Gerontology Series B: Psychological Sciences and Social Sciences gbn009 Guarente L Kenyon C 2000 Genetic pathways that regulate ageing in model organisms Nature 408 255 262 11089983 Hahm JH Kim S DiLoreto R Shi C Lee SJV Murphy CT Nam HG 2015 C. elegans maximum velocity correlates with healthspan and is maintained in worms with an insulin receptor mutation Nature communications 6 Hermann GJ Schroeder LK Hieb CA Kershner AM Rabbitts BM Fonarev P Grant BD Priess JR 2005 Genetic analysis of lysosomal trafficking in caenorhabditis elegans Molecular biology of the cell 16 3273 3288 15843430 Herndon LA Schmeissner PJ Dudaronek JM Brown PA Listner KM Sakano Y Paupard MC Hall DH Driscoll M 2002 Stochastic and genetic factors influence tissue-specific decline in ageing C. elegans Nature 419 808 814 12397350 Herskind AM McGue M Holm NV Sörensen TI Harvald B Vaupel JW 1996 The heritability of human longevity: a population-based study of 2872 danish twin pairs born 1870–1900 Human genetics 97 319 323 8786073 Hertweck M Baumeister R 2005 Automated assays to study longevity in C. elegans Mechanisms of ageing and development 126 139 145 15610772 Hosono R Sato Y Aizawa SI Mitsui Y 1980 Age-dependent changes in mobility and separation of the nematode Caenorhabditis elegans Experimental gerontology 15 285 289 7409025 Hsu AL Feng Z Hsieh MY Xu XS 2009 Identification by machine vision of the rate of motor activity decline as a lifespan predictor in C. elegans Neurobiology of aging 30 1498 1503 18255194 Huang C Xiong C Kornfeld K 2004 Measurements of age-related changes of physiological processes that predict lifespan of Caenorhabditis elegans Proceedings of the National Academy of Sciences of the United States of America 101 8084 8089 15141086 Hubbard RE 2015 Sex differences in frailty Frailty in Aging 41 41 53 Karger Publishers Hulme SE Shevkoplyas SS McGuigan AP Apfeld J Fontana W Whitesides GM 2010 Lifespan-on-a-chip: microfluidic chambers for performing lifelong observation of c. elegans Lab on a Chip 10 589 597 20162234 Ismail K Nussbaum L Sebastiani P Andersen S Perls T Barzilai N Milman S 2016 Compression of morbidity is observed across cohorts with exceptional longevity Journal of the American Geriatrics Society Johnston J Iser WB Chow DK Goldberg IG Wolkow CA 2008 Quantitative image analysis reveals distinct structural transitions during aging in Caenorhabditis elegans tissues PLoS One 3 e2821 18665238 Kassirer JP 1976 The principles of clinical decision making: an introduction to decision analysis The Yale journal of biology and medicine 49 149 941463 Kenyon C 2005 The plasticity of aging: insights from long-lived mutants Cell 120 449 460 15734678 Kenyon CJ 2010 The genetics of ageing Nature 464 504 512 20336132 Kirkwood TB Feder M Finch CE Franceschi C Globerson A Klingenberg CP LaMarco K Omholt S Westendorp RG 2005 What accounts for the wide variation in life span of genetically identical organisms reared in a constant environment? Mechanisms of ageing and development 126 439 443 15664632 Klass MR 1977 Aging in the nematode Caenorhabditis elegans: major biological and environmental factors influencing life span Mechanisms of ageing and development 6 413 429 926867 Lee A Joynt GM Ho AM Keitz S McGinn T Wyer PC Group ETSW 2009 Tips for teachers of evidence-based medicine: making sense of decision analysis using a decision tree Journal of general internal medicine 24 642 648 19247720 Mattison JA Roth GS Beasley TM Tilmont EM Handy AM Herbert RL Longo DL Allison DB Young JE Bryant M 2012 Impact of caloric restriction on health and survival in rhesus monkeys from the NIA study Nature 489 318 321 22932268 McEwen BS 2003 Interacting mediators of allostasis and allostatic load: towards an understanding of resilience in aging Metabolism 52 10 16 14577057 Melov S 2016 Geroscience approaches to increase healthspan and slow aging F1000Research 5 Montgomery H 2000 Cardiac reserve: linking physiology and genetics Intensive Care Med 26 144 Pettay JE Kruuk LE Jokela J Lummaa V 2005 Heritability and genetic constraints of life-history trait evolution in preindustrial humans Proceedings of the National Academy of Sciences of the United States of America 102 2838 2843 15701704 Piccardi M 2004 Background subtraction techniques: a review 2004 IEEE international conference on systems, man and cybernetics 4 3099 3104 IEEE Pickett CL Dietrich N Chen J Xiong C Kornfeld K 2013 Mated progeny production is a biomarker of aging in caenorhabditis elegans G3: Genes Genomes Genetics 3 2219 2232 24142929 Pincus Z Mazer TC Slack FJ 2016 Autofluorescence as a measure of senescence in C. elegans: look to red, not blue or green Aging 8 Pincus Z Slack FJ 2010 Developmental biomarkers of aging in Caenorhabditis elegans Developmental Dynamics 239 1306 1314 20151474 Pincus Z Smith-Vikos T Slack FJ 2011 Microrna predictors of longevity in Caenorhabditis elegans PLoS Genet 7 e1002306 21980307 Rea SL Wu D Cypser JR Vaupel JW Johnson TE 2005 A stress-sensitive reporter predicts longevity in isogenic populations of Caenorhabditis elegans Nature genetics 37 894 898 16041374 Rechel B Grundy E Robine JM Cylus J Mackenbach JP Knai C McKee M 2013 Ageing in the European union The Lancet 381 1312 1322 Sánchez-Blanco A Kim SK 2011 Variable pathogenicity determines individual lifespan in Caenorhabditis elegans PLoS Genet 7 e1002047 21533182 Savitzky A Golay MJ 1964 Smoothing and differentiation of data by simplified least squares procedures Analytical chemistry 36 1627 1639 Sebastiani P Sun F Andersen SL Lee J Wojczynski MK Sanders JL Yashin AI Newman AB Perls TT 2013 Families enriched for exceptional longevity also have increased health-span: findings from the long life family study Frontiers in public health 1 38 24350207 Shamir L Wolkow CA Goldberg IG 2009 Quantitative measurement of aging using image texture entropy Bioinformatics 25 3060 3063 19808878 Singson A Mercer KB L’Hernault SW 1998 The C. elegans spe-9 gene encodes a sperm transmembrane protein that contains egf-like repeats and is required for fertilization Cell 93 71 79 9546393 Stroustrup N Anthony WE Nash ZM Gowda V Gomez A López-Moyado IF Apfeld J Fontana W 2016 The temporal scaling of Caenorhabditis elegans ageing Nature Stroustrup N Ulmschneider BE Nash ZM López-Moyado IF Apfeld J Fontana W 2013 The Caenorhabditis elegans lifespan machine Nature methods 10 665 670 23666410 Vaupel JW Carey JR Christensen K Johnson TE Yashin AI Holm NV Iachine IA Kannisto V Khazaeli AA Liedo P 1998 Biodemographic trajectories of longevity Science 280 855 860 9599158 Weinstein MC Stason WB 1977 Foundations of cost-effectiveness analysis for health and medical practices New England journal of medicine 296 716 721 402576

PMC005xxxxxx/PMC5111812.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101601398 42374 Expert Opin Orphan Drugs Expert Opin Orphan Drugs Expert opinion on orphan drugs 2167-8707 27867771 5111812 10.1517/21678707.2016.1128322 NIHMS797925 Article Prospect and progress of oncolytic viruses for treating peripheral nerve sheath tumors Antoszczyk Slawomir PhD Research Associate in Neurosurgery 12 Rabkin Samuel D. PhD Professor of Neurosciences 123 1 Molecular Neurosurgery Laboratory, Department of Neurosurgery, Massachusetts General Hospital 2 Department of Neurosurgery, Harvard Medical School, Boston MA 3 Corresponding Author: Molecular Neurosurgery Laboratory, Department of Neurosurgery, MGH-Simches, 185 Cambridge St., CPZN-3800, Boston, MA 02114, Phone: 617 726-6817, Fax: 617 643-3422, rabkin@mgh.harvard.edu 25 6 2016 26 12 2015 2016 01 1 2017 4 2 129138 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Introduction Peripheral nerve sheath tumors (PNSTs) are an assorted group of neoplasms originating from neuroectoderm and growing in peripheral nerves. Malignant transformation leads to a poor prognosis and is often lethal. Current treatment of PNSTs is predominantly surgical, which is often incomplete or accompanied by significant loss of function, in conjunction with radiotherapy and/or chemotherapy, for which the benefits are inconclusive. Oncolytic viruses (OVs) efficiently kill tumor cells while remaining safe for normal tissues, and are a novel antitumor therapy for patients with PNSTs. Areas covered Because of the low efficacy of current treatments, new therapies for PNSTs are needed. Pre-clinically, OVs have demonstrated efficacy in treating PNSTs and perineural tumor invasion, as well as safety. We will discuss the various PNSTs and their preclinical models, and the OVs being tested for their treatment, including oncolytic herpes simplex virus (HSV), adenovirus (Ad), and measles virus (MV). OVs can be ‘armed’ to express therapeutic transgenes or combined with other therapeutics to enhance their activity. Expert opinion Preclinical testing of OVs in PNST models has demonstrated their therapeutic potential and provided support for clinical translation. Clinical studies with other solid tumors have provided evidence that OVs are safe in patients and efficacious. The recent successful completion of a phase III clinical trial of oncolytic HSV paves the way for oncolytic virotherapy to enter clinical practice. virotherapy MPNST HSV soft tissue sarcoma 1. Introduction To Peripheral Nerve Sheath Tumors And Oncolytic Viruses Peripheral nerve sheath tumors (PNSTs) are relatively rare neoplasms originating from neuroectoderm and growing in the peripheral nerves, causing pain, reducing nerve function and leading to disability. PNSTs are typically benign, schwannomas and neurofibromas, and sporadic or caused by genetic disorders of the nervous system, such as neurofibromatosis, and are categorized as soft tissue tumors 1, 2. Malignant peripheral nerve sheath tumors (MPNST) are very aggressive and characterized by a high mortality rate 3, 4. Benign plexiform neurofibromas (PNF) can transform to a malignant form 5. These are common tumors in patients with neurofibromatosis type 1 (NF1) and 2 (NF2) 1. The most recent classification of nerve sheath tumors is summarized in the WHO Classification of Tumors of Soft Tissue and Bone 6. We will not discuss the rarer PNSTs, such as myxoma, perineurioma, and triton tumors. Carcinoma perineural invasion also involves tumor growth in peripheral nerves and thus has treatment issues that overlap with PNSTs. Although the idea of using virus infection to induce tumor cell death, virotherapy, has been known more than 100 years 7, the advances in genetic engineering provided new possibilities to modify viruses for safety and efficacy. Oncolytic viruses (OVs) selectively replicate in tumor cells without harming normal tissue, making new infectious virus that can then spread and kill additional tumor cells 7, 8. They can kill tumor cells not only by direct cytopathic effect, oncolysis, but also by indirect mechanisms, such as inducing anti-tumor immune responses or attacking the tumor vasculature 9-11. OVs include those viruses that: (i) have a natural propensity to replicate in tumor cells, i.e. myxoma and Newcastle disease viruses (NDV); (ii) are vaccine strains that have been attenuated, i.e. measles (MV), poliovirus (PV), and vaccinia virus (VV); or (iii) are genetically engineered with mutation/deletions in pathogenic genes, genes required for replication in normal cells, and/or retargeted to tumor cell receptors, i.e. adenovirus (Ad), herpes simples virus (HSV), and vesicular stomatitis virus (VSV) 12. OVs can be ‘armed’ with therapeutic transgenes, including; immunomodulatory, anti-angiogenic, and cytotoxic genes, that are expressed in the tumor after infection 13, 14. Depending on the tumor type, OVs can be used: (i) as single agents, (ii) in combination with chemo- or/and radio-therapy, or (iii) expressing therapeutic transgenes. During the last two decades, numerous OVs have entered into clinical trials for a large variety of cancers 7, 15. However, there has been only one clinical trial to date reporting OV treatment for PNST, using oncolytic Ad 16. Recently, the first oncolytic virus, talimogene laherparepvec, an oncolytic HSV, was approved by the FDA in the US for the treatment of advanced melanoma 17, 18. Because of this lack of completed clinical trials for PNST, we will focus on OVs that have been explored preclinically for the treatment of PNSTs (Table 1), strategies to improve OV efficacy in PNSTs, safety, and future virotherapy directions. 2. Peripheral Nerve Sheath Tumors (PNST) and Perineural Invasion in Cancer 2.1. Schwannoma Normal Schwann cells form myelin, a protective sheath around peripheral nerves. Schwannomas, the most common PNST in adults and comprised of abnormal Schwann-like cells, are benign, slow-growing, circumscribed, encapsulated, solitary, and non-infiltrating tumors 2, 19. NF2 patients develop schwannomas, often multiple, in cranial, spinal, and peripheral nerves, with bilateral vestibular schwannomas a hallmark 2, 19. NF2 is an autosomal dominant disease (affects about 1:25000 births) caused by inactivating mutations of the NF2 tumor suppressor gene located on chromosome 22q12, and characterized by the development of nervous system tumors, ocular abnormalities, and skin tumors 20. The NF2 gene, encoding merlin (moesin-ezrin-radixin-like protein), is also mutated or lost in most sporadic schwannomas 21, 22. Schwannomas are often asymptomatic, although vestibular schwannomas often lead to hearing loss, and can usually be treated with surgical resection 2, 20. A broader understanding of the molecular mechanism of NF2 pathogenesis should lead to new therapies, with a number of downstream effectors of merlin being tested in clinical trials, i.e.: erlotinib (Her-1/EGFR inhibitor), lapatinib (Her-2/EGFR inhibitor), and everolimus (mTOR inhibitor) 23, 24. While the targeted therapies are promising, they have not been curative, so virotherapy is an attractive approach. 2.2. Neurofibroma Neurofibromas are benign PNSTs arising from Schwann cell precursors that include fibroblasts, mast cells, and distinct from schwannomas, a matrix of collagen fibers 2, 5. Typically they occur sporadically, however approximately 10% are associated with NF1, where they are a diagnostic criteria 25. NF1 is a frequently occurring autosomal dominant genetic disease (affects about 1:3500 births), caused by mutations in the NF1 gene located on chromosomal segment 17q11.2 and encoding neurofibromin 26. Neurofibromin contains a Ras-GTPase activating protein (Ras-GAP) domain that negatively regulates Ras activity, so that NF1 is a member of the RASopathy cancer predisposing syndromes 27. Unlike schwannoma, neurofibromas are not encapsulated and infiltrate between the nerve fascicles. There are three subtypes of neurofibroma based on their location and appearance: (i) localized cutaneous or dermal neurofibromas, which are the most common, appear at puberty and grow from small nerves in the skin or just under the skin. They have limited growth, remaining benign throughout life, and do not transform into malignant PNSTs 5; (ii) diffuse or subcutaneous neurofibromas are typically located within the subcutaneous tissues of the head and neck, are often pigmented or with melanocytic differentiation 25, 28, and in NF1 are associated with internal neurofibroma tumor burden and mortality 29, 30; and (iii) plexiform neurofibromas (PNF) are usually congenital, occurring in about a third of NF1 patients, and present anywhere in the body but often internal where they can remain aysmptomatic 26. They involve multiple nerve fascicle trunks and infiltrate surrounding tissue 25. Importantly, PNFs can undergo malignant transformation and progress to MPNST, which occurs in about 5-10% of cases 5, 31, 32. PNFs are typically debulked by surgery when indicated, but this can be associated with nerve damage and hemorrhage, and is often incomplete 5, 33. Based on our understanding of neurofibromin, a small phase II clinical trial examined imatinib mesylate (Gleevec), a kinase inhibitor, to treat PNF 34. 2.3. Malignant peripheral nerve sheath tumor (MPNST) MPNST is a soft-tissue sarcoma arising from peripheral nerves or benign PNSTs that displays nerve sheath differentiation 25, 35. They account for about 10% of soft tissue sarcomas, but half arise in NF1 patients. There are histological and clinical differences between sporadic and NF1-associated MPNSTs 33, 36, although a recent meta-analysis suggested that the survival differences have been converging in the last decade 37. MPNST frequently occurs in the extremities, often in major nerve trunks like the sciatic nerve 35. Like other malignancies, metastasis may also occur, often to the lung, liver, brain, soft tissue, bone, regional lymph nodes and retroperitoneum 32. In NF1, loss-of-heterozygosity of NF1 occurs in over 90% of tumors 38, with tumors having constitutively activated RAS and downstream activation of the Akt and MAPK pathways 39. However, the complex karyotypes in MPNSTs suggests that additional genetic alterations, such as in p53, CDKN2A, CDK4, contribute to malignancy 35. Surgical resection is the main treatment option for MPNST, however it is often incomplete, with a local recurrences ranging up to 65% depending on the tumor location, and associated with potentially significant loss of function 33, 40. Radiotherapy is recommended for high-grade lesions when surgery is not possible or incomplete, but hasn't been shown to improve survival 33, 40. Adjuvant chemotherapy, ifosfamide and doxorubicin, may have some activity in pediatric patients and non-NF1 patients 40, 41. The long-term survival of MPNST patients is poor, around 50% at 5 years 37, 40, 41, so there is a critical need for novel therapeutic approaches. 2.4. Perineural invasion in cancer Perineural invasion (PNI), the presence of tumor cells within any of the layers of the nerve sheath, is a route of metastasis distinct from vascular or lymphatic dissemination 42. It is very common and often the typical route of metastasis in head and neck squamous cell carcinoma (∼80%), prostate (∼80%), gastric (∼50%), and colorectal cancers (∼30%), and almost all pancreatic tumors 42-44. PNI is an independent prognostic factor in these tumors, with 5 year survival often decreased by half, and is also associated with severe pain, especially in pancreatic cancer 42, 43, 45. Understanding the molecular mechanisms of PNI is lagging and it is still unclear how tumor cells penetrate the nerve sheath and layers of collagen and basement membrane, or why some carcinomas have a penchant for PNI. The lack of targeted treatments and poor prognosis for PNI makes this an important target for novel therapies and potentially OVs. 3. Oncolytic Viruses 3.1. Herpes simplex virus (HSV) HSV-1 is an enveloped virus with a 152 kb double-stranded DNA genome encoding about 84 genes 46. Viral glycoproteins in the envelope bind to host attachment factors on the cell surface (heparin sulfate, nectin-1 (PVRL1), PILRα, and HVEM (TNFRSF14)) and induce membrane fusion 46. This can impact cancer cell susceptibility, for example HVEM is not expressed on most MPNST cell lines, including sensitive cells, while the levels of nectin-1 expression correlate with virus yields 47. In permissive cells, the HSV replication cycle is usually completed in less than 24 hr, and cells release viral progeny ready to infect adjacent cells. HSV is a neurotropic virus that can invade the nervous system and infect neural cells. HSV can be genetically engineered to mutate/delete viral genes that confer selective replication in tumor cells and attenuate neurovirulence, make them oncolytic. For example: (i) the thymidine kinase (TK) and ribonucleotide reductase (ICP6) genes are involved in nucleotide metabolism and necessary for virus replication in nondividing cells, but not in tumor cells; and (ii) γ34.5 is a major determinant of HSV pathogenicity, so that deletions of γ34.5 significantly reduce neurovirulence permitting treatment of nervous system tumors 48. 3.2. Oncolytic HSV (oHSV) A number of oHSVs with different mutations/deletions (ICP6, γ34.5) have been tested in a variety of PNST models (Table 1). G47Δ, a third generation oHSV (ICP6-, γ34.5Δ, ICP47Δ), was examined in a number of human xenograft and mouse transgenic schwannoma models 49, 50. In a NF2 transgenic model, spontaneous subcutaneous schwannomas were treated with a single intratumoral injection of G47Δ after MR imaging to identify the tumors 49. This led to a decrease in tumor volume within 2 weeks in G47Δ-treated tumors, however, in some cases control-treated tumors also regressed. This is an issue not only with the mouse models, but also in human patients 51, and confounds both preclinical and clinical studies of benign tumors with variable growth rates. In addition, it took about 18 months for the transgenic schwannomas to become apparent, making treatment studies difficult. To get around this problem, human schwannoma xenografts were generated by subcutaneous implantation of schwannoma tissue from patients with NF2 or schwannomatosis in immunocompromised mice 49. These tumors were histologically schwannomas and susceptible to G47Δ infection and replication 49. Treatment with two doses of G47Δ resulted in a reduction in tumor volume, while the vehicle-treated tumors continued to grow 49. In a separate study, subcutaneous tumors were established in nude mice with a human immortalized schwannoma cell line HEI193 or a mouse line (NF2S-1), and treated two successive times with G47Δ 50. Both tumors showed significant inhibition of tumor growth or regression compared with control animals, however, the NF2S-1 tumors regrew at about 7 weeks after treatment 50. These positive results in a variety of schwannoma models strongly support the use of oHSV for the treatment of schwannomas. MPNST is a lethal tumor and there are more MPNST cell lines available as models than for benign PNSTs, making it a more frequent target for OV therapy. Initial studies in vitro, showed that a set of human MPNST cell lines were sensitive to oHSV G207 and hrR3 (Table 1), regardless of Ras activation, whereas normal human Schwann cells were not permissive to virus replication 52. In contrast, the susceptibility of mouse MPNST cells isolated from transgenic mouse tumors to G207 was dependent upon elevated levels of Ras activation 53. Both G207 and hrR3 were able to inhibit tumor growth and survival after intraperitoneal injection of mice bearing intraperitoneal xenografts of human STS26T MPNST 54. rQLuc, similar to G207, but expressing luciferase (Table 1), was efficacious in subcutaneous xenografts of STS26T and S462.TY after intratumoral injection 55. Luciferase expression permits non-invasive monitoring of virus spread using bioluminescence imaging. Transcriptionally-targeting of oHSV is a way to drive replication selectively in tumor cells 48. Midkine (MDK) is overexpressed in MPNST cells compared to fibroblasts and Schwann cells 56 and thus the MDK promoter should drive MPNST-specific transcription. To enhance oHSV replication, γ34.5 was placed under the MDK promoter, generating oHSV-MDK-34.5 (Table 1) 56. oHSV-MDK-34.5 had increased viral replication and cytotoxicity in human STS26T MPNST cells in vitro, and inhibited tumor growth and increased survival in subcutaneous xenografts compared with the control virus oHSV-MDK 56. Recently, a new orthotopic MPNST model was developed by implantation of mouse NF1 transgenic MPNST cell lines or human NF1 MPNST stem-like cells (MSLCs) into the sciatic nerves of immunocompetent and immunodeficient mice, respectively 57. In this model, mice develop progressive hind limb deficits, which can be quantified to evaluate treatment efficacy. Neurologic deficits are apparent prior to palpable tumors, which provides an indication that the tumor is established and growing and thus suitable for treatment. MSLCs were isolated from an established human MPNST cell line, S462, which could self-renew, had increased expression of stem cell markers, could differentiate into cells from multiple lineages, and were more tumorigenic 58. About a third of mice bearing human MSLC-derived tumors treated with a single low dose intratumoral injection of only 2×105 plaque forming units (pfu) of G47Δ were long-term survivors, with no evidence of tumor and limited neurologic deficit, in contrast to mock-treated mice which all succumbed to tumor growth 57. In the immunocompetent mouse MPNST model, a single dose of G47Δ significantly delayed tumor growth, improved neurologic score, and significantly extended survival compared with control, with the 35% long-term survivors lacking macroscopically detectable tumor or ultrastructural abnormalities 57. Therefore, oHSV has been shown to be very effective in treating PNSTs, both benign schwannomas and malignant MPNSTs. PNI can be modeled by tumor cell implantation into sciatic nerves. The first description of oHSV treatment in such a model was the demonstration that a single intratumoral injection of G207 could extend the survival of mice bearing tumors after implantation of human neuroblastoma IMR32 cells into the sciatic nerve 59. Tumors were similarly established in the sciatic nerve with human carcinoma cell lines from pancreatic (MiaPaCa2), squamous cell (QLL2), and prostate (PC3, DU145) cancers, and treated with oHSV NV1023 (Table 1) 60, 61. While all saline-treated mice developed complete hindlimb paralysis, most NV1023-treated mice had preserved nerve function and significant tumor regression 60, 61. A similar oHSV, NV1020, has been in clinical trial for metastatic colorectal carcinoma to the liver 62. 3.3. oHSV expressing therapeutic transgenes While OVs are effective therapeutic agents, their activity can be enhanced by ‘arming’ them with therapeutic transgenes. Expression of transgenes in the tumor can target uninfected tumor cells and normal cells and extracellular matrix of the tumor microenvironment 14, 63. A number of different transgenes have been inserted into oHSV and examined in MPNST models, including those encoding antiangiogenic factors, immunomodulatory cytokines, receptor decoys, and proteinase inhibitors 48. rQT3 encodes human tissue inhibitor of metalloproteinase 3 (TIMP3) (Table 1), blocking the activity of matrix metalloproteinases, which promote tumor invasion and are expressed in MPNST cells 55. rQT3 was much more effective than G207 or rQLuc at inhibiting MPNST STS26T and S462.TY tumor growth 55. G47Δ expressing anti-angiogenic factors platelet factor 4 (PF4) and dominant-negative fibroblast growth receptor (dnFGFR) have been constructed 64, 65. Mouse MPNST cell lines expressing dnFGFR or PF4 had reduced tumor growth and angiogenesis in vivo 64, 65. Infection of human endothelial cells with G47Δ-dnFGFR or G47Δ-PF4 was more cytotoxic than control G47Δ-empty and conditioned media from infected tumor cells reduced endothelial migration in vitro 64, 65. In vivo, G47Δ expressing dnFGFR or PF4 significantly inhibited subcutaneous mouse MPNST growth in nude mice and decreased vascular density compared with control G47Δ alone 64, 65. The effects of PF4 and IL-12 expression on oHSV efficacy in an immunocompetent setting were examined using the orthotopic sciatic nerve mouse MPNST implant model. G47Δ-PF4 was not as effective in immunocompetent mice and only significantly extended survival compared to G47Δ-empty 57. In contrast, G47Δ-IL12 was significantly better than G47Δ-empty at inhibiting neurologic deficit progression and tumor growth, and increasing survival 57. 3.4. oHSV in combination with drugs MPNST is resistant to standard single agent chemotherapy 41, 66. Using oncolytic virotherapy in combination with existing chemotherapeutics may lead to synergistic interactions that increase therapeutic effects not achievable by either therapy alone 67. oHSV has been successfully combined with a number of chemotherapeutic and molecularly targeted drugs in a variety of solid tumors 67. So far this promising strategy has not been explored in PNST models, except for erlotinib (inhibitor of EGFR tyrosine kinase), and in the oncolytic Ad clinical trial. In vitro, erlotinib in combination with hrR3 or G207 was more cytotoxic to STS26T MPNST cells than either treatment alone 54. Unfortunately, combination treatment of the tumors did not enhance efficacy over oHSV alone 54. 3.5. oHSV safety Wild type HSV is highly neuroinvasive in the peripheral and central nervous system and neuropathogenic 46. After footpad inoculation, HSV-1 invades the CNS via the sciatic nerve, causing paralysis and death 68. Direct injection of wild type HSV-1 into the sciatic nerve causes nerve demyelination, Wallerian degeneration, and inflammation at early times, followed by paralysis and lethal encephalitis 59-61, 69. In contrast, sciatic nerve injection of NV1023 at a 10-fold higher dose caused no significant toxicity, with only mild Wallerian degeneration 61. Similarly, sciatic nerve injection of G47Δ was non-toxic, with mice gaining weight and exhibiting transient neurologic deficits no different than PBS-injected control mice 57. The G47Δ-injected sciatic nerves did not exhibit any ultrastructural abnormalities, with the exception of focal needle track damage that was also seen in the PBS-injected nerves 57. 3.6. Adenovirus Oncolytic Ad ONYX-015 is a hybrid Ad2/Ad5 derived from mutant of Ad dl1520 containing a deletion of E1b-55K protein 70. E1b-55K protein together with the E4orf6 protein target p53 for degradation, so that ONYX-015 selectively replicates in tumor cells with mutated p53. More recent studies suggest that E1B-55K is also involved in blocking DNA damage responses and export of late viral RNAs 71. ONYX -015 has been evaluated in a large number of clinical trials for a variety of cancers 72. In 2005, Galanis et al. 16 reported on completed results from the first clinical trial using ONYX-15 to treat patients with advanced sarcoma in combination with mitomycin-C, doxorubicin, and cisplatin chemotherapy. Among the six patients treated was one patient with metastatic MPNST (p53+, MDM2-), who received the highest ONYX-015 dose (1010 pfu) 16. This was the only patient who achieved a partial response, in both the injected bulky tumor and 2 small uninjected lesions, which lasted 11 months. Unfortunately, clinical testing with ONYX-015 was halted during a pivotal phase III trial when Onyx divested the program. A similar oncolytic Ad, H101, has been approved in China for use in head and neck cancers in combination with chemotherapy 72. Since then many new oncolytic Ad vectors have been constructed. Ad5/3-D24-GMCSF (CGTG-102) (Table 1) selectively replicates in p16/Rb-defective cells and has been evaluated in patients with advanced solid tumors 73. Ad5/3-D24-GMCSF has been examined in an immunocompetent Syrian hamster soft-tissue sarcoma (STS) model in combination with doxorubicin and ifosfamide 74. The combination drug/virus treatment was highly effective and resulted in synergistic antitumor efficacy compared with single agent treatments 74. These studies suggest that oncolytic Ad in combination with chemotherapy may be an attractive strategy for PNST. 3.7. Measles virus (MV) MV, a Morbillivirus in the family Paramyxoviridae, is an enveloped virus with a non-segmented, negative-strand RNA genome 75. Several cases of spontaneous tumor regression in individuals with lymphoma after MV infection have been reported, suggesting that MV may have oncolytic properties 75. Wild type MV uses the cellular receptor SLAM (signaling lymphocyte activation molecule) for cellular entry, which is not commonly expressed on tumor cells, however, attenuated vaccine MV strains like Edmonston are adapted for cellular entry using CD46, which is highly expressed on human tumor cells, including MPNST 76, 77. The human sodium iodide symporter (NIS) has been inserted into MV-Edmonston (MV-NIS) (Table 1) to enable non-invasive monitoring of virus infection and spread using 125I SPECT 76. NIS expression can also facilitate radiotherapy with 131I administration 76. MV-NIS is currently in clinical trials for multiple myeloma, ovarian cancer, and mesothelioma 76. MPNST cell lines were very susceptible to MV-NIS cytotoxicity, while normal Schwann cells were not, despite high-level expression of CD46 77. In vivo, intratumoral injection of MV-NIS was efficacious in inhibiting the growth of subcutaneous MPNST ST88-14 and S462TY tumors 77. NIS expression permitted the detection of MV-infected cells in mice bearing tumors 77. MV is associated with a number of serious neurologic complications in the CNS, but there is no evidence that it invades or causes pathology in the PNS 78. 3.8. Vesticular stomatitis virus (VSV) VSV is a non-segmented negative-strand RNA virus from the Rhabdoviridae family 79. VSV oncolytic selectivity is due to its extreme sensitivity to type 1 interferon (IFN) and innate antiviral responses, which are defective in most cancer cells 79. VSV can be further attenuated by disrupting the gene order, as in VSV-G/GFP, which was used to isolate VSV-rp30a (Table 1) by positive selection on glioblastoma cells 80. VSV-rp30a also grew and killed human MPNST cell lines S462-TY and STS-26T better than its parent VSV-G/GPF, and growth was not inhibited by pretreatment with INF-α 81. Primary fibroblasts and vascular endothelia cells were about 10-fold less sensitive to VSV-rp30a than the MPNST cells 81. In a human subcutaneous fibrosarcoma model, a single intravenous dose of VSV-rp30a completely inhibited tumor growth 81. A clinical trial is currently ongoing with VSV expressing IFNβ in patients with refractory hepatocellular carcinoma (http://clinicaltrials.gov/ct2/show/NCT01628640). 4. Conclusions OVs have multiple mechanisms of action, including; direct tumor cell killing, amplification in situ, and induction of anti-tumor immunity. This makes them powerful therapeutic agents. They have unique properties that derive from the biology of the viruses from which they were generated. The results we describe from preclinical studies demonstrate that OVs have robust anti-PNST activity and warrant clinical translation. Most of the studies have been conducted with oHSV, a neurotropic virus that can be genetically engineered to selectively replicate in tumor cells but not normal tissue. Safety studies with oHSV demonstrated no toxic effects or nerve damage after direct injection into the sciatic nerve. Arming oHSV with therapeutic transgenes enhances anti-tumor activity. Other OVs have also been armed, but so far none examined in PNST models. The availability of a number of MPNST preclinical models and the lack of targeted therapies or significant improvements in outcomes for this malignancy, make it a prime target for OV therapy, which has demonstrated impressive efficacy in preclinical models. Even benign PNSTs are highly susceptible to oHSV therapy. There have been fewer studies with other OVs, but oncolytic Ad ONYX-015 is the only OV to be administered to a patient with PNST. Preliminary studies with MV and VSV, suggest that they also have potential to treat PNST. 5. Expert Opinion Despite the great progress in understanding the molecular mechanisms of pathogenesis of PNSTs, surgical resection of these tumors remains the main treatment paradigm. Due to size or location, these tumors are often inoperable or entail a risk of nerve damage. On the other hand, they are not very susceptible to chemotherapy or radiotherapy, and effective molecularly targeted drugs have not been identified. Therefore, new therapeutic strategies are necessary for both benign and malignant PNSTs. OVs are one new promising approach that has been underexplored. Unfortunately, the lack of preclinical animals models for benign PNSTs has hampered progress in developing and testing new therapies, including OVs. However, where models are available, OVs have demonstrated robust anti-tumor activity. The number and types of preclinical models has increased and improved dramatically over the past decade and include models more representative of the clinical situation, such as cancer stem cells, patient-derived xenografts (PDX), and transgenic mice. All models have limitations, but having multiple options improves the likelihood of successful clinical translation. Advances in genetic engineering allows the construction of OVs that are safe and with specificity for the molecular diversity of tumors. Arming OVs with therapeutics transgenes can target the tumor microenvironment locally. The normal cells and stroma comprising the tumor are a critical component in tumorigenesis and an important target for therapy. The size of transgene to be inserted will vary depending on the virus, but for oHSV this can be between 10-30 kb, sufficient for even multiple transgenes. The choice of transgene is unlimited; be it reporter genes to monitor OV activity, cytotoxic agents or prodrug activating enzymes to kill cells, antiangiogenic factors to inhibit neovascularizaton, immune modulatory factors to enhance anti-tumor immunity, or extracellular matrix modifying agents to enhance OV spread or inhibit tumor cell migration. Importantly, expression is regulated by the selectivity of the OV, so that it is typically expressed locally within the tumor, limiting systemic toxicity. OVs derived from 12 different viruses have entered clinical trials for a large variety of cancers 15. There is no obvious reason why OVs, in addition to those described here, should not also be effective in treating PNSTs. Talimogene laherparepvec, an oHSV similar to G47Δ except expressing GMCSF, was recently approved by the FDA for the treatment of advanced melanoma. This is the first OV approved for clinical use in the US and should significantly increase the interest of the pharmaceutical industry and clinicians in the use of OVs. This is an exciting time for the field of oncolytic virotherapy and PNST is an excellent target for this new therapeutic strategy. Table 1 OVs used in PNST pre-clinical models Abbreviations: GFP, enhanced green fluorescent protein; h, human; IL, interleukin; LacZ, β-galactosidase; NIS, sodium/iodide symporter; pro, promoter; TIMP3, tissue inhibitor of metalloproteinases 3; IE4/5, immediate-early gene 4/5. Oncolytic Virus Genetic Modifications Transgene Drug Combination PNST Model Ref. oHSV G47Δ γ34.5Δ, ICP6-, ICP47Δ, LacZ+ MPNST orthotopic sciatic nerve, Transgenic schwannoma, Subcutaneous schwannoma 49, 50, 57, 65 G207 γ34.5Δ, ICP6- Subcutaneous and intraperitoneal human MPNST in athymic mice 52, 54 NV1023 UL56Δ/1 copy of ICP0, ICP4 and γ34.5 Perineural invasion model in sciatic nerve 60, 61 G207 γ34.5Δ, ICP6- Erlotinib Subcutaneous and intraperitoneal: human MPNST in athymic mice 54 rQT3 γ34.5Δ, ICP6- IE4/5pro-TIMP3, ICP6-GFP fusion in ICP6 MPNST xenograft models 55 hrR3 ICP6-, LacZ+ Subcutaneous and intraperitoneal human MPNST in athymic mice 54 oHSV-MDK-34.5 Mdk pro-γ34.5 and ICP6-GFP fusion in ICP6 Subcutaneous human MPNST in athymic mice 56 G47Δ-PF4 γ34.5Δ, ICP6-, ICP47Δ, ICP47pro-Us11 PF4 Subcutaneous MPNST xenograft, MPNST orthotopic sciatic nerve in immunocompetent mice 57 G47Δ-dnFGFR γ34.5Δ, ICP6-, ICP47Δ, ICP47pro-Us11 dnFGFR Subcutaneous MPNST xenograft, MPNST orthotopic sciatic nerve in immunocompetent mice 64 G47Δ-mIL12 γ34.5Δ, ICP6Δ mIL-12 MPNST orthotopic sciatic nerve in immunocompetent mice 65 Other OVs Ad5/3-D24-GMCSF Ad5/3 fiber, E1A 24-bpΔ, E3Δ hGMCSF doxorubicin, ifosfamide leiomyosarcoma cells in Syrian hamster 74 MV-NIS Edmonston vaccine strain NIS Subcutaneous human MPNST in athymic mice 77 VSV-rp30a Disruption of gene order In vitro human MPNST cell lines 81 Article Highlights Oncolytic viruses efficiently kill tumor cells while remaining safe for normal tissues. Oncolytic viruses, especially oncolytic HSV, have demonstrated efficacy and safety in treating PNST and perineural invasion in cancer. ‘Arming’ oncolyltic viruses with therapeutic transgenes or combining with other therapeutic agents enhances antitumor activity. A limited number of preclinical tumor models for PNST has slowed progress in the development of new therapeutic strategies. Preclinical studies with oncolytic HSV in MPNST and schwannoma support clinical translation. 1 Korf BR Neurofibromatosis Handb Clin Neurol 2013 111 333 40 23622184 2 July J Guha A Peripheral nerve tumors Handb Clin Neurol 2012 105 665 74 22230526 3 Porter DE Prasad V Foster L Dall GF Birch R Grimer RJ Survival in Malignant Peripheral Nerve Sheath Tumours: A Comparison between Sporadic and Neurofibromatosis Type 1-Associated Tumours Sarcoma 2009 756395 10 7 4 LaFemina J Qin LX Moraco NH Antonescu CR Fields RC Crago AM Oncologic outcomes of sporadic, neurofibromatosis-associated, and radiation-induced malignant peripheral nerve sheath tumors Annals of surgical oncology 2013 20 1 66 72 22878618 5 Lin AL Gutmann DH Advances in the treatment of neurofibromatosis-associated tumours Nat Rev Clin Oncol 2013 11 10 11 616 24 23939548 6 Fletcher CDM Bridge JA Hogendoorn P Mertens F WHO Classification of Tumours of Soft Tissue and Bone WHO Classification of Tumours 4th Geneva WHO Press 2013 7 Russell SJ Peng KW Bell JC Oncolytic virotherapy Nature biotechnology 2012 7 10 30 7 658 70 8 Guo ZS Thorne SH Bartlett DL Oncolytic virotherapy: molecular targets in tumor-selective replication and carrier cell-mediated delivery of oncolytic viruses Biochimica et biophysica acta 2008 4 1785 2 217 31 18328829 9 Chiocca EA Rabkin SD Oncolytic viruses and their application to cancer immunotherapy Cancer immunology research 2014 4 2 4 295 300 24764576 10 Lichty BD Breitbach CJ Stojdl DF Bell JC Going viral with cancer immunotherapy Nature reviews Cancer 2014 8 14 8 559 67 24990523 11 Angarita FA Acuna SA Ottolino-Perry K Zerhouni S McCart JA Mounting a strategic offense: fighting tumor vasculature with oncolytic viruses Trends in molecular medicine 2013 6 19 6 378 92 23540715 12 Everts B van der Poel HG Replication-selective oncolytic viruses in the treatment of cancer Cancer gene therapy 2005 2 12 2 141 61 15472714 13 Kaur B Cripe TP Chiocca EA “Buy one get one free”: armed viruses for the treatment of cancer cells and their microenvironment Current gene therapy 2009 10 9 5 341 55 19860649 14 Jeyaretna DS Kuroda T Recent advances in the development of oncolytic HSV-1 vectors: ‘arming’ of HSV-1 vectors and application of bacterial artificial chromosome technology for their construction Current opinion in molecular therapeutics 2007 10 9 5 447 66 17932809 15 Bell J McFadden G Viruses for tumor therapy Cell Host Microbe 2014 3 12 15 3 260 5 24629333 16 Galanis E Okuno SH Nascimento AG Lewis BD Lee RA Oliveira AM Phase I-II trial of ONYX-015 in combination with MAP chemotherapy in patients with advanced sarcomas Gene therapy 2005 3 12 5 437 45 * the only clinical trial so far reporting OV treatment of PNST. 15647767 17 First Oncolytic Viral Therapy for Melanoma Cancer discovery 2015 11 9 18 Andtbacka RH Kaufman HL Collichio F Amatruda T Senzer N Chesney J Talimogene Laherparepvec Improves Durable Response Rate in Patients With Advanced Melanoma Journal of clinical oncology : official journal of the American Society of Clinical Oncology 2015 9 1 33 25 2780 8 26014293 19 Hilton DA Hanemann CO Schwannomas and their pathogenesis Brain pathology 2014 4 24 3 205 20 24450866 20 Lloyd SK Evans DG Neurofibromatosis type 2 (NF2): diagnosis and management Handb Clin Neurol 2013 115 957 67 23931824 21 Sughrue ME Yeung AH Rutkowski MJ Cheung SW Parsa AT Molecular biology of familial and sporadic vestibular schwannomas: implications for novel therapeutics Journal of neurosurgery 2011 2 114 2 359 66 19943731 22 Stemmer-Rachamimov AO Xu L Gonzalez-Agosti C Burwick JA Pinney D Beauchamp R Universal absence of merlin, but not other ERM family members, in schwannomas The American journal of pathology 1997 12 151 6 1649 54 9403715 23 Plotkin SR Halpin C McKenna MJ Loeffler JS Batchelor TT Barker FG 2nd Erlotinib for progressive vestibular schwannoma in neurofibromatosis 2 patients Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology 2010 9 31 7 1135 43 24 Karajannis MA Ferner RE Neurofibromatosis-related tumors: emerging biology and therapies Current opinion in pediatrics 2015 2 27 1 26 33 25490687 25 Rodriguez FJ Folpe AL Giannini C Perry A Pathology of peripheral nerve sheath tumors: diagnostic overview and update on selected diagnostic problems Acta neuropathologica 2012 3 123 3 295 319 22327363 26 Ferner RE Gutmann DH Neurofibromatosis type 1 (NF1): diagnosis and management Handb Clin Neurol 2013 115 939 55 23931823 27 Ratner N Miller SJ A RASopathy gene commonly mutated in cancer: the neurofibromatosis type 1 tumour suppressor Nature reviews Cancer 2015 5 15 5 290 301 25877329 28 Pizem J Nicholson KM Mraz J Prieto VG Melanocytic differentiation is present in a significant proportion of nonpigmented diffuse neurofibromas: a potential diagnostic pitfall The American journal of surgical pathology 2013 8 37 8 1182 91 23715161 29 Jett K Nguyen R Arman D Birch P Chohan H Farschtschi S Quantitative associations of scalp and body subcutaneous neurofibromas with internal plexiform tumors in neurofibromatosis 1 Am J Med Genet A 2015 7 167 7 1518 24 25900062 30 Khosrotehrani K Bastuji-Garin S Riccardi VM Birch P Friedman JM Wolkenstein P Subcutaneous neurofibromas are associated with mortality in neurofibromatosis 1: a cohort study of 703 patients Am J Med Genet A 2005 1 1 132A 1 49 53 15523617 31 Brems H Beert E de Ravel T Legius E Mechanisms in the pathogenesis of malignant tumours in neurofibromatosis type 1 The Lancet Oncology 2009 5 10 5 508 15 19410195 32 Ducatman BS Scheithauer BW Piepgras DG Reiman HM Ilstrup DM Malignant peripheral nerve sheath tumors. A clinicopathologic study of 120 cases Cancer 1986 5 15 57 10 2006 21 3082508 33 Widemann BC Current status of sporadic and neurofibromatosis type 1-associated malignant peripheral nerve sheath tumors Curr Oncol Rep 2009 7 11 4 322 8 19508838 34 Robertson KA Nalepa G Yang FC Bowers DC Ho CY Hutchins GD Imatinib mesylate for plexiform neurofibromas in patients with neurofibromatosis type 1: a phase 2 trial The Lancet Oncology 2012 12 13 12 1218 24 23099009 35 Thway K Fisher C Malignant peripheral nerve sheath tumor: pathology and genetics Ann Diagn Pathol 2014 4 18 2 109 16 24418643 36 Hagel C Zils U Peiper M Kluwe L Gotthard S Friedrich RE Histopathology and clinical outcome of NF1-associated vs. sporadic malignant peripheral nerve sheath tumors Journal of neuro-oncology 2007 4 82 2 187 92 17111191 37 Kolberg M Holand M Agesen TH Brekke HR Liestol K Hall KS Survival meta-analyses for >1800 malignant peripheral nerve sheath tumor patients with and without neurofibromatosis type 1 Neuro-oncology 2013 2 15 2 135 47 23161774 38 Upadhyaya M Kluwe L Spurlock G Monem B Majounie E Mantripragada K Germline and somatic NF1 gene mutation spectrum in NF1-associated malignant peripheral nerve sheath tumors (MPNSTs) Human mutation 2008 1 29 1 74 82 17960768 39 Endo M Yamamoto H Setsu N Kohashi K Takahashi Y Ishii T Prognostic significance of AKT/mTOR and MAPK pathways and antitumor effect of mTOR inhibitor in NF1-related and sporadic malignant peripheral nerve sheath tumors Clinical cancer research : an official journal of the American Association for Cancer Research 2013 1 15 19 2 450 61 23209032 40 Grobmyer SR Reith JD Shahlaee A Bush CH Hochwald SN Malignant Peripheral Nerve Sheath Tumor: molecular pathogenesis and current management considerations Journal of surgical oncology 2008 3 15 97 4 340 9 18286466 41 Stucky CC Johnson KN Gray RJ Pockaj BA Ocal IT Rose PS Malignant peripheral nerve sheath tumors (MPNST): the Mayo Clinic experience Annals of surgical oncology 2012 3 19 3 878 85 21861229 42 Liebig C Ayala G Wilks JA Berger DH Albo D Perineural invasion in cancer: a review of the literature Cancer 2009 8 1 115 15 3379 91 19484787 43 Marchesi F Piemonti L Mantovani A Allavena P Molecular mechanisms of perineural invasion, a forgotten pathway of dissemination and metastasis Cytokine Growth Factor Rev 2010 2 21 1 77 82 20060768 44 Roh J Muelleman T Tawfik O Thomas SM Perineural growth in head and neck squamous cell carcinoma: a review Oral Oncol 2015 1 51 1 16 23 25456006 45 Bapat AA Hostetter G Von Hoff DD Han H Perineural invasion and associated pain in pancreatic cancer Nature reviews Cancer 2011 10 11 10 695 707 21941281 46 Roizman B Knipe DM Whitley RJ Herpes simplex viruses Knipe DM Howley PM Fields Virology Philadelphia Lippincott Williams & Wilkins 2013 1823 97 47 Jackson JD McMorris AM Roth JC Coleman JM Whitley RJ Gillespie GY Assessment of oncolytic HSV efficacy following increased entry-receptor expression in malignant peripheral nerve sheath tumor cell lines Gene therapy 2014 11 21 11 984 90 25119379 48 Peters C Rabkin SD Designing herpes viruses as oncolytics Molecular Therapy — Oncolytics 2015 2 15010 26462293 49 Messerli SM Prabhakar S Tang Y Mahmood U Giovannini M Weissleder R Treatment of schwannomas with an oncolytic recombinant herpes simplex virus in murine models of neurofibromatosis type 2 Human gene therapy 2006 1 17 1 20 30 * Use of novel schwannoma models to demonstrate therapeutic efficacy of oHSV in benign PNST. 16409122 50 Prabhakar S Messerli SM Stemmer-Rachamimov AO Liu TC Rabkin S Martuza R Treatment of implantable NF2 schwannoma tumor models with oncolytic herpes simplex virus G47Delta Cancer gene therapy 2007 5 14 5 460 7 17304235 51 Lee CH Chung CK Hyun SJ Kim CH Kim KJ Jahng TA A longitudinal study to assess the volumetric growth rate of spinal intradural extramedullary tumour diagnosed with schwannoma by magnetic resonance imaging Eur Spine J 2015 6 25 52 Mahller YY Rangwala F Ratner N Cripe TP Malignant peripheral nerve sheath tumors with high and low Ras-GTP are permissive for oncolytic herpes simplex virus mutants Pediatric blood & cancer 2006 6 46 7 745 54 16124003 53 Farassati F Pan W Yamoutpour F Henke S Piedra M Frahm S Ras signaling influences permissiveness of malignant peripheral nerve sheath tumor cells to oncolytic herpes The American journal of pathology 2008 12 173 6 1861 72 18988803 54 Mahller YY Vaikunth SS Currier MA Miller SJ Ripberger MC Hsu YH Oncolytic HSV and erlotinib inhibit tumor growth and angiogenesis in a novel malignant peripheral nerve sheath tumor xenograft model Molecular therapy : the journal of the American Society of Gene Therapy 2007 2 15 2 279 86 17235305 55 Mahller YY Vaikunth SS Ripberger MC Baird WH Saeki Y Cancelas JA Tissue inhibitor of metalloproteinase-3 via oncolytic herpesvirus inhibits tumor growth and vascular progenitors Cancer research 2008 2 15 68 4 1170 9 18281493 56 Maldonado AR Klanke C Jegga AG Aronow BJ Mahller YY Cripe TP Molecular engineering and validation of an oncolytic herpes simplex virus type 1 transcriptionally targeted to midkine-positive tumors The journal of gene medicine 2010 7 12 7 613 23 * Description of transcriptionally targeted OV strategy for selective growth in MPNST 20603890 57 Antoszczyk S Spyra M Mautner VF Kurtz A Stemmer-Rachamimov AO Martuza RL Treatment of orthotopic malignant peripheral nerve sheath tumors with oncolytic herpes simplex virus Neuro-oncology 2014 8 16 8 1057 66 ** Development of new orthotopic MPNST models and demonstration of oHSV efficacy in both human MPNST stem-like cells and in mouse immunocompetent tumors. First description of immunotherapeutic strategy to treat MPNST. 24470552 58 Spyra M Kluwe L Hagel C Nguyen R Panse J Kurtz A Cancer stem cell-like cells derived from malignant peripheral nerve sheath tumors PloS one 2011 6 6 e21099 21695156 59 Mashour GA Moulding HD Chahlavi A Khan GA Rabkin SD Martuza RL Therapeutic efficacy of G207 in a novel peripheral nerve sheath tumor model Experimental neurology 2001 5 169 1 64 71 11312559 60 Gil Z Rein A Brader P Li S Shah JP Fong Y Nerve-sparing therapy with oncolytic herpes virus for cancers with neural invasion Clinical cancer research : an official journal of the American Association for Cancer Research 2007 11 1 13 21 6479 85 * First description of OV to safely treat carcinoma PNI. 17975160 61 Kelly K Brader P Rein A Shah JP Wong RJ Fong Y Attenuated multimutated herpes simplex virus-1 effectively treats prostate carcinomas with neural invasion while preserving nerve function FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2008 6 22 6 1839 48 18234972 62 Geevarghese SK Geller DA de Haan HA Horer M Knoll AE Mescheder A Phase I/II study of oncolytic herpes simplex virus NV1020 in patients with extensively pretreated refractory colorectal cancer metastatic to the liver Human gene therapy 2010 9 21 9 1119 28 20486770 63 Kanai R Rabkin SD Combinatorial strategies for oncolytic herpes simplex virus therapy of brain tumors CNS Oncol 2013 3 2 2 129 42 23687568 64 Liu TC Zhang T Fukuhara H Kuroda T Todo T Canron X Dominant-negative fibroblast growth factor receptor expression enhances antitumoral potency of oncolytic herpes simplex virus in neural tumors Clinical cancer research : an official journal of the American Association for Cancer Research 2006 11 15 12 22 6791 9 17121900 65 Liu TC Zhang T Fukuhara H Kuroda T Todo T Martuza RL Oncolytic HSV armed with platelet factor 4, an antiangiogenic agent, shows enhanced efficacy Molecular therapy : the journal of the American Society of Gene Therapy 2006 12 14 6 789 97 ** First description of use of OV armed with therapeutic transgene to treat MPNST, targeting both the tumor cells and vasculature. 17045531 66 Farid M Demicco EG Garcia R Ahn L Merola PR Cioffi A Malignant peripheral nerve sheath tumors The oncologist 2014 2 19 2 193 201 24470531 67 Kanai R Wakimoto H Cheema T Rabkin SD Oncolytic herpes simplex virus vectors and chemotherapy: are combinatorial strategies more effective for cancer? Future Oncol 2010 4 6 4 619 34 20373873 68 Wildy P The progression of herpes simplex virus to the central nervous system of the mouse The Journal of hygiene 1967 6 65 2 173 92 20475878 69 Townsend JJ Collins PK Peripheral nervous system demyelination with herpes simplex virus Journal of neuropathology and experimental neurology 1986 7 45 4 419 25 3014069 70 Kirn D Oncolytic virotherapy for cancer with the adenovirus dl1520 (Onyx-015): results of phase I and II trials Expert opinion on biological therapy 2001 5 1 3 525 38 11727523 71 Blackford AN Grand RJ Adenovirus E1B 55-kilodalton protein: multiple roles in viral infection and cell transformation Journal of virology 2009 5 83 9 4000 12 19211739 72 Larson C Oronsky B Scicinski J Fanger GR Stirn M Oronsky A Going viral: a review of replication-selective oncolytic adenoviruses Oncotarget 2015 8 21 6 24 19976 89 26280277 73 Koski A Kangasniemi L Escutenaire S Pesonen S Cerullo V Diaconu I Treatment of cancer patients with a serotype 5/3 chimeric oncolytic adenovirus expressing GMCSF Molecular therapy : the journal of the American Society of Gene Therapy 2010 18 10 1874 84 20664527 74 Siurala M Bramante S Vassilev L Hirvinen M Parviainen S Tahtinen S Oncolytic adenovirus and doxorubicin-based chemotherapy results in synergistic antitumor activity against soft-tissue sarcoma International journal of cancer Journal international du cancer 2015 2 15 136 4 945 54 24975392 75 Russell SJ Peng KW Measles virus for cancer therapy Current topics in microbiology and immunology 2009 330 213 41 19203112 76 Msaouel P Opyrchal M Domingo Musibay E Galanis E Oncolytic measles virus strains as novel anticancer agents Expert opinion on biological therapy 2013 4 13 4 483 502 23289598 77 Deyle DR Escobar DZ Peng KW Babovic-Vuksanovic D Oncolytic measles virus as a novel therapy for malignant peripheral nerve sheath tumors Gene 2015 7 1 565 1 140 5 ** First description of use of MV to treat MPNST and demonstration of the utility of non-invasive imaging with NIS to follow virus spread in MPNST. 25843626 78 Griffin DE Measles virus and the nervous system Handb Clin Neurol 2014 123 577 90 25015505 79 Hastie E Grdzelishvili VZ Vesicular stomatitis virus as a flexible platform for oncolytic virotherapy against cancer The Journal of general virology 2012 12 93 Pt 12 2529 45 23052398 80 Wollmann G Tattersall P van den Pol AN Targeting human glioblastoma cells: comparison of nine viruses with oncolytic potential Journal of virology 2005 79 10 6005 22 15857987 81 Paglino JC van den Pol AN Vesicular stomatitis virus has extensive oncolytic activity against human sarcomas: rare resistance is overcome by blocking interferon pathways Journal of virology 2011 9 85 18 9346 58 21734048

PMC005xxxxxx/PMC5111816.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0413066 2830 Cell Cell Cell 0092-8674 1097-4172 27814522 5111816 10.1016/j.cell.2016.10.019 NIHMS823453 Article From Whole-Brain Data to Functional Circuit Models: The Zebrafish Optomotor Response Naumann Eva A. 13 Fitzgerald James E. 2 Dunn Timothy W. 12 Rihel Jason 3 Sompolinsky Haim 24 Engert Florian 125* 1 Department of Molecular & Cellular Biology, Harvard University, Cambridge, MA 02138, USA 2 Center for Brain Science, Harvard University, Cambridge, MA 02138, USA 3 Department of Cell and Developmental Biology, University College London, London WC1E 6BT, UK 4 Racah Institute of Physics and the Edmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem, Jerusalem 91904, Israel 5 Lead Contact * Correspondence: florian@mcb.harvard.edu 18 10 2016 3 11 2016 03 11 2017 167 4 947960.e20 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. SUMMARY Detailed descriptions of brain-scale sensorimotor circuits underlying vertebrate behavior remain elusive. Recent advances in zebrafish neuroscience offer new opportunities to dissect such circuits via whole-brain imaging, behavioral analysis, functional perturbations, and network modeling. Here, we harness these tools to generate a brain-scale circuit model of the optomotor response, an orienting behavior evoked by visual motion. We show that such motion is processed by diverse neural response types distributed across multiple brain regions. To transform sensory input into action, these regions sequentially integrate eye- and direction-specific sensory streams, refine representations via interhemispheric inhibition, and demix locomotor instructions to independently drive turning and forward swimming. While experiments revealed many neural response types throughout the brain, modeling identified the dimensions of functional connectivity most critical for the behavior. We thus reveal how distributed neurons collaborate to generate behavior and illustrate a paradigm for distilling functional circuit models from whole-brain data. Graphical Abstract INTRODUCTION How neurons across the brain collaborate to process information that ultimately guides behavior is not well understood. Most progress toward understanding comprehensive sensorimotor circuits has been made in invertebrates (Kato et al., 2015; Ohyama et al., 2015). For example, the processes by which environmental stimuli drive specific motor patterns in C. elegans (Bounoutas and Chalfie, 2007; Chalasani et al., 2007) and Drosophila (Borst et al., 2010; Ruta et al., 2010; Silies et al., 2014) have been dissected at the level of individual neurons and their synapses. In these cases, the stereotypy and small size of the invertebrate brain and the identifiability of its neurons were indispensable for precisely measuring neural circuit structure and dynamics. With a few notable exceptions (Heiligenberg and Konishi, 1991; Lisberger, 2010), isolating the neural circuits implementing vertebrate behavior has only been successful for peripheral reflexes (Fink et al., 2014; Korn and Faber, 2005). This is partly due to technical limitations, as information processing in vertebrates is typically coordinated by many neurons in various brain regions (Felleman and Van Essen, 1991). Consequently, most research has focused on microcircuits that are spatially localized and functionally coherent, e.g., within the retina (Masland, 2012) or cortex (Ko et al., 2011). This leaves the understanding of central brain mechanisms linking sensation to action incomplete. Moreover, neural response properties in vertebrates appear heterogeneous and redundant (Bianco and Engert, 2015; Rigotti et al., 2013), yet the origins and significance of this redundancy and response variability remain poorly understood (Sompolinsky, 2014). Without access to brain-scale neuronal activity in well-defined behavioral contexts, it seems unlikely that we will understand why the vertebrate brain has, despite immense energy constraints, evolved its heterogeneous, multilayered architecture. The larval zebrafish is a small and translucent vertebrate, permitting cellular resolution optical imaging of nearly all its ~100,000 neurons (Ahrens et al., 2013a). Recent studies using brain-wide imaging have revealed that sensorimotor processing is indeed widely distributed across the entire zebrafish brain (Ahrens et al., 2012, 2013a; Portugues et al., 2014). However, the overwhelming complexity of whole-brain imaging data has made it difficult to establish links between brain- and circuit-level descriptions of behavior. Such links must be established by applying both experimental and theoretical approaches to behaviors that are simple enough to characterize in mechanistic detail but sophisticated enough to engage complex sensorimotor transformations. The zebrafish optomotor response (OMR), a position-stabilizing reflex to whole-field visual motion, offers an opportunity to dissect the central circuits underlying such a behavior using closed-loop stimulus delivery, detailed behavioral analysis, whole-brain imaging, functional perturbations, and network modeling. These approaches allowed us to follow the flow of visual information through nuclei that integrate binocular retina input to motor centers that drive turning and forward swims, resulting in a functional whole-brain circuit model that describes the sensorimotor transformation in terms of experimentally observed neural response types. RESULTS The Optomotor Response Is Driven by Egocentric Processing of Optic Flow For binocular animals to perform the OMR, the brain must integrate motion signals from each eye. We thus sought to dissect this sensorimotor transformation using independent stimulus presentation to each eye. A closed-loop system allowed us to continuously update the stimulus in each monocular visual field (Figure 1A; Movie S1), ensuring that fish continually experienced a particular motion pattern relative to the body axis. For leftward or rightward motion, we labeled these component monocular stimuli egocentrically, with medial motion going toward the midline and lateral motion going away from the midline (Figure 1A, right). Following a psychophysics approach, we presented stimuli from a matrix of all possible combinations of static (no motion), medial, and lateral motion, along with forward and backward motion (Figure 1B). By comparing the behavioral responses to monocular (i.e., motion to one eye only), coherent (e.g., leftward motion comprising medial “approaching” motion to the right eye and lateral “leaving” motion to the left eye), and conflicting stimuli (i.e., “inward” medial-medial motion or “outward” lateral-lateral motion to both eyes), we quantified basic algorithmic properties by which the brain combines binocular inputs to generate a motor output. As expected, leftward and rightward coherent stimuli led to leftward and rightward turning, respectively (Figures 1C and S1A–S1C). In addition, average turning was larger to coherent motion than to monocular stimuli and were similar to the sum of the responses to each monocular component (Figure 1C). Interestingly, the contributions of medial and lateral components were not equal; medial contributed substantially more to turning than lateral. Furthermore, response latencies to stimuli containing medial motion were shorter than to stimuli containing only lateral motion (Figure S1D). Together, these observations reveal several algorithmic properties of the OMR, which are necessarily implemented in the brain: (1) the direction of whole-field coherent motion dictates the direction of behavioral output; (2) the average turning response to coherent motion approximates the sum of responses to medial plus lateral motion; and (3) despite representing environmentally equivalent motion cues, medial and lateral motion are differentially processed by the zebrafish brain to generate different behaviors. Eye- and Direction-Specific Motion Differentially Affect Forward Swimming and Turning Examination of discrete swim bouts (Figures S1E and S1F) revealed that histograms of individual bout angles in response to each stimulus were typically trimodal (Figures 1D, S1G, and S1H), with a central peak corresponding to forward swimming and left/right peaks corresponding to left/right turning. As these peaks changed dramatically across the stimulus set, we focused our analyses on their amplitude changes. Coherent stimuli enhanced turning in the direction of motion (“correct”) relative to the baseline static condition, while turning in the opposite direction of motion (“incorrect”) was suppressed. Similar to coherent stimuli, monocular medial stimuli increased forward swims and correct turns while suppressing incorrect turns (Figure 1D, inset). In contrast, monocular lateral stimuli slightly decreased forward swimming and modulated turning less than medial stimuli. Hidden by averaged turning responses (Figures S1B and S1C), this analysis revealed fundamental differences in the behavioral responses to conflicting stimuli. While each conflicting stimulus eliminated the turning responses observed under monocular stimulation (Figure 1D), inward stimuli enhanced forward swimming, resulting in behavior that looked much like that in response to forward motion, and outward stimuli suppressed forward swimming, mirroring backward motion responses. These data thus establish two additional algorithmic properties of the OMR. First, visual motion drives correct turns and suppresses incorrect turns, and these effects are both stronger for medial than for lateral motion. Second, turning and forward swimming are modulated separately, with medial motion driving swimming, akin to forward motion, and lateral motion suppressing swimming, akin to backward motion. To illustrate the minimal computational requirements for these observed sensorimotor transformations, we implemented a neural circuit architecture in which direction-selective retinal ganglion cells (DSRGCs) send monocular motion signals to premotor neurons presumed to independently drive peak frequencies of turning (T) and forward swimming (F, central diagram, Figure 1E). These premotor neurons represent ventromedial spinal projection neurons (vSPNs, T) (Huang et al., 2013) and the nucleus of the medial longitudinal fasciculus (nMLF, F) (Orger et al., 2008). Forward swimming was well modeled in this architecture as the sum of monocular components (left panels, Figure 1E), with medial motion driving swimming and lateral motion suppressing it. Although turning frequencies depended nonlinearly on monocular components, a biologically plausible input-output nonlinearity could reproduce the required transformation (right panels, Figure 1E). Thus, as few as three dedicated neurons could conceivably transform direct monocular inputs into the observed pattern of forward swimming and turning, concretely instantiating the aforementioned algorithmic properties of the OMR (Figures S1H and S1I). Whole-Brain Imaging Identifies Regions Activated during the OMR To map brain areas for response features consistent with revealed algorithmic properties, we measured neural activity to visual motion using two-photon calcium imaging in transgenic Tg(elavl3:GCaMP5G) zebrafish. Though animals were restrained during imaging, we made motor nerve recordings (Ahrens et al., 2013b) in a subset of fish and observed a pattern of bout frequency modulation that matched freely swimming fish (Figures S2A and S2B). Combined with known zebrafish neuroanatomy (Figure 2A) (Burrill and Easter, 1994), these whole-brain maps highlight the route of information flow. Because the OMR was modulated by the overall direction of motion (Figures 1C, 1D, S1A–S1C), we first examined the spatial distribution of direction-selective neural units (Figure 2B), which revealed that motion information was distributed and segregated across a handful of brain areas (Figure 2C). Of the ten arborization fields (AFs) of retinal axons (Robles et al., 2014), AF6 and 10 were strongly activated by motion stimuli (Figures 2B, S2C, and S2D). Near AF6, responses in the pretectum (Pt) (Figures 2B and 2C) were lateralized, with the left Pt responding primarily to leftward motion, and the right Pt responding to rightward motion. Responses to forward motion were bilaterally distributed across the Pt. Directional signals were demixed in the hindbrain and midbrain. In particular, forward-selective neurons were anatomically segregated from lateralized left- and right-selective neurons. This response segregation is reminiscent of the behavioral segregation of forward swimming and turning. To visualize binocular integration of coherent monocular signals, we generated maps overlaying responses to monocular medial and lateral stimuli presented sequentially to the eyes (Figures 2D and S2D). While responses in the AFs were exclusively monocular, neurons in the Pt exhibited the binocular integration required of the OMR, and downstream neurons in the anterior hindbrain (aHB) and vSPNs displayed qualitatively similar binocular responses (Figures S2D and S2E). These response maps suggest a qualitative brain-scale functional-anatomical model underlying the zebrafish OMR (Figure S2F). To move beyond map-making and characterize the local computations, we focused analyses within specific brain regions. STARting at the sensory end, we extracted fluorescence time series from motion-sensitive regions containing retinal terminals. As AF10 ablations spare the OMR (Roeser and Baier, 2003), we concentrated on AF6. Consistent with completely decussated retinal projections, responses in AF6 were exclusively monocular (Figure 2E, top). AF6 also showed localized regions of highly tuned directional responses for all direction of motion, roughly matching the size of synaptic boutons (>1 μm2) (Figure 2F, top). To test whether information arriving in AF6 is necessary for the OMR, we ablated it unilaterally and found behavioral deficits specific to contralateral eye stimulation (Figures S3A–S3C). Thus, AF6 is an important site of retinal input to the OMR circuit. To generate the OMR, neurons must integrate this monocular information to form binocular representations. We observed neurons in the Pt (Figure 2E, bottom) that showed sustained, binocular responses and sent neurites into the ipsilateral AF6 (Figure S3D). In addition, average Pt activity reflected several algorithmic properties of the behavior, including ipsilateral inhibition to medial stimuli (Figures S3E–S3G). Moreover, the Pt showed enhanced and lateralized direction selectivity (Figure 2F, bottom; Figure S3F) and receives direct retinal input in other species (Gamlin, 2006). The Pt is thus well positioned to support the sensorimotor transformation underlying the OMR. Pretectal Activity as a Neural Correlate of Optomotor Behavior The Pt exhibited considerable diversity at the neuronal level, with individual neurons responding in various patterns to monocular, coherent, and conflicting motion (Movie S2; Figure S3F). To investigate the representation, we analyzed individual neurons (3,070 Pt neurons, 12 fish). If the Pt coordinates the OMR, it must encode all directions of motion (Orger et al., 2008). Consistent with this, most neurons were strongly tuned to a particular direction of motion (Figure 3A, left), with forward-tuned neurons clustered near the midline (Figure 3A, right). Most neurons were binocularly excited (Figure S3H). These binocular neurons preferred medial to lateral motion and linearly integrated the monocular components of whole-field motion (Figure 3B). Most Pt neurons showed at least partial suppression to conflicting motion (Figure 3C, left), mirroring the reciprocal suppression of turns seen behaviorally. Finally, we found that, much like the behavior, Pt neuron responses to forward and inward motion were associated (Figure 3D). Similar associations were seen between backward and outward motion. Together, these data provide neural correlates for many OMR algorithmic properties, suggesting that the Pt links motion information from the retina to appropriate motor actions. Posterior Commissure Ablation Disrupts Binocular Integration As retinal input is exclusively monocular, binocular Pt response properties (Figures 3B and 3C) require both excitatory and inhibitory signals to re-cross the midline. In particular, coherent motion responses require interhemispheric excitation from lateral motion, and negated responses to inward stimuli require interhemispheric inhibition from medial motion. A strong anatomical candidate for these connections is the posterior commissure (PC) (Figure 4A), which passes between the left and right Pt and has been proposed to carry motion information in rodents (Giolli et al., 1984). After confirming that Pt neurons project through the PC (Figure S4A), we laser-ablated the commissure (Figure 4B, left; Figure S4B). As expected, when we imaged Pt responses before and after PC ablation, both average Pt and single neuron Pt responses to inward motion were enhanced (Figure 4B, right), and responses to stimuli with lateral components were attenuated. We also measured behavior following PC ablation and found that contributions of lateral motion to average turning diminished (Figure 4C), as responses to monocular lateral were eliminated, and differences between coherent and monocular medial were reduced (Figure 4C, inset). This was not a non-specific perturbation, as overall bout frequencies were indistinguishable from controls (Figure S4C). Instead, these deficits resulted from a reduction in both correct and incorrect turn modulation in PC-ablated fish (Figures 4D and 4E). In response to inward motion, PC-ablated fish did not exhibit behavioral alternation between leftward and rightward turning, as might be expected from two lateralized medial turning signals that could not inhibit each other. Thus, although the PC carries some medial inhibition, additional inhibition must be communicated elsewhere. Moreover, we observed that forward swimming in response to medial was reduced (Figure S4C, inset). This suggests that some forward swim drive is relayed through the PC. Simulating PC ablation in the minimal model (Figure 1E) further supported these hypotheses (Figure S4E). Combining the PC ablation results with anatomy (Figure S4F) suggests that medial signals bilaterally activate, via the PC, the downstream region known to drive swimming (i.e., the nMLF) and that the suppression of incorrect turns by medial motion should occur via additional interhemispheric connections (Figure 4F). Most Pt Neurons Can Be Classified into a Few Functional Response Types Our results suggest that Pt neurons separately drive downstream circuits that control forward swimming and turning. To investigate how response heterogeneity across the Pt population might support these parallel functions, we first classified Pt neurons into functional response types (Figures 5A and S5A–S5C; Table S1; STAR Methods). Eight bilaterally symmetric (Figure S5D) response types occurred with elevated frequency, together comprising more than 66% of the Pt population. Four of these eight classes were binocularly activated by both medial and lateral stimuli (Figures 5B, top, 5C, left), and the remaining “monocular” classes responded to at most one monocular stimulus (Figures 5B, bottom, 5C, right). Within the binocular group, all classes responded most strongly to coherent stimuli (Figure 5B), but they were distinguishable by patterns of inhibition revealed only by their responses to conflicting stimuli. For example, response type ioB, a binocular neuron (ioB) exhibiting responses to conflicting inward (ioB) and outward (ioB) motion, responded to both conflicting stimuli. On the other hand, response type B (binocular neuron) did not respond to either conflicting stimulus, despite responding to both monocular stimuli. This suggests that the left-selective B type might play a specific role in generating left turns, as conflicting stimuli do not enhance turning. Consistent with this role, the B type was selective for directions of motion associated with turning (Figure 5B, far right). The monocular classes were distinguished both by patterns of excitation in response to monocular medial or lateral stimuli and by inhibition in response to conflicting motion. For example, response class iMM was a monocular (iMM) neuron, responsive to medial motion (iMM), and activated by inward motion (iMM). Because it showed a strong preference for forward motion and responded to all stimuli that promoted forward swimming, it is a good candidate to drive this behavioral module. In order to generate all of the observed Pt response types, we predict the existence of at least three types of relay units (Figure 5D). First, excitatory oML neurons might cross the midline to generate binocular responses. Second, inhibitory oML neurons could suppress responses to conflicting outward motion in B and iB neurons. Third, inhibitory iMM neurons could project across the midline to suppress responses to conflicting inward stimuli. Finally, an excitatory iMM neuron may relay retinal signals within the Pt, as suggested by inhomogeneous retinal innervation across the Pt in other species (Gamlin, 2006). Altogether, the overrepresented response types compactly summarize the neural correlates of forward swimming and turning behaviors. Network Modeling Demonstrates Requirements of Downstream Processing With this response diversity, the Pt could, in theory, implement the minimal model derived from behavior (Figure 1E). However, no single overrepresented response class exhibited an activity profile that precisely matched the behavior. For instance, neurons directly driving turning should not respond to backward and forward motion, as neither of these stimuli significantly increased turning, yet every Pt response type was activated by backward and/or forward stimuli. Furthermore, although the B type responded most similarly to the turning behavior, many other response types reproduced features of turning, and no type predicted the suppression of incorrect turns. This reinforces our hypothesis that turning opponency occurs downstream of the Pt. The observed response heterogeneity, together with the lack of a single response type that could explain the behavior, motivated an interpretation of the data wherein a combination of several response types underlies the OMR. To quantitatively formulate this hypothesis, we applied neuronal network modeling. Since the major Pt neural types correlated strongly with each other and with behavior, it is a priori possible that any of them contributed to the behaviors. We thus included all overrepresented classes in the model network (Figure 5E, left). Because representations were lateralized, we assumed that the left and right Pt drove leftward and rightward turning, respectively. However, since none of the major Pt response types were appreciably suppressed below baseline levels by opposing motion signals, we did not force the Pt network to suppress incorrect turns (Figure 5E, right). Aside from this limitation, the model architecture was able to account for the behavior. Hindbrain Circuitry Refines Pt Representations To complete the model, we examined other brain regions revealed by whole-brain imaging (Figures 2B and S6) and investigated how Pt signals are refined. Recall that neurons preferring stimuli that promote forward swimming were anatomically segregated in the nMLF, RoL, and cHB (Figures 2A–2C andS7A). We found that monocular medial stimuli bilaterally recruited these regions (Figure S7B), explaining the reduction in forward swim frequency observed after PC ablation. Neurons responding to swim-promoting stimuli were sparsely distributed within these regions (Figure S7C) and could ultimately drive forward swimming. Elsewhere in the hindbrain, especially in the aHB and vSPNs, neurons preferred stimuli associated with turning (Figure 6A). Neurons in the aHB are anatomically positioned between visual neurons in the Pt and premotor vSPNs and were thus good candidates for mediating the additional interhemispheric inhibition necessary to suppress incorrect turns. To test this hypothesis, we first classified aHB neurons into functional response types. We found four overrepresented types that were a subset of the Pt types (Figures 6B and S7D–S7F; Table S2) and hypothesize that only this subset of Pt neurons sends projections to the aHB. Intriguingly, the aHB representation enhanced the B type, the strongest functional candidate for driving turns. Furthermore, the B type clustered within the Pt and aHB (Figure 6C1). Neurons ultimately driving turning should not respond to either forward or backward motion. We found that B neurons responding to forward and backward motion were enriched in the Pt (Figure 6C2), and B neurons lacking these responses were enriched in the vSPN region (Figure 6C3). More interestingly, anatomically organized subpopulations of aHB and anterior rhombencephalic turning region (ARTR) neurons were associated with the Pt or vSPN representation (Figures 6C2 and 6C3). We refer to these sub-regions as the early (EHB) and late (LHB) hindbrain, respectively, as the EHB is functionally associated with visual areas (i.e., Pt) and the LHB is associated with motor areas (i.e., vSPNs). Many of the B neurons in the LHB were suppressed to levels not seen in the Pt but expected from the behavioral output (Figure 6D3, right), indicating that necessary interhemispheric inhibition lies within the hindbrain. By dissecting these hindbrain circuits, we provide a critical link between visual processing in the Pt and behavior. Quantitative Model for Pt and aHB Control of Behavior These considerations lead us to a model of how the Pt generates forward swimming and the Pt and aHB cooperate to generate turning (Figure 7A). Neurons in each major Pt response category drive premotor neurons controlling forward swimming. Since brain responses associated with forward swimming are bilaterally symmetric (Figure S7A), we assume that the left and right Pt contribute symmetrically (green lines, Figure 7A). Thus, the transformation of Pt signals into forward swimming behavior consists of eight functional connection weights, from each Pt response type to premotor nMLF units. To refine turn-related signals, we suppose that a subset of Pt neurons project to aHB. Both EHB and LHB receive ipsilateral inputs from the four Pt types common to the aHB (Figure 6B). LHB also receives contralateral input from EHB. Thus, the transformation of right Pt and left EHB signals into rightward turning consists of eight functional connection weights onto neurons in right LHB, which then drives turn generation (Figure 7A). The model is symmetric across the midline for leftward turning and accounts for the missing turn suppression (Figure S7G, cf. Figure 5E). Identifying Critical Dimensions of Functional Connectivity The experimentally informed circuit architecture (Figure 7A) can account for the behavioral responses. However, one would not expect the model to predict every feature of the functional connectivity that might be revealed by future experiments, as current experiments only provide a low-dimensional view of the brain (Gao and Ganguli, 2015). For example, our stimuli did not isolate behavioral drive to individual functional response types, as multiple sets of connections (Figure S7H, top) lead to similar behavioral predictions (Figure S7H, bottom). Nevertheless, a successful model should accurately predict behavior in experiments probing functional elements well-constrained by the current data. To identify critical determinants of model performance, we examined the full ensemble of models that linearly generate sensorimotor transformations via the circuit architecture of Figure 7A (STAR Methods). The model architecture could accurately predict the behavioral data using a range of functional connections (Figure 7B), and even the signs of connections were ambiguous. Such flexibility is common in multiparameter models when correlated parameter changes compensate to avoid error (Fisher et al., 2013; O’Leary et al., 2015). Individual connections could compensate in our model (Figure 7C, left), so we identified connectivity patterns that independently affect model accuracy (Figure 7C, right). We term these patterns functional modes. Figure 7D represents the functional modes as columns in a matrix, where each row represents a response type in Pt for forward swims (Figure 7D, left) and in Pt or EHB for turns (Figure 7D, right). For example, the eighth functional mode of swimming (8F) positively weighted connections from every response type (Figure 7D, left). The modes are ordered so that parameter changes along high-numbered modes induce large errors (Figure 7D, top). The models in Figure S7H performed similarly because they differed along the first mode. The functional modes can also be interpreted as population activity patterns that contribute independently to the accuracy of the sensorimotor transformation, so we could calculate the fraction of error eliminated by each mode (Figure 7D, bottom). A few modes eliminated the vast majority of model error, and the signs of abstract functional connections from these modes were reliably determined and positive. The seventh functional mode of swimming (7F) eliminated the most error, while 6F and 8F repaired discrepancies. Similarly, the seventh mode of turning (7T) was most important, while 5T refined model predictions. Thus, one significant mode mostly explained each behavior, but secondary modes helped shape detailed behavioral patterns. Similar considerations reveal reliable optogenetic predictions. The model predicts that activation of a response type or functional mode will cause a behavioral response if there was a significant functional connection between them. The model also predicts that uniform activation of left Pt would drive forward swimming (p = 0.002), drive leftward turning (p = 2.7 × 10−13), and suppress rightward turning (p = 0.0018). Selective activation of all binocular response classes in left Pt would likely suppress swimming (p = 0.11), drive left turning (p = 4.8 × 10−12), and have no effect on right turning (p = 0.41). Selective activation of all monocular classes in left Pt would likely drive swimming (p = 0.051), suppress left turning (p = 0.0069), and have no effect on right turning (p = 0.48). Finally, our model predicts that Pt is functionally multiplexed, with different aspects of the population response driving the behaviors. For example, mode 7F weighted response types according to their preference for inward versus outward motion (Figure 7E), which made this mode a component of the population response that was activated by stimuli promoting forward swimming and suppressed by stimuli reducing forward swimming (Figure 7F). Importantly, the model also predicts that certain aspects of the population response are functionally irrelevant. The irrelevance of 8T (Figure 7D, right) means that matched responses in right Pt and left aHB do not drive behavior. Nevertheless, such activity characterized several stimulus conditions (e.g., forward motion). Unsupervised techniques for parsing population activity risk equating response frequency with relevance. Since brains must multiplex their functions in ethological conditions, it is critical that future work integrate model and experiment to align responses to specific functions. DISCUSSION Here, we combine experiment and theory to delineate a sensorimotor transformation within the zebrafish brain, from sensory input to locomotor output. By sequentially evaluating possible models with behavioral, functional, and anatomical data, we arrived at a quantitative whole-brain circuit model that accounts for the observed aspects of the OMR using experimentally identified neural response types. We note that analyses at the behavioral, whole-brain, and systems levels were critical for defining a biologically plausible and functionally relevant circuit architecture. Yet, we acknowledge that our treatment ignores processing in the retina and spinal cord. Dedicated experiments to decipher these peripheral mechanisms will complement our model of central processing. Overall, our methodology establishes a roadmap for digesting large-scale neural data from vertebrate brains and raises fundamental questions that are broadly relevant to whole-brain descriptions of function and behavior. Relevance of Eye-Specific Motion Processing We dissected the routes of sensory input by decomposing motion egocentrically. For a given eye, left- and rightward motion should, a priori, be perceptually symmetric. Yet, sensorimotor circuitry maintained asymmetries (i.e., lateral versus medial) throughout. These functional asymmetries might survive from an evolutionary past in which eye-specific pathways did not converge, reflecting vestigial functional inefficiencies: only lateral signals must re-cross the midline to drive turning behavior. The large inhibitory influence of medial motion on contralateral turning might be a corresponding consequence of the early emphasis on medial pathways. Instead, medial motion might be privileged due to the ethologically relevance of approaching motion in other behaviors, and reciprocal inhibition from opponent motion signals may help stabilize motor outputs in noisy environments (Mauss et al., 2015). Homologs of Optomotor Response Circuitry Several features of the zebrafish OMR likely translate to processing in higher-order vertebrates. The mammalian anatomy of retinal projections and the Pt suggests that binocular integration may occur via the PC (Sun and May, 2014). The zebrafish OMR is established by separate visual channels controlling multiple behavioral parameters. As visual information is organized into channels in other brains (Gollisch and Meister, 2010; Huberman et al., 2009; Yonehara et al., 2009), we hypothesize that distinct channels may activate segregated motor nuclei to generate flexible behavior in other vertebrates. In particular, the nMLF or neighboring neurons may be functional homologs of the mesencephalic locomotor region in mammals (Dunn et al., 2016; Ryczko and Dubuc, 2013; Severi et al., 2014), whereas the vSPNs are functionally and anatomically similar to neurons in the mammalian reticular formation (Armstrong, 1988). Furthermore, the prevalence of caudal interhemispheric connections in the mammalian brain (Chédotal, 2014) is reminiscent of the zebrafish hindbrain, and these connections might also refine lateralized motor streams in higher-order vertebrates. In zebrafish, OMR circuitry overlaps with circuits activated in other stabilizing reflexes, such as the optokinetic reflex (OKR) of the eyes (Kubo et al., 2014; Portugues et al., 2014). In particular, neurons in the Pt are consistently recruited by stimuli that evoke the OKR, but binocular neural activation is far less common than during the OMR, perhaps suggesting that different subpopulations coordinate each behavior. Notably, Kubo et al. also identified frequent response types in the Pt and hypothesized a Pt circuit that could generate the response types (cf.Figure 5D). Quantitative Modeling To understand the requirements of the system, we first built a minimal model to explain the measured sensorimotor transformation (Figure 1E). As this model conflicts with anatomical and functional details of the brain, we reassessed the model with each new result to eventually obtain a candidate whole-brain description (Figure 7G). This iterative migration between experiment and theory will be necessary to illuminate the mechanisms of complex sensorimotor transformations, which are underconstrained by conceivable experiments (Gao and Ganguli, 2015). Functional similarity between neurons resulted in a model whose output was insensitive to some aspects of connectivity. This will be common going forward, as neuroscientists typically characterize high-dimensional systems using a few stimuli and behaviors (Fisher et al., 2013; Gao and Ganguli, 2015; O’Leary et al., 2015). Nevertheless, a few combined parameters were critical to explain the behavior, thereby distilling the behaviorally relevant features of the population response. Thus, although we did not assign unique functional roles to each response class, we could identify model features that were well constrained by the data to make testable predictions for collective encoding. The model architecture also makes predictions for connectivity between Pt response classes, aHB classes, and premotor nuclei that can be tested with viral tracing, connectomics, channelrhodopsin-assisted circuit mapping, and electrophysiology. Validation of Microcircuitry Our model uses DSRGCs and monocular relays to construct the overrepresented Pt response types. All retinal input arrives in AF6 from the contralateral visual field, so excitatory and inhibitory relays are necessary for binocular integration and reciprocal suppression. Not all Pt neurons receive direct retinal input in other animals (Fite, 1985; Koontz et al., 1985; Scalia, 1972; Vanegas and Ito, 1983), and relays might distribute input signals throughout the Pt. Future experiments measuring coarse projections, detailed connectivity, and neurotransmitter identity of candidate relay are necessary. We predict that lateral relay neurons (oML) exist in excitatory and inhibitory forms that differ in their projection patterns. One could use transgenic lines specifically labeling glutamatergic and GABAergic populations to test this hypothesis. Our model also predicts that LHB neurons are excited by specific ipsilateral Pt response types and inhibited by contralateral EHB neurons. The latter prediction is consistent with known organization in the zebrafish hindbrain (Kinkhabwala et al., 2011; Randlett et al., 2015; Sassa et al., 2007) and with the prevalence of commissures in the rhombencephalon (Koyama et al., 2011; Lee et al., 2015), but the detailed organization proposed here must be verified. Furthermore, a rigorous definition of EHB and LHB requires further experiments to distinguish each structure anatomically and functionally. A Generalizable Framework We provide an eminently testable framework for future studies and a detailed description of the neural circuitry underlying the OMR. We chose the OMR because it is an accessible gateway to a fundamental sensorimotor transformation layered with rich complexity. While we address the feed-forward foundation of the OMR with a simple stimulus set, the circuit properties revealed here provide a basis for future studies. Indeed, complementary studies of visuomotor behaviors that aim to explain learning, decision making, and multimodal processing must first ground themselves in the circuits controlling basic behaviors. More generally, our study establishes a computational and experimental scaffold for generating testable circuit models on large scales. Our work in larval zebrafish provides a critical link between efforts in smaller model organisms and those in vertebrates, an essential step toward tackling the complexity of mammalian brains. STAR*METHODS KEY RESOURCES TABLE REAGENT or RESOURCE SOURCE IDENTIFIER Chemicals, Peptides, and Recombinant Proteins Alpha-Bungarotoxin Life Technologies B1601 Deposited Data Pretectal Response Distribution. See Table S1 This study N/A Anterior Hindbrain Response Distribution. See Table S2 This study N/A Experimental Models: Organisms/Strains Tg(elavl3:GCaMP5G) (Ahrens et al., 2013a) N/A Tg(elavl3:GCaMP2) (Ahrens et al., 2013b) N/A Tg(alpha-tubulin:C3PA-GFP) (Ahrens et al., 2013b) N/A Software and Algorithms Labview (Behavioral acquisition framework) National Instruments http://www.ni.com/en-us.html MATLAB (Behavioral and imaging analysis, modeling) MathWorks http://www.mathworks.com/?requestedDomain=www.mathworks.com Open GL (Stimulus rendering) OpenGL https://www.opengl.org/ C# (.NET Framework 3.5) (Two-photon acquisition) Microsoft/ (Ahrens et al., 2013b) http://www.microsoft.com/en-us/ FIJI (ImageJ) (Neural tracing) NIH http://fiji.sc Automatic ROI segmentation (Portugues et al., 2014) / This study N/A CONTACT FOR REAGENTS AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to the Lead Contact Florian Engert (florian@mcb.harvard.edu). EXPERIMENTAL MODEL AND SUBJECT DETAILS Zebrafish For all experiments, we used 5-7 days post-fertilization (dpf) wild-type (AB or TL) or transgenic nacre −/− zebrafish; nacre −/− mutants lack pigment in the skin but retain wild-type eye pigmentation. All measured behavioral variables in nacre −/− fish were similar to wild-type, heterozygous siblings. Zebrafish were maintained on a 14 hr. light /10 hr. dark cycle and fertilized eggs were collected and raised at 28.5 °C. Embryos were kept in E3 solution (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4). All experiments were approved by Harvard University’s standing committee on the use of animals in research and training. All imaging experiments in this study were performed on transgenic zebrafish Tg(elavl3:GCaMP5G), a generous gifts from Drs. Michael Orger, Drew Robson and Jennifer Li or in double transgenic Tg(elavl3:GCaMP5G); Tg(α-tubulin:C3PA-GFP). The Tg(α-tubulin:C3PA-GFP) was kindly provided by Dr. Michael Orger. Tg(elavl3:GCaMP2), also a gift from Drs. Michael Orger, Drew Robson and Jennifer Li was used as anatomical reference in laser ablation of the posterior commissure. METHOD DETAILS Behavior in freely swimming zebrafish Zebrafish swam freely in a clear 5 cm diameter petri dish (VWR) in filtered E3 solution at 3-5 mm height, minimizing variability of visual angle. Illumination of the fish was achieved by a circular array of infrared light-emitting diodes (810 nm) directed from below. Fish behavior to various motion stimuli was recorded at 200 Hz using an infrared-sensitive, high-speed, monochrome charge coupled device (CCD) camera (Pike F-032, 1/3,” Allied Vision Technology, Germany). A zoom lens (Edmund Optics, USA) was used with an infrared pass filter (RG72, Hoya, Japan). The center of the dish was aligned with the center of an appropriately zoomed camera image such that the edge of the dish was not visible in the field of view; this simplifies the fish detection algorithm by not having to handle sidewall reflections. Stimuli were projected onto a 12 by 12 cm diffusing screen after reflection by a 4 × 4 inch cold mirror (both Edmund optics, USA) 5 mm directly below the fish using a commercial DLP projector (Optoma, USA). Custom image processing software (Labview, National Instruments, USA and Visual C++, Microsoft, USA) extracted the position (center of mass) and orientation (vector anchored by detection of swim bladder and center of mass between eyes) of the fish at the camera acquisition frame rate (200 Hz) from a background-subtracted frame. This information was used to continuously update a stimulus rendered in real-time using OpenGL to provide consistent motion stimulation with respect to the body axis. To begin each trial, fish were induced to swim to the center of the camera field of view by concentric circular, converging sinusoidal gratings (Movie S1). Detection of the fish near the center triggered a trial. Each trial began with a 500 ms static period in which gratings (running parallel relative to the fish body axis, spatial period of 1 cm) were locked to fish orientation in a closed loop configuration. These orientation-locked gratings then began to move at 10 mm/s (or remained stationary throughout the trial, depending on the particular stimulus identity, Movie S1). Stimulus presentation continued until either the fish aborted the trial early by leaving the active area or the maximum trial duration (30 s) was reached. To independently stimulate each eye, a 0.8 cm area was blanked (black) directly underneath the fish. In control experiments, we widened the blanked area and did not observe any effect on behavioral outcome (data not shown); this is consistent with independent stimulation of each eye. Only fish that completed at least 10 repetitions in less than 3 hr were included in further analysis or ablation experiments. Extracted the position and orientation were further analyzed as described in Quantification and Statistical Analysis. Fictive behavior Transdermal motor nerve recordings of fictive behavior were performed as outlined in (Ahrens et al., 2013b). Larval zebrafish (5-7 dpf) were paralyzed by immersion in a drop of fish water with 1 mg/ml alpha bungarotoxin (Sigma-Aldrich) and embedded in a drop of 2% low melting point agarose, after which the tail was freed by cutting away the agarose around it. Two suction pipettes were placed on the tail of the fish at intersegmental boundaries, and gentle suction was applied until electrical contact with the motor neuron axons was made, usually after about 10 min. These electrodes allowed for the recording of multi-unit extracellular signals from clusters of motor neuron axons, and provided a readout of intended locomotion. Extracellular signals were amplified with a Molecular Devices Axon Multiclamp 700B amplifier and fed into a computer using a National Instruments data acquisition card. Custom software written in C# (Microsoft) recorded the incoming signals. These signals were then analyzed offline in MATLAB (Mathworks, USA). Please seeQuantification and Statistical Analysis for detailed analysis. Two-photon Ca2+ imaging In-vivo two-photon fluorescence imaging of neural activity was performed in 5-7 dpf Tg(elavl3:GCaMP5G) zebrafish larvae. Prior to imaging, larvae were paralyzed in α-Bungarotoxin (1 mg/ml, Sigma, USA) and embedded in low melting point agarose (2% w/v), which prevented movement artifacts. Viability was monitored before and after imaging by observing the heartbeat and blood flow through brain vasculature. All data acquisition and analysis was performed using custom Labview (National Instruments, USA), MATLAB (Mathworks, USA) and C# software. Imaging was performed with a custom two-photon laser-scanning microscope, using a pulsed Ti-sapphire laser tuned to 920 nm (Spectra Physics, USA). Stimuli were projected from below with a DLP or LCOS projector (Optoma, USA or AAXA Technologies, USA, respectively) using only the red channel, which allowed for simultaneous visual stimulation and detection of green fluorescence. Visual stimuli were blanked with a black bar underneath the fish, as in free behavior experiments. Stimuli were 8.5 s in duration, which was long enough for neural activity to reach characteristic peaks (for transient responses) or plateaus (for sustained responses, see Figure 5B), and were separated by 11.5 s of static (no motion) gratings. For brain-wide mapping experiments, images were acquired at 2 Hz. For pretectal imaging, frames were acquired at 3.6 Hz, as afforded by the smaller field of view. For experiments spanning smaller brain volumes (e.g., specific pretectal imaging), 10 stimulus repetitions were presented per z-plane. For experiments across large brain volumes (e.g., whole-brain experiments), 3 stimulus repetitions were presented per z-plane. The resulting image time and depth series were analyzed in MATLAB (Mathworks, USA). Subsequent analysis was performed as described in Quantification and Statistical Analysis. Tracing pretectal projection patterns Double Tg(α-tubulin:C3PA-GFP); Tg(elavl3:GCaMP5G) fish at 6 dpf were embedded in 2% agarose in a 35 mm petri dish. We modified the tracing protocol developed by Datta et al. (Datta et al., 2008) to trace projections from a subset of Pt neurons (Figure S3D). While presenting whole-field motion stimuli from below, fish were first imaged under a two-photon microscope at 925 nm to identify motion-sensitive neurons. Then, after taking anatomical stacks of the region, individual motion-sensitive Pt neurons were selected on one side of the brain. PA-GFP in selected neurons was then activated using 5-10 pulses (100 ms – 2 s) of 770 nm pulsed infrared laser light. Selective photoconversion was confirmed by switching to 925 nm and imaging the selected plane for increased fluorescence. The activation protocol was repeated until photoconverted fluorescence no longer increased significantly. Anatomical stacks of projection patterns were acquired 2 hr post-activation. The pre and post activation two-photon stacks were then aligned and analyzed using FIJI (ImageJ) software. For highlighting of the posterior commissure (Figure S4A and S4F, right), anatomical pre-activation stacks of Tg(α-tubulin:C3PA-GFP) were acquired, before activating a small region with the same protocol as described above. The pre and post activation two-photon stacks were then aligned and analyzed using FIJI (ImageJ) software in separate color channels. Two-photon laser ablations After embedding zebrafish in low melting agarose, areas were targeted using neuroanatomical landmarks, stereotypical blood vasculature, and/or neuronal activity patterns in Tg(elavl3:GCaMP2) or Tg(elavl3:GCaMP5G) larvae. Mode-locked laser power (885 nm, 120-135 mW maximum at sample) was linearly increased while the beam was scanned in a spiral pattern over the targeted region (typically 0.5 – 10 s). The laser scan was immediately terminated upon the detection of fluorescence saturation, which is presumed to result from the creation of highly localized plasma via multi-photon absorption. For single cells, this method resulted in the complete destruction of the target cell without affecting adjacent cells. Ablations in neuropil regions (like the PC) required additional validation. Due to the depth and anatomical position of the PC, nearly 50% of ablated fish showed non-specific side effects such as nearby necrotic tissue or changes in the optical density of the optic tectum, likely due to blood vessel cauterization from out-of-focus heating. These fish were excluded from post-ablation analysis. Before and after PC ablation imaging was performed in only a subset of fish. After freeing fish from agarose and a recovery period (10-120 min, until fish were swimming spontaneously), fish were tested in the behavioral assay. For the AF6 ablation, we specifically monitored fluorescence before and after ablation across AFs. Only fish that exhibited functional perturbations specific to AF6 were further analyzed. Quantitative modeling Behavior-only modeling We built the model in Figure 1E to illustrate that a simple circuit architecture could account for the observed binocular visuomotor transformation. For all behavioral model results error bars are SEM across fish (N = 38). The synaptic inputs onto the forward swimming center were 0.0395 from the medial responsive RGCs and −0.0088 from the lateral responsive RGCs. The magnitudes of synaptic inputs onto the two turning centers were 0.0133 from the medial responsive RGCs and 0.0068 from the lateral responsive RGCs. The sign of each connection onto a turning center was positive when the direction of the RGC matched the direction of the turning center. After summing the monocular components of the motion stimuli, we converted the resultant retinal input signals into the model rate through a transfer function. For forward swimming, this transfer function was the sum of the retinal input and the baseline static swimming rate (0.0836). For turning, we fit a cubic transfer function to generate the behavioral turning rates from the retinal input signals. The resulting transfer function was f(x) = 114.6x3 + 17.48x2 + 0.5168x + 0.006175, where x denotes the retinal input signal. To simulate the ablation experiment in Figure S4E, we removed those synaptic inputs that could not reach the lateralized turning centers without an interhemispheric connection within the zebrafish brain. Quantifying the visuomotor transformation Larval zebrafish responded to optomotor stimuli by modulating at least three elements of behavior: forward swimming; turning to the left; and turning to the right (Figure 1D). For each stimulus condition and element of behavior, we quantified the response magnitude as the amplitude of the associated peak of the mean bout frequency histogram. We similarly quantified the uncertainties of each mean response magnitude using the SEM of the histogram. Since neuronal fluorescence responses were measured relative to the background set by the static stimulus, we shifted the behavioral responses to also be relative to the static stimulus. We accordingly corrected the uncertainty of each stimulus response to account for the uncertainty of the static response. Since the static stimulus is zero by construction, we are left with ten non-trivial stimulus conditions. For each of the three behavioral classes, we construct a 10-dimensional vector y whose components are the experimental motor response magnitudes for the 10 stimulus conditions. We also define the 10×10 diagonal matrix C whose diagonal elements, σi2, are the squares of the response uncertainties. Suppose that y^ is a 10-dimensional vector of predicted response magnitudes by some circuit model. We quantify the error of the model as e=(y−y^)TC−1(y−y^)=∑i=110(yi−y^i)2σi2 where the superscript T denotes the matrix transpose, and i indexes the stimulus condition. Note that e is simply the squared error weighted by the uncertainty in each stimulus condition. Linear regression framework for functional connectivity As discussed in the main text and illustrated by Figure 7A, our model uses functional connections from neurons in the major Pt and aHB response categories to premotor neurons to generate each element of behavior. For example, eight functional connection strengths, from the {oB, B, iB, ioB, iMM, MM, oML, S} major Pt neuron categories (and their rightward selective counterparts) to nMLF neurons, parameterize the forward swimming component of the visuomotor transformation (green lines, Figure 7A). The model parameters for the rightward turning component of the visuomotor transformation also consists of eight functional connection weights, now from the {oB’, B’, iMM’, S’} major right-selective Pt neuron categories and the {oB, B, iMM, S} major left-selective aHB neuron categories to right LHB units. The circuit driving leftward turning is the mirror-reflection of the circuit driving rightward turning. Our task is to understand how the choice of these functional connection weights affects the accuracy of the modeled visuomotor transformation. Since the formalism does not depend on the element of behavior under consideration, we henceforth use abstract notation to describe the general method. Let X denote the n×p response matrix, which comprises the mean fluorescence response of p neurons to each of n stimuli. Note that n = 10 and p = 8 for each type of behavior considered here. For example, the columns of X correspond to the responses of the {oB’, B’, MM’, S’} major right-selective Pt neuron categories and the {oB, B, MM, S} major left-selective aHB neuron categories to the 10 stimulus conditions, when considering rightward turning behavior. We model the visuomotor transformation as a linear combination of these responses y^=Xβ, where β denotes the p-dimensional weight vector of functional connection weights. Choosing the weight vector to minimize e is mathematically equivalent to performing maximum likelihood parameter estimation with the generative model y=Xβ+η, where η is an n-dimensional vector of zero-mean Gaussian noise with covariance matrix C. In particular, the log-likelihood function for this model is l(β)=−12e(β)+A, where A is a constant that does not depend on the model parameters. We thus applied standard techniques from maximum likelihood parameter estimation to compute confidence intervals for the best-fit model parameters (Figures 7B–7D). Similarly, we computed p-values for the model’s predictions regarding the effects of optogenetic activation of a neural activity pattern, x, by considering the null hypothesis that xTβ = 0. Structure of the error function With the assumptions described in the previous sections, the error associated with the visuomotor transformation is a quadratic function of the functional connection strengths e(β)=emin+(β−β^)TH(β−β^), where β^=H−1U is the best fit functional connection strengths (colored bars in Figure 7B), H=XTC−1X is proportional to the Hessian matrix of the error function, and U is defined by U=XTC−1y. The minimal error achievable by the model architecture is emin=yTC−1y−UTH−1U, Let V denote the p×p matrix in which each column is a right eigenvector of H. Then VVT = VTV = I, where I is the identity matrix, Λ = VTHV is a diagonal matrix whose diagonal contains the eigenvalues of H, and VTβ is the weight vector in the basis of eigenvectors of H. The matrices in Figure 7D simply show V for both swimming and right turning behavior. The eigenvectors of H correspond to the principal axes of the quadratic error surface, and we refer to each eigenvector of H as a functional connectivity mode. The basis of functional modes is useful because the error function receives independent contributions from the component of the weight vector along each functional mode, e(β)=emin+∑i=1pλi(γi−γ^i)2, where the λi are eigenvalues of H, γi=(VTβ)i=∑j=1pVjiβj is the projection of β onto the ith mode, and γ^i is the projection of the optimal weight vector onto the ith mode. Note that the black bar plots in Figure 7D show the eigenvalue associated with each functional mode. Departures of β from when the optimal weight induce large errors they project onto modes that are associated with large eigenvalues. Thus, the confidence interval for γ^i is narrow for the modes associated with large eigenvalues. This basis transformation is also useful because the projections of the optimal weight vector along each mode contribute independent error reduction to model. In particular, note that we can rewrite the minimal error as emin=yTC−1y−∑i=1p(VTU)i2λi=yTC−1y(1−∑i=1pri2), where ri2=1yTC−1y(VTU)i2λi. Because the error of the fully disconnected model is yTC−1y, ri2 is the fraction of error eliminated by each mode. The bottom panels in Figure 7D shows ri2 for each mode. Interpreting the functional modes as neural activity patterns In the previous section, we defined V as the p×p matrix in which each column is a right eigenvector of H. If x is a p-dimensional pattern of neural activation across a population of behaviorally relevant major neuron categories, then VTx represents the same activity pattern in the basis defined by the columns of V, and any activity pattern can be written as a linear combination of the columns of V. Thus, we can thus think of the columns of V as defining functional modes of activity. Thus, an activity mode is the pattern of activation that maximizes the projection of the response vector onto the corresponding vector of connectivity weights. In this framework, the functional modes replace the major neuron categories as the elemental components of neural population activity. Figure 7F displays the stimulus response profiles of several functional modes (6F, 7F, 5T, 7T). Each component of the functional connectivity, γi, corresponds to a functional connection strength from one functional mode of activity to a premotor neuron driving behavior. The characterization of the error surface provided in the previous section implies that the functional modes define patterns of activity from which functional connections contribute independently to the accuracy of the modeled visuomotor transformation. The bottom panels of Figure 7D show that the functional modes provide a description of neural activity in which only a few dimensions contribute to the model’s ability to generate appropriate motor activity under the experimental stimulus conditions. QUANTIFICATION AND STATISTICAL ANALYSIS Quantification of swim kinematics Offline behavioral analysis was performed using custom MATLAB scripts (Mathworks, USA). Because we did not assume that data were normally distributed, we used non-parametric statistics for all hypothesis testing. Angular velocity for each fish was calculated as total angle turned over total stimulus presentation time. Single swim bouts were extracted by peak detection in traces obtained via multiplication of orientation change records by box-smoothed (20 ms) swim velocity records (calculated from fish center-of-mass). Latency was calculated as time between motion onset of moving gratings and first detected swim bout. Bout frequency was computed as the total number of detected swim bouts over total stimulus time. Heading angle change for each bout was calculated as orientation directly before (2 ms) and after (10 ms) a detected swim bout. By convention, negative values indicate rightward (or clockwise) and positive values indicate leftward (or counterclockwise) heading angle change. For Figure S1C, we corrected for innate turn biases by subtracting the angular velocity measured during the static stimuli from all other measurements on a fish-by-fish basis before averaging. For all behavioral bar graphs (Figures 1C, 4C, and 4E) error bars represent standard error of the mean (SEM) across fish. For histograms, shaded error is SEM across fish for angle bout type (Figures 1C and 4D). For supplemental figures, consult associated legends. Fictive behavior analysis Analysis of fictive behavior was performed on filtered signals, which consist of the standard deviation of the raw signal in 17 ms time bins. For fictive bout frequency analysis, bouts were marked at peaks in the filtered signal separated by at least 100 ms and crossing an amplitude threshold (adjusted for signal noise: 1 fish at 4.35, 2 fish at 1.95 times the peak baseline noise, as assayed by frequency histogram). Visual stimuli were presented as outlined in two-photon Ca2+ imaging, below. Because baseline fictive swim frequency was reduced relative to freely swimming experiments, we restricted our analyses only to active stimulus epochs (at least 0.1 Hz bout frequency). Functional imaging analysis To generate activity maps (Figures 2E, 2F, S3E, and S3F), fluorescence movies were averaged separately across all frames corresponding to each individual stimulus. A baseline image, the average of all frames during interstimulus periods, was then subtracted from these average images, which were masked to exclude regions outside of the brain. The resulting images, which were thresholded at 0.5* standard deviation (STD) above the mean and eroded and dilated to reduce noise, reflected normalized activity for each stimulus that was less polluted from noise in regions of low baseline fluorescence than traditional ΔF/F maps. These activity images, specific to each individual stimulus, were then combined to produce each type of map. For maps of direction selectivity, activity images for each of the 8 directions of whole-field motion were linearly combined using weighted projections into HSV color space, such that the combined color is related to the vector sum of activity elicited in response to each direction of motion: ultimately, the hue indicates directionality, the value reflects overall motion sensitivity, and the saturation represents the vector magnitude (strength of direction selectivity). For maps of medial/lateral selectivity, activity images for medial and lateral stimulus epochs were combined similarly, with the medial activity image represented in the green channel and the lateral activity image represented in the magenta channel. All other maps were generated analogously, with colors adjusted as indicated. To generate whole-brain activity maps, Figures 2D and S2D, we used a method that better captured correlated activity in neuropil regions, as follows. First, we constructed an average regressor based on strongly responding voxels, as extracted from manual ROIs in persistently active regions, which we deduced were more related to the OMR, given the persistent properties of the behavior. We used this regressor to calculate the correlation coefficient for each pixel time series in every plane of the acquired image stack for each individual stimulus. Maps were then generated by assigning the resulting correlation images to appropriate color channels and combining as in the activity maps above. Map shown in Figure 2D is a maximum intensity projection through 135 μm of brain tissue, with ventral AF10 as the most dorsal plane. The quantifications of pretectal data presented in Figure 3 were based on ROI selection using a method adapted from Ahrens et al. (2013a). For this analysis, an activity map for each imaged plane was first extracted by sweeping a square ROI roughly equal to half the size of a cell body across the spatial extent of an imaging movie. For each swept ROI, the spatially averaged fluorescence time series was normalized by the ROI’s mean fluorescence across time and the average fluorescence of all pixels across space and time. This signal was then cubed to amplify peaks, and each square’s average processed signal was assigned to each square’s spatial location. This analysis produced maps of activity that were robust to noise in regions of low baseline fluorescence. These activity maps were then overlaid on mean fluorescence images (reflecting anatomy) and used to guide manual neuron ROI selection. To make composite maps of the pretectal population across fish, all imaged planes were registered to a standard Tg(elavl3:GCaMP2) brain, which has an expression pattern indistinguishable from Tg(elavl3:GCaMP5G), using cross-correlation. Fluorescence traces from these ROIs were then divided by a stable baseline amplitude averaged across all interstimulus intervals to generate ΔF/F traces. The means of trial-averaged ΔF/F for individual stimuli were then used to calculate functional indices. Binocularity index was calculated as [(xM or xL)max – (Mx or Lx)max]/[(xM or xL)max + (Mx or Lx)max], where xM and xL are right-eye monocular medial and lateral responses, respectively, and Mx and Lx are left-eye monocular medial and lateral responses, respectively. By construction, a value of 0 corresponds to equal activation by left and right eyes, and −1 and 1 correspond to exclusive activation by the left or right eye, respectively. The inward medial-medial suppression index was computed as [(xM or Mx)max – MM]/[(xM or Mx)max + MM], where MM is the response to inward motion. By construction, an index of 0 corresponds to no suppression and an index of 1 indicates that medial motion in the contralateral eye completely suppresses the response to medial motion in the other. The outward lateral-lateral suppression index was calculated analogously. To restrict indices to the [−1 1] range, all values were thresholded at 10% ΔF/F (calculated noise floor) to exclude negative ΔF/F deflections (or suppressed responses), which we did not integrate into our indices. To calculate direction selectivity vectors, we performed a vector sum of responses to the 8 directions of motion after normalizing to the maximum response across all directions. For automatic ROI segmentation (Figures 2 and 6), we adapted a method from Portugues et al. (Portugues et al., 2014). In brief, we performed a neighborhood correlation analysis for each pixel in each plane of the fluorescence movie to generate maps of local correlation. This metric was typically high for pixels within the same cell body and discrete, local regions of high local correlation were easily segmented into individual ROIs automatically by seeding an ROI at a pixel with a local correlation maximum and growing the ROI spatially until the correlation dropped below a fixed threshold (0.15). To support ROI selection at the single cell level, ROIs were also limited in size between 30 and 200 pixels, or cells approximately 1.7 to 4.3 microns in radius at standard imaging resolution. Because this method was blind to the underlying anatomy, this automatic ROI segmentation also selected correlated regions of neuropil that met the stated size requirements. Classification of functional response types As we were confronted with complex response patterns in each brain region (Figure S6; Movie S2), we classified neural response types with barcode analysis. We assigned each neuron an 8-bit barcode (Kubo et al., 2014), which encodes whether or not the neuron significantly increased its fluorescence upon motion onset for each of the coherent, monocular, and conflicting stimuli. ΔF/F fluorescence traces for ROIs were trial-averaged for each individual stimulus (Figure 1B) but excluding forward and backward epochs, as we were primarily interested in how information from both eyes was combined. The resulting traces were then baseline corrected with a first-degree polynomial fit to the baseline period. We then assessed whether activity during stimulus periods was significantly different from its baseline. For a given neuron and stimulus, if the mean of the trial-averaged signal exceeded 1.8*STD above its baseline, then the neuron was assigned a ‘1’ for that particular stimulus; otherwise, it was assigned a ‘0.’ This resulted in a binary barcode for each neuron across all stimuli in the orthogonal set. Based on these 8 stimuli, neurons could fall into 256 (28) different response combination classes. This method, and the associated significance threshold, provided the most consistent classification across fish during visual inspection of individual neurons, and deviations from this threshold did not markedly change relative frequency distributions. While we tried other clustering methods, our barcode technique provided the most intuitive results reflecting the stimulus-behavior relationship. Other methods we tested, like k-means clustering, required arbitrary parameterizations of the data (like estimated number of clusters). We labeled a response class as over-represented if, in sorted histograms of class frequency, the class (or its mirror-symmetric partner) was separated from the others by a graphical non-linearity (see Figure 5C) and was present in all fish. These functional response types allowed us to summarize this heterogeneity in our model. Although studies have analyzed heterogeneous population responses without classifying neurons, our classification helped us in several ways. First, a model that references response types rather than individual neurons can make predictions for how activity of sub-populations of neurons will translate into behavior. Second, Kubo et al. classified Pt responses during the OKR using this scheme (Kubo et al., 2014), permitting a comparison of Pt responses across behaviors. Third, response type classification helped us to identify elements of the representation that transformed across brain areas. Fourth, certain response types had a natural interpretation (e.g., Pt relay neurons). These neuronal response classes could have subtypes that are crucial to more complex behaviors and stimuli. For analyses where relative differences in activity across stimuli were most meaningful (e.g., Figure 2F, right, Figure 3B, left, Figure 5B), we used normalized ΔF/F metrics. In these cases, we calculated trial-averaged ΔF/F traces for each ROI and then divided by the maximum ΔF/F value across all time points and stimuli. These normalized traces were then averaged across ROIs (where indicated). All ΔF/F metrics using single values from stimulus epochs (Figure 2F, right, Figure S3E, bottom, Figure S3F, bottom, Figure S3G, Figure 3B, left, Figure 3D) were calculated as mean normalized ΔF/F across all time points during each respective stimulus presentation. DATA AND SOFTWARE AVAILABILITY Data Resources We are providing two supplemental data files (Tables S1 and S2) containing information for the pretectum and anterior hindbrain for each response type, including frequency, average peak calcium response, and identifying metadata (e.g., monocular versus binocular, barcode representation). Other data on behavior or imaging will be made available upon request. Software Software for data acquisition, analysis and quantitative modeling will be made available upon request. Supplementary Material 1 2 3 4 Supplementary Material ACKNOWLEDGMENTS We thank Michael Orger for providing Tg(alpha-tubulin:PA-GFP) fish and Michael Orger, Drew Robson, and Jennifer Li for providing Tg(elavl3:GCaMP5G) fish. Misha Ahrens, Isaac Bianco, and Martin Haesemeyer helped with useful discussions. Furthermore, we thank Ruben Portugues, Mariela Petkova, Andrew Bolton, Bence Ölveczky, and Alexander Schier for reading the manuscript and providing feedback, we thank Adam Kampff for help with software development and rig design; NIH grants U01NS090449, DP1 NS082121; the Simons Foundation SCGB#325207, R24NS086601; the European Research Council; and the Marie Curie Fellowship provided financial support. Figure 1 Eye- and Direction-Specific Optic Flow Evoke Distinct Orienting and Locomotion Patterns (A) Behavioral setup. Left, freely swimming fish are monitored with a camera while moving gratings are presented. Heading (Δν) and position (Δx, Δy) are extracted to lock visual motion direction to body axis. Right, illustration of each monocular visual field (purple, green, 163°) and the small binocular overlap (dark gray, 12°). Scale bar, 500 μm. (B) Stimulus set composed of static, medial and lateral motion, and forward and backward stimuli. Arrowheads indicate motion direction. Circular icons represent eyes, and white tick marks show the direction of motion. These icons and colors are used throughout the paper. (C) Bar graph of average angular velocity (heading direction change per second) during behavioral responses. The white bar represents the linear combination of monocular medial and lateral motion. Error bars are SEM across fish (n = 38). (D) Average histograms of absolute frequency per swim bout angle. Left, conflicting stimuli affect forward swim frequency (center peak, added “swims”) relative to the static condition (inward versus static, p = 6.6 × 10−6; outward versus static, p = 7.7 × 10−8, paired Wilcoxon). Inward reciprocally suppressed turns (open circles). Middle, left/right stimuli increased correct (in motion direction, filled circles) and decreased incorrect (opposite of motion direction, open squares) turn frequency (inset). Medial enhanced but monocular lateral stimuli suppressed swimming (lateral versus static, p = 0.006). Shaded error is SEM, n = 38. Right, forward stimuli increased, backward stimuli significantly reduced forward swim frequency (versus static, p = 2.1 × 10−6). n = 30. (E) Comparison of measured bout frequencies and minimal model output. DSRGCs connect to motor centers for forward swimming (F) and left (TL) and right turning (TR). Left, forward swimming. Each point is the mean peak response to a stimulus; lighter center indicates leftward motion. Right, turning behavior. correct turns, filled circles; incorrect turns, open circles; non-directional stimuli, open squares. Model input-output functions for each behavioral component, top right and left. Dotted lines indicate baseline rates. Figure 2 Whole-Brain Activity Maps Reveal Processing Stages Underlying the OMR (A) Dorsal overview of zebrafish neuroanatomy. DSRGCs (black dots) project via the optic chiasm (OC) to ten contralateral retinal arborization fields (AFs). Pt, pretectum; nMLF, nucleus of the medial longitudinal fasciculus; aHB, anterior hindbrain; RoL, neurons in rhombomere 1; ARTR, anterior rhombencephalic turning region; vSPNs, ventromedial spinal projection neurons; cHB, caudal hindbrain; M/HB, midbrain-hindbrain border; rh1–5, rhombomeres 1–5. A, anterior; P, posterior. (B) Distribution of all motion-sensitive units (n = 76,604) across 14 Tg(elavl3:GCaMP5G) fish. Each unit is a dot, color coded for preferred motion direction (see Dir. color wheel). Bottom left, histogram of direction preference. Scale bars, 50 μm. (C) Histograms of direction preference for specific brain regions. (D) Binocular activity map generated from sequential presentation of monocular leftward motion to each eye. Pixels are colored for medial versus lateral preference (STAR Methods). Binocular regions appear white. Anatomy in gray. (E) Left, two-photon micrographs from single AF6 and Pt planes. Right, average ΔF/F for regions of interest (ROIs) (dashed white lines). Gray rectangles indicate stimulus presentation periods. Shaded areas are SEM over n = 10 stimulus repetitions. (F) Left, responses to eight whole-field motion stimuli, see color wheel, top. Right, polar plots (±SEM, shaded area) of average normalized ΔF/F in n = 7 fish. Figure 3 Population Activity in the Pretectum (A) Left, direction selectivity vectors for neurons in the Pt (see color wheel, top). Gray, magnitudes smaller than 0.4; black, population average vectors. Right, anatomical distribution of direction preference. A, anterior; P, posterior; L, left; R, right. Dashed line, brain midline. (B) Left, Linear binocular integration of normalized ΔF/F responses. *p = 1.2 × 10−37, paired Wilcoxon. Bars are mean ± SEM across all neurons. Right, scatterplot of individual Pt neuron responses. Red, least-squares regression line; dashed black, unity line. (C) Histogram of Pt neuron suppression indices (STAR Methods). (D) Normalized ΔF/F to conflicting motion for neurons preferring either backward (−135° to +135°, n = 191) or forward (−45° to +45°, n = 678) motion. Figure 4 Laser Ablation of Posterior Commissure Disrupts Binocular Integration (A) The posterior commissure (PC) connects the left and right Pt, providing a putative conduit for lateral excitation (purple arrow) and medial inhibition (green triangle). Red spiral, target for laser ablation. (OC) optic chiasm. (TL) motor center for turning. (B) Left, anatomy before (top) and after (bottom) PC ablation. Red spiral, ablation site. Red arrowheads, bottom, tissue debris (compare to top). Right, average ΔF/F before (colored) and after (gray) PC ablation for the three ROIs. n = 6 stimulus repetitions. Vertical lines represent stimulus presentation periods. (C) Average orientation change per bout in control (n = 38 fish) and ablated (post, n = 14 fish) animals. Inset, difference between coherent binocular and monocular medial responses (p = 6.2 × 10−5, unpaired Wilcoxon). C, control; P, post. Post-ablation monocular lateral versus static baseline, p = 0.2; Post-ablation coherent versus control monocular medial, p = 0.08. (*) p < 10−4, n.s., not significant, p > 0.05. (D) Average histograms of absolute frequency per bout angle post-ablation (n = 14 fish). (E) Change in peak turn probability (post-ablation minus control) for incorrect and correct turns. *p < 0.05. (F) PC ablation reveals effect of medial motion on behavior. Medial motion activates the contralateral nMLF directly and the ipsilateral nMLF indirectly via the PC to drive forward swimming (F). Medial motion suppresses incorrect turns (TL) via the PC but also via other interhemispheric connections, potentially in the hindbrain. Figure 5 Pretectal Neurons Cluster into Distinct Functional Response Types (A) Heatmap of mean normalized ΔF/F within monocular and binocular response categories (STAR Methods). Column width, relative frequency neuron types; brightness, mean normalized ΔF/F. Top types: oB, outward-responsive binocular neuron; B, binocular; iB, inward-responsive binocular; ioB, inward- and outward-responsive binocular; iMM, inward-responsive monocular (medial-selective); MM, monocular (medial-selective); oML, outward-responsive monocular (lateral-selective); S, selective for coherent motion. (‘) matching right-selective types. (B) Left, average normalized ΔF/F for the top 8 left-selective response types, ranked by frequency. Shaded area, SEM across neurons. Mirror-symmetric right-selective types in Figure S5D. Right, average direction selectivity. Vector width is bootstrapped standard error for each x and y vector component. (C) Sorted histograms of response type frequency. Major Pt types (dashed boxes) lie above the indicated discontinuity threshold (gray line, STAR Methods). (D) Hypothesized circuit for generating Pt types via monocular relay neurons. In this example, the B neuron is activated by leftward medial and lateral but inhibited by rightward medial and lateral motion. (E) Left, Model using overrepresented Pt response types generated from relay neurons that integrate DSRGCs signals. Pt neurons project ipsilaterally to left and right turning centers (TL, TR) and to a forward swimming center (F). Right, comparison of measured bout frequencies and minimal model output. Insufficient suppression of Pt neurons caused model failure for incorrect turns (open squares, inset). Figure 6 Hindbrain Circuitry Refines Turning Behavior (A) Distribution of all motion-sensitive units with higher peak ΔF/F responses to any direction over forward (n = 44,047, n = 14). Each unit is color coded for preferred direction of motion. Anatomical labels as in Figure 2. (B) Average normalized ΔF/F in response to each stimulus for the four overrepresented aHB response types (left selective). Right, average direction selectivity vectors for each respective class (cf. Figure 5B). Neurons named as in Figure 5. (C) C1 Distribution of all binocular units (B), colored for preferred motion direction. C2 B units activated by forward and backward motion. C3 Left, B units without significant responses to forward and backward motion. Right, motion-opponent B units suppressed by the opposite direction of motion. Traces show mean normalized ΔF/F traces for each population of B units. (*), active above baseline (>1.8*STD); arrowhead, suppressed below baseline (< −1.8*STD); circles, no significant response. Figure 7 Neural Network Modeling Identifies Significant Dimensions of Functional Connectivity (A) Model incorporating hindbrain circuitry. Green, F, forward premotor units. Gray circles, Pt neurons projecting to the ipsilateral early hindbrain (EHB, ovals) and late hindbrain (LHB). Red, T, turn premotor circuitry. Neurons numbered as in Figures 5B and 6B. (B) Best-fit connection strengths (95% confidence intervals) between neuron types and premotor units. Only two types predict significant non-zero connections (darker bars). (C) Left, model error as a function of connectivity strength between neuron type 1 and F and neuron type 2 and F; all others connected by best-fit connection strengths. Right, while Pt neurons do not behave independently, we identified independent dimensions of population connectivity (functional modes) by rotating the error surface. Each axis now corresponds to a pattern of activation and suppression across neuron types (STAR Methods). (D) Top, independent connectivity patterns for neuronal response types (rows) within each functional mode (columns), sorted according to the model’s sensitivity to perturbations in the direction of each mode (black bars, top). Bottom, fraction of error eliminated by each mode. (*) modes with a reliable influence on behavior (STAR Methods). (E) Illustration of how modes contribute to model output in response to stimuli (here shown only for coherent rightward motion). (F) The behavioral output associated with individual swimming modes (ModeF) and turning modes (ModeT). Top two rows show output for the two most significant swimming (left) and turning (right) modes. Bottom row shows collective output for the six most and two most significant swimming and turning modes, respectively. Bars are colored by stimulus identity. (G) Quantitative whole-brain model for the OMR, reflecting the best-fit connectivity in Figure 7B. Neuron types are represented by symbols illustrating activation and suppression by motion stimuli (see legend). Highlights Optomotor response is driven asymmetrically but linearly by visual motion to each eye Dedicated circuits differentially process eye- and direction-specific motion Neural representations are distributed over select overrepresented response types Behavior as well as neural activity is captured by realistic whole-brain circuit model In Brief Whole-brain imaging and behavioral analysis combined with network modeling reveal key circuit elements contributing to a complex sensorimotor behavior in zebrafish larvae and provide a framework for building brain-level circuit models. STAR*METHODS Detailed methods are provided in the online version of this paper and include the following: KEY RESOURCES TABLE CONTACT FOR REAGENTS AND RESOURCE SHARING EXPERIMENTAL MODEL AND SUBJECT DETAILS ○ Zebrafish METHOD DETAILS ○ Behavior in freely swimming zebrafish ○ Fictive behavior ○ Two-photon Ca2+ imaging ○ Tracing pretectal projection patterns ○ Two-photon laser ablations ○ Quantitative modeling QUANTIFICATION AND STATISTICAL ANALYSIS ○ Quantification of swim kinematics ○ Fictive behavior analysis ○ Functional imaging analysis ○ Classification of functional response types DATA AND SOFTWARE AVAILABILITY ○ Data Resources ○ Software SUPPLEMENTAL INFORMATION Supplemental Information includes seven figures, two tables, and two movies and can be found with this article online at http://dx.doi.org/10.1016/j.cell.2016.10.019. AUTHOR CONTRIBUTIONS F.E. and E.A.N. conceived of the project. E.A.N. and T.W.D. performed the experiments. E.A.N., J.E.F., and T.W.D. analyzed data. J.E.F., E.A.N., and H.S. developed the models. E.A.N., J.E.F., T.W.D., J.R., H.S., and F.E. wrote the paper. REFERENCES Ahrens MB Li JM Orger MB Robson DN Schier AF Engert F Portugues R Brain-wide neuronal dynamics during motor adaptation in zebrafish Nature 2012 485 471 477 22622571 Ahrens MB Orger MB Robson DN Li JM Keller PJ Whole-brain functional imaging at cellular resolution using light-sheet microscopy Nat. Methods 2013a 10 413 420 23524393 Ahrens MB Huang KH Narayan S Mensh BD Engert F Two-photon calcium imaging during fictive navigation in virtual environments Front. Neural Circuits 2013b 7 104 23761738 Armstrong DM The supraspinal control of mammalian locomotion J. Physiol 1988 405 1 37 3076600 Bianco IH Engert F Visuomotor transformations underlying hunting behavior in zebrafish Curr. Biol 2015 25 831 846 25754638 Borst A Haag J Reiff DF Fly motion vision Annu. Rev. Neurosci 2010 33 49 70 20225934 Bounoutas A Chalfie M Touch sensitivity in Caenorhabditis elegans Pflüg. Arch. Eur. J. Physiol 2007 454 691 702 Burrill JD Easter SS Jr. Development of the retinofugal projections in the embryonic and larval zebrafish (Brachydanio rerio) J. Comp. Neurol 1994 346 583 600 7983245 Chalasani SH Chronis N Tsunozaki M Gray JM Ramot D Goodman MB Bargmann CI Dissecting a circuit for olfactory behaviour in Caenorhabditis elegans Nature 2007 450 63 70 17972877 Chédotal A Development and plasticity of commissural circuits: From locomotion to brain repair Trends Neurosci 2014 37 551 562 25220044 Datta SR Vasconcelos ML Ruta V Luo S Wong A Demir E Flores J Balonze K Dickson BJ Axel R The Drosophila pheromone cVA activates a sexually dimorphic neural circuit Nature 2008 452 473 477 18305480 Dunn TW Mu Y Narayan S Randlett O Naumann EA Yang C-T Schier AF Freeman J Engert F Ahrens MB Brain-wide mapping of neural activity controlling zebrafish exploratory locomotion eLife 2016 5 e12741 27003593 Felleman DJ Van Essen DC Distributed hierarchical processing in the primate cerebral cortex Cereb. Cortex 1991 1 1 47 1822724 Fink AJP Croce KR Huang ZJ Abbott LF Jessell TM Azim E Presynaptic inhibition of spinal sensory feedback ensures smooth movement Nature 2014 509 43 48 24784215 Fisher D Olasagasti I Tank DW Aksay ERF Goldman MS A modeling framework for deriving the structural and functional architecture of a short-term memory microcircuit Neuron 2013 79 987 1000 24012010 Fite KV Pretectal and accessory-optic visual nuclei of fish, amphibia and reptiles: Theme and variations Brain Behav. Evol 1985 26 71 90 3907745 Gamlin PDR The pretectum: Connections and oculomotor-related roles Prog. Brain Res 2006 151 379 405 16221595 Gao P Ganguli S On simplicity and complexity in the brave new world of large-scale neuroscience Curr. Opin. Neurobiol 2015 32 148 155 25932978 Giolli RA Blanks RH Torigoe Y Pretectal and brain stem projections of the medial terminal nucleus of the accessory optic system of the rabbit and rat as studied by anterograde and retrograde neuronal tracing methods J. Comp. Neurol 1984 227 228 251 6470215 Gollisch T Meister M Eye smarter than scientists believed: Neural computations in circuits of the retina Neuron 2010 65 150 164 20152123 Heiligenberg W Konishi M Neural Nets in Electric Fish (Bradford) 1991 Huang K-H Ahrens MB Dunn TW Engert F Spinal projection neurons control turning behaviors in zebrafish Curr. Biol 2013 23 1566 1573 23910662 Huberman AD Wei W Elstrott J Stafford BK Feller MB Barres BA Genetic identification of an On-Off direction-selective retinal ganglion cell subtype reveals a layer-specific subcortical map of posterior motion Neuron 2009 62 327 334 19447089 Kato S Kaplan HS Schrödel T Skora S Lindsay TH Yemini E Lockery S Zimmer M Global brain dynamics embed the motor command sequence of Caenorhabditis elegans Cell 2015 163 656 669 26478179 Kinkhabwala A Riley M Koyama M Monen J Satou C Kimura Y Higashijima S Fetcho J A structural and functional ground plan for neurons in the hindbrain of zebrafish Proc. Natl. Acad. Sci. USA 2011 108 1164 1169 21199947 Ko H Hofer SB Pichler B Buchanan KA Sjöström PJ Mrsic-Flogel TD Functional specificity of local synaptic connections in neocortical networks Nature 2011 473 87 91 21478872 Koontz MA Rodieck RW Farmer SG The retinal projection to the cat pretectum J. Comp. Neurol 1985 236 42 59 4056090 Korn H Faber DS The Mauthner cell half a century later: A neurobiological model for decision-making? Neuron 2005 47 13 28 15996545 Koyama M Kinkhabwala A Satou C Higashijima S Fetcho J Mapping a sensory-motor network onto a structural and functional ground plan in the hindbrain Proc. Natl. Acad. Sci. USA 2011 108 1170 1175 21199937 Kubo F Hablitzel B Dal Maschio M Driever W Baier H Arrenberg AB Functional architecture of an optic flow-responsive area that drives horizontal eye movements in zebrafish Neuron 2014 81 1344 1359 24656253 Lee MM Arrenberg AB Aksay ERF A structural and genotypic scaffold underlying temporal integration J. Neurosci 2015 35 7903 7920 25995475 Lisberger SG Visual guidance of smooth-pursuit eye movements: Sensation, action, and what happens in between Neuron 2010 66 477 491 20510853 Masland RH The neuronal organization of the retina Neuron 2012 76 266 280 23083731 Mauss AS Pankova K Arenz A Nern A Rubin GM Borst A neural circuit to integrate opposing motions in the visual field Cell 2015 162 351 362 26186189 O’Leary T Sutton AC Marder E Computational models in the age of large datasets Curr. Opin. Neurobiol 2015 32 87 94 25637959 Ohyama T Schneider-Mizell CM Fetter RD Aleman JV Franconville R Rivera-Alba M Mensh BD Branson KM Simpson JH Truman JW A multilevel multimodal circuit enhances action selection in Drosophila Nature 2015 520 633 639 25896325 Orger MB Kampff AR Severi KE Bollmann JH Engert F Control of visually guided behavior by distinct populations of spinal projection neurons Nat. Neurosci 2008 11 327 333 18264094 Portugues R Feierstein CE Engert F Orger MB Whole-brain activity maps reveal stereotyped, distributed networks for visuomotor behavior Neuron 2014 81 1328 1343 24656252 Randlett O Wee CL Naumann EA Nnaemeka O Schoppik D Fitzgerald JE Portugues R Lacoste AMB Riegler C Engert F Schier AF Whole-brain activity mapping onto a zebrafish brain atlas Nat. Methods 2015 12 1039 1046 26778924 Rigotti M Barak O Warden MR Wang X-J Daw ND Miller EK Fusi S The importance of mixed selectivity in complex cognitive tasks Nature 2013 497 585 590 23685452 Robles E Laurell E Baier H The retinal projectome reveals brain-area-specific visual representations generated by ganglion cell diversity Curr. Biol 2014 24 2085 2096 25155513 Roeser T Baier H Visuomotor behaviors in larval zebrafish after GFP-guided laser ablation of the optic tectum J. Neurosci 2003 23 3726 3734 12736343 Ruta V Datta SR Vasconcelos ML Freeland J Looger LL Axel R A dimorphic pheromone circuit in Drosophila from sensory input to descending output Nature 2010 468 686 690 21124455 Ryczko D Dubuc R The multifunctional mesencephalic locomotor region Curr. Pharm. Des 2013 19 4448 4470 23360276 Sassa T Aizawa H Okamoto H Visualization of two distinct classes of neurons by gad2 and zic1 promoter/enhancer elements in the dorsal hindbrain of developing zebrafish reveals neuronal connectivity related to the auditory and lateral line systems Dev. Dyn 2007 236 706 718 17279576 Scalia F The termination of retinal axons in the pretectal region of mammals J. Comp. Neurol 1972 145 223 257 4555428 Severi KE Portugues R Marques JC O’Malley DM Orger MB Engert F Neural control and modulation of swimming speed in the larval zebrafish Neuron 2014 83 692 707 25066084 Silies M Gohl DM Clandinin TR Motion-detecting circuits in flies: Coming into view Annu. Rev. Neurosci 2014 37 307 327 25032498 Sompolinsky H Computational neuroscience: Beyond the local circuit Curr. Opin. Neurobiol 2014 25 xiii xviii 24602868 Sun W May PJ Central pupillary light reflex circuits in the cat: I. The olivary pretectal nucleus J. Comp. Neurol 2014 522 3960 3977 24706328 Vanegas H Ito H Morphological aspects of the teleostean visual system: A review Brain Res 1983 287 117 137 6315186 Yonehara K Ishikane H Sakuta H Shintani T Nakamura-Yonehara K Kamiji NL Usui S Noda M Identification of retinal ganglion cells and their projections involved in central transmission of information about upward and downward image motion PLoS ONE 2009 4 e4320 19177171

PMC005xxxxxx/PMC5112117.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9432918 8591 Immunity Immunity Immunity 1074-7613 1097-4180 27793595 5112117 10.1016/j.immuni.2016.10.005 NIHMS821381 Article Dendritic Cells Regulate Extrafollicular Autoreactive B cells via T cells Expressing Fas and Fas Ligand Ols Michelle L. 1 Cullen Jaime L. 1 Turqueti-Neves Adriana 3 Giles Josephine 23 Shlomchik Mark J. 3* 1 Department of Laboratory Medicine, Yale University School of Medicine, New Haven, CT 06519, USA 2 Department of Immunobiology, Yale University School of Medicine, New Haven, CT 06519, USA 3 Department of Immunology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261 * Correspondence and Lead Contact: Mark J. Shlomchik, mshlomch@pitt.edu 7 10 2016 25 10 2016 15 11 2016 15 11 2017 45 5 10521065 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The extrafollicular (EF) plasmablast response to self-antigens that contain Toll-like receptor (TLR) ligands is prominent in murine lupus models and some bacterial infections, but the inhibitors and activators involved have not been fully delineated. Here, we used two conventional dendritic cell (cDC) depletion systems to investigate the role of cDCs on a classical TLR-dependent autoreactive EF response elicited in rheumatoid-factor B cells by DNA-containing immune complexes. Contrary to our hypothesis, cDC depletion amplified rather than dampened the EF response in Fas-intact but not Fas-deficient mice. Further, we demonstrated that cDC-dependent regulation requires Fas and Fas ligand (FasL) expression by T cells, but not Fas expression by B cells. Thus, cDCs activate FasL-expressing T cells that regulate Fas-expressing extrafollicular helper T cells (Tefh). These studies reveal a regulatory role for cDCs in B cell plasmablast responses and provide a mechanistic explanation for the excess autoantibody production observed in Fas-deficiency. Graphical abstract INTRODUCTION B cells are pivotal in the development of systemic autoimmune diseases, which are characterized by the selective loss of tolerance to self-Ags such as DNA, RNA, and IgG (Shlomchik, 2009). In addition to autoantibody-mediated damage, B cells promote disease by stimulating T cell activation and expansion, presumably via antigen (Ag) presentation, and possibly through cytokine secretion (Shlomchik, 2009). Autoreactive B cells are often activated in extrafollicular (EF) foci, via co-ligation of B cell receptors (BCRs) and toll-like receptors (TLRs) by antigens (Ags) containing DNA and RNA. These autoreactive, EF-localized, TLR-driven B cell responses consist of short-lived plasmablasts, as found in MRL/MpJ-Faslpr/J (MRL.Faslpr), and other autoimmune mouse models, such as NZB/W, as well as in human disease (Daridon et al., 2012; Hoyer et al., 2004; Tipton et al., 2015). The EF response is not restricted to autoimmunity, and perhaps evolved to provide pathogen-specific protection. Indeed, EF B cell foci develop in response to various types of T-dependent (TD) and T-independent (TI) Ags, including viruses and bacteria (Cunningham et al., 2007; Di Niro et al., 2015; Maclennan et al., 2003). Such pathogen-induced EF responses are normally transient, indicating that they are tightly regulated. However, in the case of autoimmunity, EF responses persist for the life of the animal, becoming sites of extensive and prolonged clonal expansion, hypermutation, and affinity maturation (William et al., 2002; William et al., 2005b). How normal EF responses become dysregulated in autoimmunity remains unclear. To study how autoreactive EF B cell responses are regulated, we have used mice transgenic (Tg)—either ectopically inserted or site-directed (sd)—for an immunoglobulin (Ig) heavy chain encoding the anti-IgG2aa rheumatoid factor (RF), AM14 (Sweet et al., 2010). In autoimmune prone MRL.Faslpr and MRL/MpJ mice, AM14 B cells spontaneously expand, class switch, hypermutate and develop into antibody forming cells (AFCs) within EF foci in the spleen (Sweet et al., 2010; William et al., 2002; William et al., 2005b). This spontaneous, TLR-dependent autoactivation occurs in vivo by AM14 BCR binding to DNA- and RNA-containing immune complexes (ICs) (Herlands et al., 2008). We can also induce EF AM14 activation on the BALB/c background by administration of a model physiologic ligand, the IgG2aa anti-chromatin antibody, PL2-3 (Herlands et al., 2007), which results in a plasmablast response that is phenotypically and histologically indistinguishable from the spontaneous model. Like the spontaneous RF and anti-nuclear response, the PL2-3-induced response is TLR7, TLR9, and MyD88 dependent, as well as autoantigen-specific (Herlands et al., 2008). In contrast to the unpredictable onset and variable magnitude of the spontaneous system, the PL2-3 induced response enables precise timing and a directed approach for determination of proximal factors required for autoreactive EF B cell activation. Thus, the PL2-3 induced response has been used by a number of labs to study how a classical autoreactive B cell response is induced (Giltiay et al., 2013; Nundel et al., 2015; Sang et al., 2014). We initially hypothesized that cDCs are critical supporting cells for the EF response, in part because cDCs are prominent components of such EF foci (Maclennan et al., 2003; William et al., 2002). An activating role for cDCs in the EF response has been inferred from experiments in which anti-CD40 was used to induce cDC proliferation, and appeared to thereby increase EF plasmablast survival in response to a TI-2 Ag (Maclennan et al., 2003). Although cDCs are best known for their interactions with T cells, a small body of literature indicates that cDCs can promote B cell activation. cDCs secrete factors, such as B cell activating factor (BAFF), a proliferation inducing ligand (APRIL), and interleukin-6 (IL-6), which support B cell development, activation and differentiation (MacLennan and Vinuesa, 2002; Mohr et al., 2009). cDCs have also been shown to deliver Ag to, and directly interact with, B cells in lymph nodes (Gonzalez et al., 2010; Qi et al., 2006), and in the spleen (Balázs et al., 2002), thereby promoting B cell activation and the development of humoral immunity. In particular, via a non-degradative Ag uptake and processing pathway, cDCs can present whole Ag on their cell surface to B cells (Bergtold et al., 2005; Zelenay et al., 2012). Although these data suggest that cDCs might aid B cell responses, a more direct cDC ablation study has shown that cDCs are not required for the generation of an EF TI-2 response to 4-hydroxy-3-nitrophenylacetyl (NP)-Ficoll (Hebel et al., 2006), calling into question earlier conclusions that DCs played nonredundant roles. The autoreactive response to nucleic acids and to the ICs containing them has characteristics of TI-1, TI-2 (Herlands et al., 2008), and TD responses (Sweet et al., 2011). Such nucleic acid-driven responses are therefore likely to be regulated differently from TI-2 responses. In addition, chromatin-containing immune complexes (ICs) have a distinctive capacity to stimulate myeloid cells through both the Fc gamma receptor (FcγR) and TLRs (Boulé et al., 2004). To ask whether cDCs are required for the activation of this type of autoreactive B cell response, we employed both acute (Jung et al., 2002) and constitutive (Birnberg et al., 2008; Teichmann et al., 2010) cDC ablation systems. Both cDC-depletion systems yielded the finding that cDCs negatively regulate, rather than activate, this autoreactive B cell response. Furthermore, we found that this cDC regulatory effect was dependent on Fas and was not observed in Fas-deficiency. We further determined that cDC-dependent regulation of the EF response required Fas and FasL expression by T cells; whereas, Fas expression on B cells was not relevant for this effect. We thus propose that cDCs activate FasL-expressing T cells to regulate Fas-expressing T cells that, in turn, help B cells in EF sites. In addition to revealing a clear regulatory role for cDCs, these studies elucidate an additional mechanism whereby Fas controls autoreactive EF plasmablast responses, helping explain how Fas-deficiency promotes hypergammaglobulinemia, high autoantibody titers, and autoimmunity. RESULTS The EF AM14 Rheumatoid Factor Response Does Not Acutely Require, but is Regulated by, cDCs To determine whether cDCs are required for the generation of the EF AM14 rheumatoid factor B cell response, we made radiation chimeras with bone marrow (BM) from mice expressing the diphtheria toxin (DT) receptor (DTR) and green fluorescent protein (GFP) under a Itgax promoter (Itgax-DTR) (Jung et al., 2002), administered DT to deplete cDCs, transferred in sd-Tg AM14 B cells, and injected the activating Ag, PL2-3 (schema in Figure S1). Repeated DT treatment is well tolerated in BM chimeras (Zammit et al., 2005). Because the transferred AM14 B cells do not carry the Itgax-DTR locus, CD11clo AM14 plasmablasts are not deleted by DT in this system (Hebel et al., 2006). The AM14 cells with rheumatoid factor specificity were tracked by ELISpot analysis and flow cytometry using the anti-idiotype marker, 4-44. Notably, we observed 3, 5, 4 and 7-fold increases in the numbers of 4-44+ AM14 cells, CD22lo CD44hi plasmablasts, IgM and IgG2a antibody forming cells (AFCs), respectively, in cDC-depleted hosts as compared to controls (Figure 1). The location of the response in the cDC-depleted animals was EF (Figure S2), similar to that observed in untreated controls, and as reported (Herlands et al., 2008; Herlands et al., 2007; Sweet et al., 2011; William et al., 2002). Few, if any 4-44+ peanut agglutinin (PNA)+ germinal centers (GCs) could be detected by histology at day 6 (not depicted). These data indicate that cDCs are not required for, and—given the several-fold enhancement in their absence—may regulate, the EF AM14 response. cDCs are not Required Constitutively for the AM14 Response Although acute depletion of cDCs suggested that they are not required for the activation of AM14 B cells, it remained possible that essential cytokines produced by cDCs are retained in the extracellular matrix (Mohr et al., 2009), in which case acute cDC depletion would not reveal an effect of cDCs on B cell activation. We were also concerned that acute local DT-induced cDC death could provide an additional source of chromatin Ag, that could account for the increased AM14 response observed in the acute cDC ablation system. To address these issues, we generated recipient mice constitutively lacking cDCs by crossing mice expressing the Rosa26-flox-stop-DTA locus (Ivanova et al., 2005) to mice expressing Itgax-Cre (Caton et al., 2007), which we call ΔDC mice. We had both strains of these mice fully backcrossed onto the MRL.Faslpr background (Teichmann et al., 2010), which were suitable for our studies, because we used young, pre-diseased animals, as previously reported (Herlands et al., 2007). As had been shown for ΔDC mice, there was extensive depletion of cDCs and a substantial loss of pDCs as well as increased CD11b+ cells (Birnberg et al., 2008; Ohnmacht et al., 2009; Teichmann et al., 2010) (Figure S3). We transferred sd-Tg AM14.MRL. Faslpr B cells into these recipients on day 0, administered the activating Ag, PL2-3 on days 0, 2 and 4, and assessed the response on day 6. In contrast to our acute ablation experiments in BALB/c mice, EF AM14 B cell responses in ΔDC MRL.Faslpr mice were similar to those of control recipients (Figure 2). These results confirm that cDCs are not required either acutely or constitutively for the activation of the autoreactive AM14 B cell response. That cDCs were not required for the EF response was somewhat unexpected, because MacLennan and colleagues have concluded that cDCs are a limiting factor in plasmablast generation and survival in response to TI-2 Ags (Maclennan et al., 2003). Additionally, since we did not observe enhanced EF responses in ΔDC mice, it remained possible that acute DT-induced cDC death within the spleen of Itgax-DTR mice enhanced the AM14 B cell response. Regulation of the AM14 Response by cDCs Requires Fas However, another important difference between the acute and the constitutive cDC ablation systems was the genetic background of the mice: BALB/c in the former, and MRL.Faslpr, in the latter. Because Fas-deficient mice have prominent spontaneous autoreactive EF responses (Shlomchik, 2009; William et al., 2002), it was possible that Fas could limit this EF response in normal mice. If this role of Fas were cDC-dependent, then depletion of cDCs in Fas-intact BALB/c mice would result in an increased response, as we observed (Figure 1). However, in Fas-deficient mice, cDC elimination would have no effect on the response, as we also observed (Figure 2). This logic is shown schematically in Figure S4. To test this idea, we compared constitutive cDC depletion on the same MRL genetic background, but in the presence or absence of functional Fas. Notably, in Fas-sufficient ΔDC MRL.Faslpr/+ mice, there was a large enhancement (3, 9, 2, 5, and 4 fold, respectively) of splenic 4-44+ AM14 B cell and plasmablast numbers, as well as, IgM, IgG2a, and IgG2b AFCs, compared to control recipients (Figure 3). Again, this AM14 response was EF (Figure 3G and S5). Hence, enhancement of the EF AM14 response upon cDC depletion was seen in both acute and constitutive cDC depletion, but only when Fas was intact. Therefore, we conclude that cDCs regulate the EF AM14 plasmablast response via a Fas-dependent mechanism. Fas-Dependent cDC Regulation of the AM14 Response is AM14 B Cell Extrinsic Next, we investigated which cell type was required to express Fas for proper regulation of the response. Because EF helper T cells (Odegard et al., 2008) enhance AM14 B cell expansion (Sweet et al., 2011), it was possible that FasL-killing of Fas-expressing helper T cells could limit EF B cell responses. Alternatively, because B cells are sensitive to FasL-mediated killing (Rothstein et al., 1995), Fas-expressing AM14 B cells could be directly targeted. To distinguish these possibilities, we transferred Fas-deficient sd-Tg AM14.MRL.Faslpr B cells into Fas-sufficient ΔDC.MRL.Faslpr/+ mice or littermate controls, and induced the AM14 response with PL2-3 (Figure 4). Although caution is required in comparing across experiments, as expected, the response size of these Fas-deficient B cells appeared larger than what we had observed in prior experiments. However, if Fas expression on B cells were responsible for cDC-dependent regulation of AM14 B cells, we would have expected to see, upon transfer of Fas-deficient B cells, similar responses within the ΔDC and cDC-intact mice. Rather, we still saw a large enhancement of the AM14 response within ΔDC recipient mice as compared to controls, even when the B cells could not have been influenced by FasL. In fact, the splenic 4-44+ AM14 B cell, plasmablast, and IgM, IgG2a, and IgG2b AFC numbers were 10, 20, 4, 9, and 9 fold higher, respectively, in recipients lacking cDCs than in controls. Thus, we conclude that cDCs regulate EF autoimmune responses in a Fas-dependent, B cell-extrinsic manner. Recipient Mice Lacking Fas, cDCs, or Both Have Higher Numbers of Activated CD4+ T Cells than Control Mice Since Fas-dependent killing of B cells was not the mechanism for Fas-dependent regulation of the AM14 B cell response by cDCs, regulation must depend on a different cDC-dependent, Fas-expressing cell. T cells that help the EF response (Lee et al., 2011; Odegard et al., 2008) would be a leading candidate for such a Fas-expressing cell. Reciprocally, FasL expression could be required either on other T cells or on cDCs (Lu et al., 1997; Süss and Shortman, 1996; Van Parijs and Abbas, 1996) in order to kill the Fas-expressing EF helper T cell (Tefh). The notion that cDCs could activate both Fas-sensitive Tefh and FasL-expressing killer T cells is reasonable, because T cells are regulated by Fas-FasL interactions (Van Parijs and Abbas, 1996), and cDCs can both activate and regulate T cell responses (Reis e Sousa, 2006). Furthermore, T cells can promote AM14 activation and differentiation via CD40L and IL-21 (Sweet et al., 2011). With these possible cDC-T and T-B interactions in mind, we first assessed our three systems of AM14 B cell transfer into ΔDC mice for cDC and Fas-dependent alterations in T cell activation (Figure 5). Fas-intact ΔDC MRL mice had a higher proportion of CD4+ T cells that were CD44hi, CD62Llo than the cDC-sufficient littermate controls. Absolute numbers of this T cell subset were also elevated. These cDC-dependent differences in the T cell compartment were Fas-dependent, because the increases were only seen in Fas-sufficient and not Fas-deficient MRL recipients. Of note, similar differences were present even in mice that had not received Ag or B cell transfer, although Ag injection with B cell transfer also enhanced T cell activation above controls (Figure S6). Further analysis of DT-treated Itgax-DTR.BALB/c bone marrow chimeras revealed that the increase in CD4+ T cells observed in DC-depleted mice included an increase in the PD-1+. Bcl-6+ Tefh cell subset (Figure S6). These data imply that cDC-dependent regulation of AM14 B cells involves the elimination of Fas-expressing Tefh cells. Regulation of AM14 Plasmablasts by cDCs Requires T Cells Expressing Fas and FasL To more definitively determine cell-specific roles for Fas and FasL in limiting autoreactive EF plasmablast responses, we performed a series of mixed bone marrow (BM) chimera experiments. First, we made mice in which irradiated BALB/c recipients received a mix of donor BM that was 80% from BALB/c.Tcra−/− mice and 20% from BALB/c.Faslpr mice. Thus, all T cells within the chimeric animals lacked Fas, while other cell types had normal Fas expression on the great majority of their cells (Figure 6). Control animals received BM that was either 100% BALB/c or a mix of 80% BALB/c.Tcra−/− and 20% BALB/c. In these experiments, when T cells lacked Fas, the AM14 B cell response was enhanced by 3, 7, and 10 fold for IgM, IgG2a, and IgG2b AFCs, respectively. Finally, we made mice in which irradiated BALB/c recipients received a mix of donor BM that was 80% from BALB/c.Tcra−/− mice and 20% from BALB/c.FasLmut mice (Gregory et al., 2011). In these chimeras, no T cells had active FasL, while other cell types were normal (Figure 7A-F). Control mice received BM that was either 100% BALB/c or a mix of 80% BALB/c.Tcra−/− and 20% BALB/c. Here, when T cells could not signal via FasL, AM14 IgM, IgG2a, and IgG2b AFC numbers were 2, 3, and 8 fold higher, respectively. As expected, when T cells lacked either Fas or FasL, an activated T cell subset was elevated in frequency (Figures 6F, G & 7F, G). These experiments confirm that cDC-dependent regulation of autoreactive plasmablasts requires FasL-expressing T cells to regulate Fas-expressing Tefh cells. Loss of cDCs Reduces Cell Death of Splenic CD4+ T Cells During the AM14 B Cell Response To determine if cDCs increase the ability of FasL-expressing T cells to kill Fas-expressing helper T cells during the AM14 B cell response, we used the TUNEL assay to detect apoptotic cells in tissue sections of DT-treated Itgax-DTR.BALB/c bone marrow chimeras. Indeed, less TUNEL+ CD4+ cells were detected in cDC-depleted than control mice (Figure 7G,H). These data further support the idea that cDC-dependent regulation of AM14 B cells involves the killing of Fas-expressing helper T cells by FasL-expressing T cells. DISCUSSION In normal responses to non-replicating TI and TD Ags, the EF response is transient (Maclennan et al., 2003), though the mechanisms that constrain it are not yet defined. A limiting component is likely the inherent propensity of plasmablasts to die, which does not depend on Fas (Do et al., 2000; Ursini-Siegel et al., 2002; William et al., 2005a). Although regulatory T (Treg) cells can restrain humoral immunity, they have been implicated in GC, rather than EF responses (Sage and Sharpe, 2015), and Treg cell neutralization with the anti-CD25 Ab PC61 had no effect in our system (not depicted). Determining the identity and function of the negative regulators of EF plasmablasts is important for understanding both normal (Di Niro et al., 2015) and pathogenic, autoimmune responses. That the EF plasmablast response was greatly enhanced in the absence of cDCs was an important and unexpected finding, as it revealed a regulatory rather than stimulatory role for cDCs in the humoral immune response. That cDCs were not even required for the EF response was also somewhat unexpected based on prior reports, albeit in different contexts. MacLennan and colleagues have concluded that cDCs are a limiting factor in plasmablast generation and survival in response to TI-2 Ags (Maclennan et al., 2003). CD11c+ cells are also prominently seen in juxtaposition with dividing and differentiating EF AM14 B cells (William et al., 2002), supporting the notion that cDCs are critical in promoting the response. However, an essential role of cDCs in the EF response has been called into question by Jung, et. al., who found that cDC depletion does not alter the extent of a TI-2 response elicited by NP-Ficoll (Jung et al., 2002). Of note, the response we tested was not a TI-2 response, but one with qualities of TI-1 (given the TLR requirement), TI-2 (giventhe polymeric nature of the Ag) and TD (given the ability of T cells, when present, to enhance it). Because EF responses are typical of autoreactive and probably bacterial (Di Niro et al., 2015) and viral pathogen responses, the EF response we studied may be more physiologically relevant than idealized TI or TD protein antigens given in artificial adjuvant. There have been few reports of negative regulation of B cells by cDCs. Three groups have observed inhibition of B cell responses by cDCs in vitro via either an IL-6 and soluble CD40L-dependent mechanism (Gilbert et al., 2007), or a mechanism requiring CD22 expression by the B cells (Santos et al., 2008; Sindhava et al., 2012). None of these systems was translated in vivo. Moreover, it is unlikely that they relate directly to our in vivo observations. In the former case only anergic B cells are affected, but AM14 B cells are not anergic (Hannum et al., 1996). In the latter cases CD22 is required, but CD22 is rapidly downregulated early in plasmablast differentiation (William et al., 2005b). Use of multiple systems for cDC depletion allowed us to rule out some alternate explanations for the enhanced EF response observed in the absence of cDC. Because we saw an elevated AM14 response in both the transient and constitutive cDC depletion systems the acute release of Ag (ie. chromatin) from dying CD11c+ cells that might occur upon DT treatment in the Itgax-DTR system was not the primary cause of the enhanced AM14 responses seen in the absence of cDCs. Macrophage hyperplasia, which we observed to some degree in both cDC depletion systems and as noted by others (Birnberg et al., 2008), was also likely not the cause of enhanced EF responses, because this macrophage expansion was also present within cDC-deficient MRL.Faslpr mice, which did not have an altered frequency of AM14 AFCs. Finally, a role for pDCs is unlikely, because pDCs were not depleted in the CD11c-DTR system, in which acute cDC depletion substantially increased the AM14 AFC response. We found that the mechanism by which cDCs regulate the EF autoimmune response in vivo was dependent on Fas: cDCs did not regulate the AM14 response in Fas-deficient animals. In this case, the regulatory pathway was short-circuited because the Fas-signaling was already “off.” Depleting cDCs in addition to Fas, therefore, had no additional effect. Fas has previously been implicated in controlling autoimmunity via a number of mechanisms, but the pathway revealed by our studies appears different. For example, FasL and Fas are required for elimination of anergic B cells by CD4+ T cells in vivo (Rathmell et al., 1995). However, AM14 B cells are not anergic (but remain ignorant), and do not need to escape anergy to break tolerance (Hannum et al., 1996; Wang and Shlomchik, 1999). GC B cells upregulate Fas, which regulates the duration and extent of the GC reaction and the generation of long-lived plasma cells (Butt et al., 2015; Hao et al., 2008; Takahashi et al., 2001). In contrast, the response we are studying is EF and generates short-lived plasmablasts. cDC regulation did not depend on Fas expression by the B cell itself, even though activated B cells can be sensitive to FasL-mediated killing (Rothstein et al., 1995). Rather, we demonstrated that the FasL targets were Fas-expressing T cells that help amplify the EF B cell response, namely Tefh cells (Lee et al., 2011; Odegard et al., 2008; Sweet et al., 2011). Because Tfh cells express Fas (Rasheed et al., 2006), we would expect that the Tefh cells in our system would also be sensitive to elimination by a FasL-expressing cell: either another T cell, the activation of which is cDC-dependent, or the cDC itself (Lu et al., 1997; Süss and Shortman, 1996). We also showed that the FasL-expressing cells required in this system were indeed T cells. There is precedent for such a mechanism, because regulation of T cells by Fas-FasL interactions is well documented (Van Parijs and Abbas, 1996), and T cell restricted Fas- or FasL-deficiency leads to autoantibody production (Mabrouk et al., 2008; Stranges et al., 2007). Seo et al also noted that B cell-specific Fas expression is not required to regulate autoreactive B cell responses to strong T cell help (Seo et al., 2002). Thus, our observations are consistent with prior studies, but extend our knowledge by demonstrating that cDCs can regulate an EF plasmablast response via a mechanism that requires Fas ligation on T cells—most likely Tefh cells—rather than on B cells directly. We previously showed, using constitutive cDC-depletion, that cDCs can amplify disease in aged MRL.Faslpr mice, although they are not required for disease initiation (Teichmann et al., 2010). These studies did not reveal a regulatory role for cDCs, and in fact, certain types of autoantibodies were reduced in aged cDC-deficient MRL.Faslpr mice. Similarly, Fas expression on cDCs suppresses autoimmunity in older mice (Stranges et al., 2007). This presents the paradox that, in the short term, cDCs negatively regulate the initiation of the EF response, whereas in the long-term cDCs promote specific, chronic aspects of disease. We suggest two ways to understand this paradox. First, as in many complex cellular networks in vivo, cDCs likely have both positive and negative effects. When integrated over the life of an autoimmune-prone animal, these effects are seen as mainly promoting disease, in particular, by aiding the growth of tissue cellular infiltrates (Teichmann et al., 2015; Teichmann et al., 2010). Over time cDC-driven positive feedback cycles of T cell activation could override any short-term negative effects. Second, and most relevant to the present context, our previous study of constitutively cDC-deficient mice was on the Fas-deficient MRL.Faslpr background. In that model, the regulatory effects of cDCs would not have been observed, because the Fas-dependent mechanism was already short-circuited. These insights into the Fas-dependent regulatory role of cDCs in the generation of autoantibodies could apply to other EF plasmablast responses. The EF response can be driven by simultaneous TLR and BCR signals. In the AM14 system, chromatin-containing ICs promote the EF response in a TLR7- and TLR9-dependent manner (Herlands et al., 2008; Leadbetter et al., 2002). A more general effect of combined TLR and BCR stimulation in directing the EF response has been demonstrated in elegant studies in which a BCR signal was engineered to have a TLR9 signal associated with it on the same synthetic bead (Eckl-Dorna and Batista, 2009). Similarly, many bacteria and viruses contain Ags that can directly stimulate B cell-expressed TLR4, 7, and 9 as well as ligate relevant BCRs. This dual recognition promotes vigorous EF plasmablast responses that can undergo isotype switch, and is probably critical for the early control of pathogens like Salmonella (Cunningham et al., 2007; Di Niro et al., 2015; Neves et al., 2010). Thus, the cDC-regulated pathway we have observed could also be important during the primary response to pathogens. Initial EF responses are typically supplanted by the GC response, which provides definitive B cell memory and long-lived AFCs. GC and T-dependent responses engage multiple mechanisms to enforce self-tolerance. In contrast, the primary EF response is less selective and partially T-independent, thus making it more liable to be subverted into autoimmunity. Here we have demonstrated a Fas- and cDC-dependent pathway that reduces the magnitude of the EF response by 3-6 fold over the course of a week. cDCs are, therefore, potent regulators of the autoreactive EF response, and possibly of EF responses in general. EXPERIMENTAL PROCEDURES Mice BALB/c, BALB/c.Rag1−/−, MRL/MpJ- Faslpr/J (MRL. Faslpr), MRL/MpJ and BALB/c FVB-Tg(Itgax-DTR/EGFP)57Lan/J (Itgax-DTR.BALB/c) (Jung et al., 2002) mice were purchased from Jackson Laboratories (Bar Harbor, Maine). BALB/c sd-Tg AM14 mice were previously described (Sweet et al., 2011), and were also backcrossed greater than 10 times onto MRL.Faslpr. AM14.BALB/c mice were used at 7-9 weeks of age. AM14.MRL.Faslpr mice were used at 5-7 weeks of age. AM14.MRL.Faslpr/+ mice were used at 5-10 weeks of age. C57BL/6 CD11c-Cre BAC transgenic (Caton et al., 2007) and Rosa26-eGFP-flox-stop-DTA transgenic (Ivanova et al., 2005) mice, gifts from Boris Reizis (Columbia University) and Juan Martinez-Barbera (University College London), respectively, were backcrossed greater than 10 times onto MRL.Faslpr (Teichmann et al., 2010). These two strains were then intercrossed to obtain cDC-deficient (ΔDC MRL.Faslpr), CD11c-Cre+ Rosa26-eGFP-DTA+ mice. Single positive and WT littermates were used as controls. Mice were used at 6-7, 5-9.5 and 7-9 weeks of age in three experiments with equivalent results. To obtain Fas-sufficient cDC-deficient mice (ΔDC.MRL.Faslpr/+) mice and controls, double positive, ΔDC.MRL.Faslpr females were outcrossed to MRL/MpJ males for one generation. Mice were used at 6-7 weeks of age in three experiments and 6-10 weeks of age in two experiments, with equivalent results. BALB/c.Faslpr mice were obtained from Thomas Ferguson (Washington University) and used at 4-9 weeks of age. BALB/c.Tcra−/− mice (Sweet et al., 2011) were obtained from Kim Bottomly (Yale University) and used at 5-7 weeks of age. BALB/c.FasLmut mice, gifted from Ann Marshak-Rothstein (Gregory et al., 2011) were used at 7-9 weeks of age. All animals were housed under SPF conditions and handled according to IACUC approved protocols. B cell Purification and Adoptive Cell Transfer Splenocyte suspensions were made in 2% fetal calf serum (FCS), 1mM EDTA, phosphate buffered saline (PBS). B cells were purified using an EasySep Mouse B cell Enrichment Kit (Stemcell Technologies). 3-12 × 106 cells suspended in PBS per mouse (as indicated) were injected intravenously (IV) on day 0. Anti-Chromatin Preparation and Immunization PL2-3, an IgG2aa anti-chromatin hybridoma (Losman et al., 1992) was prepared as ascites in Rag−/− mice as described (Herlands et al., 2008; Sweet et al., 2011). PL2-3 protein concentration was determined by IgG2a ELISA. Mice were immunized intraperitoneally (IP) with 0.5 mg equivalent of PL2-3 on days 0, 2, and 4. AM14 B cell activation was assayed on day 6. Bone Marrow Chimeras and Diphtheria Toxin Treatment BM from CD11c-DTR donors or Tg-negative littermates was processed in 0.5% bovine serum albumin (BSA), 5mM EDTA, penicillin, streptomycin in PBS. After red blood cell lysis with ACK solution, BM cells were washed with BSA, PBS and resuspended in injection buffer (2.5% v/v ACD-A, 10mM HEPES, penicillin, streptomycin, PBS). Eight week old BALB/c recipients received 450 Rad of irradiation twice from a 137Cs source with a 3-4 h rest period between doses. 1-2 h after the second dose, 8-10 × 106 BM cells were injected IV. Animals were allowed to recover for 6 weeks prior to cDC depletion or B cell transfer. Diphtheria toxin (Sigma) was administered IP at 4-8 ng/g weight on days -2 and -1 prior to and on days 1, 3 and 5 after B cell transfer. Effectiveness of chimerism and depletion was assessed by flow cytometry of splenocytes (Figure S1). ELISpot Assay ELISpot assays were performed as described (Hannum et al., 1996). Spots were counted using a dissecting microsope. Flow Cytometry Splenocytes were suspended in staining media (SM: 3% FCS, 5 mM EDTA, 0.05% sodium azide, PBS) with the FcR blocking antibody, 2.4G2 and ethidium monoazide (EMA) for live/dead discrimination. Reagents used for staining were either prepared in house: 4-44-biotin, PDCA1 (927)-Alexa 647, CD62L (MEL-14)-Alexa 647; or purchased from Biolegend: CD44 (1M7)-APC/Cy7, CD11c (N418)-PE/Cy7, CD4 (GK1.5)-PE/Cy7, Gr1 (RB6.8C5)-biotin; from BD: CD22.2 (Cy34.1)-PE, TCRβ (H57-597)-PE, I-A/I-E (M5/114.15.2)-PE; from eBioscience: CD11b (MAC-1)-PE, streptavidin-PE/Cy7; or from Invitrogen: streptavidin-PacBlue. Surface stained cells were fixed with 1% PFA, washed and resuspended in SM. For intracellular staining, cells were fixed in 4% PFA in BD Perm/Wash Buffer, washed, blocked (with 5% rat serum), and stained with 4-44-Alexa 647 or GFP-FITC (Rockland) in the BD Perm/Wash Buffer, with a final wash in SM. Cytometry was performed on a LSRII (BD) and data analyzed with FlowJo software. Immunofluorescence Histology CD11c-DTR and control spleens were fixed in paraformaldehyde-lysine-periodate (PLP) solution (1% paraformaldehyde, 95 mM L-lysine, 10 mM sodium M-periodate, 0.1 M phosphate buffer (PB), pH 7.2) for 1-7 days at 4°C. They were then washed three times with PB, dehydrated stepwise in 10%, 20%, and 30% sucrose/PB and frozen in OCT (TissueTek). Alternatively, for Bcl-6 and TUNEL staining, spleens were fixed with 4% paraformaldehyde for 2-4h at 4°C and washed for 2-4h in cold PBS and frozen in OCT medium. Spleens from CD11c-cre × Rosa26-eGFP-flox-stop-DTA mice were first frozen in OCT and fixed after cryostat sectioning with 4°C acetone for 10 min. 7-10 μm cryostat sections were cut, rehydrated with PBS, blocked for 20 min with 10% rat serum, 1% BSA, 0.1% Tween-20 in PBS, stained for 1 h with respective antibodies in BSA, Tween-20, PBS, washed three times with BSA, Tween-20, PBS and once with PBS. Staining reagents were either prepared in house: 4-44-biotin, B220 (RA3-6B2)-488, B220 (RA3-6B2)-Dylight 405, B220 (RA3-6B2)-PE, CD4 (GK1.5)-Alexa 647, F4/80 (BM8)-Alexa 647, PNA-Alexa 647, CD19 (1D3.2)-Alexa 647; or purchased from Invitrogen: streptavidin-Alexa 555; from Vector Laboratories: PNA-FITC; from BD: anti-Bcl-6 (K112-91)-Alexa 647; from BioLegend: anti-CD4 (GK1.5)-PE. For the visualization of apoptotic cells by the TUNEL method, ApoAlert DNA Fragmentation Assay Kit (Clontech) was used as per manufacturer’s instructions. After air-drying, slides were mounted with Prolong Antifade (Molecular Probes). Images were captured either with a 10 × lens on a Nikon Exlipse Ti automated wide field microscope with a QImaging Retiga 200R CCD camera (Figures 3, S2, & S5) using NIS Elements software or with a 20 × lens on an Olympus IX83 microscope (Figures 7 & 6S) with an ORCA-flash 4.0 camera (Hamamatsu) using Olympus CellSens Dimension 1.15 software. Images were further processed with Adobe Photoshop software, and CellSens software assisted in TUNEL+ cell quantitation. Statistical Analysis P-values were calculated by one-way ANOVA with Tukey’s multiple comparisons test, two-tailed Mann-Whitney U-test, or unpaired T-test, as indicated, using Prism software (Graphpad). Supplementary Material 1 2 ACKNOWLEDGMENTS We thank Dr. Juan Martinez-Barbera and Boris Reizis for Rosa26-eGFP-DTA and CD11c-cre mice, respectively. Cuiling Zhang, Yuqi Zhang and Colin Smith provided outstanding technical assistance. Kevin Nickerson assisted in our experimental response to the reviewers. We thank the Yale Animal Resources Center and the Yale Cell Sorter Facility for their excellent services. This work was supported by the National Institutes of Health grant AI073722 (to MJS), a Ruth. L. Kirschstein National Research Service Award NIH T32 AI07019-30 (to MLO), and an Arthritis Foundation Postdoctoral Fellowship (to MLO). Figure 1 The EF AM14 Rheumatoid Factor Response does not Acutely Require, but is Regulated by cDCs (A) Experiment design. Itgax-DTR.BALB/c or Tg-negative littermates were used to make BM chimeras. cDCs were depleted with DT on days -1, -2, 1, 3 and 5. 9-10 × 106 AM14.BALB/c B cells were transferred into the mice on day 0, and activated with PL2-3 on days 0, 2 and 4. Spleens were harvested on day 6. (B) Representative flow cytometry gating of live AM14 B cells by surface and intracellular 4-44 (anti-idiotype) staining, and further subgating for CD22lo, CD44hi plasmablasts. (C) Number of AM14 B cells (surface and intracellular 4-44+) per spleen. (D) Number of AM14 plasmablasts (4-44+, CD22lo, CD44hi) per spleen. (E & F) ELISpot assays for the number of 4-44+, IgM+ or IgG2a+ AFCs per spleen. Data were combined from 2 experiments to obtain 4-8 mice per group. Bars represent mean and SEM. *p < 0.05; **p < 0.01 by ANOVA. See also Figures S1 and S2. Figure 2 cDCs are not Required Constitutively for the AM14 Response (A) Experiment design. 4 × 106 AM14.MRL.Faslpr B cells were transferred into MRL.Faslpr recipients that were either double positive for both the CD11c-cre and Rosa26-flox-stop-DTA Tgs, and thus lacked cDCs (ΔDC mice), or into single Tg positive littermates (Cntrl) on day 0. The AM14 B cell response was activated with PL2-3 on days 0, 2 and 4, or left untreated (no Ag). As additional controls, some mice did not receive cell transfer or Ag (no Tfr). Spleens were harvested on day 6. (B) Number of AM14 B cells (4-44+) per spleen. (C) Number of AM14 plasmablasts (4-44+, CD22lo, CD44hi) per spleen. (D, E, & F) ELISpot assays for the number of 4-44+, IgM+, IgG2a+, or IgG2b+ AFCs per spleen. Data were combined from 3 experiments to obtain 12-24 mice per group receiving PL2-3 and 2-12 mice in the untreated groups. Bars represent mean and SEM. See also Figure S3. Figure 3 Regulation of the AM14 Response by cDCs Requires Fas (A) Experiment design. 8-9 × 106 Fas-sufficient AM14.MRL.Faslpr/+ B cells were transferred into Fas-sufficient MRL.Faslpr/+ recipients that either lacked cDCs (ΔDC) or into littermate controls (Cntrl) on day 0. The AM14 B cell response was activated with PL2-3 on days 0, 2 and 4, or left untreated (no Ag). Some mice did not receive cell transfer or Ag (no Tfr). Spleens were harvested on day 6. (B) Number of AM14 B cells (4-44+) per spleen. (C) Number of AM14 plasmablasts (4-44+, CD22lo, CD44hi) per spleen. (D, E, & F) ELISpot assays for the number of 4-44+, IgM+, IgG2a+, or IgG2b+ AFCs per spleen. Data were combined from 5 experiments to obtain 15-29 mice per group receiving PL2-3 and 2-7 mice in the other groups. Bars represent mean and SEM. ****p < 0.0001 by Mann-Whitney U test. (G) Immunofluorescence histology was performed on spleens from experimental mice. Images were taken at a magnification of 100x. Red, AM14 B cells (4-44+); green, B cell follicles (B220+); blue, T cell zones (CD4+). White scale bars are 200 μm. See also Figures S4 and S5. Figure 4 Fas-Dependent cDC Regulation of the AM14 Response is AM14 B Cell Extrinsic (A) Experiment design. 8 × 106 Fas-deficient AM14.MRL.Faslpr B cells were transferred into either Fas-sufficient, cDC-deficient (ΔDC) MRL.Faslpr/+ recipients or into littermate controls (Cntrl) on day 0. The AM14 B cell response was activated with PL2-3 on days 0, 2 and 4. Some mice did not receive cell transfer or Ag (no Tfr). Spleens were harvested on day 6. (B) Number of AM14 B cells (surface and intracellular 4-44+) per spleen. (C) Number of AM14 plasmablasts (4-44+, CD22lo, CD44hi) per spleen. (D, E, & F) ELISpot assays for the number of 4-44+, IgM+, IgG2a+, or IgG2b+ AFCs per spleen. Data were combined from 2 experiments to obtain 8-15 mice per group receiving PL2-3 and 2 no Tfr controls. Bars represent mean and SEM. ***p < 0.001 by Mann-Whitney U test. See also Figure S7. Figure 5 Recipient Mice Lacking Fas and/or Dendritic Cells Have Higher Numbers of Activated CD4+ T Cells than Control Mice Splenic T cells from experimental mice shown in Figures 2-4 were analyzed by flow cytometry. (A) Representative sample showing the gating of live splenocytes for CD4+, TCRβ+ cells (top) that were further sub-gated (bottom) by their activation markers (CD44hi, CD62Llo). (B) Number of CD4+, TCRβ+ cells per spleen. (C) Percent of CD4+, TCRβ+ cells that are activated (CD44hi, CD62Llo). (D) Number of CD4+, TCRβ+, CD44hi, CD62Llo cells per spleen. ΔDC, mice that lack cDCs; Cntrl, littermate controls. Bars represent mean and SEM. **p < 0.01; ***p < 0.001 by Mann-Whitney U test. See also Figure S6. Figure 6 Regulation of AM14 Plasmablasts by cDCs Requires T Cells Expressing Fas (A) Experiment design. BALB/c mice were used to make chimeras with BM mixed from WT, Tcra−/− or Faslpr mice. 3 × 106 AM14.BALB/c B cells were transferred into the mice on day 0, and activated with PL2-3 on days 0, 2 and 4. Spleens were harvested on day 6. (B, C, & D) ELISpot assays for the number of 4-44+, IgM+, IgG2a+, or IgG2b+ AFCs per spleen. (E) Number of CD4+, TCRβ+ cells per spleen. (F) Percent of CD4+, TCRβ+ cells that are activated (CD44hi, CD62Llo). (G) Number of CD4+, TCRβ+, CD44hi, CD62Llo cells per spleen. Data were combined from 2 experiments to obtain 6-12 mice per group receiving PL2-3 and 4-8 controls. Bars represent mean and SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 by Mann-Whitney U test. See also Figure S7. Figure 7 Regulation of AM14 Plasmablasts by cDCs Requires T Cells Expressing FasL and Increases the Death of CD4 T Cells (A-F) Experiments performed as described and arranged in same order as in Figure 6, except BM was mixed from WT, Tcra−/− or FasLmut mice. Data were combined from 2 experiments to obtain 6-12 mice per group receiving PL2-3 and 3-8 controls. (G & H) As described in Figure 1, CD11c-DTR.BALB/c BM chimeras were generated and depleted of cDCs using 16 ng per g weight doses of DT. 8 × 106 AM14.BALB/c B cells were transferred and activated with PL2-3. Spleens were harvested on day 7. Data for untreated versus DT-treated mice are shown. (G) Representative histology of the spleen, taken at a magnification of 200x. Arrowheads indicate TUNEL+, CD4+ cells. Inset upper left shows a digital magnification of a TUNEL+ CD4+ cell in the white box of the larger image. Scale bars are 50 μm. (H) Quantification of TUNEL+, CD4+ cells from a total of 8 fields combined from 2 representative mice per group. Bars represent mean and SEM. *p < 0.05; **p < 0.01; ***p < 0.001 by Mann-Whitney U test. See also Figures S6 and S7. HIGHLIGHTS - cDCs are not required for extrafollicular (EF) autoreactive B cell activation - Loss of cDCs markedly enhances the EF B cell response - cDC regulation of the EF response requires B cell extrinsic Fas - Loss of Fas or FasL on T cells enhances autoantibody production IN BRIEF cDCs negatively regulate extrafollicular B activation and autoantibody production. Using two cDC ablation systems, Ols and colleagues observe elevated autoantibody production in the absence of cDCs. They further show that the regulatory effect of cDCs on B cells requires Fas and FasL expression by T cells. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. SUPPLEMENTAL INFORMATION Supplemental information includes seven figures and supplemental methods for PD-1 and Bcl-6 staining by flow cytometry. AUTHOR CONTRIBUTIONS M.L.O. designed and performed experiments, analyzed data, and wrote the manuscript. J.L.C., A.T-N., and J.G. performed experiments and analyzed data. M.J.S. designed experiments, supervised the study, and wrote the manuscript. The authors declare that they have no competing financial interests. REFERENCES Balázs M Martin F Zhou T Kearney J Blood dendritic cells interact with splenic marginal zone B cells to initiate T-independent immune responses Immunity 2002 17 341 352 12354386 Bergtold A Desai DD Gavhane A Clynes R Cell surface recycling of internalized antigen permits dendritic cell priming of B cells Immunity 2005 23 503 514 16286018 Birnberg T Bar-On L Sapoznikov A Caton ML Cervantes-Barragán L Makia D Krauthgamer R Brenner O Ludewig B Brockschnieder D Lack of conventional dendritic cells is compatible with normal development and T cell homeostasis, but causes myeloid proliferative syndrome Immunity 2008 29 986 997 19062318 Boulé MW Broughton C Mackay F Akira S Marshak-Rothstein A Rifkin IR Toll-like receptor 9-dependent and -independent dendritic cell activation by chromatin-immunoglobulin G complexes The Journal of experimental medicine 2004 199 1631 1640 15197227 Butt D Chan TD Bourne K Hermes JR Nguyen A Statham A O'Reilly LA Strasser A Price S Schofield P FAS Inactivation Releases Unconventional Germinal Center B Cells that Escape Antigen Control and Drive IgE and Autoantibody Production Immunity 2015 42 890 902 25979420 Caton ML Smith-Raska MR Reizis B Notch-RBP-J signaling controls the homeostasis of CD8-dendritic cells in the spleen The Journal of experimental medicine 2007 204 1653 1664 17591855 Cunningham AF Gaspal F Serre K Mohr E Henderson IR Scott-Tucker A Kenny SM Khan M Toellner KM Lane PJ MacLennan IC Salmonella induces a switched antibody response without germinal centers that impedes the extracellular spread of infection J Immunol 2007 178 6200 6207 17475847 Daridon C Loddenkemper C Spieckermann S Kuhl AA Salama A Burmester GR Lipsky PE Dorner T Splenic proliferative lymphoid nodules distinct from germinal centers are sites of autoantigen stimulation in immune thrombocytopenia Blood 2012 120 5021 5031 22955923 Di Niro R Lee SJ Vander Heiden JA Elsner RA Trivedi N Bannock JM Gupta NT Kleinstein SH Vigneault F Gilbert TJ Salmonella Infection Drives Promiscuous B Cell Activation Followed by Extrafollicular Affinity Maturation Immunity 2015 43 120 131 26187411 Do RK Hatada E Lee H Tourigny MR Hilbert D Chen-Kiang S Attenuation of apoptosis underlies B lymphocyte stimulator enhancement of humoral immune response The Journal of experimental medicine 2000 192 953 964 11015437 Eckl-Dorna J Batista FD BCR-mediated uptake of antigen linked to TLR9 ligand stimulates B-cell proliferation and antigen-specific plasma cell formation Blood 2009 113 3969 3977 19144984 Gilbert MR Carnathan DG Cogswell PC Lin L Baldwin AS Vilen BJ Dendritic cells from lupus-prone mice are defective in repressing immunoglobulin secretion J Immunol 2007 178 4803 4810 17404261 Giltiay NV Lu Y Cullen JL Jorgensen TN Shlomchik MJ Li X Spontaneous loss of tolerance of autoreactive B cells in Act1-deficient rheumatoid factor transgenic mice J Immunol 2013 191 2155 2163 23904159 Gonzalez SF Lukacs-Kornek V Kuligowski MP Pitcher LA Degn SE Kim Y Cloninger MJ Martinez-Pomares L Gordon S Turley SJ Carroll MC Capture of influenza by medullary dendritic cells via SIGN-R1 is essential for humoral immunity in draining lymph nodes Nat Immunol 2010 11 427 434 20305659 Gregory MS Hackett CG Abernathy EF Lee KS Saff RR Hohlbaum AM Moody KS Hobson MW Jones A Kolovou P Opposing roles for membrane bound and soluble Fas ligand in glaucoma-associated retinal ganglion cell death PLoS One 2011 6 e17659 21479271 Hannum LG Ni D Haberman AM Weigert MG Shlomchik MJ A disease-related rheumatoid factor autoantibody is not tolerized in a normal mouse: implications for the origins of autoantibodies in autoimmune disease The Journal of experimental medicine 1996 184 1269 1278 8879198 Hao Z Duncan GS Seagal J Su Y Hong C Haight J Chen N Elia A Wakeham A Li WY Fas receptor expression in germinal-center B cells is essential for T and B lymphocyte homeostasis Immunity 2008 29 615 627 18835195 Hebel K Griewank K Inamine A Chang HD Müller-Hilke B Fillatreau S Manz RA Radbruch A Jung S Plasma cell differentiation in T-independent type 2 immune responses is independent of CD11c(high) dendritic cells Eur J Immunol 2006 36 2912 2919 17051619 Herlands RA Christensen SR Sweet RA Hershberg U Shlomchik MJ T cell-independent and toll-like receptor-dependent antigen-driven activation of autoreactive B cells Immunity 2008 29 249 260 18691914 Herlands RA William J Hershberg U Shlomchik MJ Anti-chromatin antibodies drive in vivo antigen-specific activation and somatic hypermutation of rheumatoid factor B cells at extrafollicular sites Eur J Immunol 2007 37 3339 3351 18034429 Hoyer BF Moser K Hauser AE Peddinghaus A Voigt C Eilat D Radbruch A Hiepe F Manz RA Short-lived plasmablasts and long-lived plasma cells contribute to chronic humoral autoimmunity in NZB/W mice The Journal of experimental medicine 2004 199 1577 1584 15173206 Ivanova A Signore M Caro N Greene ND Copp AJ Martinez-Barbera JP In vivo genetic ablation by Cre-mediated expression of diphtheria toxin fragment A Genesis 2005 43 129 135 16267821 Jung S Unutmaz D Wong P Sano G De los Santos K Sparwasser T Wu S Vuthoori S Ko K Zavala F In vivo depletion of CD11c(+) dendritic cells abrogates priming of CD8(+) T cells by exogenous cell-associated antigens Immunity 2002 17 211 220 12196292 Leadbetter EA Rifkin IR Hohlbaum AM Beaudette BC Shlomchik MJ Marshak-Rothstein A Chromatin-IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors Nature 2002 416 603 607 11948342 Lee SK Rigby RJ Zotos D Tsai LM Kawamoto S Marshall JL Ramiscal RR Chan TD Gatto D Brink R B cell priming for extrafollicular antibody responses requires Bcl-6 expression by T cells The Journal of experimental medicine 2011 208 1377 1388 21708925 Losman MJ Fasy TM Novick KE Monestier M Monoclonal autoantibodies to subnucleosomes from a MRL/Mp(−)+/+ mouse. Oligoclonality of the antibody response and recognition of a determinant composed of histones H2A, H2B, and DNA J Immunol 1992 148 1561 1569 1371530 Lu L Qian S Hershberger PA Rudert WA Lynch DH Thomson AW Fas ligand (CD95L) and B7 expression on dendritic cells provide counter-regulatory signals for T cell survival and proliferation J Immunol 1997 158 5676 5684 9190916 Mabrouk I Buart S Hasmim M Michiels C Connault E Opolon P Chiocchia G Lévi-Strauss M Chouaib S Karray S Prevention of autoimmunity and control of recall response to exogenous antigen by Fas death receptor ligand expression on T cells Immunity 2008 29 922 933 19013083 MacLennan I Vinuesa C Dendritic cells, BAFF, and APRIL: innate players in adaptive antibody responses Immunity 2002 17 235 238 12354377 Maclennan IC Toellner KM Cunningham AF Serre K Sze DM Zúñiga E Cook MC Vinuesa CG Extrafollicular antibody responses Immunol Rev 2003 194 8 18 12846803 Mohr E Serre K Manz RA Cunningham AF Khan M Hardie DL Bird R Maclennan IC Dendritic cells and monocyte/macrophages that create the IL-6/APRIL-rich lymph node microenvironments where plasmablasts mature J Immunol 2009 182 2113 2123 19201864 Neves P Lampropoulou V Calderon-Gomez E Roch T Stervbo U Shen P Kühl AA Loddenkemper C Haury M Nedospasov SA Signaling via the MyD88 adaptor protein in B cells suppresses protective immunity during Salmonella typhimurium infection Immunity 2010 33 777 790 21093317 Nundel K Green NM Shaffer AL Moody KL Busto P Eilat D Miyake K Oropallo MA Cancro MP Marshak-Rothstein A Cell-intrinsic expression of TLR9 in autoreactive B cells constrains BCR/TLR7-dependent responses J Immunol 2015 194 2504 2512 25681333 Odegard JM Marks BR DiPlacido LD Poholek AC Kono DH Dong C Flavell RA Craft J ICOS-dependent extrafollicular helper T cells elicit IgG production via IL-21 in systemic autoimmunity The Journal of experimental medicine 2008 205 2873 2886 18981236 Ohnmacht C Pullner A King SB Drexler I Meier S Brocker T Voehringer D Constitutive ablation of dendritic cells breaks self-tolerance of CD4 T cells and results in spontaneous fatal autoimmunity The Journal of experimental medicine 2009 206 549 559 19237601 Qi H Egen JG Huang AY Germain RN Extrafollicular activation of lymph node B cells by antigen-bearing dendritic cells Science 2006 312 1672 1676 16778060 Rasheed AU Rahn HP Sallusto F Lipp M Muller G Follicular B helper T cell activity is confined to CXCR5(hi)ICOS(hi) CD4 T cells and is independent of CD57 expression Eur J Immunol 2006 36 1892 1903 16791882 Rathmell JC Cooke MP Ho WY Grein J Townsend SE Davis MM Goodnow CC CD95 (Fas)-dependent elimination of self-reactive B cells upon interaction with CD4+ T cells Nature 1995 376 181 184 7603571 Reis e Sousa C Dendritic cells in a mature age Nature reviews 2006 6 476 483 Rothstein TL Wang JK Panka DJ Foote LC Wang Z Stanger B Cui H Ju ST Marshak-Rothstein A Protection against Fas-dependent Th1-mediated apoptosis by antigen receptor engagement in B cells Nature 1995 374 163 165 7533263 Sage PT Sharpe AH T follicular regulatory cells in the regulation of B cell responses Trends Immunol 2015 36 410 418 26091728 Sang A Niu H Cullen J Choi SC Zheng YY Wang H Shlomchik MJ Morel L Activation of rheumatoid factor-specific B cells is antigen dependent and occurs preferentially outside of germinal centers in the lupus-prone NZM2410 mouse model J Immunol 2014 193 1609 1621 25015835 Santos L Draves KE Boton M Grewal PK Marth JD Clark EA Dendritic cell-dependent inhibition of B cell proliferation requires CD22 J Immunol 2008 180 4561 4569 18354178 Seo SJ Fields ML Buckler JL Reed AJ Mandik-Nayak L Nish SA Noelle RJ Turka LA Finkelman FD Caton AJ Erikson J The impact of T helper and T regulatory cells on the regulation of anti-double-stranded DNA B cells Immunity 2002 16 535 546 11970877 Shlomchik MJ Activating systemic autoimmunity: B's, T's, and tolls Curr Opin Immunol 2009 21 626 633 19800208 Sindhava VJ Tuna H Gachuki BW DiLillo DJ Avdiushko MG Onami TM Tedder TF Cohen DA Bondada S Bone marrow dendritic cell-mediated regulation of TLR and B cell receptor signaling in B cells J Immunol 2012 189 3355 3367 22942427 Stranges PB Watson J Cooper CJ Choisy-Rossi CM Stonebraker AC Beighton RA Hartig H Sundberg JP Servick S Kaufmann G Elimination of antigen-presenting cells and autoreactive T cells by Fas contributes to prevention of autoimmunity Immunity 2007 26 629 641 17509906 Süss G Shortman K A subclass of dendritic cells kills CD4 T cells via Fas/Fas-ligand-induced apoptosis The Journal of experimental medicine 1996 183 1789 1796 8666935 Sweet RA Christensen SR Harris ML Shupe J Sutherland JL Shlomchik MJ A new site-directed transgenic rheumatoid factor mouse model demonstrates extrafollicular class switch and plasmablast formation Autoimmunity 2010 43 607 618 20370572 Sweet RA Ols ML Cullen JL Milam AV Yagita H Shlomchik MJ Facultative role for T cells in extrafollicular Toll-like receptor-dependent autoreactive B-cell responses in vivo Proceedings of the National Academy of Sciences of the United States of America 2011 108 7932 7937 21518858 Takahashi Y Ohta H Takemori T Fas is required for clonal selection in germinal centers and the subsequent establishment of the memory B cell repertoire Immunity 2001 14 181 192 11239450 Teichmann LL Cullen JL Kashgarian M Dong C Craft J Shlomchik MJ Local Triggering of the ICOS Coreceptor by CD11c(+) Myeloid Cells Drives Organ Inflammation in Lupus Immunity 2015 42 552 565 25786178 Teichmann LL Ols ML Kashgarian M Reizis B Kaplan DH Shlomchik MJ Dendritic Cells in Lupus Are Not Required for Activation of T and B Cells but Promote Their Expansion, Resulting in Tissue Damage Immunity 2010 33 967 978 21167752 Tipton CM Fucile CF Darce J Chida A Ichikawa T Gregoretti I Schieferl S Hom J Jenks S Feldman RJ Diversity, cellular origin and autoreactivity of antibody-secreting cell population expansions in acute systemic lupus erythematosus Nat Immunol 2015 16 755 765 26006014 Ursini-Siegel J Zhang W Altmeyer A Hatada EN Do RKG Yagita H Chen-Kiang S TRAIL/Apo-2 ligand induces primary plasma cell apoptosis J Immunol 2002 169 5505 5513 12421926 Van Parijs L Abbas AK Role of Fas-mediated cell death in the regulation of immune responses Curr Opin Immunol 1996 8 355 361 8793992 Wang H Shlomchik MJ Autoantigen-specific B cell activation in Fas-deficient rheumatoid factor immunoglobulin transgenic mice The Journal of experimental medicine 1999 190 639 649 10477549 William J Euler C Christensen S Shlomchik MJ Evolution of autoantibody responses via somatic hypermutation outside of germinal centers Science 2002 297 2066 2070 12242446 William J Euler C Leadbetter E Marshak-Rothstein A Shlomchik MJ Visualizing the onset and evolution of an autoantibody response in systemic autoimmunity J Immunol 2005a 174 6872 6878 15905529 William J Euler C Shlomchik MJ Short-lived plasmablasts dominate the early spontaneous rheumatoid factor response: differentiation pathways, hypermutating cell types, and affinity maturation outside the germinal center J Immunol 2005b 174 6879 6887 15905530 Zammit DJ Cauley LS Pham QM Lefrançois L Dendritic cells maximize the memory CD8 T cell response to infection Immunity 2005 22 561 570 15894274 Zelenay S Keller AM Whitney PG Schraml BU Deddouche S Rogers NC Schulz O Sancho D Sousa CRE The dendritic cell receptor DNGR-1 controls endocytic handling of necrotic cell antigens to favor cross-priming of CTLs in virus-infected mice Journal of Clinical Investigation 2012 122 1615 1627 22505458

PMC005xxxxxx/PMC5112153.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8711562 6325 Oncogene Oncogene Oncogene 0950-9232 1476-5594 27181203 5112153 10.1038/onc.2016.175 NIHMS780459 Article p27T187A knockin identifies Skp2/Cks1 pocket inhibitors for advanced prostate cancer Zhao Hongling 13 Lu Zhonglei 134 Bauzon Frederick 1 Fu Hao 15 Cui Jinhua 1 Locker Joseph 2 Zhu Liang 1 1 Department of Developmental and Molecular Biology, and Ophthalmology & Visual Sciences, and Medicine, The Albert Einstein Comprehensive Cancer Center and Liver Research Center, Albert Einstein College of Medicine, Bronx, NY 10461, USA 2 Department of Pathology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261, USA Corresponding author: Dr. Liang Zhu, Department of Developmental and Molecular Biology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Room U-521, Bronx, NY 10461, USA, Phone: 718-430-3320, Fax: 718-430-8975, liang.zhu@einstein.yu.edu 3 Co-first authors. 4 Present address: College of Biological Science and Engineering, Fuzhou University, Fuzhou, Fujian, 350108, China. 5 Present address: Department of Biochemistry, Shenyang Medical College, Shenyang, Liaoning, 110034, China. 22 4 2016 16 5 2016 05 1 2017 05 7 2017 36 1 6070 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. SCFSkp2/Cks1 ubiquitinates Thr187-phosphorylated p27 for degradation. Over-expression of Skp2 coupled with under-expression of p27 are frequent characteristics of cancer cells. When the role of SCFSkp2/Cks1 mediated p27 ubiquitination in cancer was specifically tested by p27 Thr187-to-Ala knockin (p27T187A KI), it was found dispensable for KrasG12D induced lung tumorigenesis but essential for Rb1 deficient pituitary tumorigenesis. Here we identify pRb and p53 doubly deficient (DKO) prostate tumorigenesis as a context in which p27 ubiquitination by SCFSkp2/Cks1 is required for p27 down-regulation. p27 protein accumulated in prostate when p27T187A KI mice underwent DKO prostate tumorigenesis. p27T187A KI or Skp2 knockdown (KD) induced similar degrees of p27 protein accumulation in DKO prostate cells, and Skp2 KD did not further increase p27 protein in DKO prostate cells that contained p27T187A KI (AADKO prostate cells). p27T187A KI activated an E2F1-p73-apoptosis axis in DKO prostate tumorigenesis, slowed disease progression, and significantly extended survival. Querying co-occurrence relationships among RB1, TP53, PTEN, NKX3-1, and MYC in TCGA of prostate cancer identified co-inactivation of RB1 and TP53 as the only statistically significant co-occurrences in metastatic castration resistant prostate cancer (mCRPC). Together, our study identifies Skp2/Cks1 pocket inhibitors as potential therapeutics for mCRPC. Procedures for establishing mCRPC organoid cultures from contemporary patients were recently established. An Skp2/Cks1 pocket inhibitor preferentially collapsed DKO prostate tumor organoids over AADKO organoids, which spontaneously disintegrated over time when DKO prostate tumor organoids grew larger, setting the stage to translate mouse model findings to precision medicine in the clinic on the organoid platform. p27 phosphorylation and accumulation Skp2/Cks1 pocket inhibitors advanced prostate cancer prostate cancer organoids Introduction Levels of p27, a cyclin-dependent kinase inhibitor, can be regulated by mRNA expression, protein synthesis, and protein degradation. One down-regulation mechanism is its ubiquitination by SCFSkp2/Cks1 ubiquitin ligase, which prepares p27 for degradation in the proteasomes (1). SCFSkp2/Cks1-mediated p27 ubiquitination requires its phosphorylation on Thr187. Replacing Thr187 with alanine (p27T187A) to render this position unphosphorylable prevents p27 binding to and ubiquitination by SCFSkp2/Cks1 (2, 3). This is because T187p and a conserved E185 together mediate p27 binding to a pocket formed jointly by Skp2 and the accessory protein Cks1 (4–6). To test the physiological significance of this biochemical mechanism, p27T187A/T187A (p27T187A KI) mice were generated (7). p27T187A KI mice do not show p27 protein accumulation, demonstrating that p27 ubiquitination by SCFSkp2/Cks1 is dispensable for p27 degradation in normal physiology. p27 is a haplo-insufficient tumor suppressor (8), whose reduced expression is frequently associated with aggressive cancers (9). Knockout of p27 induces spontaneous pituitary tumorigenesis and can accelerate tumorigenesis induced by a large number of oncogenic events in numerous tissues, suggesting that stabilizing p27 can be therapeutic for many types of cancer. However, when p27 KO accelerated lung tumorigenesis in KrasG12D/+ mice, p27T187A KI did not have a protective effect (10), suggesting that p27 ubiquitination by SCFSkp2/Cks1 is dispensable for p27 degradation in lung tumorigenesis by KrasG12D. On the other hand, p27 KO accelerated pituitary melanotroph tumorigenesis in Rb1+/− mice (11) and p27T187A KI blocked it (12). Thus, the role of SCFSkp2/Cks1 in mediating p27 degradation can be highly critical in promoting tumorigenesis but only in specific contexts. Finding additional tumorigenic contexts in which SCFSkp2/Cks1 plays a critical role in promoting tumorigenesis could be challenging, since most tissues resist tumorigenesis by deletion of Rb1 alone. It is however evident that, once the susceptible contexts are identified, inhibition of SCFSkp2/Cks1 is technically feasible (13, 14), affects a highly select sub-group of SCFSkp2 substrates, and is forecasted by the p27T187A KI mice to be non-toxic. Here we identify pRb and p53 doubly deficient (DKO) prostate tumorigenesis as an additional susceptible context. We further make the case for translating mouse model findings to clinical metastatic castration resistant prostate cancer (mCRPC) on the organoid platform. Results p27T187A KI accumulates p27 in pRb and p53 doubly deficient prostate Based on our previous findings that Skp2 KO and p27T187A KI accumulated p27 in Rb1 KO melanotroph tumorigenesis (15) and Skp2 KO accumulated p27 in Rb1 and Trp53 DKO prostate tumorigenesis (16), we hypothesized that p27T187A KI could also accumulate p27 in DKO prostate tumorigenesis. To test this hypothesis, we generated mice of p27+/+, p27T187A/T187A, and PB-Cre4;Rb1lox/lox;Trp53lox/lox on p27+/+ or p27T187A/T187A background to determine p27 expression in their prostate epithelium by IHC. At 5–6 months of age, co-deletion of Rb1 and Trp53 induced neoplastic lesions only in proximal prostate acini. We show p27 staining in distal prostate acini to compare the four genotypes in normal appearing tissues (Figure 1Aa,c,e,g). Similar p27 staining were observed in p27+/+ and p27T187A/T187A prostates (Figure 1Ab,d). When Rb1 and Trp53 were co-deleted, p27 staining reduced in p27+/+ prostate (Figure 1Af), but increased in p27T187A/T187A prostates (Figure 1Ah). p27 Western blot of the prostate ventral and anterior lobes provided similar findings (Figure 1B). This Western blot also revealed increased Skp2 protein levels in DKO prostate in p27+/+ and p27T187A/T187A mice. This finding resembles similar findings in human prostate cancer and explains in part the reduction of p27 in p27+/+ prostate but increase of p27 in p27T187A/T187A prostate (please see Discussion for more details). To learn the molecular basis for this requirement for p27 Thr187 phosphorylation, we established early passage prostate cells from DKO prostate in p27+/+ mice (called DKO cells) and in p27T187A/T187A mice (AADKO cells). We found that Skp2 knockdown in DKO cells increased p27 protein levels while AADKO cells intrinsically contained similarly increased levels of p27 (Figure 1C). RT-qPCR and CHX experiments suggest that p27 accumulation in AADKO prostate cells was not due to increased expression of p27 mRNA (Figure 1D) but was due to increased protein stability (Figure 1E). If p27T187A KI prevented p27 ubiquitination by SCFSkp2/Cks1 to accumulate p27, knockdown of Skp2 in p27T187A KI cells would not further increase p27, and this proved true (Figure 1C). These genetic and molecular findings suggest that ubiquitination of p27T187p by SCFSkp2/Cks1 became rate-limiting in preventing p27 accumulation in prostate when Rb1 and Trp53 were co-deleted in it. p27T187A KI activates a p27-E2F1-p73-apoptosis axis in DKO prostate tumorigenesis Skp2 KO or p27T187A KI inhibited pRb deficient pituitary melanotroph tumorigenesis by E2F1 induced apoptosis (Figure S1A) (15, 17). To determine whether this E2F1 induced apoptosis could remain effective when Trp53 is additionally deleted, we determined the contribution of p73, an E2F1 target gene and a p53 family member, to this apoptosis. As shown in Figure S1B. Rb1 deletion did not increase apoptosis, measured as sub-G1 cells, in Skp2 WT MEFs, but increased them to 19% of the population in Skp2 KO MEFs. Combining knockdown of p73 reduced sub-G1 cells to 7–8% of the population (Figure S1C), demonstrating that p73 contributed about half of the E2F1 induced apoptosis in this context. We next investigated whether p27T187A KI activated this p27-E2F1-p73 axis in AADKO prostate tumor cells. We found that Skp2 knockdown in DKO cells increased p73 protein. AADKO cells intrinsically contained a similarly increased level of p73 protein, which Skp2 knockdown did not further increase (Figure 2A). Thus, effects of Skp2 knockdown and p27T187A KI on p73 protein mirrored those on p27 protein. At mRNA levels however, p27T187A KI significantly increased expression of p73 mRNA in AADKO cells compared to DKO cells (Figure 2B), different from p27 (Figure 1D). The increased p73 mRNA expression was accompanied by significantly increased occupancy of E2F1 and p73 promoters (both are typical E2F target genes) but not GAPDH promoter (a housekeeping gene) by E2F1 in AADKO cells compared to DKO cells (Figure 2C, left). E2F4 occupancy on E2F1 and p73 promoters did not increase in AADKO cells (Figure 2C, middle), which is consistent with the fact that E2F4 does not bind cyclin A. ChIP with normal IgG did not enrich E2F1 or p73 promoter sequences in AADKO cells (Figure 2C, right). We next determined p73 protein expression in prostate tissues. Of the four genotypes examined, we found robust anti-p73 staining only in p27T187A/T187A;PB-Cre4;Rb1lox/lox;Trp53lox/lox prostate (Figure 2D). Western blot of prostate tissues confirmed the staining findings (Figure 2E). p73 protein similarly increased in Skp2−/−;PB-Cre4;Rb1lox/lox;Trp53lox/lox prostate (Figure S2). The Western blot also revealed accumulation of cleaved caspase 3 in p27T187A/T187A;PB-Cre4;Rb1lox/lox;Trp53lox/lox prostate tissues (Figure 2E), suggesting that the increased p73 in this genotype induced apoptosis in the absence of p53. We next used TUNEL staining to demonstrate that DKO neoplastic lesions contained more apoptosis in p27T187A/T187A mice than in p27+/+ mice (Figure 3A,B,C). p73 can induce apoptosis independent of p53 likely because the TAp73 isoform activates similar target genes for cell cycle arrest (p21) or apoptosis (PUMA, BAX) as p53 (see (18) for a recent review). To determine whether p73 contributed to this apoptosis, we determined the effects of p73 knockdown (Figure 3D) in AADKO prostate tumor cells. In untreated AADKO cell culture, 8.8% of the population showed sub-G1 DNA content. Knockdown of p73 reduced the sub-G1 population to 1.4%, which was confirmed by a 2nd shp73 (Figure 3E). We further show that DKO prostate tumor cells contained little sub-G1 cells (0.6%), which was increased to 17% by Skp2 knockdown, additional knockdown of p73 reduced it to background levels (Figure 3F). These findings delineate a p27-E2F1-p73-apoptosis pathway activated by p27T187A KI in DKO prostate tumorigenesis (Figure S3). p27T187A KI significantly delays progression of DKO prostate tumorigenesis to lethality DKO prostate tumorigenesis is invasive, metastatic, castration resistant, and becomes lethal within 7–10 months (19), but cannot progress beyond PIN stages in Skp2 KO mice (16). Skp2 KO mice are significantly smaller and less fertile (20), which are not seen in p27T187A KI mice (7). We next determined how DKO prostate tumorigenesis progressed in p27T187A KI mice. In addition to inducing p53-independent apoptosis via a p27-E2F1-p73 axis, p27T187A KI inhibited cell proliferation in DKO prostate tumorigenesis, as measured by Ki67 (Figure S4A,C) and pHH3 (Figure S4B,D) staining. At 5–7 months of age, 50% of PB-Cre4;Rb1lox/lox;Trp53lox/lox mice contained gross prostate tumors but none of the p27T187A/T187A;PB-Cre4;Rb1lox/lox;Trp53lox/lox mice did so; by 8–9 months, 80% of PB-Cre4;Rb1lox/lox;Trp53lox/lox mice contained gross prostate tumors compared to 50% of p27T187A/T187A;PB-Cre4;Rb1lox/lox;Trp53lox/lox mice (Figure 4A,B). When characterized by survival, 2 of 30 PB-Cre4;Rb1lox/lox;Trp53lox/lox mice were alive pass 9 months, while 13 of 38 p27T187A/T187A;PB-Cre4;Rb1lox/lox;Trp53lox/lox mice survived pass 10 months (Figure 4C). Thus, remarkably, the innocuous p27T187A mutation can delay progression of the aggressive DKO prostate tumorigenesis to significantly extend survival. Co-occurrences of RB1 and TP53 inactivation is statistically significant in mCRPC To determine how often both RB1 and TP53 are genetically inactivated in human prostate cancer, we queried the genetic status of RB1, TP53, PTEN, NKX3-1, and MYC (five known prostate cancer drivers) in TCGA of prostate cancer on cBioPortal (21) (Table 1). Inactivation of RB1 and TP53 did not co-occur in three cohorts of primary prostate cancer but co-occurred with statistical significance in two of the three metastatic castration resistant prostate cancer (mCRPC) cohorts. Many of the co-occurred inactivation are likely biallelic since they were detected together with shallow deletions of RB1 and TP53 (Figure S5). None of the other possible pairs among these five genes showed statistically significant co-occurrences for their alterations. Inactivation of TP53 and activation of MYC co-occurred with statistical significance in one primary prostate cancer cohorts but the co-occurrences lost statistical significance with disease progression to mCRPC. About one quarter of the patients develop metastatic prostate cancer after prostatectomy or at presentation. These cases are treated with androgen deprivation therapies, but invariably the cancer would relapse in the form of mCRPC. Inhibitors of residual androgen synthesis, such as abiraterone acetate, 2nd generation AR antagonist, such as enzalutamide, and chemotherapies with taxane can be effective but the efficacies are not durable. Currently, mCRPCs kill 27,000 men annually in US alone. Our TCGA analyses suggest that DKO prostate tumorigenesis could serve as a mouse model for mCRPC to identify pressingly needed new therapy strategies. A small molecule Skp2/Cks1 pocket inhibitor inhibits DKO and DU 145 cells Using a structure-based in silico screen, small molecules that occupy Skp2/Cks1 pocket to inhibit its interaction with p27T187p were identified (13) (Figure 5A). We employed one of such inhibitors, Compound 1 (C1), to determine the pharmacologic effects of inhibiting the Skp2/Cks1-p27T187p interaction in DKO prostate tumor cells. We also tested C1 on AADKO prostate tumor cells, in which the Skp2/Cks1-p27T187p interaction is not present, to genetically interrogate its mechanism of action. There are six human metastatic prostate cancer cell lines in CCLE (Cancer Cell Line Encyclopedia (22)); one of them, DU145, is hormone-insensitive and contains truncating mutation K715* in RB1 and inactivating mutation V274F in TP53. We included DU145 as a model for human mCRPC to test the effects of C1 (Figure 5B). At 1.25 μM, C1 reduced DKO cell proliferation by 80% relative to DMSO vehicle control but reduced AADKO cell proliferation by only 20%. This selectivity for DKO cells continued when C1 concentration was increased to 2.5 μM. AADKO cells proliferated significantly more slowly than DKO cells in the absence of C1 (Figure 5C), but C1 at 5.0 μM still reduced AADKO proliferation by 75% relative to DMSO control (Figure 5B). We found that C1 treatment increased p27 only in DKO cells but increased p21 in both DKO and AADKO cells (Figure 5D). These findings support the designed mechanism of action of C1 and reveal a therapeutic advantage in pharmacological targeting of the Skp2/Cks1 pocket over p27T187A KI (Figure 5A). In DU145 cells, C1 inhibited proliferation and increased p27 and p21 to a degree comparable to DKO prostate tumor cells (Fig. 5B,C,D), suggesting that inhibition of the Skp2/Cks1 pocket is a potential therapeutic strategy for mCRPC (also see Discussion). Skp2/Cks1 pocket inhibitor inhibits DKO prostate tumor organoids Recently, six organoid lines from mCRPC biopsies from contemporary patients were reported, 5 of them contain mutations and/or gene deletions in both RB1 and TP53 (23). Unlike monolayer cultures, organoids respond to therapeutics in acini-like structures. To translate mouse model findings to clinical mCRPC on the same organoid platform, we next generated DKO and AADKO prostate tumor organoids (Figure 6A,B). At day 15 after plating, DKO and AADKO cells generated organoids with similar efficiencies and size ranges (Figure 6E,G,H). Longitudinal monitoring revealed most DKO organoids grew larger while most AADKO organoids spontaneously disintegrated into debris piles over a six-day period (Figure 6C). As such, a prominent difference between DKO and AADKO organoid cultures is the 1.9 fold increase in debris piles in the latter in untreated cultures (Figure 6F,H). These characteristics differed from monolayer cultures, where AADKO cells proliferated much more slowly than DKO cells (Figure 5C), and demonstrate the ability of DKO and AADKO organoid cultures to more closely model autochthonous DKO and AADKO prostate tumors. We then tested the effects of C1 on DKO and AADKO prostate tumor organoids. At 1.25 μM, C1 reduced organoid forming efficiency by 1.7 fold for DKO organoids compared to 1.1 fold for AADKO organoids (Figure 6G); and increased debris piles 2.9 fold for DKO organoids compared to 1.1 fold for AADKO organoids (Figure 6H). This selectivity for DKO organoids largely disappeared when C1 concentrations were increased to 5.0 μM, although about 20% of small AADKO organoids remained in AADKO culture. These organoid findings provide further genetic and functional validation for the therapeutic potential of Skp2/Cks1 pocket inhibitors for pRb and p53 deficient prostate cancer. Discussion Prostate cancer is one of the major cancer types that showed Skp2 overexpression coupled with p27 under-expression. A tissue microarray study of 4 normal prostate samples, 74 high grade prostatic intraepithelial neoplasm (HGPIN), and 622 primary prostate cancer specimens (24) documented significant Skp2 overexpression in HGPIN over normal tissue, in primary prostate cancer over HGPIN, and Skp2 overexpression significantly correlated with p27 under-expression (Table S1). pRb inhibits Skp2 binding to p27 (25), represses Skp2 mRNA expression via E2F (26, 27), and promotes Skp2 protein degradation via APC/CCdh1 (28). Pten also inhibits Skp2 expression but the mechanisms are not known. Pten overexpression in human glioblastoma cells decreased Skp2 expression while Pten KO MEFs show increased Skp2 expression (29). To determine whether these findings could explain Skp2 overexpression in primary prostate cancer, we queried RB1, TP53, PTEN; and Skp2 in a cohort of 333 primary prostate cancer specimens in cBioPortal (Table S2). Inactivation of TP53 or PTEN tend to co-occur with activation of Skp2, but neither reached statistical significance. Inactivation of RB1 does not co-occur with Skp2 activation, likely because RB1 is inactivated in only 0.9% of the samples in this cohort,. When we made the same queries to a cohort of 150 mCRPC specimens, in which inactivation of RB1 and activation of Skp2 both become more frequent, they co-occurred with statistical significance. These TCGA findings provide one explanation for overexpression of Skp2 in prostate cancer and might explain the inhibition of Rb1 and Trp53 DKO prostate tumorigenesis or inhibition of Pten KO prostate tumorigenesis by Skp2 KO (16, 30). Skp2 has a growing list of substrates in addition to p27; p27T187A KI can test the contribution of lacking p27 ubiquitination by SCFSkp2/Cks1 to the effects of Skp2 KO. In Rb1 deficient pituitary tumorigenesis, p27T187A KI largely phenocopied Skp2 KO to block this tumorigenesis (12, 15). On the other hand, p27T187A KI did not inhibit KrasG12D/+ induced lung tumorigenesis. In this study, we determined the effects of p27T187A KI on DKO prostate tumorigenesis, which represents a more challenging tumorigenic context, since co-deletion of Rb1 and Trp53 wound disable most of the antitumor mechanisms in the cells. Co-deletion of Rb1 and Trp53 confers dependency on SCFSkp2/Cks1 to prevent p27 protein accumulation in prostate p27 can be ubiquitinated by several ubiquitin ligases, including SCFSkp2/Cks1, Pirh2 (31), KPC1 (32), Cul4 (33), and WWP1 (34). Combined activities of these ligases, rather than any single ligase, likely determine the stability of p27 protein. pRb inhibits Skp2 via at least three mechanisms (described in the previous section); pRb loss may thereby greatly enhance Skp2 activity, leading to greatly enhanced p27 ubiquitination by SCFSkp2/Cks1 in Rb1 deficient pituitary melanotroph tumorigenesis. In this context, p27 protein degradation became dependent on Thr187 phosphorylation; p27T187A KI therefore can inhibit Rb1 deficient pituitary tumorigenesis (12, 15). In another context, DNA damage-activated p53 stimulates expression of Pirh2 and KPC1 as typical p53 target genes to degrade p27 in mouse embryonic fibroblasts (16). Deletion of Trp53 in prostate however did not induce p27 protein accumulation (our unpublished results), likely because p53 is not activated in prostate development and small reductions in Pirh2 and KPC1 can be compensated by other p27 ubiquitin ligases. Co-deletion of Trp53 and Rb1 reduced p27 protein (Fig. 1Ae,f), likely because the other p27 ubiquitin ligases now included increased (Fig. 1B) and activated Skp2 due to pRb loss. In this context of increased and activated Skp2 and reduced Pirh2 and KPC1, p27T187A KI accumulated p27 protein (Figure 1A and B). Our unpublished results further show that deletion of Rb1 in Skp2 KO or p27T187A KI prostate also did not accumulate p27, likely because p53 is activated by Rb1 deletion to stimulate expression of Pirh2 and KPC1. These findings, together with the results from cultured DKO and AADKO prostate cells (Figure 1C to E) provide the first combined genetic and biochemical evidence for the relevance of SCFSkp2/Cks1-mediated p27 ubiquitination in preventing p27 accumulation in vivo (please also see Introduction). p27T187A KI identifies substrate-specific inhibitors of protein degradation for mCRPC With FDA approval for the use of a proteasome inhibitor, bortezomib, to treat multiple myeloma, targeting ubiquitin-mediated protein degradation has become a validated cancer therapy strategy (35, 36). The current challenge is to add specificity to this strategy, since unspecific inhibition of protein degradation generates serious side-effects, which could explain at least in part the negative outcome of a phase II trial of bortezomib for castration resistant metastatic prostate cancer (37). In the ubiquitination mediated protein degradation hierarchy, inhibitors of the Skp2/Cks1 pocket are perhaps the most substrate selective, affecting only p27 and p21 (and possibly p57 by sequence prediction). Our findings that p27T187A significantly inhibited progression of DKO prostate tumorigenesis to extend survival provide a basis for identifying prostate cancer patients for treatment with such inhibitors. Indeed, in the above mentioned phase II trial of bortezomib, one of 24 evaluable patients with castration resistant metastatic prostate cancer achieved the primary end point (defined as no increase in PSA from baseline and no radiographic progression at 12 weeks) (37). It is regretful that the genetic status of RB1 and TP53 in this patient’s prostate cancer was not determined and documented in the pre-TCGA era. It is highly likely that many mCRPC patients could be indicated for therapy with Skp2/Cks1 pocket inhibitors since RB1 and TP53 inactivation co-occurred statistically significantly in mCRPC based on the most recently available TCGA of primary prostate cancer and mCRPC (Table 1). Our findings with p27T187A KI also provided genetic support for the mechanisms of action of the currently available inhibitor compound 1 (C1), since it inhibited DKO prostate tumor cells far more effectively than AADKO prostate tumor cells (in which the interaction between Skp2/Cks1 pocket and p27 is already prevented) and, furthermore, demonstrated the superiority of such inhibitors over p27T187A KI, since C1 at higher concentration can still inhibit AADKO prostate tumor cells. Since higher concentrations of C1 induced similar p21 accumulation in DKO and AADKO cells, C1 may increase p27 and p21, likely p57 and other yet to be identified substrates of the Skp2/Cks1 pocket to exert stronger inhibition of the Skp2/Cks1 pocket than p27T187A KI. However, it is important to note that regulation of p21 and p27 protein degradation are by different as well as shared mechanisms. While SCFSkp2/Cks1 can promote degradation of p27 and p21, p21 degradation is also promoted by MDM2/MDMX (38), APC/CCdc20 (39), CRL4Cdt2 (40), and CRL2LRR1 (41), which are not known to promote p27 degradation. Furthermore, although the comparison between DKO and AADKO prostate tumor cells support the designed action mechanism of C1, the likelihood of its having other substrates, off-target effects, or both, still exists. This is especially true in human prostate cancer cell lines, such as DU145, which likely contain far more genetic and epigenetic alterations and overall far more heterogeneous than targeted mouse prostate cancer cells, such as DKO and AADKO cells. Inhibition of DKO prostate tumorigenesis by p27T187A KI is notably weaker than that by Skp2 KO, the latter blocked DKO prostate tumorigenesis inside the PIN stage (16). It is therefore of great interest to test inhibitors that target Skp2 as a whole, such as the more recently reported inhibitor Compound 25, which blocks Skp2-Skp1 interaction in the SCFSkp2 ligase (42). The broadening of substrate spectrum would however likely result in undesirable side effects. In this respect, Skp2 KO mice are significantly smaller and less fertile while p27T187A KI mice are virtually normal. Translating mouse model findings to clinical precision medicine on organoid platform DU145 cell line was established about 40 years ago (43), and is the only prostate cancer cell line containing inactivating mutations in RB1 and TP53. Extended monolayer culture of established cancer cell lines can encourage secondary genomic alterations that are not present in the original cancer specimens (44, 45), and monolayer culture bears no resemblance to the prostate when testing therapeutic effects of potential therapeutics. Organoid cultures are emerging as an alternative since cancer cell organoids demonstrate stable genomic landscapes faithful to the original cancer specimens, and respond to therapeutics in acinus-like structures (46, 47). Prostate cancer organoid cultures can be generated from mouse models (48, 49), and we have now established organoid cultures from DKO and AADKO prostate tumor models. We provide evidence that the proliferation and survival characteristics of DKO and AADKO organoids more closely resemble those in DKO and AADKO autochthonous prostate tumors than DKO and AADKO monolayer cells. While monolayer AADKO cells simply proliferated much more slowly than DKO monolayer cells, their respective organoids formed with similar efficiency. The difference is in the expansion of the organoids: when DKO organoids were growing larger, AADKO organoids spontaneously disintegrated into debris piles. In this scenario, the inhibitor C1 collapsed DKO organoids into debris piles, mimicking AADKO organoids. Six recently established mCRPC organoid lines from contemporary patients contained frequent inactivation of RB1 and TP53 while maintaining stable genomic landscapes faithful to the biopsy specimens that they were derived from (23). Based on this success, more mCRPC organoid lines from contemporary patients are being generated potentially reaching the scale of a biobank, as recently reported for colorectal cancer (47). Testing potential therapeutics on prostate cancer organoids established from mouse prostate cancer models and human clinical cancer specimens side-by-side could increase the predictive value of the observed effects in selecting candidates for further development and in guiding clinical trial designs. Materials and Methods Mice PB-Cre4 mice (50), Rb1lox/lox mice (51), Trp53lox/lox mice (52), and p27T187A/T187A mice (7) were described previously. All male mice used for experiments are on FVB, C57BL6J, and 129Sv hybrid background. Animals were allocated to experimental groups based on their genotypes and ages. Investigators were not blinded to the genotypes and ages. We did not estimate sample sizes before the experiments. All samples were included in analysis when control samples validated the experiments. Sample sizes were chosen when we obtained a p value, which was based on at least three biological replicates. Mice were maintained under pathogen-free conditions in the Albert Einstein College of Medicine animal facility. Mouse experiments protocols were reviewed and approved by Einstein Animal Care and Use Committee, conforming to accepted standards of humane animal care. IHC, IF, TUNEL Assay and microscopy Tissue sections and staining have been described previously (16). Briefly, paraffin embedded prostate tissues were sectioned at 5 μm thickness. Slides were deparaffinized, hydrated, and incubated in a steamer for 20 minutes in sodium citrate buffer (Vector Labs, H3301) for antigen retrieval. Sections were first treated with 3% H2O2 to quench endogenous peroxidase, washed several times, blocked with 10% normal goat serum, and then incubated in primary antibodies at 4°C overnight. For immunohistochemistry (IHC) and immunofluorescence (IF) staining, antibodies included: phospho-histone H3 (Cell Signaling Technology, #9701), Ki67 (Vector Labs, SP6), rabbit anti-p27 (Abcam, ab92741) and p73 (Abcam, ab40658). TUNEL staining was performed with an Apoptosis Detection Kit (Millipore, S7100). IHC staining was counterstained with Harris Hematoxylin (Poly Scientific R&D Corp, S212), and IF staining were counterstained with DAPI (Sigma-Aldrich, D-9564). Images were visualized with a Nikon Eclipse Ti-U microscope and captured with Olympus DP71 camera and DP Controller software (3.2.1.276), and saved with DP manager software (3.1.1.208). The images were further processed by Adobe Photoshop. For staining quantification, pictures ere taken under 400× magnificence, about 300–400 cells were counted in each image and 3 images from each mouse were counted. The data were presented as means of 3 mice. Early Passage Prostate Cancer Cell Cultures, DU145 Cell Line, Cell Proliferation Measurement, Organoid Cultures, and Treatments Primary prostate tumor cells were prepared from 0.3 cm × 0.3 cm tumor tissues, which were minced and dissociated in collagenase A in DMEM for 2 hours at 37°C. Primary prostate tumor cells were cultured in DMEM containing 10% FBS and their genotypes were confirmed by PCR genotyping of Rb1, Trp53, p27, and PB-Cre4. DU145 cell line was obtained from ATCC and we did not perform STR profiling on this line. DU145 was cultured in DMEM containing 10% FBS. Measurement of cell proliferation was performed as previously described (16). Prostate tumor cell organoid cultures were generated based on published protocols (48, 49). Quantification in Figure 5 was based on photographs of ten 10x lens fields, each field yielding 3–5 photographs focusing on consecutive planes inside Matrigel layers. Two duplicate wells were counted in this manner. Knockdown of Skp2 or p73 was by lenti-shRNA vectors obtained from the Einstein shRNA core facility (shSkp2: 5′-GCAAGACTTCTGAACTGCTAT-3′; shp73 (12753): 5′-GCGCCTGTCATCCCTTCCAAT-3′ and shp73 (12757): 5′-CAGCCTTTGGTTGACTCCTAT-3′); controlled by scrambled shRNA (shScrmbl). Generation of lentiviral stocks and transduction of cells were as described (53). Successful lentiviral transduction was ensured by puromycin (#BP2956-100, Fisher Scientific) resistance, followed by mRNA (RT-qPCR) and protein (Western blots) measurements. Compound 1 (C1, which inhibits interaction between Skp2/Cks1 and p27T187p (13)) was purchased from Xcess Biosciences Inc. (#432001-69-9). Prostate cancer cells were plated overnight and then treated with C1 for 2 days at the indicated concentrations. Western Blot, CHX, RT-qPCR, ChIP Assays Western blot, CHX, RT-qPCR, ChIP experiments were described previously (16, 17). Antibodies for Western blots are Skp2 (Santa Cruz Biotechnology, H435), p21 (Santa Cruz Biotechnology, SC-397), p27 (BD Bioscience, #610242), p73 (Abcam, ab40658), activated Caspase 3 (Cell Signaling Technology, #9661), and α-tubulin (Sigma-Aldrich, T6074). RT-qPCR primers for p27 are sense: 5′-GCGGTGCCTTTAATTGGGTCT and antisense: 5′-GGCTTCTTGGGCGTCTGC T; p73 sense: 5′-AACGCCGAGCATCAATCC) and antisense: 5′-AGCCCAGACTCTGAGCACTT; GAPDH sense (5′-GGTTGTCTCCTGCGACTTCA and antisense 5′-GGTGGTCCAGGGTTTCTTAC. ChIP antibodies are E2F1 (Santa Cruz Biotechnology, SC-193X), E2F4 (Santa Cruz Biotechnology, SC-866X), IgG (Santa Cruz Biotechnology, SC-2027). ChIP primer sequences are: E2F1: sense 5-CTTTGGAGGTGAGCCTGAAGAG-3′, antisense 5′-GGGTCTGGCGAAGCGAACA-3′; p73: sense 5′-TGAGAGTGCGGTTCTATTGGC-3′, antisense 5′-GCCCTGAACATCTGCGTCTC-3′; GAPDH: sense 5′-GAGTTCTGGGAGTCTCGTGG-3′, antisense 5′-CTCTTCGGGTGGTGGTTCA-3′. Flow Cytometry Analysis for sub-G1 Cell Populations Cells were plated for 16 hours. Cells were washed with PBS, trypsinized with 0.25% trypsin-EDTA at 37°C for 3–5 minutes, and resuspended with 0.5 ml PBS. Then cells were fixed with ice-cold 80% EtOH at 4°C overnight. The fixed cells were spun down and resuspended in 0.5 ml 0.25 mg/ml RNAse A and 30 μg/ml propidium iodide in PBS. Cells were filtered using 40 μm cell strainer before DNA content analysis using DXP10 Calibur at Einstein FACS core facility. Data were analyzed using Flowjo software. Statistical Analyses In Kaplan-Meier survival analysis, p value, hazard ratio and 95% confidence interval of hazard ratio were analyzed by log-rank test (GraphPad Prism 6 Software). p values for differences in Ki67, phospho-histone H3, and TUNEL positive cells, cell proliferation, RT-qPCR, and ChIP between indicated samples were analyzed by Student’s t-test. All statistical analyses are two-sided. p<0.05 is considered as statistically significant. We have not determined whether test data conform to normal distribution and we did not determine variance for any group of data. Supplementary Material Suppl This work was supported by NIH grants RO1CA127901and RO1CA131421 (LZ), Albert Einstein Comprehensive Cancer Research Center (5P30CA13330) and Liver Research Center (5P30DK061153) provided core facility support. HZ was a recipient of DOD PCRP Postdoctoral Fellowship (PC121837), and LZ was a Irma T. Hirschl Career Scientist Award recipient. We thank Dr. James Roberts for providing the p27T187A KI mice. Figure 1 p27T187A/T187A mice accumulated p27 protein in DKO prostate (A) Representative (from ten mice of each genotype) H&E and p27 IHC staining of consecutive prostate sections of the four genotypes as marked. (B) Protein levels were determined by Western blot of prostate ventral and anterior lobes of the indicated genotypes. A/A is abbreviated from T187A/T187A. (C) DKO and AADKO prostate cells were transduced with lentiviruses expressing the indicated shRNA (Scrmbl, scrambled sequences). Following antibiotic selection, the transduced cells were subject to Western blot for the indicated proteins. The same cells were subject to RT-qPCR to determine levels of p27 mRNA relative to GAPDH mRNA (D) and CHX chase to determine p27 protein stability compared to α-tubulin (E). Error bars indicate s.e.m. of the means of three samples. p value is by two-sided t test. Figure 2 p27T187A KI activated an E2F1-p73 axis (A) DKO and AADKO prostate cells treated with the indicated shRNA (as in Figure 1C) were subject to Western blot. (B) The same cells were used for RT-qPCR to determine levels of p73 mRNA relative to GAPDH mRNA. (C) ChIP assay using antibody for E2F1 (left), E2F4 (middle), or normal IgG as control, to determine recruitment of E2F1 and E2F4 onto E2F1, p73, and GAPDH promoters in DKO and AADKO prostate cells. Error bars indicate s.e.m. of the means of three samples. p values are by two-sided t test. ns, not significant (p > 0.05). (D) Representative (of three, see also Figure S2) p73 IHC staining of prostate sections of the four genotypes as indicated. (E) Protein levels were determined by Western blot with prostate ventral and anterior lobes of the indicated genotypes. Figure 3 p27T187A KI increased apoptosis in DKO prostate tumorigenesis and p73 played a major role in the apoptosis (A and B) Consecutive prostate sections of four genotypes, as indicated, stained with H&E or TUNEL. (C) Quantification of apoptosis. Arrows in the photograph demonstrate how total apoptosis was quantified. About 200 cells on each section/mouse were counted. The bar graph was based on the average of quantifications from three mice. p value is by two-sided t test comparing the combined apoptosis in p27+/+ and p27T187A/T187A mice. (D) RT-qPCR to determine knockdown efficiencies by two shp73 constructs. Error bars are s.e.m. (E and F) Propidium iodide based DNA content FACS (representative of two experiments) to detect and quantify apoptosis as sub-G1 cell% in the population, as marked above the brackets. Figure 4 Progression of DKO prostate tumorigenesis was inhibited in p27T187A/T187A mice (A) Representative H&E stained prostate sections of the indicated genotypes at the indicated ages. (B) Pathological diagnoses of prostate lesions as PIN of four stages, invasive carcinomas, and gross tumors. Numbers of mice examined in each age group are indicated below the chart. (C) Kaplan-Meier survival analysis comparing the p27+/+ and p27T187A/T187A cohorts of mice undergoing DKO prostate tumorigenesis. Hazard ratio = 2.514, 95% confidence interval = 2.077 to 6.222, p is by log-rank test. Figure 5 A specific inhibitor of SCFSkp2/Cks1 selectively inhibited DKO prostate tumor cells and DU145 cells in monolayer cultures (A) The designed inhibition mechanism for SCFSkp2/Cks1 inhibitor Compound 1 (C1) compared to p27T187A KI (p57 is a predicted substrate). (B) Proliferation chart showing cell numbers relative to the vehicle (DMSO) following treatment with Skp2/Cks1 pocket inhibitor Compound 1 (C1) at three concentrations after 2 days in monolayer cultures of the indicated cells. (C) p27T187A KI inhibited proliferation of DKO prostate tumor cells in monolayer culture (representative of three independent experiments). Error bars are s.e.m. of the means of triplet plates. p value is by two-sided t test. Cell numbers were counted in triplicate plates every two days for six days. (D) Western blot following treatment with C1 (5 μM) for 2 days. Figure 6 A specific inhibitor of SCFSkp2/Cks1 selectively inhibited DKO prostate tumor cell in organoid cultures. (A and B) Organoid cultures of DKO and AADKO prostate tumor cells after 15 days in culture. (C and D) The same areas of the organoid cultures were photographed on day 15 and day 21. Solid red arrows point to some examples of organoids growing larger, dashed red arrows point to some examples of organoids disintegrated into debris piles. (E) Organoids of various sizes in (A, a, b, and c in red) were cropped out and shown at the original photograph size. (F) Debris piles in (B, a and b in blue) were cropped out and shown. (G) Counting of organoids and debris piles of various sizes following treatment with C1 at various concentrations for 15 days with vehicle (DMSO) control. (H) Fractions of organoids of the indicated sizes and debris piles. Table 1 Pairwise co-occurrence relationships among RB1, TP53, PTEN, NKX3-1 and MYC in primary prostate cancer and metastatic castration-resistant prostate cancer (mCRPC) Primary prostate cancer mCRPC Studies MSKCC, 20101 Broad/Cornell 2012 TCGA, 2015 MSKCC, 20101 Michigan, 20122 Robinson et al, 2015 Specimen #s 157 109 333 28 502 150 RB1 inactivation3 3.2% 0.0% 0.9% 10% 29% 8.6% TP53 inactivation3 1.9% 6.4% 7.5% 10% 54% 50% PTEN inactivation3 5.7% 7.3% 17% 39% 50% 40% NKX3-1 inactivation3 3.8% 0.0% 16% 3.6% 18% 3.3% MYC activation4 40% 1.8% 13% 50% 25% 19% Co-occurrence5 PPt, RM, PM, PtM PPt PM, RM, RPt, PPt RP, RN, PN, RPt, RM, PPt, NM, RN, PN, RPt, RP RP, PPt, NM, PM, RPt, PN, Statistic significance6 p = 0.033 p = 0.023 p = 0.039 We retrieved published data (54–58) and analyzed them on cBioPortal. 1 This study included both primary prostate cancer and mCRPC. 2 mCRPC specimens were obtained at autopsy. 3 Inactivation of RB1, TP53, PTEN, and NKX3-1 is queried for HOMDEL MUT. 4 Activation of MYC is queried for AMP MUT EXP > 2 (larger than 2 SD from the mean). 5 Tendency for co-occurrence is by Log Odds Ratio; R, P, Pt, N, and M are short for RB1, TP53, PTEN, NKX3-1, and MYC, respectively, to indicate the pairs. 6 p value is by Fisher Exact Test. p < 0.05 is considered statistically significant, which is highlighted by bold font. Other pairs show tendencies with p values between 0.083 and 0.575. Tendency pairs with p values between 0.631 (the next higher value) to 0.907 (the highest) are not shown. RB1 and TP53 often incur Shallow Deletions, suggesting biallelic inactivation for some Mutation samples, as shown by the Oncoprints for two studies in Figure. S5. Conflict of Interest The authors declare no conflict of interest. 1 Starostina NG Kipreos ET Multiple degradation pathways regulate versatile CIP/KIP CDK inhibitors Trends in cell biology 2012 1 22 1 33 41 22154077 2 Vlach J Hennecke S Amati B Phosphorylation-dependent degradation of the cyclin-dependent kinase inhibitor p27 EMBO J 1997 9 1 16 17 5334 44 9311993 3 Montagnoli A Fiore F Eytan E Carrano AC Draetta GF Hershko A Ubiquitination of p27 is regulated by Cdk-dependent phosphorylation and trimeric complex formation Genes Dev 1999 5 1 13 9 1181 9 10323868 4 Ganoth D Bornstein G Ko TK Larsen B Tyers M Pagano M The cell-cycle regulatory protein Cks1 is required for SCFSkp2-mediated ubiquitinylation of p27 Nat Cell Biol 2001 3 321 4 11231585 5 Spruck C Strohmaier H Watson M Smith APL Ryan A Krek W A CDK-independent function of mammalian Cks1: targeting of SCFSkp2 to the CDK inhibitor p27Kip1 Mol Cell 2001 7 639 50 11463388 6 Hao B Zheng N Schulman BA Wu G Miller JJ Pagano M Structural basis of the Cks1-dependent recognition of p27(Kip1) by the SCF(Skp2) ubiquitin ligase Mol Cell 2005 10 7 20 1 9 19 16209941 7 Malek NP Sundberg H McGrew S Nakayama K Kyriakidis TR Roberts JM A mouse knock-in model exposes sequential proteolytic pathways that regulate p27Kip1 in G1 and S phase Nature 2001 9 20 413 6853 323 7 11565035 8 Fero ML Randel E Gurley KE Roberts JM Kemp CJ The murine gene p27Kip1 is haplo-insufficient for tumour suppression Nature 1998 396 177 80 9823898 9 Chu IM Hengst L Slingerland JM The Cdk inhibitor p27 in human cancer: prognostic potential and relevance to anticancer therapy Nature reviews Cancer 2008 4 8 4 253 67 18354415 10 Timmerbeul I Garrett-Engele CM Kossatz U Chen X Firpo E Grunwald V Testing the importance of p27 degradation by the SCFskp2 pathway in murine models of lung and colon cancer Proc Natl Acad Sci U S A 2006 9 19 103 38 14009 14 16966613 11 Park MS Rosai J Nguyen HT Capodieci P Cordon-Cardo C Koff A p27 and Rb are on overlapping pathways suppressing tumorigenesis in mice Proc Natl Acad Sci (USA) 1999 96 6382 7 10339596 12 Zhao H Bauzon F Bi E Yu JJ Fu H Lu Z Substituting Threonine187 With Alanine in p27Kip1 Prevents Pituitary Tumorigenesis By Two-Hit Loss Of Rb1 And Enhances Humoral Immunity In Old Age J Biol Chem 2015 1 12 published online Jan 12, 2015 13 Wu L Grigoryan AV Li Y Hao B Pagano M Cardozo TJ Specific small molecule inhibitors of Skp2-mediated p27 degradation Chemistry & biology 2012 12 21 19 12 1515 24 23261596 14 Skaar JR Pagan JK Pagano M SCF ubiquitin ligase-targeted therapies Nature reviews Drug discovery 2014 12 13 12 889 903 25394868 15 Wang H Bauzon F Ji P Xu X Sun D Locker J Skp2 is required for survival of aberrantly proliferating Rb1-deficient cells and for tumorigenesis in Rb1+/− mice Nat Genet 2010 1 42 1 83 8 19966802 16 Zhao H Bauzon F Fu H Lu Z Cui J Nakayama K Skp2 Deletion Unmasks a p27 Safeguard that Blocks Tumorigenesis in the Absence of pRb and p53 Tumor Suppressors Cancer Cell 2013 11 11 24 5 645 59 24229711 17 Lu Z Bauzon F Fu H Cui J Zhao H Nakayama K Skp2 suppresses apoptosis in Rb1-deficient tumours by limiting E2F1 activity Nat Commun 2014 5 3463 24632684 18 Candi E Agostini M Melino G Bernassola F How the TP53 family proteins TP63 and TP73 contribute to tumorigenesis: regulators and effectors Human mutation 2014 6 35 6 702 14 24488880 19 Zhou Z Flesken-Nikitin A Corney DC Wang W Goodrich DW Roy-Burman P Synergy of p53 and Rb deficiency in a conditional mouse model for metastatic prostate cancer Cancer Res 2006 8 15 66 16 7889 98 16912162 20 Nakayama K Nagahama H Minamishima YA Matsumoto M Nakamichi I Kitagawa K Targeted disruption of Skp2 results in accumulation of cyclin E and p27(Kip1), polyploidy and centrosome overduplication EMBO J 2000 19 2069 81 10790373 21 Gao J Aksoy BA Dogrusoz U Dresdner G Gross B Sumer SO Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal Science signaling 2013 4 2 6 269 pl1 23550210 22 Barretina J Caponigro G Stransky N Venkatesan K Margolin AA Kim S The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity Nature 2012 3 29 483 7391 603 7 22460905 23 Gao D Vela I Sboner A Iaquinta PJ Karthaus WR Gopalan A Organoid cultures derived from patients with advanced prostate cancer Cell 2014 9 25 159 1 176 87 25201530 24 Yang G Ayala G De Marzo A Tian W Frolov A Wheeler TM Elevated Skp2 protein expression in human prostate cancer: association with loss of the cyclin-dependent kinase inhibitor p27 and PTEN and with reduced recurrence-free survival Clin Cancer Res 2002 11 8 11 3419 26 12429629 25 Ji P Jiang H Rekhtman K Bloom J Ichetovkin M Pagano M An Rb-Skp2-p27 pathway mediates acute cell cycle inhibition by Rb and is retained in a partial-penetrance Rb mutant Mol Cell 2004 10 8 16 1 47 58 15469821 26 Zhang L Wang C F-box protein Skp2: a novel transcriptional target of E2F Oncogene 2005 4 27 25 18 2615 27 27 Yung Y Walker JL Roberts JM Assoian RK A Skp2 autoinduction loop and restriction point control J Cell Biol 2007 8 27 178 5 741 7 17724117 28 Binne UK Classon MK Dick FA Wei W Rape M Kaelin WG Jr Retinoblastoma protein and anaphase-promoting complex physically interact and functionally cooperate during cell-cycle exit Nat Cell Biol 2007 2 9 2 225 32 17187060 29 Mamillapalli R Gavrilova N Mihaylova VT Tsvetkov LM Wu H Zhang H PTEN regulates the ubiquitin-dependent degradation of the CDK inhibitor p27(KIP1) through the ubiquitin E3 ligase SCF(SKP2) Curr Biol 2001 2 20 11 4 263 7 11250155 30 Lin HK Chen Z Wang G Nardella C Lee SW Chan CH Skp2 targeting suppresses tumorigenesis by Arf-p53-independent cellular senescence Nature 2010 3 18 464 7287 374 9 20237562 31 Hattori T Isobe T Abe K Kikuchi H Kitagawa K Oda T Pirh2 promotes ubiquitin-dependent degradation of the cyclin-dependent kinase inhibitor p27Kip1 Cancer Res 2007 11 15 67 22 10789 95 18006823 32 Kamura T Hara T Matsumoto M Ishida N Okumura F Hatakeyama S Cytoplasmic ubiquitin ligase KPC regulates proteolysis of p27(Kip1) at G1 phase Nat Cell Biol 2004 12 6 12 1229 35 15531880 33 Miranda-Carboni GA Krum SA Yee K Nava M Deng QE Pervin S A functional link between Wnt signaling and SKP2-independent p27 turnover in mammary tumors Genes Dev 2008 11 15 22 22 3121 34 19056892 34 Cao X Xue L Han L Ma L Chen T Tong T WW domain-containing E3 ubiquitin protein ligase 1 (WWP1) delays cellular senescence by promoting p27(Kip1) degradation in human diploid fibroblasts J Biol Chem 2011 9 23 286 38 33447 56 21795702 35 Bassermann F Eichner R Pagano M The ubiquitin proteasome system – implications for cell cycle control and the targeted treatment of cancer Biochimica et biophysica acta 2014 1 1843 1 150 62 23466868 36 Orlowski RZ Kuhn DJ Proteasome inhibitors in cancer therapy: lessons from the first decade Clinical cancer research : an official journal of the American Association for Cancer Research 2008 3 15 14 6 1649 57 18347166 37 Morris MJ Kelly WK Slovin S Ryan C Eicher C Heller G A phase II trial of bortezomib and prednisone for castration resistant metastatic prostate cancer The Journal of urology 2007 12 178 6 2378 83 discussion 83–4 17936848 38 Jin Y Lee H Zeng SX Dai MS Lu H MDM2 promotes p21waf1/cip1 proteasomal turnover independently of ubiquitylation EMBO J 2003 12 1 22 23 6365 77 14633995 39 Amador V Ge S Santamaria PG Guardavaccaro D Pagano M APC/C(Cdc20) controls the ubiquitin-mediated degradation of p21 in prometaphase Molecular cell 2007 8 3 27 3 462 73 17679094 40 Havens CG Walter JC Docking of a specialized PIP Box onto chromatin-bound PCNA creates a degron for the ubiquitin ligase CRL4Cdt2 Molecular cell 2009 7 10 35 1 93 104 19595719 41 Starostina NG Simpliciano JM McGuirk MA Kipreos ET CRL2(LRR-1) targets a CDK inhibitor for cell cycle control in C. elegans and actin-based motility regulation in human cells Dev Cell 2010 11 16 19 5 753 64 21074724 42 Chan CH Morrow JK Li CF Gao Y Jin G Moten A Pharmacological Inactivation of Skp2 SCF Ubiquitin Ligase Restricts Cancer Stem Cell Traits and Cancer Progression Cell 2013 8 1 154 3 556 68 23911321 43 Stone KR Mickey DD Wunderli H Mickey GH Paulson DF Isolation of a human prostate carcinoma cell line (DU 145) International journal of cancer Journal international du cancer 1978 3 15 21 3 274 81 631930 44 Gillet JP Varma S Gottesman MM The clinical relevance of cancer cell lines Journal of the National Cancer Institute 2013 4 3 105 7 452 8 23434901 45 Domcke S Sinha R Levine DA Sander C Schultz N Evaluating cell lines as tumour models by comparison of genomic profiles Nat Commun 2013 4 2126 23839242 46 Sato T Clevers H SnapShot: Growing Organoids from Stem Cells Cell 2015 6 18 161 7 1700 e1 26091044 47 van de Wetering M Francies HE Francis JM Bounova G Iorio F Pronk A Prospective derivation of a living organoid biobank of colorectal cancer patients Cell 2015 5 7 161 4 933 45 25957691 48 Karthaus WR Iaquinta PJ Drost J Gracanin A van Boxtel R Wongvipat J Identification of multipotent luminal progenitor cells in human prostate organoid cultures Cell 2014 9 25 159 1 163 75 25201529 49 Chua CW Shibata M Lei M Toivanen R Barlow LJ Bergren SK Single luminal epithelial progenitors can generate prostate organoids in culture Nat Cell Biol 2014 10 16 10 951 61 1 4 25241035 50 Wu X Wu J Huang J Powell WC Zhang J Matusik RJ Generation of a prostate epithelial cell-specific Cre transgenic mouse model for tissue-specific gene ablation Mechanisms of development 2001 3 101 1–2 61 9 11231059 51 Sage J Miller AL Perez-Mancera PA Wysocki JM Jacks T Acute mutation of retinoblastoma gene function is sufficient for cell cycle re-entry Nature 2003 7 10 424 6945 223 8 12853964 52 Jonkers J Meuwissen R van der Gulden H Peterse H van der Valk M Berns A Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer Nat Genet 2001 12 29 4 418 25 11694875 53 Sun D Melegari M Sridhar S Rogler CE Zhu L A multi-miRNA hairpin method that improves gene knockdown efficiency and provides linked multi-gene knockdown BioTechniques 2006 41 59 63 16869514 54 Taylor BS Schultz N Hieronymus H Gopalan A Xiao Y Carver BS Integrative genomic profiling of human prostate cancer Cancer Cell 2010 7 13 18 1 11 22 20579941 55 Barbieri CE Baca SC Lawrence MS Demichelis F Blattner M Theurillat JP Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer Nat Genet 2012 6 44 6 685 9 22610119 56 TCGA The Molecular Taxonomy of Primary Prostate Cancer Cell 2015 11 5 163 4 1011 25 26544944 57 Grasso CS Wu YM Robinson DR Cao X Dhanasekaran SM Khan AP The mutational landscape of lethal castration-resistant prostate cancer Nature 2012 7 12 487 7406 239 43 22722839 58 Robinson D Van Allen EM Wu YM Schultz N Lonigro RJ Mosquera JM Integrative clinical genomics of advanced prostate cancer Cell 2015 5 21 161 5 1215 28 26000489

PMC005xxxxxx/PMC5112159.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8806104 2772 Comput Med Imaging Graph Comput Med Imaging Graph Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society 0895-6111 1879-0771 27373749 5112159 10.1016/j.compmedimag.2016.05.003 NIHMS800935 Article Stain Normalization using Sparse AutoEncoders (StaNoSA): Application to digital pathology Janowczyk Andrew 1 Basavanhally Ajay 2 Madabhushi Anant 1 1 Case Western Reserve University, Cleveland, Ohio 2 Inspirata, Inc., Tampa, Florida 12 7 2016 16 5 2016 4 2017 01 4 2018 57 5061 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Digital histopathology slides have many sources of variance, and while pathologists typically do not struggle with them, computer aided diagnostic algorithms can perform erratically. This manuscript presents Stain Normalization using Sparse AutoEncoders (StaNoSA) for use in standardizing the color distributions of a test image to that of a single template image. We show how sparse autoencoders can be leveraged to partition images into tissue sub-types, so that color standardization for each can be performed independently. StaNoSA was validated on three experiments and compared against five other color standardization approaches and shown to have either comparable or superior results. digital histopathology stain normalization deep learning image processing 1. Introduction Digital pathology (DP) is the process by which histology slides are digitized to produce high resolution images via whole slide digital scanners [1]. These digitized slides afford the possibility of applying image analysis techniques to tissue images for the purpose of object detection, segmentation, and tissue classification. These automated image analysis algorithms are very relevant for nuclei detection, mitosis quantification, and tubule counting. Additionally these image analysis algorithms provide the ability of performing higher level, supervised learning tasks such as disease grading, thereby enabling the development of decision support algorithms for pathologists [2]. Prior to evaluation of the specimen under the microscope by a pathologist, the tissue is invariably treated by artificial or natural agents, to stain the various cellular structures. Hemotoxylin and Eosin (H&E) is one of the the most routinely used stains for evaluating disease morphology, an example of which is shown in Figure 1. The hemotoxylin provides a blue or purple appearance to the nuclei while the eosin renders eosinophilic structures (e.g., cytoplasm, collagen, and muscle fibers) a pinkish hue. Since the staining process is a chemical one, there are many variables which can drastically change the overall appearance of the same tissue. For example, the specimen thickness, concentration of the stain, manufacturer, time, and temperature at which the stain is applied can have significant implications for the appearance of the final tissue specimen. Figure 1a, shows an H&E stained gastrointestinal (GI) biopsy tissue image. Figure 1b shows a sample taken from the same specimen but stained using a slightly different protocol (i.e. a different concentration of H&E respectively), and as such, appears significantly darker. The staining process is not the only source of visual variability in imaging of tissue specimens. The digitization process also could potentially induce changes and variability in tissue image appearance. For example Figure 1 shows the same physical specimen scanned using two different scanners. Differences in the scanning platform (e.g., bulbs, ambient illumination, sensor chips), image stitching algorithms, and acquisition technologies (e.g., compression, tiling, whiteness correction) can also induce substantive differences in appearance of the resulting tissue image. Pathologists are specifically trained to be able to cope with these variations and for the most part do not struggle with diagnostic decision making or object counting/quantification (e.g., nuclear or mitosis counting). On the other hand, image analysis methods, especially supervised learning/classification algorithms for nuclei segmentation or tissue partitioning, typically find it more difficult to cope with variations in image appearance and staining [3]. For instance if an algorithm is trained to identify nuclei based off chromatic cues from a singe site, the variations in staining might cause the algorithm to have a large number of errors for slightly differently stained images from a different site. This is further compounded when we consider extremely large datasets that are curated from many different sites, such as The Cancer Genome Atlas (TCGA). These variations in stain and tissue appearance have spurred recent research in development of color standardization and normalization algorithms to help improve performance of subsequent image analysis algorithms [3, 4]. Often, this occurs by identifying a single image with the most optimal tissue staining and visual appearance, and designating this image as the ”template”. Subsequently all other images to be standardized have their intensity distributions mapped to match the distribution of the template image. Previous works [5, 6, 7] have suggested that partitioning the image into constitute tissue subtypes (i.e., epithelium, nuclei, stroma, etc.) and attempting to match distributions on a tissue-per-tissue basis is more optimal compared to an approach which involves simply aligning global image distributions between the target and template images. In the context of histopathology this process might involve first identifying stromal tissue, nuclei, lymphocytes, fatty adipose tissue, cancer epithelium both within the target and template images and then specifically establishing correspondences between the tissue partitions in the two images. Subsequently the tissue specific distributions could then be aligned between the target and template images. While these tissue specific alignment procedures [8] have had more success compared to global intensity alignment approaches [9], successfully identifying the partitions remains an open challenge. For example, nuclei segmentation on its own is a large area of research [10, 11, 12, 13], yet represents only a single histologic primitive. It is therefore clear that more powerful and flexible approaches are needed for automated partitioning of the entire tissue image into distinct tissue compartments. Our approach, Stain Normalization using Sparse AutoEncoders (StaNoSA), is based off the intuition that similar tissue types will be clustered close to each other in a learned feature space. This feature space is derived in an unsupervised manner, releasing it from the requirement of domain specific knowledge such as having to know the ”true” color of the tissue stains, a requirement for a number of other approaches [14]. Our approach employs sparse-auto encoders (SAE), a type of deep learning approach which through an iterative process learns filters which can optimally reconstruct an image. These filters provide the feature space for our approach to operate in. Once the pixels are appropriately clustered in this deep learned feature space into their individual tissue sub-types, tissue distribution matching (TDM) can occur on a per channel, per cluster basis. This TDM step allows for altering the target image to match the template image's color space. The main contribution of this work is a new TDM based algorithm for color standardization for digital pathology images and which employs sparse autoen-coders for automated tissue partitioning and establishing tissue specific correspondences between the target and template images. Autoencoding is the unsupervised process of learning filters which can most accurately reconstruct input data when transmitted through a compression medium. By performing this procedure as a multiple layer architecture, increasingly sophisticated data abstractions can be learned [15]. Additionally as part of our approach we perturb the input data with noise and attempt to recover the original unperturbed signal, an approach termed denoising auto-encoders [15], that has been shown to yield robust features. StaNoSA is thus a fully automated way of transforming images of the same stain type to the same color space so that the amount of variance from (a) technicians, (b) protocols, and (c) equipment could be minimized. The rest of the paper is organized as follows. Section 2 involves a review of the previous work in the field. Section 3 describes the approach (StaNoSA) and associated algorithms. Section 4 rigorously evaluates the method across 2 different datasets and compares the approach with a state of the art approach and 4 other common methods. Section 5 contains the discussion. Finally, in Section 6 we present our concluding remarks. 2. Previous Work and Novel Contributions Previous approaches [8] to color normalization for digital histopathology images tend to fall into one of two categories. The first set of approaches exploit staining characteristics directly, such as Beer-Lambert's law through color de-convolution [16]. They attempt to divide the image color space into individual stain contributions and normalize these individually. The second category of algorithms take a statistical approach which rely on finding clusters of pixels belonging to a similar tissue-types (e.g., nuclei, epithelium, stroma). Once these partitions are identified, and correspondences across different images is established, the color distributions can be matched to a template distribution, so that each histologic entity can be manipulated separately. While a thorough review of the state of the art is out of the scope of this paper, below, we briefly discuss and differentiate approaches most relevant to StaNoSA. Specifically, while many previous works operate solely in the gray color space [17, 18, 19], we limit our discussion primarily to color based approaches since the specific application considered in this work is digital pathology (and specifically H&E stained imagery). We do however direct the interested reader to a very recent review of state of the art [8] approaches for color normalization. 2.1. Stain Specific Algorithms The most popular and well cited backbone for stain specific normalization algorithms is color deconvolution (CD) [16]. CD is based on the idea that the individual stains can be separated and subsequently normalized independently. Subsequently, the normalized color channels can be recombined to recreate an improved version of the original color image. The underlying stain signals are identified and separated by taking advantage of the Beer-Lambert's law [16] which is then used to linearly estimate the contribution of each stain to the final pixel color. The stain matrix describes, through coefficients for each stain, how much each stain contributes to the final pixel color value. This approach has been shown to work well when the color matrix can be reliably and definitively determined [16]. On the other hand, identifying this matrix requires either (a) images which are scanned on the same scanner using only a single stain to compute the deconvolution matrix (a type of color calibration approach) or (b) a semi-supervised approach wherein users are required to select pixels corresponding to the relevant classes as shown in [20]. While in [16] the authors were able to obtain accurate estimations of the stain matrix on their specific dataset, the same stain matrix may not adequately generalize to images from other sites and scanning platforms. They describe how ideal matrices can be estimated but the process requires the usage of control tissue and a single stain scanning process which may not always be possible. It may also not be possible to identify a stain matrix that simultaneously addresses variations due to scanning platforms. Magee et al. in [21] discussed this limitation as well by alluding to the difficulty in identifying the appropriate parameters for the deconvolution matrix. 2.2. Clustering Type Algorithms A second category of approaches aims to identify distinct tissue clusters in a feature space and aims to perform a cluster-to-cluster distribution alignment across the template and target images. Some approaches have attempted to decompose an image using non-negative matrix factorization [22], only requiring knowledge of the number of stains used. Other approaches aim to classify every pixel in the image to one of multiple different tissue classes. In [23] the Expectation Maximization approach was employed to create a fuzzy labeling of the image into distinct tissue classes. However, these approaches [8] may not work optimally if the individual tissue clusters are not proportionately represented in the target and template images. For example, the EM approach in [23] could end up mapping similar appearing regions from distinct tissue classes to the same cluster. During the the TDM process this can lead to distorted color density distributions. 2.3. Novel Contributions In this paper we present a novel technique, StaNoSA, for fully unsupervised normalization of images to a template image. StaNoSA solely requires as input a template image, as opposed to domain specific information such as mixing coefficients or stain properties. This approach therefore enables the ability to shift a target image in the color domain to more accurately resemble the template image. We extend the approach initially presented in [23] by considering not solely chromatic value of pixels for clustering. Instead, we employ a fully unsupervised deep learned bank of filters. These filters represent optimal [24] filters for image reconstruction via compression. By operating in this filtered space, we obtain significantly more robust pixel classes. These classes are not tightly coupled to individual stain classes, and thus the approach has a greater likelihood of success in the case of challenging images, images where one might find new tissue classes or a disproportionate tissue distribution compared to the template image. A representative result showing the application of StaNoSA is presented in Figure 1, where we normalized the target image (Figure 1b) into the color space of the template image (Figure 1a) to produce a result (Figure 1c), which has very similar color characteristics as the template color space. This process was applied by clustering the pixels in a sparse auto-encoded feature space so that respective tissue partitions could be aligned using their respective color distributions. By using individual tissue partitions, StaNoSA is able to more sensitively modify the color space as compared to a global method where all pixels are considered concurrently. We can see the resulting density functions of the pixels in 3 channel RGB, from which we can see that the target image's distribution (Figure 1e) is heavily skewed towards the left. After normalization (Figure 1f) the probability distribution of the 3 channels more closely resembles that of the template image (Figure 1d). The primary novel contributions of this work may be summarized as follows. StaNoSA is a domain agnostic approach, in that it does not rely on specifically knowing stain values or tissue values allowing the user to choose the template to standardize to. StaNoSA leverages a sparse auto-encoder (SAE) [15, 25] as the core method to automatically partition the image into distinct tissue categories. DL methods have been shown to be robust and accurate for image partitioning [14]. First of its kind study in which we quantitatively measure intra-scanner and inter-scanner variances of digitized H&E histology images. 3. Methods 3.1. Notation For all methods, we define the dataset Z={C1,C2,…,CM} of M images, where an image C=(C,ψ) is a 2D set of pixels c ∈ C and ψ is the associated vectorial function which assigns RGB values. T=Ca∈Z is chosen from Z as the template image by which all other images in the dataset will be normalized to. Without loss of generality we chose S=Cb∈Z to be the ”target image”, which is to be normalized into the color space of T. See Table 1 for additional notation used in this manuscript. 3.2. Deep Learning of Filters from Image Patches Denoising auto-encoders are leveraged in this work to learn representative feature spaces for optimizing tissue partitioning. We present them briefly below. 3.2.1. One Layer Autoencoder From T we randomly select p∈Rv×v×3 sub-images, or patches, of v × v dimension in 3-tuple color space (RGB) (see Figure 3). We set V = v × v × 3 to simplify notation. These values are reshaped into a data matrix X∈Rp×V of x∈R1×V samples. This matrix X forms the basis from which the filters will be learned. A simple one layer auto-encoder can be defined as having both an encoding and decoding function. The encoding function encodes a data sample from its original dataspace of size V to a space of size h. Consequently, the decoding function decodes a sample from h space back to V space. We use notation from [15] where they show a typical encoding function for a sample x is (1) y=fθ(x)=s(Wx+b), parameterized by θ = {W, b}. W is a h × V weight matrix, b∈R1,V is a bias vector, and s is an activation function (in this work s is assumed to be the hyperbolic tangent function). The reconstruction of x, termed z, proceeds similarly using a decoding function z = gθ′ (y) = s(W′y + b′) with θ′ = {W′, b′}. Here W′ is a V × h weight matrix, and b′∈R1,h is a bias vector. We use stochastic gradient descent [26] to optimize both θ and θ′ relative to the average reconstruction error, θ*. To further use notation from [15], this error is defined as: (2) θ⋆,θ′⋆=argminθ,θ′1p∑i=1pL(x(i),z(i)) where the loss function L is a simple squared error L(x, z) = ||x – z||2. 3.2.2. Expansion to Multiple Layers and Denoising It has been shown that by applying these auto-encoders in a greedy layer wise fashion, higher level abstractions, in a lower dimensional space, may be learned. In particular, this means taking the output from layer l, i.e., y(l), and directly using that as the input (x(l+1)) at the next layer to learn a further abstracted output y(l+1) by re-applying Equation 2. Layer 1 has input x(1) of size R1×V and output y(1) of size R1×h(1). Layer 2 thus has x(2) = y(1) of size R1×h(1) and output y(2) of size R1×h(2). This layering can continue as deemed necessary. Additionally, by intentionally adding noise to the input values of X, more robust features across all levels may be learned. Briefly, we formalize this by saying X̂ = ϵ(X), where ϵ is a binomial corrupter which sets elements in X to 0 with probability ϕ. Using x̂ in place of x in Equation 1, results in the creation of a noisy lower dimensional version ẑ. This reconstruction is then used in Equation 2 in place of z, while the original x remains in place. 3.2.3. Generating feature space representations for image Once the filters are learned for all levels, we apply the full hierarchy of encoders on both the template image, T, and a ”moving image”, S. The main assumption of our approach is that regardless of visual appearance, pixels belonging to the same tissue class will have similar responses to the learned filters. Figure 4 shows an example of this procedure using two images of the same tissue stained, but with different stain concentrations. It can be seen that although the visual appearance of these two images is quite different, the filters appear to elicit a similar response to similar tissue classes in the image. As such, this feature space allows for an unsupervised clustering process to identify tissue partitions. In Figure 4(c) and 4(f), each unique grayscale value represents a different cluster, suggesting that pixels of similar textural appearance tend to be grouped into the same cluster. However we note that there is no obvious connection between the individual clusters and the tissue classes present in the image. This therefore suggests that we need a subsequent unsupervised clustering step in order to align the tissue partitions. 3.3. Unsupervised Clustering Once we obtain the filter responses for T and S, i.e., T. and S. respectively, we aim to cluster them into individual partitions. To this end, we employ a standard k-means approach on T. to identify K cluster centers. Subsequently, we assign each of the pixels in S. to its nearest cluster, without performing any updating. This process yields S˚, a cluster indicator variable. This approach helps to assuage some of the typical instability issues commonly associated with k-means. We note that although the clustering is taking place on a pixel level, the feature space has been computed on a 32 × 32 window around each pixel, providing the necessary context for our clusters to be better informed. In [23], for instance, the K clusters loosely corresponded to individual tissue classes such as nuclei, stroma or lymphocytes. However, the maximum K that could be chosen in [23] was implicitly limited. In the case of StaNoSA, we use a much larger K, on the order of 50. A larger number of clusters potentially allows for more precisely tuned clusters. 3.4. Histogram Shifting Once the clusters are defined, a more precise color standardization process can take place. For each K, operating on a per channel basis, we standardize the distribution of the moving images to that of the template images [9]. We present this approach in Algorithm 2 which is the basis for the implementation of the imhistmatch function in Matlab. As was mentioned in Section 2, a common problem with global normalization techniques is the inability to account for both tissue class proportions and in cases where the color distributions are already similar, StaNoSA is able to minimize the overall error (see Experiment 1 in Section 4). By assigning the pixels in S to a larger number of clusters, but not performing updating, the disproportional class representation is effectively managed. 4. Experimental Evaluation To rigorously evaluate our approach, we perform three experiments, each focused on directly addressing a different reason for variance in color and appearance of pathologic tissue slides. Specifically we attempted to address variations induced by (a) differences in platform and scanners, and (b) staining. Additionally we also evaluated the performance of a nuclear detection algorithm on the pre- and post-standardized images to evaluate the role of color standardization methods in facilitating object detection. Using three different datasets, as shown in Table 2, we attempted to evaluate StaNoSA when (a) different concentrations of H&E were deliberately varied in the same tissue section and (b) when the same slide was scanned multiple times on two different platforms. Additionally we also compare our approach to 5 other color normalization approaches. 4.1. Datasets 4.1.1. Dual Scanner Breast Biopsies The S1 dataset consists of 5 breast biopsy slides. Each slide was scanned at 40x magnification 3 times on a Ventana whole slide scanner and once on a Leica whole slide scanner, resulting in 20 images of dimension 100,000 × 100,000 pixels. Each set of 4 images (i.e., 3 Ventana and 1 Leica), were mutually co-registered so that from each biopsy set, 10 sub-regions of 1,000 × 1,000 could be extracted. This resulted in 200 images: 10 sub-images from 4 scans across 5 slides. The slides were formalin fixed para n embedded and stained with Hematoxylin and Eosin (H&E), each slide having some evidence of cancer presence. Since the sub-images are all produced from the same tissue slide, this allows for a rigorous examination of intra- and inter-scanner variation. 4.1.2. Gastro-Intestinal Biopsies of differing protocols The S2 dataset consists of slices taken from a single cancer positive Gastro Intestinal (GI) biopsy. The specimen was formalin fixed paraffin embedded and had 7 adjacent slices sectioned from the same biopsy tissue and subjected to different staining protocols: HE, H↓E, H↑E, ↓HE, ↓H↓E, ↑HE, and ↑H↑E, where ↑ and ↓ indicate over- and under-staining of the specified dye. These intentional staining differences were intended to simulate the typical variability seen in clinical settings, especially across different sites. Each slide was then digitized using an Aperio whole-slide scanner at 40× magnification (0.25 μm per pixel), from which 25 random 1,000 × 1,000 resolution images were cropped at 20× magnification. Representative images within S2 can be seen in Figure 6. 4.1.3. Gastro-Intestinal Biopsies of differing protocols with annotations The S3 dataset is a subset of S2 manual annotations of the nuclei. From each of the 7 different protocols, as discussed above, a single image patch of about 1,000 × 1,000 pixels was cropped at 40× magnification. Nuclear boundaries were then delineated by an expert pathologist. 4.2. Comparative Color Standardization Schemes 4.2.1. DL Normalization The parameters used for each experiment associated with our approach (StaNoSA) are as follows: 250,000 patches of size (v) 32 × 32, were extracted from only template image. A 2 layer SAE was created with the first layer containing 100 hidden nodes (h1) and the second layer ten (h2). The denoising variable was set to ϵ = .2. Histogram equalization took place using Q = 128 bins. Additionally, the following pre-processing steps were applied to each cohort of images: ZCA whitening and global contrast normalization [27]. 4.2.2. Direct This ”approach” only employs the raw image without any modifications to quantify what would happen if no normalization was undertaken at all. 4.2.3. Global Standardization This approach is similar to Algorithm 2, except assuming that K = 1, in which all pixels in the image belong to a single cluster. Again, Q = 128 bins were used for the histogram matching process. 4.2.4. Four Additional Approaches We also compared StaNoSA against the publicly available stain normalization toolbox presented in [8]. This toolbox also comprises four additional approaches, Reinhard et al. (RH) [20], Macenko et al. (MM) [28], Histogram Specification (HS) [9], and Khan et al. (DM) [8]. We direct the reader to their respective papers for implementation details. 4.3. Implementation Details It took 5 hours to train the deep learning network using a Nvidia M2090 GPU with 512 cores at 1.3ghz and under 3 minutes to generate each output image needed for the clustering mechanism. Subsequently, for images of size 1,000 × 1,000, the entire clustering and shifting process takes under 1 minute on a 4 core 2.5ghz laptop computer. All deep learning was developed and performed using the popular open source library Pylearn2 [29] which uses Theano [30] for its backend graph computation. 4.4. Experiment 1: Standardization across scanners 4.4.1. Design We aim to evaluate the extent of differences between colors for the same slide scanned multiple times on the same scanner (Cj,iin, where i ∈ {1, 2, 3} represents the scan number and j ∈ {1, . . . , N} represents the image number) to compute intra-scanner error (μi,kin=∑jSSD(Cj,1in,Cj,2in)∕N, i ∈ {1, 2, 3}, k ∈ {1, 2, 3})) and the error in scanning the same slide across a different platform (Cjit,j∈ {1, ..., N}) to compute the inter-scanner error (μiit=∑jSSD(Cj,iin,Cj,1it)∕N, i ∈ {1, 2, 3})). The SSD error involves taking each color channel and computing a 128 bin histogram and use the sum of the squared difference of the bins for each of the color channels. The optimal error would of course be 0 if both images were identical. Our aim in this experiment was to evaluate whether StaNoSA could help bring μit into μin range. More specifically the goal was to assess whether |μit – μin| < γ where γ is a predefined threshold that represents σin, typical intra-scannar variance. While ideally we would have liked to perform a pixel level mean squared error difference between a template and target image, this was not possible on account of two issues. Firstly, the image resolution for slides digitized on the scanners is not identical. This in turn would actually require image re-scaling and interpolation which in turn would likely induce additional error. Secondly, there are visible tiling artifacts visible on both the intra/inter scanner images, making the pixel level error nearly impossible to determine. Instead, we use the SSD measure (introduced above) as a surrogate. Using the 200 images from dataset S1, we perform two experiments. First, we compute the mean and variance SSD per image (μin, σin) across the 3 Ventana scans (S1,Vi, i ∈ {1, 2, 3}) and apply StaNoSA to determine if it is possible to reduce μit, σit. Second, we use the co-registered Leica scan (S1,L) from each set and apply StaNoSA to S1,Vi, i ∈ {1, 2, 3}, measuring μit before and after StaNoSA is performed. 4.4.2. Results After comparing errors, we note that the SSD μin is about .03 (see Figure 5). The global normalization (GL) approach when applied to S1 seems to exacerbate the existing error. On the other hand, StaNoSA is clearly shown to not only reduce μin (.01) but also substantially reduce the σin seen within samples. We additionally also examined if and how variations in image appearance on account of the use of different scanners could be reduced (see Figure 5). In this instance, we see that the GL technique does indeed reduce μit from about .14 to .096, but our StaNoSA approach reduces the error down even more substantially to .047 which is on the order of μin as shown in Figure 2(b) which has a μin of .0473. In all cases, we can see that StaNoSA reduces both μin, μit and σin, σit. 4.5. Experiment 2: Standardization Across Stain Protocols 4.5.1. Design We aim to determine how well StaNoSA can succeed at minimizing SSD by bringing S into the color space of T. In this experiment we use S2 as a way to quantify how well the mean SSD error (μp,q=∑iSSD(Cp,Cqin)∕N,i∈{1,…,N}, where p, q represent any of the 7 stain proctols (e.g., HE, H↓E, H↑E, ↓HE, ↓H↓E, ↑HE, and ↑H↑E) can be reduced across S2. In each instance we arbitrarily select T from the group and attempt to standardize the remaining 6 images to that image and compute μp,q and the variance σp,q. We do this for all protocol pairings (i.e., p, q) and images: 7 protocols versus the remaining 6 with 25 images each resulting in 1,050 normalization operations. We report both μp,q and σp,q across all protocols. 4.5.2. Results The confusion matrix shown in Table 3 contains μp,q and σp,q for all p, q. Here we stress that it is impossible for the μp,q to be zero as images are adjacent slices, not replicates as in S1. Hence there are guaranteed to be slight differences in the visual appearance of the slides. It can be seen that StaNoSA consistently provides the smallest μp,q. In Figure 6 we present images of S2 for qualitative evaluation, choosing specifically the most extreme of the images to normalize: ↓H↓E and ↑H↑E. We can see that although the stains are notably different in S2, our approach can successfully alter S to match the color space of T. 4.6. Experiment 3: Evaluate standardization via object detection 4.6.1. Design In this experiment we evaluate the effect of color standardization on a particular object detection task, namely nuclei detection. Our experiment involves selecting two H&E images to be used as templates (Figure 7), one which is not anomalous (Figure 7a) and one which is (see Figure 7b). The large proportion of red blood cells in the image tends to affect the balance of representation of tissue classes and hence can affect the performance of various color standardization algorithms. We perform color deconvolution using the H&E stain matrix as presented in [16]. Subsequently, we identify the optimal threshold (ψ = .914 in this instance) on the T image, by which to separate the nuclei stained pixels from other pixels in the resultant H channel. Lastly we normalize the 7 images of S3 to T, and process them in similar fashion: (a) color deconvolution followed by (b) thresholding. To evaluate the results, we compute the Dice coefficient [31] (μϕ=∑iϕ(i)∕N, i ∈ {i, ..., N}) and its variance (σϕ) of the resulting nuclei as compared to the manually annotated ground truth for all approaches (e.g., Direct, GL, StaNoSA, DM, HS, MM, and RH). We chose an H&E image from S2 to act as T, but one in particular which does not have balanced class proportions (Figure 7b). We specifically selected the template image in Figure 7b to determine if the methods we compare against (Direct, GL, StaNoSA, DM, HS, MM, and RH) are robust against imbalanced tissue class representations. To provide a comparison, T shown in Figure 7a does not have any unusual artifacts which disrupt the relative class proportionality. 4.6.2. Results As we can see from Figure 8 and Figure 9, the StaNoSA is able to improve μϕ by 10% while reducing σϕ. One of the difficulties with color deconvolution, as discussed in [8], is the fact that the approach needs careful selection of the appropriate stain matrix in order to achieve satisfactory results. In this case, because we have seven different staining protocols, it is unlikely that the same matrix would consistently work well. Instead, using StaNoSA, we find T which works well and then shift S3 to that image. In the cases of ↑H and ↑H↑E we see significant improvements in color constantcy across S3 as a result of StaNoSA. As expected, GL and HS (global normalization techniques) perform poorly because of the large red blood cell artifacts. On the other hand, when these artifacts are not present, the GL, HS, and StaNoSA normalization technique perform comparably. 5. Discussion Color standardization of digital histopathology images is critical to reducing stain variability and improving the robustness of computer assisted diagnostic and image quantification algorithms such as nuclei and mitoses detection. Previous approaches have potentially been handicapped by the necessity of accurately defining a stain matrix or requiring images to have similar tissue type representations in the image (i.e., similar proportions of stroma, nuclei). StaNoSA is able to circumvent the need for a stain matrix by identifying tissue sub-types within the image in an unsupervised manner. The large number of tissue sub-types produced affords the opportunity of managing class imbalances better than previous approaches, by providing greater specificity through more tightly defined clusters. In this manner, only very similar pixels in the template image are used to normalize a target pixel, as opposed to all pixels at a histologic primitive level (e.g., all epithelial pixels). The larger number of clusters helps manage situations where there is a significant proportional difference in the presence of individual tissue classes in the target and template images. Traditionally, the most important aspect of a normalization process is choosing an optimal template image. While it makes sense to develop algorithms with as much robustness as possible, inferring cohort parameters from a poor template image will typically be a challenging proposition. The most important quality associated with a good template image, especially in the context of color standardization approaches, is that it should be representative of the other cases within the cohort. Theoretically one would ideally wish to have a correspondence from every pixel value (or color) within the template image to a corresponding pixel color in each of the cohort images, so that the color distributions can be brought into alignment. Since having such an idealized template image is highly unlikely, we attempt to select one which comes as close as possible to such an ideal image. This therefore implies the need for template images which proprtionately cover the various tissue classes of interest - nuclei, stroma, epithelium, blood vessels, etc. The presence of these tissue classes are imperative within the target and template images in order to be able to identify and align tissue clusters. In general, normalization of images becomes increasingly difficult when the variations in staining of the slides increases. Our results show that in many cases, when the moving and template images are similar in terms of their staining characteristics, less sophistocated approaches tend to do very well. It is in scenarios where there is large discrepancy in staining variations between slides that the StaNoSA approach is able to demonstrate a substantial improvement over the state of the art. As StaNoSA is computationally more burdensome, it may be possible to selectively invoke StaNoSA in only those settings where it is determined that the tissue class representation for the target and template images differ substantially. 6. Concluding Remarks In this work, we present a new color standardization approach called Stain Normalization using Sparse AutoEncoders (StaNoSA) which attempts to address the limitations of previous related approaches. We leverage the intuition that filters learned via deep learning tend to respond similarly to tissue sub-types having similar characteristics, even across images. This invariance allows for accurate, and unsupervised, partitioning of the tissue compartments for subsequent histogram matching and alignment. In a comparison of StaNoSA with 4 different color standardization approaches (one of them being a recent state of the art scheme), StaNoSA was able to (1) reduce inter scanner color variance to within the range observed in images scanned multiple times on the same scanner, (2) reduce extreme variations in color induced by differential staining, slide preparation, and slide digitization, and (3) improve performance of subsequent image processing algorithms, specifically nuclei detection. In a majority of cases, StaNoSA outperformed the other comparative approaches, in many cases by over 50%. In conclusion, StaNoSA appears to be able to handle situations wherein tissue classes are disproportionately represented between target and template images. Future work will entail a more rigorous and automated way of identifying the algorithmic parameters (e.g. number of clusters K) and larger scale validation studies. Acknowledgments Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under award numbers 1U24CA199374-01, R21CA167811-01, R21CA179327-01; R21CA195152-01 the National Institute of Diabetes and Digestive and Kidney Diseases under award number R01DK098503-02, the DOD Prostate Cancer Synergistic Idea Development Award (PC120857); the DOD Lung Cancer Idea Development New Investigator Award (LC130463), the DOD Prostate Cancer Idea Development Award; the Ohio Third Frontier Technology development Grant, the CTSC Coulter Annual Pilot Grant, the Case Comprehensive Cancer Center Pilot Grant VelaSano Grant from the Cleveland Clinic the Wallace H. Coulter Foundation Program in the Department of Biomedical Engineering at Case Western Reserve University. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Figure 1 Visual appearance of the H&E slides can differ greatly depending on protocols and equipment used for staining. A GI stained H&E sample (a) acts as a template image, while the target image (b) is another H&E GI image in which we can see a strong visual difference due to staining protocol. Finally, after applying StaNoSA we can see (c) a shifted version of (b) which is in the same color space as (a). Their respective intensity distributions for each of the R, G, B color channels are presented in panels (d)-(f) respectively, which in turn show that the color distributions for the target images histogram (e) is heavily skewed towards the left. After applying StaNoSA (f) the intensity distribution for each of the 3 channels begins to more closely resemble that of the template image (d). Panels (g) and (h) show the same tissue slide but digitized with a (g) Leica scanner and (h) Ventana scanner, respectively. We can see that even though the only difference between the two images is the fact that they were digitized on two different scanners, panel (g) reveals more of a pink hue while (h) presents with a darker purple appearance. These differences are reflected in the corresponding color distributions ((i), (j)) for the two images in (g) and (h) respectively. Figure 2 An illustrative flowchart showing the StaNoSA process of color standardization. Figure 3 Original image (a) used as T with a sampling of (b) random training patches. For a sparse autoencoder to learn representative filters well, we need to ensure an image is well represented from the patches which are extracted. We can see in (b) that our sampling procedure provides a nice distribution of the various classes which are present in the image such as background (red arrow), nuclei (green arrow), lymphocytes (blue arrow), and stroma regions (yellow arrow). Figure 4 Illustration of different filter responses via the StaNoSA algorithm on H&E images. We can see that both the (a) template image and (d) moving image respond to an arbitrarily selected (b) & (e) filter in similar ways. Extracted from the green boxes of (a) and (d), the assignment of each pixel to one of 50 clusters can be viewed in gray scale shown in (c) and (f), respectively. Figure 5 Box and Whisker plots showing μin and σin for intra scanner differences ((a)-(c)). We compare (a) S1,V1 to S1,V2, and (b) S1,V3 and then (c) S1,V2 to S1,V3. Red line indicates μin, the blue box bounds the 25th percentile and the black whiskers extend to the 75th percentile. It can be seen quite clearly that our StaNoSA approach not only reduces μin, but also reduce variance across σin. Additionally, we compare ((d)-(f)) S1,L to S1,Vi i ∈ {1, 2, 3} and see a similar trend for μit and σit. Figure 6 The original ↓H↓E and ↑H↑E images in the top row. In the middle row we normalize to the others color space, such that ↑H↑E gets normalized to ↓H↓E and ↓H↓E gets normalized to ↑H↑E. Lastly, we show both of them being normalized to the standard HE image. In all cases, we can see that the images strongly resemble their template image color characteristics. Figure 7 Typical (a) T versus a (b) T used which was specifically chosen because of the class dis-proportionality. In particular we see the large pink object, of red blood cells in (b), on the right side of the screen. It would be expected that this affects the GL and HS approachs as non-red blood cells in S are aligned with the red blood cells histogram components. Figure 8 Box and Whisker plots showing μϕ and σϕ before normalization and normalization using Direct, Global Histogram (GL), Reinhard et al. (RH) [20], Macenko et al. (MM) [28], Histogram Specification (HS) [9], and Khan et al. (DM)[8]. Panel (a) shows the instances where the template and moving image share similar tissue class proportions, while (b) illustrates the consequences of having imbalanced class proportions. The Red line indicates the mean, the blue box bounds the 25th percentile and the black whiskers extend to the 75th percentile. Figure 9 Bar plots showing μϕ on a per staining protocol basis (HE, H↓E, H↑E, ↓HE, ↓H↓E, ↑HE, and ↑H↑E, where ↑ and ↓ indicate over- and under-staining of the specified dye) before normalization and after normalization using Direct, Global Histogram, Reinhard et al. (RH) [20], Macenko et al. (MM) [28], Histogram Specification (HS) [9], and Khan et al. (DM)[8] Table 1 List of common mathematical notation in this paper. Symbol Description T Template Image T. Feature space representation of T T˚ A pixel-wise cluster indicator of T S Target Image p Number of patches v Width and height of patches V Number of elements per patch (i.e., v × v × 3 in RGB) x Vector version of p of size 1 × V X Data matrix of size p × V W Weight Matrix of size h × V b Bias vector of size 1 × V θ Encoding function paramterised by {W,b} θ* Optimal θ found by gradient descent ε Bionomial corrupter X̂ A corrupted version of X (i.e., ε(X)) l Layer indexing variable x (l) Representation of x at layer l Table 2 The three unique datasets used in the evaluation of Stanosa Name Organ Stain Images Importance S 1 Breast HE 25 at 40× Same slides scanned on different equipment S 2 GI HE 175 at 40× Adjacent slices stained using different protocols S 3 GI HE 7 at 40× A subset of S2 containing manual annotations of nuclei boundaries Table 3 Confusion matrix showing μp,q and σp,q across all 7 protocols of 25 images for Direct, Global Histogram (GL), Reinhard et al. (RH) [20], Macenko et al. (MM) [28], Histogram Specification (HS) [9], and Khan et al. (DM) [8]. Lowest μp,q for each group is bolded. In almost all cases, StaNoSA has the lowest μp,q. HE ↑H↑E ↓H↓E ↓HE H↓E H↑E ↑HE HE N/A 0.35±0.02 0.43±0.03 0.43±0.03 0.45±0.03 0.46±0.03 0.54±0.03 Direct N/A 0.09±0.00 0.12±0.00 0.12±0.00 0.11±0.00 0.11±0.00 0.10±0.00 GL N/A 0.05±0.00 0.06±0.00 0.05±0.00 0.04±0.00 0.04±0.00 0.04±0.00 StaNoSA N/A 0.07±0.00 0.10±0.00 0.10±0.00 0.08±0.00 0.09±0.00 0.07±0.00 DM N/A 0.41±0.02 0.34±0.02 0.37±0.02 0.33±0.02 0.35±0.02 0.38±0.02 HS N/A 0.35±0.05 0.42±0.03 0.42±0.03 0.45±0.03 0.45±0.03 0.45±0.03 MM N/A 0.48±0.02 0.43±0.03 0.43±0.03 0.48±0.02 0.47±0.02 0.55±0.02 RH ↑H↑E 0.35±0.02 N/A 0.39±0.01 0.36±0.01 0.29±0.01 0.27±0.01 0.25±0.02 Direct 0.28±0.01 N/A 0.14±0.00 0.12±0.00 0.10±0.00 0.10±0.00 0.10±0.00 GL 0.17±0.00 N/A 0.07±0.00 0.07±0.00 0.05±0.00 0.05±0.00 0.04±0.00 StaNoSA 0.28±0.01 N/A 0.13±0.00 0.12±0.00 0.10±0.00 0.10±0.00 0.09±0.00 DM 0.31±0.03 N/A 0.18±0.02 0.17±0.01 0.16±0.01 0.15±0.01 0.17±0.02 HS 0.96±0.14 N/A 0.22±0.01 0.22±0.01 0.20±0.02 0.20±0.02 0.21±0.02 MM 0.31±0.01 N/A 0.24±0.01 0.24±0.01 0.25±0.01 0.24±0.01 0.29±0.01 RH ↓H↓E 0.43±0.03 0.39±0.01 N/A 0.10±0.01 0.19±0.01 0.24±0.01 0.41±0.01 Direct 0.33±0.02 0.16±0.01 N/A 0.10±0.00 0.10±0.00 0.10±0.00 0.09±0.00 GL 0.28±0.01 0.16±0.01 N/A 0.03±0.00 0.04±0.00 0.05±0.00 0.06±0.00 StaNoSA 0.34±0.02 0.15±0.01 N/A 0.07±0.00 0.08±0.00 0.08±0.00 0.07±0.00 DM 0.37±0.02 0.22±0.02 N/A 0.12±0.02 0.11±0.01 0.15±0.01 0.22±0.03 HS 1.04±0.17 0.37±0.18 N/A 0.09±0.01 0.14±0.01 0.16±0.01 0.23±0.02 MM 0.38±0.00 0.43±0.00 N/A 0.32±0.01 0.36±0.00 0.34±0.01 0.45±0.00 RH ↓HE 0.43±0.03 0.36±0.01 0.10±0.01 N/A 0.14±0.00 0.20±0.00 0.39±0.01 Direct 0.34±0.02 0.16±0.01 0.11±0.00 N/A 0.09±0.00 0.09±0.00 0.09±0.00 GL 0.27±0.01 0.15±0.01 0.06±0.00 N/A 0.05±0.00 0.05±0.00 0.06±0.00 StaNoSA 0.34±0.02 0.16±0.01 0.10±0.00 N/A 0.08±0.00 0.08±0.00 0.07±0.00 DM 0.39±0.02 0.21±0.02 0.11±0.01 N/A 0.11±0.01 0.11±0.01 0.18±0.01 HS 1.02±0.14 0.35±0.17 0.10±0.01 N/A 0.12±0.00 0.15±0.00 0.22±0.01 MM 0.36±0.00 0.42±0.00 0.29±0.00 N/A 0.34±0.00 0.32±0.00 0.43±0.01 RH H↓E 0.45±0.03 0.29±0.01 0.19±0.01 0.14±0.00 N/A 0.11±0.00 0.30±0.01 Direct 0.37±0.02 0.18±0.01 0.15±0.00 0.13±0.00 N/A 0.09±0.00 0.09±0.00 GL 0.28±0.01 0.15±0.01 0.07±0.00 0.06±0.00 N/A 0.03±0.00 0.04±0.00 StaNoSA 0.36±0.02 0.18±0.02 0.15±0.00 0.13±0.00 N/A 0.08±0.00 0.08±0.00 DM 0.27±0.02 0.22±0.02 0.12±0.01 0.11±0.01 N/A 0.12±0.01 0.20±0.02 HS 1.25±0.14 0.36±0.18 0.17±0.00 0.16±0.00 N/A 0.09±0.00 0.17±0.01 MM 0.34±0.00 0.36±0.00 0.26±0.01 0.26±0.01 N/A 0.26±0.01 0.36±0.01 RH H↑E 0.46±0.03 0.27±0.01 0.24±0.01 0.20±0.00 0.11±0.00 N/A 0.28±0.01 Direct 0.37±0.02 0.18±0.01 0.15±0.00 0.13±0.00 0.10±0.00 N/A 0.10±0.00 GL 0.27±0.01 0.14±0.01 0.07±0.00 0.06±0.00 0.03±0.00 N/A 0.04±0.00 StaNoSA 0.36±0.02 0.17±0.01 0.14±0.00 0.12±0.00 0.08±0.00 N/A 0.08±0.00 DM 0.32±0.02 0.18±0.02 0.14±0.01 0.12±0.01 0.11±0.01 N/A 0.17±0.02 HS 1.24±0.12 0.38±0.20 0.21±0.01 0.19±0.00 0.09±0.00 N/A 0.17±0.01 MM 0.31±0.00 0.35±0.00 0.23±0.00 0.22±0.00 0.26±0.00 N/A 0.34±0.00 RH ↑HE 0.54±0.03 0.25±0.02 0.41±0.01 0.39±0.01 0.30±0.01 0.28±0.01 N/A Direct 0.38±0.02 0.20±0.02 0.17±0.00 0.15±0.00 0.11±0.00 0.12±0.00 N/A GL 0.28±0.01 0.15±0.02 0.09±0.00 0.08±0.00 0.05±0.00 0.05±0.00 N/A StaNoSA 0.38±0.02 0.19±0.02 0.16±0.00 0.14±0.00 0.10±0.00 0.11±0.00 N/A DM 0.27±0.02 0.19±0.02 0.15±0.02 0.13±0.01 0.14±0.01 0.13±0.01 N/A HS 1.50±0.12 0.44±0.26 0.27±0.01 0.26±0.01 0.16±0.01 0.16±0.01 N/A MM 0.29±0.00 0.22±0.01 0.22±0.00 0.22±0.00 0.18±0.00 0.18±0.00 N/A RH Algorithm 1 LearningFilters Input: A template image T, a target image S, patch matrix X, number of levels L, architecture configuration h Output: T.∈R∣T∣×h(L),S.∈R∣S∣×h(L)     1: T.=ψ(c),∀c∈(T)     2: S.=ψ(c),∀c∈(S)     3: for l = 1 to L do     4:     Find θ(l)*, θ′(l)* using Equation 2     5:     X = fθ(l)* (X)     6:         T.=fθ(l)⋆(T.)     7:         S.=fθ(l)⋆(S.)     8: end for     9: return T.,S. Algorithm 2 HistogramMatching Input: T,T˚,S,S˚,K, number of bins Q Output: final normalized image S~     1: for k = 1 : K do     2:     T^=subset of T which has T˚=k     3:     S^=subset of S which has S˚=k     4:     for h = {R, G, B} do     5:         fT = Cumulative Sum of ψT^,h(T^) using Q bins     6:         fS = Cumulative Sum of ψS^,h(S^) using Q bins     7:         Δ is a function which minimizes ∣fS(Δ(q))−fT(q)∣∀q∈{1,…,Q}     8:                 ψS,h(S^)=Δ(S^)     9:         end for     10: end for     11: return R(q) Highlights Digital histopathology slides have many sources of variance These variances can cause algorithms to perform erratically Stain Normalization using Sparse AutoEncoders (StaNoSA) in introduced It standardizes color distributions of a test image to a single template image Validated using 3 experiments with 5 other color standardization approaches This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. References 1 Gurcan M Boucheron L Histopathological image analysis: a review. IEEE Rev Biomed Eng 2009 2 147 171 20671804 2 Veta M Pluim J Breast cancer histopathology image analysis: a review. IEEE Trans Biomed Eng 2014 61 5 1400 1411 24759275 3 Monaco J Hipp J Image segmentation with implicit color standardization using spatially constrained expectation maximization: detection of nuclei. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2012 15 365 72 4 Khan AM El-Daly H Rajpoot N Ranpec: Random projections with ensemble clustering for segmentation of tumor areas in breast histology images, in: Medical Image Understanding and Analysis (MIUA) British Machine Vision Association (BMVA) 2012 17 23 5 Hipp J Cheng J Spatially invariant vector quantization: A pattern matching algorithm for multiple classes of image subject matter including pathology. J Pathol Inform 2011 2 13 21383936 6 Kong J Sertel O Computer-aided grading of neuroblastic differentiation: Multi-resolution and multi-classifier approach. ICIP (5) 2007 525 528 IEEE 7 Wang Y Chang S A color-based approach for automated segmentation in tumor tissue classification. Conf Proc IEEE Eng Med Biol Soc 2007 2007 6577 6580 18003532 8 Khan A Rajpoot N A nonlinear mapping approach to stain normalization in digital histopathology images using image-specific color de-convolution. IEEE Trans Biomed Eng 2014 61 6 1729 1738 24845283 9 Jain AK Fundamentals of Digital Image Processing 1989 Prentice-Hall, Inc. Upper Saddle River, NJ, USA 10 Irshad H Veillard A Methods for nuclei detection, segmentation, and classification in digital histopathology: A review – current status and future potential Biomedical Engineering, IEEE Reviews in 2014 7 97 114 11 Sahirzeeshan A Madabhushi A An integrated region-, boundary, shape-based active contour for multiple object overlap resolution in histological imagery. IEEE transactions on medical imaging 2012 31 1448 60 22498689 12 Fatakdawala H Xu J Expectation-maximization-driven geodesic active contour with overlap resolution (emagacor): application to lymphocyte segmentation on breast cancer histopathology. IEEE Transactions on Biomedical Engineering 2010 57 7 1676 1689 20172780 13 Xu J Xiang L Stacked sparse autoencoder (ssae) for nuclei detection on breast cancer histopathology images. IEEE Transactions on Medical Imaging 14 Tian F Gao B Learning deep representations for graph clustering 2014 1293 1299 Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence Québec City, Québec, Canada July 27 -31, 2014 15 Vincent P Larochelle H Extracting and composing robust features with denoising autoencoders Proceedings of the 25th International Conference on Machine Learning, ICML ’08 2008 1096 1103 ACM New York, NY, USA 16 Ruifrok AC Johnston DA Quantification of histochemical staining by color deconvolution. Anal Quant Cytol Histol 2001 23 4 291 299 11531144 17 Hamilton PW Bartels PH Automated location of dysplastic fields in colorectal histology using image texture analysis. J Pathol 1997 182 1 68 75 9227344 18 Qureshi H Sertel O Adaptive discriminant wavelet packet transform and local binary patterns for meningioma subtype classification. Med Image Comput Comput Assist Interv 2008 11 Pt 2 196 204 18982606 19 Ruiz A Sertel O Pathological image analysis using the gpu: Stroma classification for neuroblastoma Bioinformatics and Biomedicine, 2007. BIBM 2007. IEEE International Conference on 2007 78 88 20 Reinhard E Ashikhmin M Color transfer between images IEEE Comput. Graph. Appl 2001 21 5 34 41 21 Magee D Treanor D Colour normalisation in digital histopathology images Proc. Optical Tissue Image analysis in Microscopy, Histopathology and Endoscopy (MICCAI Workshop) 2009 100 111 22 Rabinovich A Agarwal S Thrun S Saul L Schölkopf B Unsupervised color decomposition of histologically stained tissue samples Advances in Neural Information Processing Systems 2004 16 667 674 MIT Press 23 Basavanhally A Madabhushi A Em-based segmentation-driven color standardization of digitized histopathology Proc. SPIE 2013 8676 86760G 86760G–12 24 Bengio Y Lamblin P Greedy layer-wise training of deep networks In NIPS 2007 MIT Press 25 Bengio Y Learning deep architectures for ai Foundations and Trends in Machine Learning 2009 2 1 1 127 26 Zhang T Solving large scale linear prediction problems using stochastic gradient descent algorithms ICML 2004: Proceedings of the 21st International Conference on Machine Learning 2004 919 926 27 Coates A Lee H Ng A Gordon G Dunson D Dudk M An analysis of single-layer networks in unsupervised feature learning Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, Vol. 15 of JMLR Workshop and Conference Proceedings 2011 215 223 28 Macenko M Niethammer M A method for normalizing histology slides for quantitative analysis Biomedical Imaging: From Nano to Macro, 2009. ISBI ’09. IEEE International Symposium on 2009 1107 1110 29 Goodfellow IJ Warde-Farley D Lamblin P Dumoulin V Mirza M Pascanu R Bergstra J Bastien F Bengio Y Pylearn2: a machine learning research library arXiv preprint arXiv:1308.4214 30 Bergstra J Breuleux O Theano: a CPU and GPU math expression compiler Proceedings of the Python for Scientific Computing Conference (SciPy) 2010 oral Presentation 31 Dice L Measures of the amount of ecologic association between species Ecology 1945 26 3 297 302

PMC005xxxxxx/PMC5112176.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985117R 4816 J Immunol J. Immunol. Journal of immunology (Baltimore, Md. : 1950) 0022-1767 1550-6606 25786692 5112176 10.4049/jimmunol.1401220 NIHMS664582 Article The TNF-family ligand TL1A and its receptor DR3 promote T-cell mediated allergic immunopathology by enhancing differentiation and pathogenicity of IL-9 producing T cells1 Richard Arianne C. *2 Tan Cuiyan †2 Hawley Eric T. * Gomez-Rodriguez Julio ‡ Goswami Ritobrata § Yang Xiang-ping ¶ Cruz Anthony C. * Penumetcha Pallavi * Hayes Erika T. * Pelletier Martin * Gabay Odile * Walsh Matthew ‖ Ferdinand John R. *# Keane-Myers Andrea ** Choi Yongwon ‖ O'Shea John J. ¶ Al-Shamkhani Aymen # Kaplan Mark H. § Gery Igal † Siegel Richard M. *2 Meylan Françoise *2 * Immunoregulation Section, Autoimmunity Branch, NIAMS, NIH † Experimental Immunology Section, NEI, NIH ‡ Genetic Disease Research Branch, NHGRI, NIH § Department of Pediatrics and Microbiology and Immunology, Indiana University School of Medicine ¶ Molecular Immunology and Inflammation Branch, NIAMS, NIH ‖ Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA # Cancer Sciences Academic Unit, Faculty of Medicine, University of Southampton, Southampton, UK ** Biological Defense Research Directorate, Biological Naval Medical Research Center, Fort Detrick, MD Corresponding author: Richard M. Siegel, M.D, Ph.D. Bldg 10 Rm 13C103A, NIH Bethesda MD, 20892, rsiegel@nih.gov, Ph: 301-496-3761, FAX: 301-451-5394 2 Equal Contributors 28 3 2015 18 3 2015 15 4 2015 16 11 2016 194 8 35673582 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The TNF family cytokine TL1A (Tnfsf15) costimulates T cells and type 2 innate lymphocytes (ILC2) through its receptor DR3 (Tnfrsf25). DR3-deficient mice have reduced T cell accumulation at the site of inflammation, and reduced ILC2-dependent immune responses in a number of models of autoimmune and allergic diseases. In allergic lung disease models, immunopathology and local Th2 and ILC2 accumulation is reduced in DR3 deficient mice despite normal systemic priming of Th2 responses and generation of T cells secreting IL-13 and IL-4, prompting the question of whether TL1A promotes the development of other T cell subsets that secrete cytokines to drive allergic disease. Here we find that TL1A potently promotes generation of murine T cells producing IL-9 (Th9) by signaling through DR3 in a cell-intrinsic manner. TL1A enhances Th9 differentiation through an IL-2 and STAT5-dependent mechanism, unlike the TNF-family member OX40, which promotes Th9 through IL-4 and STAT6. Th9 differentiated in the presence of TL1A are more pathogenic, and endogenous TL1A signaling through DR3 on T cells is required for maximal pathology and IL-9 production in allergic lung inflammation. Taken together, these data identify TL1A-DR3 interactions as a novel pathway that promotes Th9 differentiation and pathogenicity. TL1A may be a potential therapeutic target in diseases dependent on IL-9. Introduction Cytokines in the TNF superfamily have important roles in host defense and amplification of both innate and adaptive immune responses. The TNF family cytokine TL1A costimulates T cells and innate lymphocytes through its unique receptor DR3 (Tnfrsf25). TL1A is synthesized in response to TLR ligands and other proinflammatory stimuli and optimizes T cell responses to bacterial pathogens such as salmonella and viral infections (1, 2). TL1A may also be involved in autoimmune disease pathogenesis, as TL1A levels are elevated in chronic inflammatory diseases such as rheumatoid arthritis and inflammatory bowel disease (3-5). Genetic variants in the TL1A locus are associated with human inflammatory bowel disease (6, 7), and a duplication in the DR3 locus has been linked to rheumatoid arthritis (8). TL1A/DR3 interactions are required for maximal pathology in diverse autoimmune disease models including collagen-induced arthritis (CIA), experimental autoimmune encephalomyelitis (EAE), and multiple forms of induced colitis (9-13). In allergic disease models, TL1A plays an important role in T cell-dependent airway hypersensitivity induced by ovalbumin (10, 14) and in ILC2-driven allergic lung disease triggered by papain exposure (15, 16). While TL1A has been shown to activate ILC2s and promote IL-13 production in allergic disease (15, 16), the mechanism by which TL1A enhances the adaptive arm of allergic responses remains less clear. IL-9 is an important driver of allergic disease (17-21), and although its production was historically attributed to the Th2 subset of CD4+ T cells, recent work has highlighted another T cell subset, named Th9, as a major producer of IL-9 (22). Th9 can be generated from naïve T cell precursors by activation in the presence of TGFβ and IL-4, with IL-4 acting through Signal Transducer and Activator of Transcription 6 (STAT6) (23-25). Alternative cytokines and signaling molecules such as IL-25, IL-1, TSLP, Jagged-2, calcitonin gene-related peptide and the TNF-family member OX40 ligand (OX40L) have been proposed to enhance Th9 differentiation and/or T-cell production of IL-9 (17, 26-30), while cyclooxygenase-2-derived prostaglandins, programmed cell death ligand 2, and 1,25-dihidroxyvitamin D3 inhibit Th9 polarization (31-33). In addition to murine allergic lung disease, T cell production of IL-9 has been shown to promote T-cell mediated inflammatory eye disease, EAE, colitis, and clearance of intestinal nematode infection (23, 24, 34-38). In the lung, IL-9 acts on mast cells, epithelial cells and smooth muscle cells to promote production of cytokines and chemokines such as IL-8, IL-13 and eotaxin, which enhance recruitment and activation of neutrophils, eosinophils and mast cells in allergic inflammation. IL-9 also induces smooth muscle hyperplasia and goblet cell metaplasia, important factors in the airway remodeling seen in asthma and allergic airway disease (19, 39, 40). Importantly, IL-9 has been shown to induce airway remodeling in an IL-13 dependent manner, placing IL-9 ‘upstream’ of IL-13 in these disease models (37, 41). We investigated the role of TL1A in promoting IL-9 production in allergic and autoimmune disease, and found that during T cell-dependent induced airway hypersensitivity and the T cell phase of allergic lung disease, total IL-9 and the number of IL-9-producing T cells are reduced in concert with decreased allergic pathology in Tnfrsf25-/- mice, leading to the hypothesis that TL1A may drive Th9 differentiation. Indeed, TL1A specifically and potently enhances the differentiation of IL-9 secreting T cells from naïve T cell precursors. Enhancement of IL-9 secretion depends on cell-intrinsic signaling through DR3, as well as TGFβ. TL1A can act independently of STAT6, but IL-2 signaling through STAT5 is absolutely required for its enhancement of Th9 differentiation. TL1A also acts independently of the ETS family transcription factor PU.1, which has previously been found to be important for Th9 generation (18). In addition to promoting Th9 differentiation, TL1A enhances pathology induced by Th9 in models of experimental eye inflammation and allergic airway disease. DR3-deficient Th9 transferred into a wild-type host are defective in inducing allergic airway inflammation, indicating that endogenous TL1A signaling through DR3 is necessary for Th9 to promote allergic pathology. These results define TL1A as a cytokine that can promote the generation and pathogenicity of IL-9 secreting T cells through a pathway distinct from those previously defined for this T helper subset. Materials and Methods Mice Wild-type C57BL/6 and CD45.1+ mice were obtained from Taconic. Tnfrsf25-/- mice were generated as previously described (42), and were back-crossed to the C57BL/6 background for at least ten generations. This line was subsequently crossed to OT-II Rag-deficient mice. Stat5a/bfl/fl CD4Cre OT-II and Stat5a/bfl/fl OT-II mice were from the NIAMS Taconic mouse colony, and Stat6-/- mice were from Jackson Labs. Sfpi1fl/fl LckCre mice were produced as previously described (18). Mice used for the analysis of ocular inflammation were (FVB/N×B10.BR) F1 hybrids, transgenically expressing either HEL in their eyes (“HEL-Tg”), or HEL-specific TCR by their T- cells (“3A9”) (43). T cell specific TRAF6-deficient mice were previously described (44). T cell differentiation assays Lymph nodes and spleens were harvested from mice of the appropriate genotypes and cells were passed through a 40 μm strainer. Red blood cells were lysed with Ack lysis buffer, and cells were sorted for CD4+ T cells using the EasySep Mouse CD4+ T Cell Enrichment Kit (Stemcell Technologies), according the manufacturer's protocol. CD4+ T cells were then stained with anti-mouse CD4 PerCP-Cy5.5, anti-mouse CD44 APC, anti-mouse CD62L PE, and anti-mouse CD25 FITC (eBioscience and BD Biosciences). Naïve CD4+ T cells identified as CD4+, CD44lo, CD62Lhi, and CD25lo were separated by fluorescence-activated cell sorting on a FACSAria Flow Cytometer (BD Biosciences). For certain experiments, cells were CFSE-labeled. Cells were cultured in complete RPMI medium (RPMI with 10% fetal calf serum, 10 mM HEPES, 1 mM sodium pyruvate, 10 U/ml penicillin, 10 U/ml streptomycin, 2 mM glutamine and 0.05 mM β-mercaptoethanol). Cells were plated at 50,000-100,000 cells per well on 96-well plate or 100,000-400,000 cells per well on a 48-well plate. For activation and costimulation, plates were either coated with anti-CD3 (clone BE0001-1) and anti-CD28 (clone 37.51) or cells were cultured in the presence of T-depleted splenic antigen-presenting cells (APC) (5:1, APC:T cells) and soluble anti-CD3 and anti-CD28. APC were prepared from total mouse splenocytes that were strained, treated with Ack lysis buffer, and depleted of T cells by staining with biotin-conjugated Thy1.1 and using magnetic biotin beads in an AutoMACS Depletes sort (Miltenyi Biotec). APC were then irradiated at 1000 rad to prevent growth. Polarization conditions were as follows: for Th0, 10 μg/mL anti-IFNγ (clone XMG1.2) and 10 μg/mL anti-IL-4 (clone 11B11); for Th1, 20 ng/mL murine IL-12 and 10 μg/mL anti-IL-4 (clone 11B11); for Th2, 20 ng/mL murine IL-4 and 10 μg/mL anti-IFNγ; for Th9, 20 ng/mL murine IL-4 and 5 ng/mL human TGFβ; for Th17, 20 ng/mL murine IL-6, 5 ng/mL human TGFβ, 10 μg/mL anti-IFNγ (clone XMG1.2), 10 μg/mL anti-IL-4 (clone 11B11), and 10 μg/mL anti-IL-2 (clone S4B6); and for iTreg, 100 units/mL human IL-2, 10 ng/mL human TGFβ, 10 μg/mL anti-IL-4 (clone 11B11), and 10 μg/mL anti-IFNγ (clone XMG1.2). In presence of APC, anti-IL-12 was added for Th0, Th2 and iTreg conditions. Additional conditions included the following as indicated: 10 ng/mL TL1A, 10 ng/mL OX40L, 10 μg/mL anti-IL-9 (clone 222622), 10 μg/mL anti-IL-13 (ratIgG1) obtained from Centocor/Johnson and Johnson (Horsham, PA), 10 μg/mL anti-IL-2 (clone S4B6), 10 μg/mL anti-CD25 (clone PC61), 20 ng/mL murine IL-6, or 100 units/mL human IL-2. Cells were cultured for 3 days. For Ova-specific Th9 cells, OT-II Rag-deficient T cells were purified and cultured with T-depleted APC with 10 ng/mL murine IL-4, 2 ng/mL human TGFβ, 0.5 μg/ml anti-CD28 (clone 37.51) 10 μg/ml of anti-IFNγ (clone XMG1.2), and 1 μM ovalbumin in complete IMDM medium for 3 days. To generate HEL-specific TCR transgenic Th9 cells we used the culture system described in detail elsewhere (45). Surface DR3 was stained with anti-mouse DR3 biotin (R&D Systems) followed by streptavidin-conjugated eFluor 450 (eBioscience). For analysis of cytokine production, cells were restimulated with 10 nM PMA, 1 μM ionomycin, and monensin for 4 hours, harvested and washed for staining. After labeling with surface antibodies, cells were fixed overnight with Cytofix/Cytoperm (BD Biosciences) or FoxP3 Fixation/Permeabilization buffer (eBioscience). Cells were blocked and stained with the following intracellular antibodies as indicated: anti-mouse/rat FoxP3 eFluor 450 (eBioscience), anti-mouse IL-9 APC (BioLegend), anti-mouse IL-17A FITC (eBioscience), anti-mouse IL-13 PE (eBioscience), anti-mouse IFNγ PE (BD Biosciences) and/or anti-mouse IL-4 FITC (eBioscience). Fluorphore-conjugated isotype control antibodies were used to check for non-specific binding. Cells were then washed and analyzed by flow cytometry. For phosphorylated STAT5 staining, cells were fixed in 4% paraformaldehyde, washed in PBS, and permeabilized with methanol overnight at -20 °C. Cells were stained in PBS with 0.5% Triton X-100 and 0.1% bovine serum with anti-phospho-STAT5 PE (BD Biosciences) and anti-CD4 at room temperature in the dark for 60 minutes. Fluorphore-conjugated isotype control antibodies were used to check for non-specific binding. Cells were then washed and analyzed by flow cytometry. Induction and quantification of Th9-mediated ocular inflammation HEL-specific Th9 cells generated in culture, at the indicated numbers were injected via the tail vein into syngeneic mice transgenically expressing HEL in their lens. The recipient eyes were collected at day 7 (or an alternative indicated time point) and analyzed by histological examination for inflammatory changes, using a scale of 0-9, as previously described (46). To test the effect of local TL1A, mice were intraocularly injected with TL1A (2 μg/eye) or PBS on days 2 and 3 post cell transfer, and eyes were examined for histological changes on day 4. To test TL1A blockade, mice were intraperitoneally injected with 20 mg/kg of the blocking anti-mouse TL1A monoclonal antibody 5.4G6 or hamster Ig control on days -1 and 3 post cell transfer. To test the effect of adding IL-10, recipient mice were treated intraperitoneally with a daily dose of 50 μg/kg (2 experiments) or 100 μg/kg (2 experiments) murine IL-10 starting 3 hours before cell injection until day 6 post cell injection. For measuring proliferation of transferred cells, Th9 cells were labeled with Cell Proliferation Dye eFluor 670, as recommended by the manufacturer (eBioscience). Five million labeled cells were injected into HEL-Tg recipient mice and spleens of these recipients were collected for flow cytometric analysis of the proportion of donor (1G12+) cells, as indicated. Ovalbumin and Papain Induced Lung Inflammation For ovalbumin-induced lung inflammation, mice were sensitized systemically on days 0 and 7 via a 200 μl intraperitoneal (i.p.) injection containing either 100 μg Chicken Ova (Sigma) or PBS emulsified in an equal volume mixture with alum (Thermo Scientific). For assessment of pulmonary inflammation, mice were challenged with 100 μg Ova or PBS/30 μl inoculum intratracheally (i.t.) on day 14 and intranasally (i.n.) on day 15. Mice were euthanized 24-72 hours after the final challenge to evaluate cell infiltration, cellular inflammation in the lung, and cytokine levels in the sera and bronchoalveolar lavage (BAL). BAL fluid was obtained by direct cannulation of the lungs with a 20-gauge intravenous catheter and lavage with 500 μl 1% fetal bovine serum (FBS) in PBS (for cytokine analysis) and with 750 μl 1% FBS in PBS (for analysis of cellular infiltration). Samples for cellular analysis were prepared as a cytospin (Thermo-Shandon, Pittsburgh, PA) for differential cellular analysis after staining with HEMA 3 stain set (Fisher Scientific), and a portion was used to determine total cell counts. Lung tissues were fixed in 4% neutral buffered formaldehyde, embedded in paraffin, sectioned, and stained with hematoxilin and eosin (H&E) or periodic acid-Schiff (PAS) stain. Cells were isolated from lungs by incubating lung fragments with 100U collagenase for 1 hour. Cells were stained for surface antigens and intracellular cytokines after stimulation with PMA/ionomycin for 4 hours with LIVE/DEAD Fixable Blue Dead Cell Stain (Life Technologies) and the following antibodies: TCRβ APC-eFluor 780 (eBioscience), CD45.2 PeCy7 (BioLegend), CD44 AlexaFluor 700 (eBioscience), CD4 V500 (BD Biosciences), IL-9 PE (BioLegend), IL-13 eFluor 660 (eBioscience), IL-17 FITC (eBioscience), IL-10 PerCP-Cy5.5 (eBioscience), IFNγ eFluor 450 (eBioscience). Cells from the BAL were stained using antibodies against CD45, SiglecF, F4/80, Ly6G and CD11b. Eosinophils were identified as CD45+ F4/80- Ly6G- CD11b+ SiglecF+. For RNA extraction, lung tissue was stored in Trizol and homogenized with a Precellys 24 (Bertin Laboratories) before chloroform extraction and purification with the PureLink RNA Mini Kit (Life Technologies). mRNA transcripts were measured by Taqman gene expression assays described below. For analysis of the pathogenic potential of Th9 in allergic lung disease, ovalbumin-specific wild-type or Tnfrsf25-/- Th9 cells (106 T cells in 200 μl PBS) were adoptively transferred into CD45.1 congenic wild-type recipient mice via tail vein injection. Mice were challenged twenty-four hours after cell transfer i.t. and forty-eight hours after cell transfer i.n. with 100 μg Ova. Mice were then sacrificed 12 hours after last challenge for further analyses described above. For papain-induced lung inflammation, mice were anaesthetized with isoflurane and exposed intranasally to 25 μg papain (Calbiochem) in 30 μL PBS on day 0, 3, 6 and 14. 12-16 hours after the last challenge, bronchoalveolar lavage was performed as described above. Lung-isolated cell analyses, RNA extraction and gene expression assays were performed as above. Lung histology was scored on H&E and Periodic Acid Schiff (PAS) stained sections by a reader with experimental conditions masked, using a scoring system modified from that described previously (47). Perivascular and peri-bronchiolar cuffing (PVC and PBC) were each scored as follows: 0: No visible infiltrate. 1: Patchy infiltrate in <25% of bronchioles or vessels, 2: Patchy infiltrate in <50% of bronchioles or vessels, 3: Widespread infiltrate >50% of bronchioles or vessels with circumferential infiltrates in most bronchioles or vessels. 4. Criteria for score of 3 plus vascular obliteration (for PVC) or bronchiolar plugging (for PBC). Interstitial inflammation was graded from 0-3 depending on the extent of cellular infiltrate into alveoli. Goblet cell hyperplasia was scored for small airways as follows: 0: No visible hyperplasia or mucous production, 1: patchy hyperplasia and/or PAS staining in <25% of bronchioles or vessels, not circumferential, 2: patchy hyperplasia and/or PAS staining of <50% of bronchioles, 3: widespread hyperplasia and >50% PAS staining in most bronchioles, 4: criteria for 3 plus bronchiolar plugging or obliteration. Scores reported were the total score for each lung (0-15). Lung immunohistochemistry Lungs were perfused with PBS and filled with OCT via the trachea. Lungs were subsequently embedded in OCT and frozen using an iso-pentane bath cooled with dry ice. Sections were cut to 5 microns using a cryostat (Leica CM1850) cooled to -23 °C and mounted on superfrost plus slides (Fisher). All subsequent steps were carried out at room temperature. Sections were dried for at least 1 hour and fixed in acetone for 10 minutes. Sections were blocked with 2.5% (v/v) normal goat serum (Vector laboratories) for 30 minutes and stained with 10 μg/ml Tan 2-2 (Rat anti mouse TL1A, was generated as described previously (9)) in PBS plus 0.05% Tween-20 (Sigma) for 2 hours. Slides were incubated with in 0.3% H2O2 in MeOH for 20 minutes to suppress endogenous peroxidases. Sections were incubated with ImmPRESS anti Rat Ig peroxidase (Vector laboratories) for 30 minutes and developed using DAB peroxidase substrate kit (Vector laboratories) for 7 minutes. Tissue was counterstained using Hematoxylin (Vector laboratories) for 2.5 minutes, washed with 2% glacial acetic acid and blued with 0.45% NaOH in 70% ethanol for 30 seconds. Slides were mounted using VectaMount AQ (Vector laboratories) and imaged using a Keyence microscope (Digital Microscopes). Image backgrounds were adjusted to white using Adobe Photoshop. Measurement of allergic airway reactivity Bronchial reactivity was determined 12h after the last challenge of Ova in the Ova-induced lung inflammation model or of papain in the papain-induced model. Mice were anesthetized by i.p. administration of ketamine/xylazine mixture (1 ml ketamine [100 mg/ml], 0.5 ml xylazine [20 mg/ml], and 8.5 ml PBS). A 18-gauge blunt-end needle was inserted into the trachea, and the animals then were ventilated mechanically. Baseline measurements were recorded after the aerosol administration of saline, followed by doubling doses of methacholine (6.25–100 mg/ml) using flexiVent (Scireq Scientific Respiratory Equipment). Cytokine measurement and analysis Supernatants from CD4+ T cell differentiation cultures were collected at day 3 after activation and analyzed for cytokines by a multiplex, bead-based protein detection assay. Supernatants were probed for 23 different cytokines and chemokines using BioRad's Bio-Plex Pro™ Mouse Cytokine 23-plex Assay according to the manufacturer's protocol. Protein concentrations were detected and calculated using a Bio-Plex 200 instrument and Bio-Plex Manager 6.0 (BioRad). Ratios were calculated for each cytokine produced in the presence of TL1A versus in the absence of TL1A. The natural logarithms of these ratios were hierarchically clustered using uncentered correlation in Cluster 3.0 (Michael Eisen) and visualized in Java TreeView (Alok Saldanha). For measurement of IL-9 from the supernatant of wild-type and PU.1-deficient T cells, cells were stimulated with plate bound anti-CD3 (2 μg/ml) for 24 hours. Cell-free supernatant was used to detect IL-9 by ELISA using anti-IL-9 capture antibody (D8402E8; BD Biosciences) and biotin-labeled anti-IL-9 detection antibody (D9302C12; Biolegend). For measurement of IL-9 and IL-10 in HEL-specific T cell cultures, supernatant cytokine concentrations were measured with ELISA kits from RayBiotech and R&D Systems, respectively. For cytokine measurements in lung homogenates, the middle right lung lobes were snap frozen in liquid nitrogen and stored at -80 until analysis. The frozen lungs were weighed and immediately transferred to tubes containing 500 uL Tissue Protein Extraction Reagent (ThermoScientific) with 5 uL HALT Protease Inhibitor Cocktail (ThermoScientific, 100X). Lung tissues were homogenized using a Precellys® instrument, and then centrifuged at 9,000 × g for 10 minutes at 4 degrees. Supernatants were transferred to clean microcentrifuge tubes. Total protein concentrations in the lung tissue homogenates were determined using a BCA kit. Cytokines were measured using the MAGPIX ® multiplexing system with BioPlex reagents (BioRad). Retroviral transduction Naïve CD4+ T cells were activated with plate-bound anti-CD3, anti-CD28 and anti-IFNγ for 24 hours. The activated cells were transduced with supernatants containing hNGFR retrovirus or hNGFR-caSTAT5 retrovirus in the presence of polybrene (8 μg/ml) by centrifuge at 2500 rpm for 90 minutes at 30°C. The same procedure of transduction was repeated 24 hours later and differentiation cytokines were added for 2 days. Flow cytometry analysis gated live, NGFR+ cells before examining cytokine production. Chromatin Immunoprecipitation (ChIP) ChIP for STAT5 binding was performed as described previously (48). Naïve CD4+ T cells were activated and polarized for 3 days followed by cross-linking for 8 minutes with 1% formaldehyde. The cells were harvested and lysed by sonication. After pre-clearing with protein A agarose beads (Upstate, VA), cell lysates were immunoprecipitated with anti-STAT5A/STAT5B antibody (R&D Systems) overnight at 4°C. After washing and elution, crosslinks were reversed at 65°C for 4 hours. The eluted DNA was purified, and samples were analyzed by quantitative-PCR with customer-designed primers and probes (Supplemental Table 1) using a 7500 real time PCR system (both Applied Biosystems, CA). The Ct value for each sample was normalized to corresponding input value. ChIP for PU.1 and IRF4 binding was performed as previously described (25). Gene expression analysis Where indicated, mRNA levels were measured by Taqman Gene Expression Assays (Life Technologies): IL-9 (Mm00434305_m1), IL-13 (Mm00434204_m1), IL-2 (Mm00434256_m1), IL-10 (Mm00439614_m1), IRF4 (Mm00516431_m1), B2M (Mm00437762_m1), Tbx21 (Mm00450960_m1), Gata3 (Mm00484683_m1), Rorc (Mm01261022_m1). Cytokine RNA expression measurements were made using the iTaq Universal Probes One-Step Kit (BioRad) on a CFX96 Real-Time PCR Detection System (BioRad). For transcription factor expression measurements, RNA was first converted to cDNA using the SuperScript VILO cDNA Synthesis Kit (Life Technologies) and then qPCR reactions using Taqman Gene Expression Master Mix (Life Technoliges) were run on a CFX384 Touch Real-Time PCR Detection System (Bio-Rad). All reactions were run in triplicate. For low-expressed genes, RNA input was increased until detectable and consistent Ct values were obtained across technical replicates. Several PBS-treated mice expressed no detectable IL-9 in the lung even with increased RNA input, and these were excluded from analysis. Replicate Ct values were normalized to replicate reference gene (B2M) Ct values (deltaCt), and fold-changes were calculated with respect to the indicated reference sample (2 raised to the power of delta-deltaCt). Results DR3 is required for optimal production of IL-9 and IL-13 in allergic airway disease DR3 knockout (Tnfrsf25-/-) mice or animals treated with blocking anti-TL1A antibodies are resistant to allergic disease induced by sensitization and airway challenge with ovalbumin (Ova), and less IL-13 mRNA is detected in the lung (10, 14). We examined the kinetics of IL-9 and IL-13 production in the lung of wild-type (WT) and Tnfrsf25-/- mice in this T-cell-dependent model (Figure 1A). We found that in WT mice IL-9 mRNA was elevated on day 16, one day after the last challenge with Ova, whereas IL-13 production was higher at day 18. Strikingly, in Tnfrsf25-/- mice, IL-9 mRNA was not induced at either time point, and IL-13 was significantly reduced at day 18 (Figure 1B). To determine whether T cells infiltrating the lung also depend on DR3 for efficient IL-9 and IL-13 secretion, we isolated cells from the lung at day 16 and determined the frequency of IL-9 and IL-13 producing T cells after brief restimulation in vitro. Although total lung IL-13 mRNA is very low at this time-point, IL-13 detection by flow cytometry in individual T cells enables examination of differences in this small population of infiltrating cells. As shown in Figure 1C, numbers of IL-9 and IL-13 secreting cells increased approximately 10-fold in Ova-sensitized WT mice compared to PBS-sensitized controls, whereas no such changes in cells secreting IL-9 nor IL-13 could be seen in Tnfrsf25-/- mice. This was not due to a general failure of the T response in the lung, because total numbers of CD4+CD44hi cells were still significantly increased by Ova priming in Tnfrsf25-/- mice. Despite correlated changes in IL-9- and IL-13-producing T cells, no double-producers were observed, and CD4+CD44hi cells did not secrete IL-10, IL-17, or IFNγ (Supplemental Figure 1A). Histological examination revealed reduced goblet cell hyperplasia and cellular infiltrates of Tnfrsf25-/- lungs compared to those from WT Ova-sensitized mice at day 16 (Figure 1D). Lung pathology in Ova-challenged Tnfrsf25-/- mice was significantly reduced compared to that in WT mice at day 16 (Figure 1D, lower panel), similar to what we previously observed at day 18 (10). Measurements of airway resistance in response to methacholine challenge revealed less airway hyperreactivity (AHR) in Tnfrsf25-/- mice challenged with Ova as compared to WT mice (Figure 1E). Bronchoalveolar lavage (BAL) of Tnfrsf25-/- mice challenged with Ova also showed reduced numbers of eosinophils and neutrophils at day 16 (Figure 1F), and as expected, IL-5 protein from whole lung homogenates paralleled eosinophil reduction (Supplemental Figure 1B). These data suggest that TL1A-DR3 interactions are important for IL-9 and IL-13 production by T cells early in the lung hypersensitivity response, which may shape the ensuing allergic pathology. To test whether DR3 is required to support Th9 development and lung inflammation in the setting of a local allergic response, we exposed WT or Tnfrsf25-/- mice to inhaled papain, an environmental protease known to induce pulmonary allergic responses and induce IL-9 along with other Th2 cytokines (Figure 2A). We chose to analyze responses after two weeks of exposure to papain, the time at which T cells play a role in disease (49-51). As with mice sensitized and challenged with Ova, the BAL fluid of Tnfrsf25-/- mice sensitized to papain contained reduced numbers of eosinophils and neutrophils compared to WT mice (Figure 2B), and lung IL-5 protein expression was again reduced (Supplemental Figure 2A). Of note, immunohistochemical staining revealed that papain induced the expression of TL1A on cells localized near blood vessels (Figure 2C). WT mice treated with papain developed severe immunopathology with goblet cell hyperplasia and cellular infiltration in the perivascular, interstitial and peribroncheal areas, while Tnfrsf25-/- mice were substantially less affected (Figure 2D). Similar to Ova-induced lung disease, papain-exposed Tnfrsf25-/- mice showed less AHR than WT in response to methacholine challenge (Figure 2E). Transcription of IL-9 and IL-13 was highly induced after 2 weeks of papain exposure, and DR3-deficient mice were defective in induction of both IL-9 and IL-13 mRNA, with the reduction in IL-9 even more pronounced than that in IL-13 (Figure 2F). Because ILC2s have been identified as the main producers of papain-induced IL-9 in the early allergic response (51), we restimulated total lung cells with PMA and ionomycin and analyzed protein production by flow cytometry to specifically determine whether T cell production of allergic cytokines was dependent on DR3. We found that numbers of T cells expressing either IL-9 or IL-13 were strikingly reduced in DR3 deficient mice (Figure 2G). Similar to ovalbumin-induced lung disease, there were no IL-9 and IL-13 double-producing T cells, and we found very little induction of IL-10 and IFNγ in response to papain. However, in contrast to the ovalbumin model, papain exposure did induce IL-17-producing effector T cells separate from IL-9-producing cells, which were also reduced in antigen-challenged Tnfrsf25-/- mice (Supplemental Figure 2B). These results show that DR3 also governs the production of cytokine secreting T cells and disease pathology after exposure to an inhaled allergen. TL1A promotes Th9 differentiation in the presence of TGFβ and either IL-4 or IL-2 TL1A may exert the effects we observed in the lung through enhancing differentiation of naïve T cells or promoting proliferation or survival of activated T cells. To investigate the effects of TL1A on T cell differentiation, we activated T cells under conditions optimized for differentiation into various T cell subsets in the presence or absence of TL1A. We previously found that DR3 was not required for generation of Th1, Th2 or Th17 cells, but that TL1A-DR3 signaling suppressed generation of FoxP3-expressing iTreg (10, 13). Consistent with previous reports (52), TL1A increases the percent of IFNγ-producing T cells in Th1 differentiation conditions to a moderate extent, but we do not see any consistent effects on Th2 or Th17 differentiation (Supplemental Figure 3A). Expression of lineage-defining transcription factors generally paralleled flow cytometry measurements of signature cytokines (Supplemental Figure 3B). When T cells were cultured under iTreg-inducing conditions in the presence of IL-2 and TGFβ, we found that in addition to reducing the percentage of FoxP3-expressing cells, TL1A promoted the production of IL-9 (Figure 3A). The effect was particularly strong in the presence of antigen presenting cells (APC), which enhanced T cell costimulation as evidenced by increased secretion of many cytokines and chemokines (Supplemental Figure 3C). Suppression of FoxP3 and induction of IL-9 was dependent on DR3, as Tnfrsf25-/- T cells were unaffected by TL1A treatment (Figure 3A). TL1A did not induce IL-9 production in T cells cultured for 3 days under non-polarizing Th0, Th1 or Th2 conditions, and TL1A did not enhance the production of cells expressing IL-4 or IL-13 under Th2 conditions (Supplemental Figure 3A), suggesting that the TGFβ and IL-2 included in iTreg polarization conditions are important for the ability of TL1A to induce IL-9 production. Interestingly, under Th17 conditions (TGFβ, IL-6 and blocking antibodies to alternative lineages), TL1A-induced IL-9 production was suppressed compared with that of iTreg cultures, which differ only in their lack of IL-6 and presence of IL-2 (1.6 % vs. 17.5 % in Figure 3B), while IL-17 production remained largely unaffected. This suggests that IL-6-derived signals antagonize the ability of TL1A to promote Th9 differentiation, or that IL-2 signaling enhances it. In the presence of IL-4 and TGFβ, conditions previously found to direct T cells to become IL-9-producing effector cells (23, 24, 53), TL1A significantly enhanced the development of IL-9-producing cells, both in the presence and absence of APC. In the presence of APC, some cells producing both IL-13 and IL-9 were induced by TL1A (Figure 3C). As with iTreg conditions, the enhancement of IL-9 production by TL1A was completely dependent on DR3. Although DR3 deficient T cells differentiated normally into Th9, endogenous TL1A produced under these culture conditions may not be sufficient to influence Th9 differentiation, as the bulk of TL1A is produced by APC induced by inflammatory stimuli such as TLR ligands, TNF and IL-1b (54). Titration of TGFβ showed that TL1A could enhance Th9 differentiation across a wide range of concentrations of TGFβ above 0.1 ng/ml, and TL1A lowered the amount of TGFβ necessary to induce Th9 differentiation by approximately 10-fold (Figure 3D). TL1A was extremely potent at enhancing Th9 differentiation, doubling the percentage of IL-9-producing T cells at concentrations as low as 0.1 ng/ml, and sharply augmenting IL-9 production with increasing concentration before plateauing between 1 and 10 ng/ml (Figure 3E). T cell subset-specific responses to TL1A may reflect differential expression of the TL1A receptor DR3. To test this possibility we measured DR3 surface expression under various polarization conditions (Figure 3F). DR3 expression was upregulated under all conditions with TGFβ (Figure 3F), indicating that TGFβ signaling enhances cellular responsiveness to TL1A. However, DR3 expression on Th9-polarized cells decreased with addition of TL1A (Figure 3G), suggesting downregulation of the receptor by its ligand. These results show that TL1A potently promotes differentiation of IL-9-producing T cells in a TGFβ-dependent manner but does not induce a positive feedback loop by inducing DR3 expression. Since TL1A can costimulate T cell proliferation, the increase in IL-9-producing T cells we observed might be due to enhanced proliferative responses. To explore this possibility, we examined IL-9 production in cultures of CFSE-labeled T cells to simultaneously track cell division and IL-9 production. Addition of exogenous TL1A only slightly enhanced proliferation in the absence of APC (data not shown), while no difference was detected in the presence of APC (Figure 3H). This is consistent with previous data showing that TL1A affects proliferation mainly under suboptimal stimulatory conditions (10). TL1A enhanced IL-9 production in all cells that had divided at least once, increasing IL-9 production in the first division by a factor of eight whether APC were absent or present (data not shown, and Figure 3H). TL1A enhanced the percentage of IL-9 producing cells at least 10-fold in cells that had divided twice. These data show that the TL1A primarily induces Th9 differentiation and IL-9 production as opposed to increasing T cell proliferation. TL1A can co-stimulate cytokine production and may therefore promote Th9 differentiation by inducing production of other cytokines that act in an autocrine manner. We measured cytokines in the supernatants of T cells activated in the presence or absence of TL1A under a number of polarization conditions by multiplex cytokine detection assays. In addition to IL-9, TL1A enhanced the production of IL-2, IL-3, IL-10, IL-13, and GM-CSF by purified T cells activated under Th9 conditions (Figure 3I, top panel). In the presence of APC, IL-5 and IL-6 were also upregulated by TL1A (Figure 3I, bottom panel). Unsupervised clustering of the TL1A-induced changes in cytokines secreted by various T helper subset differentiation cultures showed that while TL1A may be generally considered a costimulator, its effect depends on the polarization culture conditions. For example, IL-9 and IL-13 upregulation appeared strongest in Th9 and iTreg conditions in purified T cell culture. It is also interesting to note that the cytokines induced by TL1A under Th2 and Th9 conditions were closely correlated, which may reflect a common differentiation pathway of these two subsets (24). Induction of Th9 by TL1A depends on IL-2 and STAT5 To determine which cytokines may be responsible for enhanced Th9 differentiation, we considered those cytokines upregulated by TL1A that activate JAK/STAT signaling, which is well known to program T cell differentiation. IL-2, IL-3, IL-5, IL-6, IL-9, and GM-CSF fit these criteria and are upregulated by TL1A in both iTreg and Th9 conditions. To test whether IL-9 itself might positively influence Th9 differentiation, we added an IL-9 blocking antibody to T cells cultured under Th9 conditions in the presence and absence of TL1A. As shown in Figure 4A, blocking IL-9 did not reduce the frequency of IL-9-producing cells (Figure 4A, left panel). As in Th17 differentiation cultures containing IL-6 (Figure 3B), addition of IL-6 suppressed the ability of TL1A to induce Th9 (Figure 4A, middle panel), whereas blocking IL-6 antibodies enhanced Th9 differentiation in the presence of TL1A (Figure 4A, left panel). Addition of exogenous IL-2 did not increase Th9 producing cells over the TGFβ and IL-4 condition (Figure 4A, middle panel). However, a combination of anti-IL-2 and anti-CD25 antibodies, which completely blocks IL-2 signaling, dramatically suppressed Th9 differentiation in T cell activation cultures containing TGFβ and IL-4 and potently blocked the ability of TL1A to enhance Th9 differentiation (Figure 4A, right panel). Thus IL-2 appears to be a key autocrine factor in Th9 differentiation and its enhancement by TL1A, while IL-6 appears to antagonize this process, possibly though competition between IL-2-induced STAT5 and IL-6-induced STAT3 (48). To probe the relevant signaling pathways downstream of candidate cytokines induced by TL1A that might aid Th9 polarization, we used T cells deficient in the relevant STAT transcription factors that mediate their action. STAT6 is responsible for IL-4 and IL-13 signal transduction and is required for Th9 polarization induced by IL-4 (23-25). As expected, Stat6-deficient T cells were defective in IL-9 production under Th9 conditions, but, importantly, TL1A still enhanced numbers of IL-9-producing T cells by 10-fold (Figure 4B). Under iTreg conditions, deviation towards Th9 differentiation by TL1A was completely independent of STAT6 (Figure 4B). This shows that although signaling through STAT6 is required for conventional Th9 differentiation mediated by IL-4 and TGFβ, STAT6 is not required for the enhancement of Th9 differentiation mediated by TL1A. As a major signal transducer for IL-2, STAT5 is a good candidate for mediating the effects of TL1A. To determine the role of STAT5 in the enhancement of Th9 differentiation by TL1A, we activated naïve CD4+ T cells with conditional deletion of the STAT5A/B locus (Stat5CD4-/-) or control T cells (Stat5fl/fl) under Th9 polarization conditions with or without TL1A. Due to the requirement for STAT5 in T cell development (55), both STAT5-deficient and control T cells carried the OT-II transgenic TCR, but T cells were activated polyclonally. Under Th9 conditions, STAT5-deficient T cells were unable to differentiate into IL-9-producing cells, and importantly, IL-9 production by STAT5 deficient T cells could not be rescued by the addition of TL1A (Figure 4C). Only IL-13 production is increased by TL1A in the absence of STAT5 signaling (Figure 4C). To determine whether STAT5 is sufficient to drive Th9 development, we activated naïve T cells transduced with a retrovirus encoding a constitutively active STAT5 protein. These cells efficiently differentiated into IL-9-producing cells in the absence of TL1A or other T cell polarizing cytokines (Figure 4D), indicating that activated STAT5 is sufficient to drive Th9 differentiation. T cells expressing activated STAT5 differentiated into Th9 even in the presence of IL-6, suggesting that activated STAT5 can overcome the inhibitory effects of IL-6 acting through STAT3 on Th9 differentiation. Taken together, these data confirm that STAT5 is necessary and sufficient for Th9 differentiation, and show that TL1A, acting through autocrine IL-2 and STAT5, mediates a distinct pathway from IL-4 and TGFβ for inducing Th9 from naïve T cell precursors. To determine whether DR3 plays a cell-intrinsic role in Th9 differentiation beyond promoting autocrine IL-2 production, we mixed CD45 congenic WT naïve CD4+ T cells with DR3-deficient naïve CD4+ T cells, and activated them together under Th9 and iTreg conditions. Only wild-type cells responded to TL1A with an increase in IL-9 production under either iTreg or Th9 conditions (Figure 5A), indicating that T cells must receive a signal through DR3 in addition to IL-2 for optimal Th9 differentiation. These data suggested that DR3 signaling may potentiate the action of other transcription factors on the IL-9 promoter. PU.1 (Sfpi1), an ETS-family transcription factor that binds to the IL-9 promoter, enhances the generation of Th9 both in vitro and in vivo (18). As previously described, activating PU.1-deficient T cells under Th9 conditions resulted in a lower percentage of IL-9-producing cells and reduced IL-9 in the culture supernatant. However, TL1A enhanced IL-9 production in both WT and PU.1-deficient T cells to a similar degree (Figure 5B). Chromatin immunoprecipitation (ChIP) experiments also showed that TL1A did not significantly enhance PU.1 binding to the IL-9 promoter (Figure 5C). IRF4, which is induced by STAT6 and coordinately binds the IL-9 promoter with SMAD2/3 (25, 56), has also been identified as a key transcription factor in Th9 differentiation (57). ChIP experiments revealed a trend toward increased binding of IRF4 to the IL-9 promoter in the presence of TL1A (Figure 5D), but addition of TL1A did not increase IRF4 expression at the mRNA level (data not shown). Along with the data from STAT6-deficient T cells (Figure 4B), these results suggest that TL1A makes the IL-9 promoter more accessible to IRF4 but its effect is not directly through the STAT6-IRF4 pathway. Alternatively, we reasoned that DR3 signaling may enhance the ability of STAT5 to transactivate the IL-9 promoter. In support of this possibility, T cells activated under Th9 conditions in the presence of TL1A had higher baseline levels of phosphorylated STAT5 and increased STAT5 binding to the IL-9 promoter as assayed by ChIP (Figure 5E). These data support a role for TL1A, acting through DR3, in enhancing IL-2 signaling through STAT5 and increasing STAT5 promoter binding to promote differentiation of T cells capable of producing IL-9. OX40L, another member of the TNF superfamily, was recently reported to enhance Th9 differentiation through its receptor OX40, signaling via the non-canonical NF-κB pathway and TNF-receptor associated factor 6 (TRAF6) (26). To determine whether TL1A may induce Th9 through a TRAF6-dependent pathway, we activated naïve Traf6-/- T cells with TGFβ and IL-4. Unlike OX40L, we found that TL1A was able to enhance Th9 generation with similar efficiency in WT and Traf6-/- T cells (Figure 5F). Th9 differentiation by OX40L was dependent on IL-4 acting through STAT6, as OX40L only slightly increased the percentage of IL-9-producing cells differentiating from STAT6-deficient naïve T cell precursors (Figure 5G). These data show that TL1A mediates Th9 differentiation through mechanisms that are distinct from those of OX40L with respect to STAT6 and non-canonical NF-kB signaling through TRAF6. TL1A enhances the pathogenicity of antigen-specific Th9 in ocular and allergic lung inflammatory disease To determine whether TL1A enhances the pathogenicity of antigen-specific Th9 cells, we turned to a model system where T cells specific for hen egg lysozyme (HEL) trigger ocular inflammation when transferred into transgenic mice expressing HEL in the eye. HEL-specific T cells differentiated into Th1, Th17 and Th9 can all trigger ocular pathology when transferred into HEL-expressing recipients, and pathogenicity of Th9 correlates with their ability to produce IL-9 (45). We activated HEL-specific TCR transgenic T cells with antigen and APC under Th9 conditions in the presence or absence of TL1A. TL1A enhanced the differentiation of HEL-specific Th9 to an even greater extent than in polyclonal T cell cultures, increasing the percentage of T cells capable of producing IL-9 to more than 90% after 3 days of activation, the peak of IL-9 production, compared with 40% without TL1A (Figure 6A). TL1A also increased the percentage of IL-9-producing T cells after 4 days of activation (30% vs 8%), but after 6 days of activation, the percentage of IL-9-producing T cells fell to less than 3% with or without TL1A. IL-9 production was enhanced by TL1A at the RNA level as well, with a two-fold enhancement of IL-9 RNA at the peak of expression at day 3. IL-9 RNA expression steeply declined at day 4 with or without TL1A (Figure 6B). TL1A also greatly enhanced the amount of IL-9 in culture supernatants on days 3 and 4, with a ten-fold enhancement of IL-9 in the supernatant (Figure 6C). Interestingly, a significant fraction of T cells produced IL-10 after 6 days of culture, and TL1A almost completely extinguished IL-10 production by these ‘ex-Th9’ cells (Figure 6A). Measurement of IL-10 on days 3 and 4 by intracellular staining, in culture supernatants, and in RNA showed a smaller reduction with addition of TL1A, demonstrating that downregulation of IL-10 by TL1A occurred later after activation (Figures 6A-C). To assess the effects of TL1A on the pathogenic potential of Th9 cells, HEL-specific T cells differentiated under Th9 conditions for 3 or 4 days in the presence or absence of TL1A were transferred into mice expressing HEL in the eye. Examination of ocular pathology 7 days after transfer of Th9 cells revealed that TL1A added during T cell differentiation enhanced the ability of HEL-specific cells to induce ocular inflammation, even when the increased efficiency of Th9 generation in the presence of TL1A was taken into account by transferring proportionally fewer cells (Figure 6D). Histological changes in the recipients of Th9 cells included accumulation of inflammatory cells at the optic nerve head and limbus, as well as in the anterior chamber and throughout the vitreous. These changes were markedly more severe in the recipients of Th9 cells generated with the addition of TL1A (Figure 6E, left panels). Higher magnification showed the limbus and ciliary body area, with higher numbers of invading cells as compared with the control Th9 eye section (Figure 6E, right panels). In addition to intra-ocular changes, transferred Th9 cells uniquely invaded the conjunctiva, evoking changes reminiscent of allergic conjunctivitis. These changes were remarkably more severe in recipients of TL1A-stimulated donor cells (Figure 6F). To determine whether TL1A influences the proliferation of HEL-specific Th9 prior to their invasion of the recipient mouse eye, we monitored transferred cells on days 1, 2, 3, 4 and 7 post transfer in the spleens of recipient mice, where Th9 initially migrate and proliferate (45). We enumerated the percentage of TCR transgenic T cells in the CD4+ T cell pool using the anti-clonotypic monoclonal antibody 1G12. HEL-specific transgenic T cells exposed to TL1A during Th9 differentiation were present in greater numbers than control Th9 after transfer to HEL-expressing recipients (Figure 6G). Th9 cells generated in the presence of TL1A proliferated more rapidly than control Th9 cells (Figure 6H), demonstrating that while TL1A does not greatly enhance proliferation of Th9 during differentiation (as shown in Figure 3H), it does prime cells for more rapid expansion in vivo, which may contribute to their increased pathogenicity. To determine whether TL1A can also enhance the pathogenicity of Th9 at the site of inflammation we injected TL1A or PBS into the eyes of HEL-expressing recipient mice 2 and 3 days after transfer of HEL-specific Th9. Intra-ocular injection of TL1A into Th9 recipients slightly increased inflammation on day 4 post-transfer compared to those receiving PBS (Figure 6I). It is important to note that the ocular injection itself induced some inflammation and may have obscured the effects of TL1A, as evidenced by increased basal histological scores seen with Th9 transferred cells followed by intra-ocular PBS injection. Injection of TL1A alone induced very little ocular pathology. TL1A and DR3 have previously been found upregulated in experimental autoimmune uveitis, suggesting an endogenous role for this signaling pathway in ocular inflammation (58). To determine whether endogenous TL1A/DR3 interactions are important in vivo to enhance ocular pathology mediated by Th9, we treated recipient HEL-transgenic mice receiving HEL-specific Th9 cells with a blocking anti-TL1A antibody (13). Anti-TL1A significantly reduced ocular inflammation compared to control immunoglobulin (Figure 6J), with reduced cellular infiltration (Figure 6K). Anti-TL1A reduced the number of donor HEL-specific T cells in the eye (1.0% with isotype versus 0.6% with anti-TL1A), while the percentages of donor HEL-specific T cells in the spleen remained unchanged. Percentages of host cells in both organs only slightly decreased with anti-TL1A treatment (data not shown). We also examined the possibility that the effect of TL1A is related to the suppression of IL-10 production that we observed in vitro, but injection of IL-10 did not reduce ocular pathology induced by Th9 differentiated in the presence of TL1A (data not shown). Thus, TL1A appears to enhance Th9 pathogenicity independently of reducing IL-10 production. Taken together, these data show that in addition to improving the efficiency of Th9 differentiation, endogenously produced TL1A can increase pathology mediated by antigen-specific Th9. To determine whether endogenous TL1A can enhance the pathogenicity of Th9 cells in Ova-induced lung inflammation, we transferred WT or Tnfrsf25-/- OT-II Ova-specific T cells activated under Th9 conditions into congenic hosts. Recipient mice were then given two inhaled challenges of Ova and responses were measured 12 hours after the last challenge, the time point at which we observed maximal IL-9 expression (Figure 7A and 7B). At this time point, neutrophils predominated in the BAL, and there was a trend towards lower neutrophil counts in recipients of Tnfrsf25-/- OT-II cells (Figure 7C). Histological analysis revealed a significant reduction in pathology in recipients of Tnfrsf25-/- OT-II Th9 cells after challenge with Ova compared to mice receiving WT OT-II Th9 cells, including reduced perivascular and peribronchial infiltrates (Figure 7D). IL-9, IL-13 and IL-2 mRNA were significantly reduced in Ova-treated recipients of Tnfrsf25-/- OT-II T cells compared with mice that received WT OT-II T cells (Figure 7E). Expansion of donor OT-II Th9 was observed locally in the lung (Figure 7F) and the mediastinal lymph nodes but not in the spleen after Ova challenge (data not shown). Expansion of the total number of donor OT-II cells, as well as the number of OT-II cells producing IL-9, was significantly reduced when the cells lacked DR3, along with reduction in a much smaller number of OT-II cells producing IL-13 (Figure 7F). The mean fluorescence of IL-9 upon restimulation and intracellular cytokine staining was lower in Tnfrsf25-/- T cells (Figure 7F), suggesting that TL1A may potentiate their IL-9-producing phenotype. Because only T cells lacked DR3 in these experiments, these results demonstrate a requirement for interaction between endogenous TL1A and DR3 expressed on Th9 cells to promote the expansion and effector function of Th9 and immunopathology in this model of allergic lung disease. Discussion The importance of TL1A in animal models of allergic disease has been well characterized (10, 11, 14), but the mechanism by which TL1A promotes T cell-mediated immunopathology in allergic disease has remained elusive, as systemic Th2 polarization is unaffected in DR3-deficient mice despite reduced lung pathology. Here we find that TL1A acts on CD4+ T cells both to inhibit iTreg differentiation and to promote Th9 differentiation. The effects of TL1A are independent of IL-4 and STAT6, but require IL-2 acting through STAT5, as well as TGFβ signaling. TL1A also enhances the pathogenic potential of Th9, and TL1A-DR3 interactions in vivo are required for allergic lung inflammation and the accumulation of IL-9 secreting cells in the lung early in the inflammatory response. These findings provide a more specific mechanism by which TL1A can promote T cell mediated immunopathology and identify an alternate pathway for generation of IL-9-secreting T cells that is more efficient than TGFβ and IL-4 alone and generates higher numbers of pathogenic T cells. TL1A/DR3 interactions are not required for generation of Th1 or Th2 cells in vitro or during primary polarized T cell responses, and although TL1A has been shown to promote or inhibit Th17 differentiation depending on culture conditions, its interaction with DR3 is not necessary for Th17 formation (10, 11, 59). We here show that TL1A can divert T cells in iTreg differentiation cultures to become producers of IL-9 and enhance differentiation of Th9 in the presence of TGFβ and IL-4, conditions previously shown to bias naïve T cells towards IL-9 production (23, 24). However, TL1A can induce Th9 in the complete absence of STAT6 signaling, indicating that the action of TL1A is mechanistically distinct from Th9 differentiation induced by TGFβ and IL-4, which depends on STAT6 (24). Other cytokines, such as the IL-17 family member IL-25 (60) and the IL-1 family members IL-1β, IL-18, and IL-33 (28), have also been shown to enhance T cell production of IL-9, but unlike these cytokines, TL1A also enhanced differentiation of IL-9 producing cells from naïve precursors. More recently the TNF-family member OX40L was shown to limit iTreg differentiation and bias T cell differentiation towards IL-9 production (26). Like OX40L, TL1A-induced Th9 differentiation does not require the transcription factor PU.1, which independently enhances the generation of Th9 (18). However, unlike OX40L, TL1A promotion of Th9 differentiation is independent of IL-4 and STAT6. Rather than IL-4, we have found that IL-2 signaling through STAT5 is critical for the ability of TL1A to enhance Th9 differentiation. IL-2 has been previously identified as an enhancer of IL-9 production by T cells (53, 61). Mechanistically, TL1A enhances STAT5 activation and binding to the IL-9 promoter. Yao et al. have recently observed that TSLP can enhance IL-9 production by directly activating STAT5 in T cells (17), and STAT5 was necessary for increased IL-9 production by mice deficient in the SOCS protein CIS (62). The fact that TL1A was unable to induce IL-9 production under conditions with blocked IL-2 signaling suggests that DR3 signaling alone is not sufficient to promote IL-9. Rather, IL-2-STAT5 signaling is required in addition to the signal mediated by DR3 on the same cell to promote Th9 differentiation. TL1A enhancement of Th9 differentiation in the presence of IL-2 (iTreg cultures) was most efficient when T cells were activated in the presence of APC, suggesting that APC-derived costimulatory signals or cytokines are likely required for optimal effects of TL1A. As suggested by the global increase in cytokine production in the presence of APC (Supplemental Figure 3C), it is possible that the enhanced effect of TL1A is due to extra costimulatory capacity from B cells, macrophages, and dendritic cells present in these cultures. Future studies of separated APC subsets may help to identify specific cell types and factors responsible for this effect. DR3 signals similarly to TNFR1 through TRADD, RIP, and TRAF2, ultimately resulting in activation of MAP-kinases and NF-κB, or alternatively through FADD and Caspase-8 to activate apoptosis (63-67). TL1A signaling through DR3 is unlikely to directly induce STAT5 activation, but may enhance STAT5 activity indirectly. MEK and ERK kinases have been found to phosphorylate STAT5 (68, 69), and NF-kB and STAT proteins can bind coordinately to the promoters of a number of genes, including TLR2 (70), FcɛRII (71), and NOS2 (72). The IL-9 promoter contains binding sites for NF-κB and STAT5 making such transcriptional coordination possible (17, 73). Xiao et al. (26) have suggested that signaling through TRAF6 and the non-canonical NF-κB pathway is responsible for the ability of OX40 to bias T cell differentiation towards Th9. Interestingly, the same mechanism does not apply for DR3, suggesting that different TNF family members have unique ways to modulate T cell differentiation. In addition to enhancing the efficiency of Th9 differentiation, TL1A also enhances the pathogenicity of Th9 in animal models of ocular inflammation and allergic lung disease. In ocular inflammation, antigen-specific Th9 generated in the presence of TL1A induced significantly more pathology than conventional Th9, even when the increased percentage of IL-9-producing cells differentiated in the presence of TL1A was taken into account. In allergic lung disease, antigen-specific Th9 cells capable of DR3 signaling exhibited significantly greater pathogenicity than DR3-deficient Th9. Detection of TL1A-expressing cells in the perivascular region of inflamed lungs, as well as the reduction of ocular pathology seen after local injection of anti-TL1A, supports the notion that endogenous TL1A is likely the trigger for enhanced expansion of antigen-specific T cells which we observed in these models. We and others have found that TL1A costimulation of ILC2s plays a role in allergic lung disease (15, 16), and it has recently been shown that ILC2s can influence T cell responses in allergic diseases (49, 74). Thus, defective ILC2 activity in DR3-deficient mice may play a role in the reduced T cell accumulation and allergic response in the T-cell dependent allergic inflammation models that we studied. However, our results with transfer of DR3-deficient T cells show that TL1A-DR3 signaling on T cells plays an independent role in amplifying allergic lung disease mediated by Th9, as shown in this work, and as shown previously for Th2 cells (10). As IL-2 produced by T cells can promote ILC2 expansion (51, 74) and IL-9 can promote ILC2 survival (75), the reduced IL-2 and IL-9 production seen in the lungs of recipients of DR3-deficient Th9 suggests a mechanism by which T cells could reciprocally activate ILC2 in a DR3 dependent manner. Taken together, these results show that TL1A-DR3 interactions coordinately amplify both innate and adaptive immune responses in allergic disease. Our results predict that blockade of TL1A-DR3 interactions would be beneficial in diseases in which IL-9 plays an important role, including allergic asthma. Given the prominent role of STAT5 in mediating the effects of TL1A on Th9, blockade of STAT5 activity with small molecules directly targeting STAT5 or essential upstream kinases for STAT5 activation such as JAK3 (76) could also act to limit the pathological consequences of IL-9 production triggered by TL1A. Supplementary Material 1 We would like to thank Arian Laurence and Haydeé Ramos for discussions, provision of reagents and performing pilot experiments that led to this work, Jim Simone, Jeff Lay and Kevin Tinsley from the NIAMS flow cytometry core for cell sorting, Crystal Brobst-Wormell, Joe Woo and the NIAMS animal facility for excellent support in management of the mouse colonies. Figure 1 DR3 is required for optimal pathology and IL-9 and IL-13 production in T-cell driven allergic lung inflammation (A)Time line of antigen sensitization and challenge in this model; IP (intraperitoneal), IN (intranasal), and IT (intratracheal). (B) mRNA levels of IL-9 and IL-13 measured by qRT-PCR are shown for lung samples harvested 16 or 18 days after initial challenge with ovalbumin. Values were normalized to the average level of RNA in the PBS-treated wild-type (WT) mice at each time point (t-test statistics comparing Ova-treated WT and Ova-treated Tnfrsf25-/- mice, *p<0.05, **p<0.01). (C) Frequency of IL-9 and IL-13 producing CD4+CD44hi T cells in the lung are shown from cells harvested at day 16 from mice challenged as in (A). Right panels, absolute numbers with total number of CD4+CD44hi cells shown at the top. (D) PAS-stained sections of lungs harvested at day 16 from mice challenged as in (A); airways (aw), and blood vessels (bv) marked; scale bar = 50 μm. Bottom panel shows histopathology scores. (E) Airway resistance was measured in response to increasing doses of aerosolized methacholine in mice challenged as in (A). Results are a combination of 2 independent experiments (Two-way ANOVA comparing Ova-treated WT and Tnfrsf25-/- mice, **genotype term p<0.01). (F) Absolute numbers of eosinophils and neutrophils in the bronchoalveolar lavage (BAL) of mice treated as in (A) at day 16 are shown (Mann-Whitney test: *p<0.05, **p<0.01). (A-D and F) are representative of at least 2 independent experiments. Figure 2 DR3 is required for optimal pathology and IL-9 and IL-13 production by T cells in response to an inhaled allergen (A) WT or Tnfrsf25-/- mice were given intranasal PBS or papain according to the timeline. All mice were euthanized 12 hours after the last challenge. (B) Total cell counts, neutrophils and eosinophils present in the BAL. (C) Localization of TL1A expression in the lungs of WT mice treated with PBS or papain is demonstrated by immunohistochemistry. (D) Lung histology of WT and Tnfrsf25-/- mice (PAS staining, left panel). Scale bar= 50 μm. Histopathology scores of lung sections from the indicated groups of mice (right panel). (E) Airway resistance was measured in response to increasing doses of aerosolized methacholine in mice challenged as in (A) (Two-way ANOVA comparing papain-treated WT and Tnfrsf25-/- mice, ****genotype term p<0.0001). (F) IL-9 and IL-13 mRNA expression in the lung relative to β2m in each condition. Values were normalized to the average level of RNA in PBS-treated WT mice. (G) Frequency and total number of IL-9- or IL-13-producting CD45+TCRβ+CD4+CD44+ T cells, as well as total CD44high T cells, were measured by flow cytometry. In all panels, averages are indicated for each group and statistical significances between papain-treated WT and Tnfrsf25-/- mice are shown (Mann-Whitney, *p< 0.05, **p<0.01, ***p< 0.001, **** p<0.0001). (A-G) are pooled data from 2 experiments. Figure 3 TL1A enhances IL-9 production by T cells differentiated under Th9 and iTreg conditions (A) Naïve CD4+ T cells from WT and Tnfrsf25-/- mice were differentiated for 3 days under iTreg conditions with or without TL1A, in the absence or presence of T-depleted antigen presenting cells (APC). Intracellular staining shows a representative experiment, with a compilation of 4 independent experiments depicted below (Mann-Whitney test, *p<0.05). (B) Naïve CD4+ T cells from WT mice were differentiated into Th17 or iTreg with and without TL1A in the presence of APC. Intracellular staining is representative of 2 independent experiments. (C) Naïve CD4+ T cells isolated from WT and Tnfrsf25-/- mice were differentiated into Th9 in the absence or presence of APC with or without TL1A. Representative intracellular staining is shown with a compilation of 8 independent experiments (-APC) and 4 independent experiments (+APC) (Mann-Whitney test, *p<0.05, ***p<0.001). (D) Th9 differentiation was performed as in (C) in the absence of APC, with varying concentrations of TGFβ. Percentage of IL-9-producing cells by intracellular staining is representative of 2 independent experiments. (E) Th9 differentiation was performed as in (C) without APC, with varying amounts of TL1A. Percentage of IL-9-producing cells by intracellular staining is representative of 2 independent experiments. (F) DR3 surface expression vs. isotype control was measured by flow cytometry in naïve CD4+ T cell cultures differentiated in the absence of APC under various polarization conditions. Results are a compilation of 2 independent experiments; error bars represent +/- SEM. (G) DR3 surface expression was measured by flow cytometry in Th9 cultures with and without TL1A. Results are representative of 2 independent experiments. (H) Naïve WT CD4+ T cells were CFSE-labeled and differentiated under Th9 conditions as in (C) in the presence of APC: top row, CFSE dilution with the indicated expansion index; bottom row, CFSE dilution versus intracellular IL-9 with the percentage of cells producing IL-9 after each division shown above each box. Results are representative of 2 independent experiments. (I) Cytokines were measured in the supernatants from Th9 differentiation cultures without (top) or with T-depleted splenocytes (bottom). Heatmaps show TL1A-induced cytokine changes within each condition (red, increased; green, decreased; black, no change; grey, undetectable in one or both conditions). Results show average changes from two independent experiments. Only those cytokines detectable with and without TL1A in at least 4 polarization conditions are shown. Figure 4 TL1A-induced Th9 differentiation depends on IL-2 and STAT5 (A) Naïve CD4+ T cells were polarized under Th9 conditions with or without TL1A, with the indicated blocking antibodies or added cytokines; left and middle panels without APC, right panel with APC. IL-9 production assayed by flow cytometry is representative of 2 independent experiments. (B) WT and STAT6-deficient (Stat6-/-) naïve CD4+ T cells were differentiated in the presence of APC under Th9 and iTreg conditions with and without TL1A. Intracellular staining is representative of 3 independent experiments. (C) Control Stat5fl/fl and STAT5-deficient (Stat5CD4-/-) naïve CD4+ T cells were differentiated in the presence of APC under Th9 conditions with and without TL1A. Intracellular staining is representative of 2 independent experiments. (D) WT naïve CD4+ T cells were transfected with a constitutively active caSTAT5-retrovirus or control virus and differentiated under Th0, Th17 in the presence of IL-2, and regular Th17 (blocking IL-2) conditions. Intracellular staining is representative of 2 independent experiments. Figure 5 Cell-intrinsic effects of TL1A in Th9 differentiation are independent of PU.1 and Traf6 (A) Naïve CD4+ T cells from CD45.1+ congenic WT mice were mixed with an equal number of naïve CD4+ T cells from CD45.1- Tnfrsf25-/- mice and polarized under Th9 conditions without APC, or under iTreg conditions with APC, with and without TL1A. IL-9 intracellular staining versus CD45.1 surface staining with the percentages of CD45.1+ and CD45.1- cells expressing IL-9 is representative of 2 independent experiments. (B) WT and PU.1-deficient (Sfpi1lck-/-) naïve CD4+ T cells were differentiated without APC under Th9 conditions with and without TL1A. Left panel shows intracellular staining. Right panel shows IL-9 production in the supernatant of restimulated cells 5 days after polarization by ELISA. Results are representative of 2 independent experiments. (C) Naïve CD4+ T cells were differentiated without APC under Th9 conditions for 5 days and analyzed by ChIP assay for PU.1 binding to the IL-9 promoter. Error bars represent combined results from 3 mice, representative of 2 independent experiments. (D) As (C), cells were analyzed by ChIP assay for IRF4 binding to the IL-9 promoter. Error bars represent combined results from 4 mice in two independent experiments. (E) WT naïve CD4+ T cells were differentiated without APC under Th9 conditions for 3 days and stained for intracellular phospho-STAT5 (left panel), representative of 2 independent experiments. Cells were also analyzed by ChIP assay for STAT5 binding to the IL-9 promoter (right panels). Error bars represent SEM of technical qPCR replicates. Data is representative of 2 independent experiments. (F) WT and Traf6-/- naïve CD4+ T cells were polarized under Th9 conditions in the absence of APC, with or without TL1A. Intracellular staining is representative of 2 independent experiments. (G) WT and Stat6-/- naïve CD4+ T cells were polarized in the presence of APC under Th9 conditions with or without TL1A or OX40L. Intracellular staining is representative of 3 independent experiments. Figure 6 TL1A enhances ocular inflammation induced by Th9 (A) Naïve anti-HEL CD4+ TCR transgenic T cells were activated with HEL+APC under Th9 conditions, with or without TL1A. Cells were analyzed for intracellular IL-9 and IL-10 production at the indicated time points (Student's T test: * p<0.05, **p<0.01); combined results from 3 independent experiments. (B) IL-9 and IL-10 mRNA was measured relative to β-actin in cultures described in (A); combined results from 2 independent experiments. (C) Cytokine production was measured in the supernatants of cultures described in (A); combined results from 3 independent experiments. (D) Th9 cells generated as in (A) (5 million or 1.7 million to compensate for increased percentage of cells expressing IL-9 in the presence of TL1A) were adoptively transferred into syngeneic recipients expressing HEL in the lens. Inflammatory changes in recipients of Th9 cells activated for 3 or 4 days in culture were evaluated by histological analysis 7 days after transfer (Mann-Whitney *p<0.05, ****p<0.0005); combined results from 3 independent experiments. (E) H&E stained eye sections are examples of mice described in (D) receiving 5 million Th9 cells, differentiated for 3 days, and a mouse with no treatment (naïve); scale bars: 400 μM for low magnification and 50 μM for high magnification enlarged from the indicated boxes; representative of 3 independent experiments. (F) Histological sections showing conjunctivitis in recipients of Th9 and TL1A-stimulated Th9 cells are representative of 3 independent experiments. (G) As in (D), 5 million transgenic T cells were transferred into recipients after 3 days of differentiation. Recipient spleens were collected at the indicated time points and donor cells identified by flow cytometry with an anti-clonotypic monoclonal antibody (1G12), specific to the donor cell TCR. Graph shows mean +/- SEM of transgenic T cell yields from 2 independent experiments. (H) Th9 cells transferred as in (D) were labeled with Cell Proliferation Dye eFluor 670 and their division in vivo determined by dye dilution at the indicated times after transfer: black line, Th9 control; gray line and shaded graph, Th9+TL1A; representative of 2 independent experiments. (I) Naïve mice or recipients of Th9 transfer as in (D) were injected with PBS or TL1A intraocularly 2 and 3 days after transfer. Eye pathology was examined at day 4 (Mann-Whitney *p<0.05, **p<0.01); combined results from at least 2 independent experiments per condition. (J) Recipients of Th9 transfer as in (D) were administered anti-TL1A antibody or hamster Ig control on days -1 and 3. Recipient eyes were analyzed on day 7 (Mann-Whitney *p<0.05); combined results from 3 independent experiments. (K) Representative examples of ocular histological sections from transfer experiments in (J). Figure 7 TL1A enhances allergic lung inflammation induced by antigen-specific Th9 (A) Cytokine expression was measured in restimulated Th9-polarized OT-II T cells or Tnfrsf25-/- OT-II T cells. (B) Timeline of transfer experiment: 106 T cells were transferred into CD45.1+ congenic recipient mice and the recipient mice challenged with IT Ova 24 hours after transfer and IN Ova 48 hours after transfer. Control recipient mice were challenged with PBS. All mice were euthanized 12 hours after the last challenge. (C) Cell counts of neutrophils and eosinophils present in the BAL from each mouse are shown. (D) Representative lung histology of each group (left panel; H&E stain, 20x original magnification, black bar represents 50 μm), and summary of lung histopathology scores from the same experiment (right panel) are shown. (E) IL-9, IL-13 and IL-2 mRNA expression in the lung relative to β2M in each condition was measured by qRT-PCR. (F) Yields of total CD45.2+ donor T cells and IL-9- and IL-13-producing donor T cells per lung (left panel) are shown with representative FACS plots of IL-9 and IL-13 expression by restimulated T cells from each group (right panel). In all graphs, each point represents one mouse with the mean as a black line and significances of differences between Ova-treated groups indicated (Mann-Whitney *p<0.05, **p<0.01). A-D and F are representative of 2 independent experiments. E shows compiled results of 2 independent experiments. 1 This work was supported in part by the intramural research programs of NIAMS and NEI, and the NIH-Oxford-Cambridge and NIH-Wellcome Trust Ph.D. programs. 1 Buchan SL Taraban VY Slebioda TJ James S Cunningham AF Al-Shamkhani A 2012 Death receptor 3 is essential for generating optimal protective CD4(+) T-cell immunity against Salmonella European journal of immunology 42 580 588 22259035 2 Twohig JP Marsden M Cuff SM Ferdinand JR Gallimore AM Perks WV Al-Shamkhani A Humphreys IR Wang EC 2012 The death receptor 3/TL1A pathway is essential for efficient development of antiviral CD4(+) and CD8(+) T-cell immunity FASEB journal : official publication of the Federation of American Societies for Experimental Biology 26 3575 3586 22593543 3 Cassatella MA Pereira-da-Silva G Tinazzi I Facchetti F Scapini P Calzetti F Tamassia N Wei P Nardelli B Roschke V Vecchi A Mantovani A Bambara LM Edwards SW Carletto A 2007 Soluble TNF-like cytokine (TL1A) production by immune complexes stimulated monocytes in rheumatoid arthritis Journal of immunology 178 7325 7333 4 Bamias G Siakavellas SI Stamatelopoulos KS Chryssochoou E Papamichael C Sfikakis PP 2008 Circulating levels of TNF-like cytokine 1A (TL1A) and its decoy receptor 3 (DcR3) in rheumatoid arthritis Clinical immunology 129 249 255 18757243 5 Bamias G Martin C 3rd Marini M Hoang S Mishina M Ross WG Sachedina MA Friel CM Mize J Bickston SJ Pizarro TT Wei P Cominelli F 2003 Expression, localization, and functional activity of TL1A, a novel Th1-polarizing cytokine in inflammatory bowel disease Journal of immunology 171 4868 4874 6 McGovern DP Gardet A Torkvist L Goyette P Essers J Taylor KD Neale BM Ong RT Lagace C Li C Green T Stevens CR Beauchamp C Fleshner PR Carlson M D'Amato M Halfvarson J Hibberd ML Lordal M Padyukov L Andriulli A Colombo E Latiano A Palmieri O Bernard EJ Deslandres C Hommes DW de Jong DJ Stokkers PC Weersma RK Consortium NIG Sharma Y Silverberg MS Cho JH Wu J Roeder K Brant SR Schumm LP Duerr RH Dubinsky MC Glazer NL Haritunians T Ippoliti A Melmed GY Siscovick DS Vasiliauskas EA Targan SR Annese V Wijmenga C Pettersson S Rotter JI Xavier RJ Daly MJ Rioux JD Seielstad M 2010 Genome-wide association identifies multiple ulcerative colitis susceptibility loci Nature genetics 42 332 337 20228799 7 Jostins L Ripke S Weersma RK Duerr RH McGovern DP Hui KY Lee JC Schumm LP Sharma Y Anderson CA Essers J Mitrovic M Ning K Cleynen I Theatre E Spain SL Raychaudhuri S Goyette P Wei Z Abraham C Achkar JP Ahmad T Amininejad L Ananthakrishnan AN Andersen V Andrews JM Baidoo L Balschun T Bampton PA Bitton A Boucher G Brand S Buning C Cohain A Cichon S D'Amato M De Jong D Devaney KL Dubinsky M Edwards C Ellinghaus D Ferguson LR Franchimont D Fransen K Gearry R Georges M Gieger C Glas J Haritunians T Hart A Hawkey C Hedl M Hu X Karlsen TH Kupcinskas L Kugathasan S Latiano A Laukens D Lawrance IC Lees CW Louis E Mahy G Mansfield J Morgan AR Mowat C Newman W Palmieri O Ponsioen CY Potocnik U Prescott NJ Regueiro M Rotter JI Russell RK Sanderson JD Sans M Satsangi J Schreiber S Simms LA Sventoraityte J Targan SR Taylor KD Tremelling M Verspaget HW De Vos M Wijmenga C Wilson DC Winkelmann J Xavier RJ Zeissig S Zhang B Zhang CK Zhao H International IBDGC Silverberg MS Annese V Hakonarson H Brant SR Radford-Smith G Mathew CG Rioux JD Schadt EE Daly MJ Franke A Parkes M Vermeire S Barrett JC Cho JH 2012 Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease Nature 491 119 124 23128233 8 Osawa K Takami N Shiozawa K Hashiramoto A Shiozawa S 2004 Death receptor 3 (DR3) gene duplication in a chromosome region 1p36.3: gene duplication is more prevalent in rheumatoid arthritis Genes and immunity 5 439 443 15241467 9 Bull MJ Williams AS Mecklenburgh Z Calder CJ Twohig JP Elford C Evans BA Rowley TF Slebioda TJ Taraban VY Al-Shamkhani A Wang EC 2008 The Death Receptor 3-TNF-like protein 1A pathway drives adverse bone pathology in inflammatory arthritis The Journal of experimental medicine 205 2457 2464 18824582 10 Meylan F Davidson TS Kahle E Kinder M Acharya K Jankovic D Bundoc V Hodges M Shevach EM Keane-Myers A Wang EC Siegel RM 2008 The TNF-family receptor DR3 is essential for diverse T cell-mediated inflammatory diseases Immunity 29 79 89 18571443 11 Pappu BP Borodovsky A Zheng TS Yang X Wu P Dong X Weng S Browning B Scott ML Ma L Su L Tian Q Schneider P Flavell RA Dong C Burkly LC 2008 TL1A-DR3 interaction regulates Th17 cell function and Th17-mediated autoimmune disease The Journal of experimental medicine 205 1049 1062 18411337 12 Takedatsu H Michelsen KS Wei B Landers CJ Thomas LS Dhall D Braun J Targan SR 2008 TL1A (TNFSF15) regulates the development of chronic colitis by modulating both T-helper 1 and T-helper 17 activation Gastroenterology 135 552 567 18598698 13 Meylan F Song YJ Fuss I Villarreal S Kahle E Malm IJ Acharya K Ramos HL Lo L Mentink-Kane MM Wynn TA Migone TS Strober W Siegel RM 2011 The TNF-family cytokine TL1A drives IL-13-dependent small intestinal inflammation Mucosal immunology 4 172 185 20980995 14 Fang L Adkins B Deyev V Podack ER 2008 Essential role of TNF receptor superfamily 25 (TNFRSF25) in the development of allergic lung inflammation The Journal of experimental medicine 205 1037 1048 18411341 15 Meylan F Hawley ET Barron L Barlow JL Penumetcha P Pelletier M Sciume G Richard AC Hayes ET Gomez-Rodriguez J Chen X Paul WE Wynn TA McKenzie AN Siegel RM 2013 The TNF-family cytokine TL1A promotes allergic immunopathology through group 2 innate lymphoid cells Mucosal immunology 16 Yu X Pappu R Ramirez-Carrozzi V Ota N Caplazi P Zhang J Yan D Xu M Lee WP Grogan JL 2014 TNF superfamily member TL1A elicits type 2 innate lymphoid cells at mucosal barriers Mucosal immunology 7 730 740 24220298 17 Yao W Zhang Y Jabeen R Nguyen ET Wilkes DS Tepper RS Kaplan MH Zhou B 2013 Interleukin-9 is required for allergic airway inflammation mediated by the cytokine TSLP Immunity 38 360 372 23376058 18 Chang HC Sehra S Goswami R Yao W Yu Q Stritesky GL Jabeen R McKinley C Ahyi AN Han L Nguyen ET Robertson MJ Perumal NB Tepper RS Nutt SL Kaplan MH 2010 The transcription factor PU.1 is required for the development of IL-9-producing T cells and allergic inflammation Nature immunology 11 527 534 20431622 19 Kearley J Erjefalt JS Andersson C Benjamin E Jones CP Robichaud A Pegorier S Brewah Y Burwell TJ Bjermer L Kiener PA Kolbeck R Lloyd CM Coyle AJ Humbles AA 2011 IL-9 governs allergen-induced mast cell numbers in the lung and chronic remodeling of the airways American journal of respiratory and critical care medicine 183 865 875 20971830 20 Kung TT Luo B Crawley Y Garlisi CG Devito K Minnicozzi M Egan RW Kreutner W Chapman RW 2001 Effect of anti-mIL-9 antibody on the development of pulmonary inflammation and airway hyperresponsiveness in allergic mice American journal of respiratory cell and molecular biology 25 600 605 11713102 21 Cheng G Arima M Honda K Hirata H Eda F Yoshida N Fukushima F Ishii Y Fukuda T 2002 Anti-interleukin-9 antibody treatment inhibits airway inflammation and hyperreactivity in mouse asthma model American journal of respiratory and critical care medicine 166 409 416 12153980 22 Kaplan MH 2013 Th9 cells: differentiation and disease Immunological reviews 252 104 115 23405898 23 Dardalhon V Awasthi A Kwon H Galileos G Gao W Sobel RA Mitsdoerffer M Strom TB Elyaman W Ho IC Khoury S Oukka M Kuchroo VK 2008 IL-4 inhibits TGF-beta-induced Foxp3+ T cells and, together with TGF-beta, generates IL-9+ IL-10+ Foxp3(-) effector T cells Nature immunology 9 1347 1355 18997793 24 Veldhoen M Uyttenhove C van Snick J Helmby H Westendorf A Buer J Martin B Wilhelm C Stockinger B 2008 Transforming growth factor-beta ‘reprograms’ the differentiation of T helper 2 cells and promotes an interleukin 9-producing subset Nature immunology 9 1341 1346 18931678 25 Goswami R Jabeen R Yagi R Pham D Zhu J Goenka S Kaplan MH 2012 STAT6-dependent regulation of Th9 development Journal of immunology 188 968 975 26 Xiao X Balasubramanian S Liu W Chu X Wang H Taparowsky EJ Fu YX Choi Y Walsh MC Li XC 2012 OX40 signaling favors the induction of T(H)9 cells and airway inflammation Nature immunology 13 981 990 22842344 27 Angkasekwinai P Srimanote P Wang YH Pootong A Sakolvaree Y Pattanapanyasat K Chaicumpa W Chaiyaroj S Dong C 2013 Interleukin-25 (IL-25) promotes efficient protective immunity against Trichinella spiralis infection by enhancing the antigen-specific IL-9 response Infection and immunity 81 3731 3741 23897610 28 Uyttenhove C Brombacher F Van Snick J 2010 TGF-beta interactions with IL-1 family members trigger IL-4-independent IL-9 production by mouse CD4(+) T cells European journal of immunology 40 2230 2235 20540113 29 Elyaman W Bassil R Bradshaw EM Orent W Lahoud Y Zhu B Radtke F Yagita H Khoury SJ 2012 Notch receptors and Smad3 signaling cooperate in the induction of interleukin-9-producing T cells Immunity 36 623 634 22503540 30 Mikami N Miyagi Y Sueda K Takatsuji M Fukada S Yamamoto H Tsujikawa K 2013 Calcitonin gene-related peptide and cyclic adenosine 5′-monophosphate/protein kinase A pathway promote IL-9 production in Th9 differentiation process Journal of immunology 190 4046 4055 31 Li H Edin ML Bradbury JA Graves JP DeGraff LM Gruzdev A Cheng J Dackor RT Wang PM Bortner CD Garantziotis S Jetten AM Zeldin DC 2013 Cyclooxygenase-2 inhibits T helper cell type 9 differentiation during allergic lung inflammation via down-regulation of IL-17RB American journal of respiratory and critical care medicine 187 812 822 23449692 32 Kerzerho J Maazi H Speak AO Szely N Lombardi V Khoo B Geryak S Lam J Soroosh P Van Snick J Akbari O 2013 Programmed cell death ligand 2 regulates TH9 differentiation and induction of chronic airway hyperreactivity The Journal of allergy and clinical immunology 131 1048 1057 1057 e1041 1042 23174661 33 Palmer MT Lee YK Maynard CL Oliver JR Bikle DD Jetten AM Weaver CT 2011 Lineage-specific effects of 1,25-dihydroxyvitamin D(3) on the development of effector CD4 T cells The Journal of biological chemistry 286 997 1004 21047796 34 Tan C Gery I 2012 The unique features of Th9 cells and their products Critical reviews in immunology 32 1 10 22428852 35 Li H Nourbakhsh B Ciric B Zhang GX Rostami A 2010 Neutralization of IL-9 ameliorates experimental autoimmune encephalomyelitis by decreasing the effector T cell population Journal of immunology 185 4095 4100 36 Nowak EC Weaver CT Turner H Begum-Haque S Becher B Schreiner B Coyle AJ Kasper LH Noelle RJ 2009 IL-9 as a mediator of Th17-driven inflammatory disease The Journal of experimental medicine 206 1653 1660 19596803 37 Licona-Limon P Henao-Mejia J Temann AU Gagliani N Licona-Limon I Ishigame H Hao L Herbert DR Flavell RA 2013 Th9 Cells Drive Host Immunity against Gastrointestinal Worm Infection Immunity 39 744 757 24138883 38 Gerlach K Hwang Y Nikolaev A Atreya R Dornhoff H Steiner S Lehr HA Wirtz S Vieth M Waisman A Rosenbauer F McKenzie AN Weigmann B Neurath MF 2014 TH9 cells that express the transcription factor PU.1 drive T cell-mediated colitis via IL-9 receptor signaling in intestinal epithelial cells Nature immunology 15 676 686 24908389 39 Hauber HP Bergeron C Hamid Q 2004 IL-9 in allergic inflammation International archives of allergy and immunology 134 79 87 15133304 40 Goswami R Kaplan MH 2011 A brief history of IL-9 Journal of immunology 186 3283 3288 41 Steenwinckel V Louahed J Orabona C Huaux F Warnier G McKenzie A Lison D Levitt R Renauld JC 2007 IL-13 mediates in vivo IL-9 activities on lung epithelial cells but not on hematopoietic cells Journal of immunology 178 3244 3251 42 Wang EC Thern A Denzel A Kitson J Farrow SN Owen MJ 2001 DR3 regulates negative selection during thymocyte development Molecular and cellular biology 21 3451 3461 11313471 43 Shi G Cox CA Vistica BP Tan C Wawrousek EF Gery I 2008 Phenotype switching by inflammation-inducing polarized Th17 cells, but not by Th1 cells Journal of immunology 181 7205 7213 44 King CG Kobayashi T Cejas PJ Kim T Yoon K Kim GK Chiffoleau E Hickman SP Walsh PT Turka LA Choi Y 2006 TRAF6 is a T cell-intrinsic negative regulator required for the maintenance of immune homeostasis Nature medicine 12 1088 1092 45 Tan C Aziz MK Lovaas JD Vistica BP Shi G Wawrousek EF Gery I 2010 Antigen-specific Th9 cells exhibit uniqueness in their kinetics of cytokine production and short retention at the inflammatory site Journal of immunology 185 6795 6801 46 Kim SJ Zhang M Vistica BP Chan CC Shen DF Wawrousek EF Gery I 2002 Induction of ocular inflammation by T-helper lymphocytes type 2 Investigative ophthalmology & visual science 43 758 765 11867595 47 McConchie BW Norris HH Bundoc VG Trivedi S Boesen A Urban JF Jr Keane-Myers AM 2006 Ascaris suum-derived products suppress mucosal allergic inflammation in an interleukin-10-independent manner via interference with dendritic cell function Infection and immunity 74 6632 6641 16966410 48 Yang XP Ghoreschi K Steward-Tharp SM Rodriguez-Canales J Zhu J Grainger JR Hirahara K Sun HW Wei L Vahedi G Kanno Y O'Shea JJ Laurence A 2011 Opposing regulation of the locus encoding IL-17 through direct, reciprocal actions of STAT3 and STAT5 Nature immunology 12 247 254 21278738 49 Halim TY Steer CA Matha L Gold MJ Martinez-Gonzalez I McNagny KM McKenzie AN Takei F 2014 Group 2 innate lymphoid cells are critical for the initiation of adaptive T helper 2 cell-mediated allergic lung inflammation Immunity 40 425 435 24613091 50 Kamijo S Takeda H Tokura T Suzuki M Inui K Hara M Matsuda H Matsuda A Oboki K Ohno T Saito H Nakae S Sudo K Suto H Ichikawa S Ogawa H Okumura K Takai T 2013 IL-33-mediated innate response and adaptive immune cells contribute to maximum responses of protease allergen-induced allergic airway inflammation Journal of immunology 190 4489 4499 51 Wilhelm C Hirota K Stieglitz B Van Snick J Tolaini M Lahl K Sparwasser T Helmby H Stockinger B 2011 An IL-9 fate reporter demonstrates the induction of an innate IL-9 response in lung inflammation Nature immunology 12 1071 1077 21983833 52 Papadakis KA Prehn JL Landers C Han Q Luo X Cha SC Wei P Targan SR 2004 TL1A synergizes with IL-12 and IL-18 to enhance IFN-gamma production in human T cells and NK cells Journal of immunology 172 7002 7007 53 Schmitt E Germann T Goedert S Hoehn P Huels C Koelsch S Kuhn R Muller W Palm N Rude E 1994 IL-9 production of naive CD4+ T cells depends on IL-2, is synergistically enhanced by a combination of TGF-beta and IL-4, and is inhibited by IFN-gamma Journal of immunology 153 3989 3996 54 Meylan F Richard AC Siegel RM 2011 TL1A and DR3, a TNF family ligand-receptor pair that promotes lymphocyte costimulation, mucosal hyperplasia, and autoimmune inflammation Immunological reviews 244 188 196 22017439 55 Yao Z Cui Y Watford WT Bream JH Yamaoka K Hissong BD Li D Durum SK Jiang Q Bhandoola A Hennighausen L O'Shea JJ 2006 Stat5a/b are essential for normal lymphoid development and differentiation Proceedings of the National Academy of Sciences of the United States of America 103 1000 1005 16418296 56 Tamiya T Ichiyama K Kotani H Fukaya T Sekiya T Shichita T Honma K Yui K Matsuyama T Nakao T Fukuyama S Inoue H Nomura M Yoshimura A 2013 Smad2/3 and IRF4 play a cooperative role in IL-9-producing T cell induction Journal of immunology 191 2360 2371 57 Staudt V Bothur E Klein M Lingnau K Reuter S Grebe N Gerlitzki B Hoffmann M Ulges A Taube C Dehzad N Becker M Stassen M Steinborn A Lohoff M Schild H Schmitt E Bopp T 2010 Interferon-regulatory factor 4 is essential for the developmental program of T helper 9 cells Immunity 33 192 202 20674401 58 Qin T 2011 Upregulation of DR3 expression in CD4(+) T cells promotes secretion of IL-17 in experimental autoimmune uveitis Molecular vision 17 3486 3493 22219644 59 Jones GW Stumhofer JS Foster T Twohig JP Hertzog P Topley N Williams AS Hunter CA Jenkins BJ Wang EC Jones SA 2011 Naive and activated T cells display differential responsiveness to TL1A that affects Th17 generation, maintenance, and proliferation FASEB journal : official publication of the Federation of American Societies for Experimental Biology 25 409 419 20826539 60 Angkasekwinai P Chang SH Thapa M Watarai H Dong C 2010 Regulation of IL-9 expression by IL-25 signaling Nature immunology 11 250 256 20154671 61 Monteyne P Renauld JC Van Broeck J Dunne DW Brombacher F Coutelier JP 1997 IL-4-independent regulation of in vivo IL-9 expression Journal of immunology 159 2616 2623 62 Yang XO Zhang H Kim BS Niu X Peng J Chen Y Kerketta R Lee YH Chang SH Corry DB Wang D Watowich SS Dong C 2013 The signaling suppressor CIS controls proallergic T cell development and allergic airway inflammation Nature immunology 14 732 740 23727894 63 Kitson J Raven T Jiang YP Goeddel DV Giles KM Pun KT Grinham CJ Brown R Farrow SN 1996 A death-domain-containing receptor that mediates apoptosis Nature 384 372 375 8934525 64 Bodmer JL Burns K Schneider P Hofmann K Steiner V Thome M Bornand T Hahne M Schroter M Becker K Wilson A French LE Browning JL MacDonald HR Tschopp J 1997 TRAMP, a novel apoptosis-mediating receptor with sequence homology to tumor necrosis factor receptor 1 and Fas(Apo-1/CD95) Immunity 6 79 88 9052839 65 Chinnaiyan AM O'Rourke K Yu GL Lyons RH Garg M Duan DR Xing L Gentz R Ni J Dixit VM 1996 Signal transduction by DR3, a death domain-containing receptor related to TNFR-1 and CD95 Science 274 990 992 8875942 66 Screaton GR Xu XN Olsen AL Cowper AE Tan R McMichael AJ Bell JI 1997 LARD: a new lymphoid-specific death domain containing receptor regulated by alternative pre-mRNA splicing Proceedings of the National Academy of Sciences of the United States of America 94 4615 4619 9114039 67 Pobezinskaya YL Choksi S Morgan MJ Cao X Liu ZG 2011 The adaptor protein TRADD is essential for TNF-like ligand 1A/death receptor 3 signaling Journal of immunology 186 5212 5216 68 Maki K Ikuta K 2008 MEK1/2 induces STAT5-mediated germline transcription of the TCRgamma locus in response to IL-7R signaling Journal of immunology 181 494 502 69 Pircher TJ Petersen H Gustafsson JA Haldosen LA 1999 Extracellular signal-regulated kinase (ERK) interacts with signal transducer and activator of transcription (STAT) 5a Molecular endocrinology 13 555 565 10194762 70 Musikacharoen T Matsuguchi T Kikuchi T Yoshikai Y 2001 NF-kappa B and STAT5 play important roles in the regulation of mouse Toll-like receptor 2 gene expression Journal of immunology 166 4516 4524 71 Tinnell SB Jacobs-Helber SM Sterneck E Sawyer ST Conrad DH 1998 STAT6, NF-kappaB and C/EBP in CD23 expression and IgE production International immunology 10 1529 1538 9796920 72 Farlik M Reutterer B Schindler C Greten F Vogl C Muller M Decker T 2010 Nonconventional initiation complex assembly by STAT and NF-kappaB transcription factors regulates nitric oxide synthase expression Immunity 33 25 34 20637660 73 Stassen M Muller C Arnold M Hultner L Klein-Hessling S Neudorfl C Reineke T Serfling E Schmitt E 2001 IL-9 and IL-13 production by activated mast cells is strongly enhanced in the presence of lipopolysaccharide: NF-kappa B is decisively involved in the expression of IL-9 Journal of immunology 166 4391 4398 74 Mirchandani AS Besnard AG Yip E Scott C Bain CC Cerovic V Salmond RJ Liew FY 2014 Type 2 innate lymphoid cells drive CD4+ Th2 cell responses Journal of immunology 192 2442 2448 75 Turner JE Morrison PJ Wilhelm C Wilson M Ahlfors H Renauld JC Panzer U Helmby H Stockinger B 2013 IL-9-mediated survival of type 2 innate lymphoid cells promotes damage control in helminth-induced lung inflammation The Journal of experimental medicine 210 2951 2965 24249111 76 Kontzias A Kotlyar A Laurence A Changelian P O'Shea JJ 2012 Jakinibs: a new class of kinase inhibitors in cancer and autoimmune disease Current opinion in pharmacology 12 464 470 22819198

PMC005xxxxxx/PMC5112770.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101525337 37346 ACS Chem Neurosci ACS Chem Neurosci ACS chemical neuroscience 1948-7193 25822288 5112770 10.1021/acschemneuro.5b00013 NIHMS798448 Article Azaphilones inhibit tau aggregation and dissolve tau aggregates in vitro Paranjape Smita R. 1 Riley Andrew P. 2 Somoza Amber D. 3 Oakley C. Elizabeth 1 Wang Clay C. C. 34 Prisinzano Thomas E. 5 Oakley Berl R. 1 Gamblin T. Chris 1 1 Department of Molecular Biosciences, Univ. of Kansas, Lawrence, KS 2 Department of Chemistry, University of Kansas, Lawrence, KS 3 Department of Chemistry, University of Southern California, Los Angeles, CA 4 Department of Pharmacology and Pharmaceutical Sciences, School of Pharmacy, University of Southern California, Los Angeles, CA 5 Department of Medicinal Chemistry, University of Kansas, Lawrence, KS 9 11 2016 15 4 2015 20 5 2015 17 11 2016 6 5 751760 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The aggregation of the microtubule-associated protein tau is a seminal event in many neurodegenerative diseases, including Alzheimer’s disease. The inhibition or reversal of tau aggregation is therefore a potential therapeutic strategy for these diseases. Fungal natural products have proven to be a rich source of useful compounds having wide varieties of biological activities. We have previously screened Aspergillus nidulans secondary metabolites for their ability to inhibit tau aggregation in vitro using an arachidonic acid polymerization protocol. One aggregation inhibitor identified was asperbenzaldehyde, an intermediate in azaphilone biosynthesis. We therefore tested 11 azaphilone derivatives to determine their tau assembly inhibition properties in vitro. All compounds tested inhibited tau filament assembly to some extent, while four of the 11 compounds had the advantageous property of disassembling preformed tau aggregates in a dose-dependent fashion. The addition of these compounds to the tau aggregates reduced both the total length and numbers of tau polymers. The most potent compounds were tested in in vitro reactions to determine whether they interfere with tau’s normal function of stabilizing microtubules (MTs). We found that they did not completely inhibit MT assembly in the presence of tau. These derivatives are very promising lead compounds for tau aggregation inhibitors and, more excitingly, for compounds that can disassemble pre-existing tau filaments. They also represent a new class of anti-tau aggregation compounds with a novel structural scaffold. Graphical Abstract Tau microtubule-associated protein aggregation inhibitor Alzheimer’s disease azaphilone natural products Aspergillus Aspergillus nidulans Introduction Alzheimer’s disease (AD) is the most common form of dementia. This devastating condition is made worse by the relative lack of therapies available for its treatment. Current therapeutics largely target cholinergic pathways and do little to slow or reverse the accumulation of aggregates of either beta amyloid or the microtubule-associated protein tau into senile plaques or neurofibrillary tangles respectively. The location and amount of tau aggregation into neurofibrillary tangles directly correlates with the type and severity of the disease progression1. Therefore, there is great interest in identifying small molecules that may inhibit or reverse tau aggregation. Recently a tau aggregation inhibitor, a stable, reduced form of methylthioninium chloride, has reached phase 3 clinical trials2, validating the potential of tau aggregation as a target, but there is certainly a need for additional lead anti-tau aggregation compounds for further development into therapeutics. An ideal tau aggregation inhibitor should inhibit the assembly of tau aggregates and disassemble pre-formed tau aggregates as well. Inhibitors of tau aggregation also should not impair the normal function of tau to bind and stabilize microtubules. Previously identified tau aggregation inhibitors, including molecules belonging to the anthraquinone class, have had widely diverse structures, with fused ring structures being a commonly occurring structural motif3. Recent emphasis has been placed on identifying natural products with novel scaffolds that may have useful properties for the treatment of AD by inhibiting tau aggregation4, 5. Fungi have historically been a good source of secondary metabolites, which have useful pharmacological properties such as antibiotics, immunosuppressants, and cholesterol-lowering drugs, among others6. We have identified numerous biosynthetic pathways in Aspergillus nidulans that lead to production of a wide range of secondary metabolites6–9. In a previous study, we tested several A. nidulans secondary metabolites for their ability to inhibit tau aggregation in vitro and found that several were active inhibitors at micromolar concentrations, although they did not have tau disaggregation properties10. Among these, two, ω-dihydroxyemodin and asperthecin, belonged to the anthraquinone class of compounds, a class that includes compounds shown to inhibit tau aggregation. A third compound, asperbenzaldehyde, however, was structurally distinct from previously identified tau aggregation inhibitors. Asperbenzaldehyde is also interesting in that it is an intermediate in the biological synthesis of azaphilone compounds11. Azaphilones are known to exhibit a great variety of biologically important activities including inhibitions of gp120–CD4 binding12 and heat shock protein 90 (Hsp90)13–15, among others. Several azaphilones have been shown to have lipoxygenase inhibitor activity11. Inhibition of lipoxygenases may help reduce fatty acid metabolite levels that are elevated in AD16. Azaphilones, including lipoxygenase-inhibiting azaphilones, can be obtained from asperbenzaldehyde using a 2–3 step semisynthetic route11. We therefore sought to determine whether azaphilones derived from asperbenzaldehyde inhibit tau aggregation, hoping that they might be a useful step in finding compounds with two biological targets relevant to treating AD. Beginning with asperbenzaldehyde, which was purified from a fungal strain engineered to overproduce this compound, the azaphilones were prepared as previously described using two schemes11. The first employs p-toluenesulfonic acid to form the 2-benzopyrilium salt followed by oxidation by lead tetraacetate with or without halogenation. The second scheme employs the hypervalent-iodine-mediated phenol oxidative dearomatization of the 2-benzopyrilium salt with o-iodoxy-benzoic acid followed by halogenation and/or a Wittig olefination with carbethoxymethylenetriphenylphosphorane. Using standard biochemical assays, we investigated the ability of these compounds to alter the aggregation of tau and its stabilization of microtubules. We found that while all compounds inhibited tau aggregation, a smaller subset had the added activity of disassembling pre-formed tau aggregates. The compounds most effective at inhibiting tau aggregation and disassembling pre-formed tau filaments also allowed tau to retain the majority of its microtubule stabilizing functions. Results Eleven compounds with the same azaphilone backbone differing at three points of diversity (R1, R2, and R3) were used in this study (Figure 1). Tau polymerization was initiated in vitro using a standard arachidonic acid induction assay17. To determine whether the compounds could inhibit assembly of tau filaments, each of the compounds, at a final concentration of 200 µM, was preincubated with 2 µM tau for 20 min before the addition of 75 µM arachidonic acid. The degree of tau aggregation inhibition for each compound was determined using a membrane filter assay18. This assay has been used previously to screen A. nidulans secondary metabolites including anthraquinones, xanthones, polyketides, a benzophenone and the asperbenzaldehyde compound that was the parent compound for the synthesis of the azaphilones used in this study10. A mixture of antibodies to the amino terminal region, central region and carboxy terminal region of tau (tau 12, tau 5, and tau 7 respectively) was used to detect tau aggregates. In this assay, only compound aza-11 significantly reduced the amount of tau aggregation detected (Figure 2A). Compounds aza-13 and aza-15 significantly increased the amount of tau aggregation and the remaining compounds had no significant effect (Figure 2A). However, when antibodies against toxic species of tau were employed for detection, very different results were observed. All aza compounds completely abolished recognition by the TOC1 antibody, which recognizes toxic oligomers in vitro and in Alzheimer’s disease tissue19, as compared to controls without compound (Figure 2B). Similarly, significant reductions in recognition by TNT1, an antibody that recognizes the phosphatase-activating domain of tau and is exposed in pathological forms of tau20, were observed for all aza compounds as compared to controls without compound (Figure 2C). When the resulting tau aggregates from the inhibition reactions were visualized by electron microscopy, there were abundant numbers of long filaments in the absence of added compound (Figure 3). Surprisingly, all the azaphilones inhibited the formation of long tau filaments that were observed in the absence of compounds. Instead, amorphous small aggregates were observed after treatment with the compounds (Figure 3). This degree of tau aggregation inhibition was similar to what was observed by electron microscopy for asperbenzaldehyde and asperthecin, and stronger than what was observed for 2,ω-dihydroxyemodin in a prior study10. Because tau aggregation inhibitors that inhibit filament formation have previously been shown to stabilize off-pathway soluble oligomers that are large enough to be trapped in the membrane filter assay21, we believe that the mixture of tau antibodies to normal tau was detecting these aggregates in the filter trap assay (Figure 2A). These aggregates do not seem to be toxic because of their lack of reactivity to TOC1 (Figure 2B) and TNT1 (Figure 2C). To determine whether these compounds can disassemble pre-formed tau aggregates, tau aggregation was allowed to proceed for six hours before the addition of compounds to a final concentration of 200 µM. After 12 hours the effect of compounds on the tau aggregation were examined by filter trap assay using the mixture of antibodies against normal tau (Figure 4). All compounds reduced the amount of preformed tau filaments, with compounds aza-8, aza-9, aza-11, aza-12, and aza-13 having the greatest activity. Electron microscopy was used to validate and extend the results from the filter trap assay. Compounds aza-8, aza-9, aza-12 and aza-13 substantially reduced the preexisting filament mass while the other compounds had less effect (Figure 5). To test whether compounds aza-8, aza-9, aza-12, and aza-13 were not simply blocking the adherence of tau filaments to the electron microscopy grids, compounds were added to pre-formed tau filaments and were immediately prepared for electron microscopy without allowing time for disassembly to occur. Under these conditions, none of the compounds blocked the binding of the filaments to the grid (Supplemental Figure S2). Quantitative analysis of the filament lengths in the presence and absence of compounds confirmed that compounds aza-8, aza-9, aza-12 and aza-13 had fewer aggregates overall compared to reactions without compound and virtually no filaments remaining greater than 200 nm in length (Figure 6). The IC50 of the four most potent compounds was determined using the filter trap assay. The amount of pre-formed filaments remaining following treatment with compounds for 12 hours was reduced in a concentration dependent manner for all four compounds tested (Figure 7). Compound aza-9 had an IC50 of 56 ± 14 µM, compared to 118 ± 19 µM, 98 ± 16 µM and 216 ± 18 µM for compounds aza-8, aza-12 and aza-13 respectively, indicating that aza-9 has the most activity for dissolving pre-formed tau filaments in vitro (Figure 7). In a number of studies heparin has been used to induce tau aggregation. Because heparin-induced tau filaments might be different from arachidonic acid induced filaments, we wished to determine if aza 9 inhibited heparin-induced tau aggregation or disassembled heparin assembled tau aggregates. We chose aza 9 because it was the most potent compound among the 11 azaphilones. Filter trap assays were performed using the mixture of antibodies against normal tau, the TOC1 antibody and the TNT1 antibody. Aza 9 significantly reduced the assembly of TOC1 and TNT1 positive aggregates (Figure 8A). The addition of aza 9 also resulted in the significant disassembly of pre-formed filaments recognized by the TOC1 and TNT1 antibodies (Figure 8B). We chose the most potent azaphilone derivatives aza-8, aza-9, aza-12 and aza-13 to determine their effects on the normal function of tau to stabilize microtubules. Tubulin was mixed with tau in the presence or absence of 90 µM compound and the resulting microtubule formation was monitored by turbidity. All polymerization curves were fit using a Gompertz growth equation (Figure 9). While all four compounds affected the apparent rate and maximum amount of microtubule formation at the concentration tested (Table 1), tau still retained a significant ability to stabilize microtubule formation. Discussion Tau based therapeutic strategies have recently been gaining additional attention largely due to the major role tau pathology plays in many neurological disorders including Alzheimer’s disease. Several tau-directed therapeutic strategies with disease-modifying potential have been identified including modulating tau phosphorylation, microtubule stabilization, tau aggregation inhibitors and tau clearance using antibodies22–30. Conversion of soluble monomeric tau into insoluble tau aggregates could, potentially, result in both loss of function and gain of function toxicities31. Therefore, inhibiting aggregation of tau might prevent formation of the toxic oligomers or tangles. Inhibiting aggregation could also increase the levels of monomeric tau, thereby increasing the chances for its clearance through chaperone mediated processes32. Previous studies have identified several tau aggregation inhibitor (TAI) molecules including those belonging to the class of anthraquinones, phenothiazines, and a benzothiazolidine derivatives among others3, 33, 34. One TAI, a stable, reduced form of methylthioninium chloride, is currently in Phase III clinical trials2, indicating this approach has promise and it is, consequently, worthwhile to identify additional structural backbones with this activity. Many of the previously identified TAIs are comprised of fused ring structures believed to be capable of interacting with the β-sheet structures formed in tau aggregates, thereby inhibiting formation of tau filaments10, 21. Fungal extracts are known to include pharmaceutically important secondary metabolites35. We therefore previously screened 17 secondary metabolites obtained from the fungus Aspergillus nidulans for TAIs due to their structural similarity to previously identified TAIs10. From this screen we identified 3 compounds that inhibited tau aggregation at micromolar concentrations. Two of these compounds belonged to the anthraquinone class of compounds and one was structurally unique from all the previously identified TAIs. We were particularly interested in this compound – asperbenzaldehyde. Asperbenzaldehyde is a precursor to an important class of natural products called azaphilones. Azaphilones are a structurally diverse group of polyketides that share a highly oxygenated bicyclic core and chiral quaternary center36. The azaphilones used in this study were obtained by semi-synthetic diversification of asperbenzaldehyde. All eleven azaphilones inhibited the formation of tau filaments but some of them produced small amorphous tau aggregates, which can be seen in the electron micrographs in Figure 3. These aggregates were not recognized by TOC119 and TNT120 antibodies, which bind to toxic forms of tau, therefore we believe these compounds promote the formation of small off-pathway aggregates of tau which are not toxic and do not act as seeds for further tau filament assembly. The induction of these aggregates could be similar to soluble aggregates of tau induced by porphyrin pthalocyanine tetrasulfonate that have a different conformation from insoluble toxic tau oligomers21. From a therapeutic point of view, TAIs would be more useful if they could also dissolve pre-formed tau filaments because they could theoretically be beneficial to patients that already demonstrate cognitive impairments. We found that a subset of the compounds, aza-8, aza-9, aza-12, and aza-13, showed this property. These four compounds have Br or Cl at position R1, while the other compounds have either I or H at R1 (Figure 1). Therefore, halogenation at position R1 may be not necessarily important for inhibition of tau filament formation, but electron-withdrawing groups at R1 specifically seem to enhance disassembly of tau filaments. Cl and Br are more electronegative (3.0 and 2.8 respectively), than I (2.5) and H (2.1) indicating that increased electronegativity at position R1 could have a significant impact on the activity of tau aggregation inhibitors with this scaffold. The four disassembly causing compounds have a ketone at position R3, while presence of the CHCO2Et moiety at the same position seems to virtually eliminate disassembly, even with halogenation. The impact of the chemical groups at the R2 position seems to be dependent upon the substitution at R1. Compounds aza-8 and aza-12 both have Cl at R1, but possess acetate and hydroxyl groups at R2, respectively. Despite this structural difference, there is no significant difference in their activity levels. However compounds aza-9 and aza-13 both containing Br at R1, and acetate and hydroxyl groups at R2, respectively differ in their levels of activity. Aza-9 is more potent than aza-13, therefore positioning of the acetate group at R2 in the presence of a Br at R1 might be important for compound activity. Additionally, all four disassembly causing compounds have lipoxygenase-1 inhibitory activity in the low micromolar range (IC50 of 2–8µm)11. Inhibition of LOX-1 may help reduce fatty acid metabolites of arachidonic acid and docasahexanoic acid that are elevated in Alzheimer’s disease16. These compounds could therefore have two positive therapeutic activities in tau dementias. The relatively high IC50 values of our compounds indicate that they are unlikely to be of therapeutic value and we do not know if they have suitable bioavailability or pharmacokinetic properties. It is important, however, to identify new scaffolds with the appropriate biological activity for further development. We believe these compounds provide a novel TAI scaffold with the added features that they inhibit LOX-1 and some of them disassemble pre-formed tau aggregates. In conclusion, this study shows that these compounds inhibit assembly of tau aggregates, disassemble preformed tau aggregates, and partially preserve tau’s ability to bind to tubulin and promote microtubule assembly. These compounds provide a promising novel scaffold for TAI molecules. The structure-activity relationship studies give us several leads for the probable important chemical groups required in this scaffold structure required for the anti-tau aggregation activity of the compounds. Further studies on the interaction between the compounds and tau will help to determine the precise mechanism of action of these compounds. Methods Chemicals and Reagents Full length 2N4R tau (441 amino acids) was expressed in E. coli and purified as described previously37. Arachidonic acid (ARA) was purchased from Cayman Chemicals (Ann Arbor, MI). Heparin sodium salt was purchased from SIGMA (St. Louis, MO). Asperbenzaldehyde was purified from Aspergillus nidulans and converted to the compounds aza-7 through aza-17 (Figure 1) as described previously11 with the following minor modifications. We constructed a number of strains with various promoter combinations and used a variety of induction conditions to maximize asperbenzaldehyde production. Our best yields were obtained with strain LO8355 (pyrG89, pyroA4, riboB2, nkuA::argB, stcJ::AfriboB, AN1029(p):: AfpyrG-alcA(p), AN1033::AfpyroA, alcR(p)::ptrA-gpdA(p)). In this strain, the promoter of asperfuranone biosynthesis transcription factor AN1029 (using the AspGD nomenclature (http://aspergillusgenomes.org/)) is replaced with the highly inducible alcA promoter and the alcR promoter is replaced with the strong, constitutive gpdA promoter (−1241 to −1). AN1033 is replaced with the AfpyroA gene to interrupt the asperfuranone biosynthesis pathway causing asperbenzaldehyde to accumulate. Growth was in lactose minimal medium (20 g/L lactose, 6 g/L NaNO3, 0.52 g/L KCl, O.52 g/L MgSO4,·7H2O, 1.52 g/L KH2PO4, 1 mL/L trace elements solution)38. Spores were inoculated at 106/mL into 500 mL of medium in a 2L flask. Incubation was at 37 °C on a gyratory shaker and induction was with 30 mM methyl-ethyl-ketone, added 55 hours after inoculation. Cultures were harvested six days after inoculation. Yields of purified asperbenzaldehyde were greater than 2 g/L representing an approximate conversion of 10% of the carbon source to final product. Additional information on the purity of the compounds can be found in supplemental materials. Inhibition of tau aggregation 75 µM arachidonic acid was used to initiate the aggregation of 2 µM tau in polymerization buffer (PB, 10 mM HEPES (pH 7.64), 5 mM DTT, 100 mM NaCl, 0.1 mM EDTA, and 3.75% ethanol) in vitro as previously described17. Compounds dissolved in DMSO were added to a final concentration of 200 µM and incubated with tau protein in PB 20 minutes prior to the addition of arachidonic acid. The reactions were allowed to proceed for 16 h at room temperature before analysis10. For heparin induced tau assembly inhibition reactions, 0.6 µM heparin was used to initiate aggregation on 2 µM tau in polymerization buffer (PB, 10mM HEPES (pH 7.64), 5 mM DTT, 17.7 mM NaCl) in vitro. Compounds dissolved in DMSO were added to a final concentration of 200 µM and incubated with tau protein in PB 20 minutes prior to the addition of heparin. The reactions were allowed to proceed for 16 h at 37 °C before analysis. Disassembly of pre-formed filaments Pre-formed tau filaments were generated with 2 µM tau and 75 µM ARA in PB as described above for 6 h at room temperature. Compounds dissolved in DMSO were added to the tau solution at final concentrations indicated in the Results section and figure legends. The reactions were allowed to proceed at room temperature for 12 h before analysis. For heparin induced tau aggregation assays, pre- formed tau filaments were generated with 2 µM tau and 0.6 µM heparin in PB as described above for 12 h at 37 °C. Compounds dissolved in DMSO were added to the tau solution at final concentrations indicated in the Results section and figure legends. The reactions were allowed to proceed at 37 °C for 24 h before analysis. Filter trap assay The amount of tau aggregates following assembly or disassembly reactions was determined by filter trap assay as described previously10. Reactions were diluted into TBS such that they contained 20 ng protein in 300 µL. For heparin induced tau aggregation reactions, the reactions were diluted into TBS such that they contained 60 ng protein in 300 µL. Solutions were passed through a nitrocellulose membrane using house vacuum in a dot-blot apparatus. The aggregates trapped on the membrane were detected by either general antibodies (a mixture of Tau 539 at 1:50,000 dilution, Tau 1240 at 1:250,000 dilution and Tau 741 at 1:250,000 dilution), or antibodies to toxic conformations (TNT119 at 1:200,000 or TOC120 at 1:7,000). All antibodies were a kind gift from Drs. Nick Kanaan and Lester I. Binder. HRP-linked Goat anti-mouse IgG (general antibodies and TNT1) or HRP-linked Goat anti-mouse IgM (TOC1) (Thermo Scientific, Rockford, IL) were used as the secondary antibodies and blots were developed using ECL (enhanced chemiluminescence) Western Blotting Analysis System (GE Healthcare, Buckinghamshire, UK). Images were captured with a Kodak Image Station 4000R or ChemiDoc-It2 Imager and were quantified using the histogram function of Adobe Photoshop 7.0. Statistical analyses were performed using unpaired t-tests to compare the triplicate values to control values. Transmission electron microscopy Polymerization reaction samples were diluted 1:10 in PB and fixed with 2% glutaraldehyde for 5 min. Fixed samples were placed on formvar carbon-coated grids and stained with uranyl acetate as previously described10. Images were captured with a Technai F20 XT Field emission transmission electron microscope (FEI Co., Hillsboro, OR) and Gatan Digital Micrograph imaging system (Gatan, Inc., Pleasanton, CA). The filaments were quantified using Image-Pro Plus 6.0 software (Media Cybernetics, Inc., Rockville, MD) as previously described10. For quantitative analysis, filament lengths were placed into bins as described in results. Statistical analyses were performed using unpaired t-tests to compare 4 or 5 replicates for each bin size with the no compound data serving as reference values. Tubulin Polymerization Assay Polymerization of tubulin was measured using a Tubulin Polymerization Assay kit BK006P (Cytoskeleton, Inc., Denver, CO) following the manufacturer’s protocol. Briefly, 2 mg/ml porcine tubulin was incubated with 1.5 µM tau and 90 µM aza-8, aza-9, aza-12 or aza-13 compounds. Tubulin polymerization was monitored by turbidity at 340 nm in a Varian 50 MPR microplate reader at 37 °C for 1 h. Experiments were performed in triplicate, averaged and fit to a Gompertz growth equation as previously described42 using the equation: y=ae−e−((t−ti)b) Where y is the amount of microtubule polymerization measured at time t, a is the maximum amount of microtubule polymerization observed at an absorbance of 340 nm (Max), ti is the point of inflection of the curve at the time of maximum growth rate in minutes, and b is inversely proportional to the apparent rate of polymerization (kapp, min−1). The average values for the parameters a, b and ti were determined and compared to the no-compound control using a paired t-test to determine statistical significance *p≤ 0.05; **p≤ 0.01; ***p≤ 0.001. Supplementary Material Supplemental Material We thank Adam Miltner for assistance with protein expression and purification. Funding was provided in part by P01-GM084077 from the National Institute of General Medical Sciences (CCCW and BRO), T32 GM008545 from the National Institute of General Medical Sciences (APR), R01-NS083391 from the National Institute of Neurological Disorders and Stroke (TCG) and by the H.L. Snyder Medical Foundation (BRO and TCG). We thank Drs. Nick Kanaan and Lester I. Binder for their kind gift of the antibodies used in this study. Figure 1 Compounds used in this study The core structure of the azaphilone compounds is shown with the positions of modications indicated by R1, R2 and R3. A.-K.: The structures of the 11 compounds used in this study: A) aza-7, B) aza-8, C) aza-9, D) aza-10, E) aza-11, F) aza-12, G) aza-13, H) aza-14, I) aza-15, J) aza-16, and K) aza-17. Figure 2 Filter trap assay of tau filament formation Tau polymerization reactions in A), B), and C) were performed with 2 µM tau and 75 µM arachidonic acid either with or without 200 µM compound. The compounds used are listed on the X- axis. The resulting amount of tau filament formation was determined using a filter trap assay. The values for tau filament formation were normalized to the amount of aggregation in the absence of compound (dashed line). Negative values indicate that there was less detectable tau on the filter after treatment with a compound than was observed with monomeric tau in the absence of arachidonic acid. The amount of tau on the filter was detected using A) a mixture of antibodies tau 5, tau 7 and tau 12, B) antibody TOC1, and C) antibody TNT1. The observation that the values are lower in B) and C) suggests that the tau aggregates that are retained on the filters are not in the toxic oligomeric form recognized by the TOC1 antibody and the phosphatase domain recognized by TNT1 antibody is not exposed. Values are the average of three trials ± SD. *P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001 Figure 3 Electron microscopy of tau filament formation in the presence of azaphilone derivatives Tau polymerization reactions were performed with 2 µM tau and 75 µM arachidonic acid either with or without 200 µM compound. Aliquots of the reactions were prepared for negative stain electron microscopy. Representative images are shown for A) no compound control, B) aza-7, C) aza-8, D) aza-9, E) aza-10, F) aza-11, G) aza-12, H) aza-13, I) aza-14, J) aza-15, K) aza-16, and L) aza-17. The scale bar in the lower right panel represents 1 µm and is applicable to all images. Figure 4 Filter trap assay of tau filament disassembly Tau polymerization reactions were performed with 2 µM tau and 75 µM arachidonic acid at room temperature. After 6 hours, 200 µM compound or an equal volume of DMSO was added to the reactions. The compounds used are listed on the X-axis. The resulting amount of tau filament formation was determined using a filter trap assay. The values for tau filament formation were normalized to the amount of aggregates detected in the absence of compound (dashed line). Negative values indicate that there was less detectable tau on the filter after treatment with a compound than was observed with monomeric tau in the absence of arachidonic acid. The amount of tau on the filter was detected using a mixture of antibodies tau 5, tau 7 and tau 12. Values are the average of three trials ± SD. *P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001 Figure 5 Electron microscopy of tau filament disassembly in the presence of azaphilone derivatives Tau polymerization reactions were performed with 2 µM tau and 75 µM arachidonic acid at room temperature. After 6 hours, 200 µM compound or equal volume of DMSO was added to the reactions. Aliquots of the reactions were prepared for negative stain electron microscopy. Representative images are shown for A) no compound control, B) aza-7, C) aza-8, D) aza-9, E) aza-10, F) aza-11, G) aza-12, H) aza-13, I) aza-14, J) aza-15, K) aza-16, and L) aza-17. The scale bar in the lower right panel represents 1 µm and is applicable for all images. Figure 6 Filament length distributions Tau disassembly reactions were performed and viewed by electron microscopy. The filaments remaining following incubation with or without compound were measured, and placed into bins according to their length (30–50 nm, 50–100 nm, 100–150 nm, 150–200 nm and > 200 nm). The lengths within a bin were summed to determine the total amount of filament length in each bin. A) Represents the length distributions for filaments in the control reaction without compound. The length distributions are also shown for filaments remaining following incubation in the presence of B) aza-7, C) aza-8, D) aza-9, E) aza-10, F) aza-11, G) aza-12, H) aza-13, I) aza-14, J) aza-15, K) aza-16, and L) aza-17. Each point is the average distribution for images of 5 different fields ± SD.*P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001. Figure 7 IC50 determination tau filament disassembly Polymerization reactions at 2 µM tau and 75 µM arachidonic acid were performed at room temperature. After 6 hours, compounds were added to these reactions at several different concentrations and incubated an additional 12 hours. The resulting amount of tau filaments in the reaction were determined by a filter trap assay detected by a mixture of antibodies to normal tau (Tau 5, Tau 7 and Tau 12). The amount of polymerization was normalized to controls in the absence of compound (100%). The normalized data was plotted against the log of the inhibition concentration and fit to a dose-response curve to determine the IC50 for A) Aza-8, B) Aza-9, C) Aza-12 and D) Aza-13. Data points are the average of three trials ± SD. Figure 8 Filter trap assay for filament assembly and disassembly of heparin-induced tau filaments A) For assembly inhibition, 100 µM aza 9 was incubated with 2 µM tau for 20 minutes before the addition 0.6 µM heparin. 16 hours after induction, the degree of aggregation was determined using the filter trap assay detected by a mixture of antibodies to normal tau (Tau 5, Tau 7 and Tau 12), an antibody to oligomeric tau (TOC1) and an antibody to a toxic conformation of tau (TNT1). The average of three independent trials ± SD is shown for no compound controls (white bars) and 100 µM aza 9 (black bars). B) 2 µM tau and 0.6 µM heparin were incubated together for 21 hours prior to the addition of 100 µM aza 9 or an equal volume of DMSO. Disassembly reactions proceeded for an additional 12 hours and the degree of aggregation was determined using the filter trap assay detected by a mixture of antibodies (Tau 5, Tau 7 and Tau 12), TOC1 and TNT1. The average of three independent trials ± SD is shown for no compound controls (white bars) and 100 µM aza 9 (black bars). *P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001. Figure 9 Microtubule assembly in the presence of the most potent azaphilone tau aggregation inhibitors Tubulin was incubated with either tau protein alone or tau in the presence of A) aza-8, B) aza-9, C) aza-12, or D) aza-13 at a concentration of 90 µM. Microtubule assembly was monitored by absorbance at 340 nm (y- axis) over time (x-axis). Each point is the average of three independent trails ± SD. All data were fit to Gompertz growth curve (dashed and solid lines for no compound and 90 µM azaphilones respectively). For clarity, only every second time point is shown on the graph.*P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001. Table 1 Statistical analysis of the effects of compounds aza-8, aza-9, aza-12, and aza-13 on the stabilization of microtubules. ti (min) kapp (min−1) Max (A340) Aza-8 − 3.08 ± 1.28 0.87 ± 0.14 0.76 ± 0.36 + 13.42 ± 6.59 0.12 ± 0.02*** 0.36 ± 0.14* Aza-9 − 2.26 ± 0.65 0.62 ± 0.08 0.65 ± 0.38 + 5.03 ± 1.93 0.26 ± 0.19* 0.38 ± 0.10* Aza-12 − 3.11 ± 1.22 0.78 ± 0.09 0.75 ± 0.10 + 12.83 ± 1.35*** 0.09 ± 0.02*** 0.50 ± 0.04* Aza-13 − 2.58 ± 1.29 0.86 ± 0.17 0.72 ± 0.06 + 15.65 ± 7.72* 0.12 ± 0.05** 0.30 ± 0.07** Max (A340) is the maximum amount of microtubule polymerization, ti is the point of inflection of the curve at the time of maximum growth rate in minutes, and b is inversely proportional to the apparent rate of polymerization (kapp, min−1) as determined from the fit of three individual microtubule polymerization curves for each condition to a nonlinear Gompertz growth function (see Methods). Competing Interests The authors declare that they have no competing interests. Author Contributions SRP carried out the tau assembly and disassembly experiments and drafted the manuscript. APR purified asperbenzaldehyde, generated and purified azaphilone derivatives, and assisted with analysis of structure/activity relationships. ADS purified asperbenzaldehyde, generated and purified derivatives of azaphilones. CEO constructed the A. nidulans strains for asperbenzaldehyde over-production and cultured the strains for overproduction. CCCW and BRO conceived of the generation and purification of derivatives of asperbenzaldehyde and azaphilones. TEP assisted with analysis of structure/activity relationships. BRO and TCG conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript. References 1 Arriagada PV Growdon JH Hedley-Whyte ET Hyman BT Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer's disease Neurology 1992 42 631 639 1549228 2 Wischik CM Harrington CR Storey JM Tau-aggregation inhibitor therapy for Alzheimer's disease Biochemical pharmacology 2014 88 529 539 24361915 3 Pickhardt M Gazova Z von Bergen M Khlistunova I Wang Y Hascher A Mandelkow EM Biernat J Mandelkow E Anthraquinones inhibit tau aggregation and dissolve Alzheimer's paired helical filaments in vitro and in cells The Journal of biological chemistry 2005 280 3628 3635 15525637 4 Calcul L Zhang B Jinwal UK Dickey CA Baker BJ Natural products as a rich source of tau-targeting drugs for Alzheimer's disease Future medicinal chemistry 2012 4 1751 1761 22924511 5 Fang L Gou S Fang X Cheng L Fleck C Current progresses of novel natural products and their derivatives/ analogs as anti-Alzheimer candidates: an update Mini reviews in medicinal chemistry 2013 13 870 887 23305400 6 Yaegashi J Oakley BR Wang CC Recent advances in genome mining of secondary metabolite biosynthetic gene clusters and the development of heterologous expression systems in Aspergillus nidulans Journal of industrial microbiology & biotechnology 2014 41 433 442 24342965 7 Nayak T Szewczyk E Oakley CE Osmani A Ukil L Murray SL Hynes MJ Osmani SA Oakley BR A versatile and efficient gene-targeting system for Aspergillus nidulans Genetics 2006 172 1557 1566 16387870 8 Oakley CE Edgerton-Morgan H Oakley BR Tools for manipulation of secondary metabolism pathways: rapid promoter replacements and gene deletions in Aspergillus nidulans Methods in molecular biology 2012 944 143 161 23065614 9 Szewczyk E Nayak T Oakley CE Edgerton H Xiong Y Taheri-Talesh N Osmani SA Oakley BR Fusion PCR and gene targeting in Aspergillus nidulans Nature protocols 2006 1 3111 3120 17406574 10 Paranjape SR Chiang YM Sanchez JF Entwistle R Wang CC Oakley BR Gamblin TC Inhibition of Tau Aggregation by Three Aspergillus nidulans Secondary Metabolites: 2,omega-Dihydroxyemodin, Asperthecin, and Asperbenzaldehyde Planta medica 2014 80 77 85 24414310 11 Somoza AD Lee KH Chiang YM Oakley BR Wang CC Reengineering an azaphilone biosynthesis pathway in Aspergillus nidulans to create lipoxygenase inhibitors Organic letters 2012 14 972 975 22296232 12 Matsuzaki K Tahara H Inokoshi J Tanaka H Masuma R Omura S New brominated and halogen-less derivatives and structure-activity relationship of azaphilones inhibiting gp120-CD4 binding The Journal of antibiotics 1998 51 1004 1011 9918393 13 Clark RC Lee SY Searcey M Boger DL The isolation, total synthesis and structure elucidation of chlorofusin, a natural product inhibitor of the p53-mDM2 protein-protein interaction Natural product reports 2009 26 465 477 19642417 14 Musso L Dallavalle S Merlini L Bava A Nasini G Penco S Giannini G Giommarelli C De Cesare A Zuco V Vesci L Pisano C Castorina M Milazzo F Cervoni ML Dal Piaz F De Tommasi N Zunino F Natural and semisynthetic azaphilones as a new scaffold for Hsp90 inhibitors Bioorganic & medicinal chemistry 2010 18 6031 6043 20655237 15 Nam JY Kim HK Kwon JY Han MY Son KH Lee UC Choi JD Kwon BM 8-O-Methylsclerotiorinamine, antagonist of the Grb2-SH2 domain, isolated from Penicillium multicolor Journal of natural products 2000 63 1303 1305 11000046 16 Chu J Pratico D Pharmacologic blockade of 5-lipoxygenase improves the amyloidotic phenotype of an Alzheimer's disease transgenic mouse model involvement of gamma-secretase The American journal of pathology 2011 178 1762 1769 21435457 17 Carlson SW Branden M Voss K Sun Q Rankin CA Gamblin TC A complex mechanism for inducer mediated tau polymerization Biochemistry 2007 46 8838 8849 17608454 18 Chang E Kuret J Detection and quantification of tau aggregation using a membrane filter assay Analytical biochemistry 2008 373 330 336 17949677 19 Ward SM Himmelstein DS Ren Y Fu Y Yu XW Roberts K Binder LI Sahara N TOC1: a valuable tool in assessing disease progression in the rTg4510 mouse model of tauopathy Neurobiology of disease 2014 67 37 48 24631720 20 Kanaan NM Morfini GA LaPointe NE Pigino GF Patterson KR Song Y Andreadis A Fu Y Brady ST Binder LI Pathogenic forms of tau inhibit kinesin-dependent axonal transport through a mechanism involving activation of axonal phosphotransferases The Journal of neuroscience : the official journal of the Society for Neuroscience 2011 31 9858 9868 21734277 21 Akoury E Gajda M Pickhardt M Biernat J Soraya P Griesinger C Mandelkow E Zweckstetter M Inhibition of tau filament formation by conformational modulation Journal of the American Chemical Society 2013 135 2853 2862 23360400 22 Bhat RV Budd Haeberlein SL Avila J Glycogen synthase kinase 3: a drug target for CNS therapies J Neurochem 2004 89 1313 1317 15189333 23 Lagoja I Pannecouque C Griffioen G Wera S Rojasdelaparra VM Van Aerschot A Substituted 2-aminothiazoles are exceptional inhibitors of neuronal degeneration in tau-driven models of Alzheimer's disease European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences 2011 43 386 392 21664968 24 Brunden KR Yao Y Potuzak JS Ferrer NI Ballatore C James MJ Hogan AM Trojanowski JQ Smith AB 3rd Lee VM The characterization of microtubule-stabilizing drugs as possible therapeutic agents for Alzheimer's disease and related tauopathies Pharmacological research : the official journal of the Italian Pharmacological Society 2011 63 341 351 25 Zhang B Carroll J Trojanowski JQ Yao Y Iba M Potuzak JS Hogan AM Xie SX Ballatore C Smith AB 3rd Lee VM Brunden KR The microtubule-stabilizing agent, epothilone D, reduces axonal dysfunction, neurotoxicity, cognitive deficits, and Alzheimer-like pathology in an interventional study with aged tau transgenic mice The Journal of neuroscience : the official journal of the Society for Neuroscience 2012 32 3601 3611 22423084 26 Chang E Congdon EE Honson NS Duff KE Kuret J Structure-activity relationship of cyanine tau aggregation inhibitors Journal of medicinal chemistry 2009 52 3539 3547 19432420 27 Honson NS Jensen JR Darby MV Kuret J Potent inhibition of tau fibrillization with a multivalent ligand Biochemical and biophysical research communications 2007 363 229 234 17854770 28 Pickhardt M Larbig G Khlistunova I Coksezen A Meyer B Mandelkow EM Schmidt B Mandelkow E Phenylthiazolyl-hydrazide and its derivatives are potent inhibitors of tau aggregation and toxicity in vitro and in cells Biochemistry 2007 46 10016 10023 17685560 29 Asuni AA Boutajangout A Quartermain D Sigurdsson EM Immunotherapy targeting pathological tau conformers in a tangle mouse model reduces brain pathology with associated functional improvements The Journal of neuroscience : the official journal of the Society for Neuroscience 2007 27 9115 9129 17715348 30 Krishnamurthy PK Deng Y Sigurdsson EM Mechanistic Studies of Antibody-Mediated Clearance of Tau Aggregates Using an ex vivo Brain Slice Model Frontiers in psychiatry 2011 2 59 22025915 31 Sahara N Maeda S Takashima A Tau oligomerization: a role for tau aggregation intermediates linked to neurodegeneration Current Alzheimer research 2008 5 591 598 19075586 32 Blair LJ Zhang B Dickey CA Potential synergy between tau aggregation inhibitors and tau chaperone modulators Alzheimer's research & therapy 2013 5 41 33 Chirita C Necula M Kuret J Ligand-dependent inhibition and reversal of tau filament formation Biochemistry 2004 43 2879 2887 15005623 34 Wischik CM Edwards PC Lai RY Roth M Harrington CR Selective inhibition of Alzheimer disease-like tau aggregation by phenothiazines Proc Natl Acad Sci U S A 1996 93 11213 11218 8855335 35 Keller NP Turner G Bennett JW Fungal secondary metabolism - from biochemistry to genomics Nature reviews. Microbiology 2005 3 937 947 16322742 36 Gao JM Yang SX Qin JC Azaphilones: chemistry and biology Chemical reviews 2013 113 4755 4811 23581812 37 Rankin CA Sun Q Gamblin TC Pseudo-phosphorylation of tau at Ser202 and Thr205 affects tau filament formation Brain research. Molecular brain research 2005 138 84 93 15913837 38 Vishniac W Santer M The thiobacilli Bacteriological reviews 1957 21 195 213 13471458 39 LoPresti P Szuchet S Papasozomenos SC Zinkowski RP Binder LI Functional implications for the microtubule-associated protein tau: localization in oligodendrocytes Proceedings of the National Academy of Sciences of the United States of America 1995 92 10369 10373 7479786 40 Horowitz PM Patterson KR Guillozet-Bongaarts AL Reynolds MR Carroll CA Weintraub ST Bennett DA Cryns VL Berry RW Binder LI Early N-terminal changes and caspase-6 cleavage of tau in Alzheimer's disease The Journal of neuroscience : the official journal of the Society for Neuroscience 2004 24 7895 7902 15356202 41 Horowitz PM LaPointe N Guillozet-Bongaarts AL Berry RW Binder LI N-terminal fragments of tau inhibit full-length tau polymerization in vitro Biochemistry 2006 45 12859 12866 17042504 42 Combs B Voss K Gamblin TC Pseudohyperphosphorylation has differential effects on polymerization and function of tau isoforms Biochemistry 2011 50 9446 9456 21942206

PMC005xxxxxx/PMC5113143.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985117R 4816 J Immunol J. Immunol. Journal of immunology (Baltimore, Md. : 1950) 0022-1767 1550-6606 27742832 5113143 10.4049/jimmunol.1601427 NIHMS817240 Article Small molecule inhibition of Rab7 impairs B cell class-switching and plasma cell survival to dampen the autoantibody response in murine lupus Lam Tonika *1 Kulp Dennis V. *1 Wang Rui *12 Lou Zheng * Taylor Julia * Rivera Carlos E. * Yan Hui * Zhang Qi * Wang Zhonghua § Zan Hong * Ivanov Dmitri N. § Zhong Guangming * Casali Paolo *3 Xu Zhenming *3 * Department of Microbiology, Immunology and Molecular Genetics, University of Texas School of Medicine, UT Health Science Center at San Antonio, TX 78229 § Department of Biochemistry, University of Texas School of Medicine, UT Health Science Center at San Antonio, TX 78229 Correspondence: Paolo Casali: Phone: (210) 567-3925; Fax: (210) 567-6612; pcasali@uthscsa.edu. Zhenming Xu: Phone: (210) 567-3964; Fax: (210) 567-6612; xuz3@uthscsa.edu 1 These authors contributed equally. 2 Current address: Xiangya School of Medicine, Central South University of China, Changsha, Hunan, China. 3 These authors jointly directed the study. 17 9 2016 14 10 2016 15 11 2016 15 11 2017 197 10 37923805 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. IgG autoantibodies mediate pathology in systemic lupus patients and lupus-prone mice. Here we showed that the class-switched IgG autoantibody response in MRL/Faslpr/lpr and C57/Sle1Sle2Sle2 mice was blocked by the CID 1067700 compound, which specifically targeted Rab7, an endosome-localized small GTPase that was upregulated in activated human and mouse lupus B cells, leading to prevention of disease development and extension of life-span. These were associated with decreased IgG-expressing B cells and plasma cells, but unchanged numbers and functions of myeloid cells and T cells. The Rab7 inhibitor suppressed T cell-dependent and T cell-independent antibody responses, but did not affect T cell-mediated clearance of Chlamydia infection, consistent with a B cell-specific role of Rab7. Indeed, B cells and plasma cells were inherently sensitive to Rab7 gene knockout or Rab7 activity inhibition in class-switching and survival, respectively, while proliferation/survival of B cells and generation of plasma cells were not affected. Impairment of NF-κB activation upon Rab7 inhibition, together with the rescue of B cell class-switching and plasma cell survival by enforced NF-κB activation, indicated that Rab7 mediates these processes by promoting NF-κB activation, likely through signal transduction on intracellular membrane structures. Thus, a single Rab7-inhibiting small molecule can target two stages of B cell differentiation to dampen the pathogenic autoantibody response in lupus. Rab7 NF-κB AID CSR plasma cells autoantibodies lupus small molecule compound Introduction Class switch DNA recombination (CSR) in the IgH locus substitutes the IgH constant (CH) region, such as Cμ of IgM expressed in naïve B cells, with Cγ, Cα or Cε, thereby giving rise to IgG, IgA or IgE antibodies. Together with somatic hypermutation (SHM), CSR is central to the maturation of the antibody response to pathogens, as class-switched antibodies display wide tissue distributions and possess diverse biological effector functions (1). These traits also make IgG and IgA autoantibodies highly pathogenic and capable of mediating multiple organ damage in systemic lupus erythematosus (SLE) (2, 3). Un-switched IgM autoantibodies are mostly nonpathogenic and, rather, may mediate frontline immune protection as natural polyreactive antibodies (4). Class-switched B cells further differentiate into memory B cells or plasma cells, which secrete antibodies at a high rate. A hallmark of active lupus is the high number of IgG+ antibody-secreting cells (ASCs) that produce autoantibodies. These are diverse but generally target nuclear antigens, such as DNA (5, 6), making the mechanisms underlying B cell class-switching and plasma cell generation/maintenance targets of new lupus therapeutics. Like SHM, CSR requires activation-induced cytidine deaminase (AID, encoded by AICDA in humans and Aicda in mice). AID expression is mainly restricted in peripheral B cells activated by CD154 engagement of CD40 on the B cell surface or by complex antigens that engage both a Toll-like receptor (TLR) and the B cell receptor (BCR) (7). AID is elevated in B cells of lupus patients and lupus mice, consistent with the heightened CSR/SHM in these B cells (8), and AID deficiency abrogates IgG autoantibodies in lupus-prone MRL/Faslpr/lpr mice (8, 9). Inhibitors of AID deaminase activity are yet to be developed, thereby emphasizing the need for molecules that target the mechanisms underlying AID induction in order to dampen the class-switched pathogenic autoantibody response. Rab7 (encoded by RAB7A in humans and Rab7 in mice) is a small GTPase that, when bound to its GTP substrate, promotes endosome maturation and autophagy. As we have shown (10), Rab7 is induced in activated B cells in vivo (i.e., in PNAhi germinal center B cells) and in vitro, e.g., by CD154 and TLR ligands, the same stimuli that induce AID expression and CSR. It plays a B cell-intrinsic role in antibody responses, as mice that conditionally knockout Rab7 in activated B cells cannot mount mature antibody responses to T cell-dependent or -independent antigens (10). Rab7 promotes CSR (to IgG, IgE and IgA) and does so by mediating AID induction, as enforced expression of AID rescues CSR in Rab7 knockout B cells. Further, Rab7 plays an important role in CD40- or TLR-triggered activation of NF-κB, which directly regulates Aicda gene transcription by binding to the promoter and enhancers of this gene (1, 11, 12). Rab7 is, however, dispensable for Erk1/Erk2 activation and expression of Blimp-1, both of which critically mediate plasma cell generation (13, 14), and, as a consequence, B cell differentiation into plasma cells, suggesting that Rab7 and its associated intracellular membrane structures (i.e., endosomes) specify receptor-triggered signaling for selective gene expression and B cell differentiation processes (15). Whether Rab7 plays a role in the maintenance of plasma cells remains unclear. Here we hypothesized that the lupus autoantibody response can be suppressed by inhibition of CSR in B cells and impairment of generation or maintenance of plasma cells, ideally by a single molecule that can target both cell types. To test this hypothesis, we have used a high-affinity and specific Rab7 inhibitor, CID 1067700. This has been identified by high-throughput screening as the only compound to affect the binding of purified recombinant Rab7 to GTP and GDP (16). By analyzing the level of activated Rab7 form (GTP-bound Rab7, Rab7-GTP) in B cells treated with CID 1067700 and using B cell-specific Rab7 knockout mice as well as retroviruses that enforced specific gene expression, we have verified the specific targeting of Rab7 by this small molecule in B cells and the consequent impairment in NF-κB activation. By using our defined in vitro B cell and plasma cell culture systems, we have further analyzed the impact of Rab7 inhibition on B cell class-switching and plasma cell generation/survival as well as the role of Rab7-dependent NF-κB activation in these processes. Finally, by analyzing – for the first time – the in vivo effect of the Rab7 inhibition on antibody and autoantibody responses in normal C57BL/6 (C57) mice and two widely used lupus mouse strains, female MRL/Faslpr/lpr and C57/Sle1Sle2Sle3 mice (17, 18), we have outlined the potential of Rab7 as a therapeutic target in lupus and possibly other NF-κB-, CSR- and/or plasma cell-mediated disease conditions. Materials and Methods Mice, drug treatment, immunization and infection Female mice were used in all experiments. C57 (Stock # 000664), MRL/Faslpr/lpr (Stock #000485) and C57/Sle1Sle2Sle3 (Stock #007228) mice were from The Jackson Laboratory and maintained in a pathogen-free vivarium. Conditional Rab7 knockout Igh+/Cγ1-creRab7fl/fl mice and their “wildtype” Igh+/Cγ1-creRab7+/fl littermates on the C57 background were as described (10). Rosa26+/fl-STOP-fl-Ikkβca mice on the C57 background (19) were from The Jackson Laboratory (Stock #008242). For in vivo treatment, CID 1067700 (MLS000673908; SID 24798006; SID 57578339; Glixx Laboratories) dissolved in DMSO (stock concentration 40 mM, 16 mg/ml) was diluted with the solvent to the final volume of 50 μl and injected intraperitoneally (i. p.) once per week at the dose of 16 mg/Kg body weight. This dose was well tolerated by mice at all ages (data not shown); it is within the dose range used in patients and animal models for several drugs with comparable EC50 (10–100 nM) to their respective targets (higher doses led to variable mouse death, perhaps due to putative off-targeting effect of the drug). C57, MRL/Faslpr/lpr and C57/Sle1Sle2Sle3 mice injected i. p. with the vehicle DMSO (nil; 50 μl) showed no difference in examined parameters, as compared to their respective counterparts without any injection (data not shown). For survival studies and skin lesion analyses, MRL/Faslpr/lpr mice were treated with nil or CID 1067700 for 10 weeks and maintained until moribund (e.g., showing signs of severe loss of mobility, hunched back, piloerection, ruffled fur, dyspnea, gasping and weight loss), at which point they were euthanized. For other clinical, serological, cellular and molecular analyses (see below), mice were treated for 7 weeks in a “double-blind” manner, as their identities were made unknown to investigators who performed these assays. For immunization, C57 mice were treated with nil or CID 1067700 7 d before i. p. injection with 100 μg of NP-CGG (in average 16 molecules of 4-hydroxy-3-nitrophenyl acetyl, NP, conjugated to one molecule of chicken γ-globulin; Biosearch Technologies) in 100 μl of alum (Imject® Alum, Pierce) or 25 μg of NP-LPS (0.2 NP molecule conjugated to one LPS molecule; Biosearch Technologies) in 100 μl of PBS. Mice were treated with nil or CID 1067700 once every three days until being sacrificed. For Chlamydia muridarum infection, the Nigg strain was propagated in Hela cells for the isolation of the Nigg3G0.10.1 clone, as previously described (20). Purified Nigg3G0.10.1 elementary bodies (EB) were used to infect 6-week old C57 mice intravaginally with 2 × 105 inclusion-forming units (IFUs). Mice were injected with 2.5 mg of medroxyprogesterone (Depo-Provera; Pharmacia Upjohn) subcutaneously at d −5 to increase the susceptibility and nil or CID 1067700, starting at d −7, once a week for 5 weeks. Mice were monitored for vaginal live organism shedding up to d 63. All protocols were in accordance with rules and regulations of the Institutional Animal Care and Use Committee of the University of Texas Health Science Center at San Antonio (UTHSCSA). Human and mouse primary B cells Human PBMCs were prepared, following a standard protocol, from peripheral blood of de-identified SLE patients and healthy subjects, as collected by venipuncture. SLE patients, who were recruited with informed consent under the protocol approved by the Institutional Review Board of UTHSCSA, fulfilled at least four 1982 American Rheumatism Association revised criteria for SLE and had the SLE Disease Activity Index (SLEDAI) above 3. All donors were free of infection by HBV, HCV, HPV, HIV and EBV. To prepare human B cells for stimulation, PBMCs (5 × 107) were subjected to negative selection (against CD2, CD14, CD16, CD27, CD36, CD43 and CD235a) using Human Naïve B Cell Isolation Kit II (Miltenyl) following the manufacturer’s instructions, resulting in over 95% IgD+ B cells (generally over 3 × 106). Cells were analyzed or seeded at 3 × 105 cell/ml for culturing in RPMI 1640 medium (Invitrogen) supplemented with FBS (10% v/v, Hyclone), penicillin-streptomycin/amphotericin B (1% v/v) and 50 μM β-mercaptoethanol (RPMI-FBS). Mouse immune cells were isolated from single cell suspensions prepared from the spleen and pooled axillary, inguinal and cervical lymph nodes. Lymph node cells were directly lysed for immunoblotting studies. Spleen cells were resuspended in ACK Lysis Buffer (Lonza) to lyse red blood cells and, after quenching with RPMI-FBS, were resuspended in PBS for analysis or further preparation. To isolate B cells, spleen cells were subjected to negative selection (against CD43, CD4, CD8, CD11b, CD49b, CD90.2, Gr-1 or Ter-119) using EasySepTM Mouse B cell Isolation Kit (StemCellTM Technologies) following the manufacturer’s instructions, resulting in preparations of more than 98% IgM+IgDhi B cells. After pelleting, B cells were resuspended in RPMI-FBS before further analysis or stimulation. For cell sorting, spleen cells were stained with fluorophore-labeled anti–CD19 mAb, anti–CD138 mAb, anti–TCR mAb and/or peanut agglutinin (PNA) to purify B cells, activated B cells and plasma cells, as indicated, resulting preparations that were at least 98% pure by post-sorting analysis. Bone marrow cells were isolated form tibia and fibula for myeloid cell differentiation and ELISPOT experiments. B cell culture, stimulation and drug treatment Human B cells were stimulated with CD154 (3 U/ml) in the presence of recombinant IL-4 (20 ng/ml; BioLegend) and IL-21 (50 ng/ml; R&D Systems) for 48 h for transcript analyses or 120 h for flow cytometry analysis of class-switched IgG1 and other markers. CD154 was prepared as membrane fragments isolated from Sf21 insect cells infected by CD154-expressing baculovirus, as we described (10, 21) – membrane fragments from non-infected Sf21 cells failed to stimulate B cells. For CSR induction in mouse B cells, spleen B cells were cultured (3 × 105 cell/ml) in RPMI-FBS were stimulated for 96 h (or 48 h for transcript and protein analyses) by the following primary stimuli: CD154 (3 U/ml), LPS (3 μg/ml, from E. coli, serotype 055:B5, deproteinized by chloroform extraction; Sigma-Aldrich) or R-848 (30 ng/ml; Invivogen), a ligand of TLR7, which promotes lupus autoantibody responses in a B cell-intrinsic manner. Recombinant IL-4 (4 ng/ml), for CSR to IgG1 and IgE, and TGF-β (2 ng/ml; R&D Systems), for CSR to IgA, and anti–Igδ mAb (clone 11-26c conjugated to dextran, anti–δ/dex, 100 ng/ml; Fina Biosolutions), which crosslinks IgD BCR to potentiate CSR to IgA in primary B cells, were added as indicated. To induce CSR to IgA in the mouse IgM+ CH12F3 lymphoma cell line, these B cells (105 cell/ml) were cultured in RPMI-FBS and stimulated with CD154 (3 U/ml) or, as first shown here, by LPS at low concentrations (100 ng/ml), plus IL-4 and TGF-β for 48 h. To generate plasma cells in vitro, purified B cells were stimulated with LPS plus IL-4, TGF-β, anti–Igδ mAb conjugated to dextran (anti–δ/dex) and retinoic acid (RA, 10 μM; Sigma) for 66 h. This condition, as we first found here, resulted in CD19loCD138hi plasma cells representing over 70% of cells in the culture. To treat human and mouse B cells in vitro with the Rab7 inhibitor, CID 1067700 was diluted in DMSO and added to cell cultures to the final concentration of 40 μM (or as indicated). CID 1067700 or DMSO (nil) was added either at the time when B cell stimulation started, or 66 h after B cells were stimulated with LPS plus IL-4, TGF-β, anti–δ/dex and RA, as indicated, for analysis of plasma cell survival (see below). Analysis of CSR, cell proliferation and survival To analyze IgG-expressing B cells and plasma cells in vivo, spleen cells (2 × 106) were first stained with fluorophore-labeled mAbs to surface markers and 7–AAD (Supplemental Table 1). After washing, cells were resuspended in the BD Cytofix/CytopermTM buffer (250 μl, BD Biosciences) and incubated at 4°C for 20 m. After washing twice with the BD Perm/WashTM buffer, cells were stained in the same buffer with fluorophore-labeled mAb to IgM, IgG1, IgG2a or IgG2b followed by washing for flow cytometry analysis. Dead (7–AAD+) cells were excluded from analysis. For CSR analysis of B cells stimulated in vitro, B cells were analyzed by flow cytometry for surface expression of IgG and IgA as well as intracellular expression of IgE, as we have described (10, 21). For B cell proliferation analysis in vivo, mice were injected i. p. with 2 mg of bromodeoxyuridine (BrdU) in 200 μl PBS twice, with the first and second injection at 24 h and 20 h prior to sacrificing, respectively. Spleen B cells were analyzed for BrdU incorporation (by anti–BrdU mAb staining) and DNA contents (by 7–AAD staining after cells were permeablized) using a BrdU Flow Kit® (BD Biosciences) following the manufacturer’s instructions. For B cell proliferation analysis in vitro, naïve B cells were labeled with CFSE (Invitrogen) following the manufacturer’s instructions and then cultured for 96 h in the presence of appropriate stimuli. Cells were analyzed by flow cytometry for CFSE intensity, which was reduced by half after completion of each cell division, as CFSE-labeled cell constituents were equally distributed between daughter cells. The number of B cell divisions was determined by the “Proliferation Platform” of the FlowJo® software. To analyze B cell and plasma cell survival in MRL/Faslpr/lpr mice treated with CID 1067700 or DMSO in vivo, spleen cells were stained with anti–CD19 mAb, anti–CD138 mAb, anti–TCR mAb (to identify TCR−CD19+CD138− B cells and TCR−CD19−/loCD138+ plasma cells) in the presence of mAb Clone 2.4G2, which blocks the FcγIII and FcγII receptors, and 7–AAD without permeabilization (to identify apoptotic/necrotic cells) and analyzed by flow cytometry. To analyze in vitro survival of immune cells isolated from MRL/Faslpr/lpr mice, spleen cells were cultured in RPMI-FBS for 24 or 48 h in the presence of CID 1067700 or DMSO (nil). After staining with surface markers to identify different cell types and 7–AAD, cells were analyzed by flow cytometry. To analyze survival of in vitro generated plasma cells (after B cell stimulation with LPS plus IL-4, IL-5, TGF-β, anti–δ/dex and RA for 66 h), cells were washed with RPMI-FBS twice and cultured in RPMI-FBS, without any additional stimuli to block further plasma cell generation, for 24, 48 and 72 h, in the presence of nil or CID 1067700 (40 μM). Cells were collected and stained with mAbs to CD138 and CD19 (to mark plasma cells and B cells) as well as with Annexin V and 7–AAD to distinguish live cells (Annexin V−7–AAD−) from cells undergoing early (Annexin V+7–AAD−) and late (Annexin V+7–AAD+) apoptosis. Retrovirus transduction Retroviral vectors pMIG and pMIG-AID (encoding AID) were as described (10); pMIG-Cre (encoding the Cre recombinase) was from Addgene; pMIG-Rab7 and pMIG-Rab2a (encoding Rab7 and Rab2a, respectively) were constructed using specific oligonucleotides (Supplemental Table 2). These vectors were co-transfected with the packaging plasmid into 293T cells to produce retroviruses, as described (10, 22). For transduction, B cells isolated from C57 or Rosa26+/fl-STOP-fl-Ikkβca mice were stimulated with LPS for 24 h in the presence of nil or CID 1067700, as indicated, incubated with viral particles that were pre-mixed with 6 μg/ml DEAE-dextran at 25°C for 30 m (Sigma-Aldrich). After incubation at 37°C for 5 h with gentle mixing every hour, cells were centrifuged at 500 g for 1 h and then 1,000 g for 4 m. Transduced B cells were cultured in virus-free FBS-RPMI in the presence of LPS plus IL-4 for 96 h and then harvested for flow cytometry analysis of expression of GFP (indicating expression of exogenous genes) and IgG1. Rosa26+/fl-STOP-fl-Ikkβca B cells expressed IKKβCA (as well as the GFP from the same bi-cistronic transcript) from the Rosa26 locus upon expression of the Cre recombinase and, consequently, deletion of the “STOP” cassette. Dead (7–AAD+) cells were excluded from analysis. To enforce expression of IKKβCA in plasma cells pre-generated from B cells stimulated with LPS plus IL-4, TGF-β, anti–δ/dex and RA for 66 h, cells were transduced with pMIG or pMIG-Cre virus, as above, and harvested after 72 h to analyze by flow cytometry the expression of GFP (indicating transduced cells) and plasma cell proportion and viability. Clinical and pathological analysis of MRL/Faslpr/lpr mice Skin lesions, lymphadenopathy and urine protein contents were assessed weekly. Skin lesions were scored on a scale of 0 to 4, with 0 = none, 1 = mild (snout only), 2 = moderate (< 2 cm in snout, ears and back), and 3 = severe (> 2 cm in snout and ears). Lymphadenopathy, as indicated by the appearances of lumps, was scored on a scale of 0 to 4, with 0 = none; 1 = small (at one site), 2 = moderate (at two sites), 3 = large at three or more different sites, and 4 = extremely large (at more than three sites). It was further confirmed by the increase in size of brachial and inguinal lymph nodes when mice were sacrificed. Proteinuria was assessed and scored using semi-quantitative Albustix® strips (Bayer), which, despite giving a range of serum albumin levels for each scale (scale 0 for negative or trace, and 1 through 4), generated few false-positive results when the score was above 3, as observed in virtually all 17-week old MRL/Faslpr/lpr mice. Splenomegaly was assessed when mice were sacrificed by the spleen size. To assess kidney pathology, kidneys were fixed in Tissue-Tek® OCTTM compound (Sankura Finetek) on dry ice. Seven-μm sections, as prepared by a Cryostat (Leica®), were fixed in cold acetone and stained with FITC-labeled anti–IgG1/anti–IgG2a mAbs. Sections were mounted using ProLong® Gold with DAPI (Invitrogen) for confocal microscopic analysis. For periodic acid–Schiff (PAS) staining, kidneys were fixed in paraformaldehyde (3.6%) at RT for 72 h and, after washing twice with PBS, embedded in paraffin. Five-μm sections were prepared and processed with periodic acid–Schiff (PAS) stain. Glomerular change severity was graded based upon glomerular activity, including glomerular cell proliferation (particularly mesangial matrix expansion), leukocyte infiltration and cellular crescents. Mesangial matrix expansion was grades as follows: 0 = no increase (matrix occupied up to 10% of the glomerulus), 1 = mild (10 – 25%), 2 = moderate (25 – 50%), and 3 = marked (50 – 100%). Leukocyte infiltration is graded as: 1 = mild, 2 = moderate, 3 = extensive. The severity of interstitial mononuclear cell infiltration was based on the ratio of infiltrated areas over the entire section (the value of area was determined by the ImageJ® software), quantified as the average of three sections 200 μM apart, and graded as: 1 = mild, 2 = moderate, 3 = severe. ELISA, ELISPOT and ANA assays To determine titers of total IgM, IgG1, IgG2a and IgG2b, sera were first diluted 4 to 128-fold with PBS (pH 7.4) plus 0.05% (v/v) Tween-20 (PBST). Two-fold serially diluted samples and standards for each Ig isotypes were incubated in 96-well plates pre-treated with sodium carbonate/bicarbonate buffer (pH 9.6) and coated with pre-adsorbed goat anti–IgM or anti–IgG (to capture IgG1, IgG2a and IgG2b) Abs (all 1 mg/ml; Supplemental Table 1). After washing with PBST, captured Igs were detected with biotinylated anti–IgM, –IgG1, –IgG2a or –IgG2b Abs (Supplemental Table 1), followed by reaction with horseradish peroxidase (HRP)-labeled streptavidin (Sigma-Aldrich), development with o-phenylenediamine and measurement of absorbance at 492 nm. Ig concentrations were determined using Prism® (GraphPad) or Excel® (Microsoft). To analyze titers of antigen-specific antibodies (high-affinity NP-binding or anti–dsDNA Abs), sera were diluted 1,000-fold in PBST. Twofold serially diluted samples were incubated in a 96-well plate pre-blocked with BSA and coated with NP4-BSA (in average 4 NP molecules on one BSA molecule) or dsDNA (10 μg/ml sonicated herring DNA). Captured Igs were detected with biotinylated Ab to IgM, IgG1, IgG2a or IgG3. Data are relative values based on end-point dilution factors. For ELISPOT analysis of total, NP-binding and dsDNA-binding ASCs, Multi-Screen® filter plates (Millipore) were activated with 35% ethanol, washed with PBS and coated with anti–IgM, anti–IgG, NP4-BSA or dsDNA (all 5 μg/ml) in PBS. Single spleen or bone marrow cell suspensions were cultured at 50,000 cells/ml in FBS-RPMI supplemented with 50 μM β-mercaptoethanol at 37°C for 16 h. After supernatants were removed, plates were incubated with biotinylated goat anti–mouse IgM, –IgG1, –IgG2a or –IgG3 Ab, as indicated, for 2 h and, after washing, incubated with HRP-conjugated streptavidin. Plates were developed using the Vectastain AEC peroxidase substrate kit (Vector Laboratories). The stained area in each well was quantified using the CTL Immunospot software (Cellular Technology) and depicted as the percentage of total area of each well for ASC quantification. For semi-quantitative ANA assays of sera from MRL/Faslpr/lpr and C57/Sle1Sle2Sle3 mice, sera were serially diluted in PBS (from 1:40 to 1:160), incubated on antinuclear Ab substrate slides (HEp-2 cell-coated slides, MBL-BION) and detected with a 1:1 mixture of FITC-labeled anti–IgG1 and anti–IgG2a mAbs (Supplemental Table 1) following manufacturer’s instructions. Images were acquired with an Olympus CKX41 fluorescence microscope. Cytokine intracellular staining Spleen cells were cultured in RPMI-FBS and stimulated with phorbol 12-myristate 13-acetate (PMA, 20 ng/ml) plus ionomycin (1 μg/ml; both from BioLegend) for 4 h at 37°C in the presence of brefeldin A (BFA). After pelleting, cells were stained with anti–CD4 mAb and Fixable Viability Dye eFluo®450 (FVD eFluo®450) followed by incubation with the BD Cytofix/CytopermTM buffer at 4°C for 20 m. After washing twice with the BD Perm/WashTM buffer, cells were resuspended in Hank’s Buffered Salt Solution (HBSS) with 1% BSA and stored overnight at 4°C. Cells were then stained with anti–IFN-γ and anti–IL-17 mAbs conjugated to different fluorophores in Perm/WashTM buffer and analyzed by flow cytometry. Dead (FVD eFluo® 450+) cells were excluded. Myeloid cell differentiation and function Bone marrow cells (107) were cultured for 10 d in RPMI-FBS in the presence of conditioned media from L-929 cells that express macrophage colony-stimulating factor (M-CSF, which promotes differentiation into CD11b+ macrophages) or recombinant murine Flt3-ligand (which promotes differentiation into CD11c+ DCs, Peprotech), resulting in generation of over 90% of CD11b+ or 20% of CD11c+ cells. After stimulation with TLR9 ligand CpG ODN for 1 h, cells were collected for RNA extraction and transcript analysis. GST-RILP pull-down RILP is an effector protein of activated Rab7 (Rab7-GTP). The E. coli strain BL21 harboring GST or GST-RILP expressing vector, as previously described (23), was grown at 37°C to an O.D. of 0.6 before induced by isopropyl-1-thio-β-D-galactopyranoside (IPTG, 0.5 mM) at 30°C for 4 h to express GST or GST-RILP. After pelleting and washing with cold PBS, bacterial cells were resuspended in 5 ml of cold lysis buffer (25 mM Tris-HCl pH 7.4, 1 M NaCl, 0.5 mM EDTA, 1 mM dithiothreitol and 0.1% Triton X-100) supplemented with a protease inhibitor cocktail (Sigma) and sonicated. Lysates were cleared and, after 5 ml of cold lysis buffer was added, incubated with 300 μl of pre-equilibrated slurry (50% packed volume) of glutathione-Sepharose 4B beads (GE Healthcare) at RT for 30 m. After washing with lysis buffer, beads (30 μl) with immobilized GST or GST-RILP were resuspended as slurry for protein quantification by the Bradford assay and pre-equilibrated in pull-down buffer (20 mM HEPES, 100 mM NaCl, 5 mM MgCl2 and 1% Triton X-100 plus protease inhibitor cocktail). Beads were then incubated by rotating with lysates (300 μg) prepared from stimulated B cells in the pull-down buffer at 4°C overnight. After washing twice with cold pull-down buffer, bound proteins were eluted by incubation of beads in the SDS-PAGE sample buffer at 95°C for 10 m. Immobilized GST-RILP was analyzed by SDS-PAGE/immunoblotting using anti–GST mAb; Rab7-GTP was analyzed by immunoblotting using anti–Rab7 mAb. Purification of recombinant Rab7 and NMR spectroscopy analysis DNA sequence encoding Rab7 (human RAB7A amino acid residues 2 – 207) was synthesized using codon optimization for E.coli expression and cloned into the pET30a vector, which allows for expression of an N-terminal Strep-tag followed by a SUMO tag. After plasmid transformation, BL21(DE3) cells were grown at 37°C in minimum media until the culture OD600 reached 0.6. After the culture was transferred to 18°C, protein expression was induced by IPTG (1 mM) for 15 h. For protein purification, lysates were first subjected to streptactin affinity chromatography and then removal of the Strep-SUMO tag by digestion of SUMO protease at 4°C overnight, resulting in release of cleaved Rab7 (with an additional N-terminal Thr residue) into the supernatants. Trace amounts of cleaved tag were removed by a new round of strep-tactin affinity chromatography. Protein purity was > 90%. For NMR spectroscopy analysis of free RAB7A, RAB7A (0.3 mM) was exchanged into the buffer containing 20 mM NaPO4 (20 mM, pH 6.9), NaCl (100 mM) and TCEP (1 mM). with 10% D2O. 2D 1H-15N HSQC was acquired with 2048 and 128 complex points in the direct and indirect dimensions, respectively. For acquisition of the NMR spectrum of RAB7A in complex with CID 1067700, 2.5 μl of CID 1067700 stock solution (40 mM in DMSO) was added to 40 μl of RAB7A to yield final 1:8 protein:compound ratio, followed by NMR spectrum acquisition using the same parameters. Docking analysis All modeling studies were performed using the Schrodinger 2015-3 software suite and its graphical interface, Maestro. The crystal structure of the GTP-bound Rab7 (PDBID: 1T91) was used as the docking target for CID 1067700. The bound GTP molecule and all water molecules were removed from the structure prior to docking. Docking was performed with Glide using first standard precision protocol to identify all possible poses with favorable Glide scores. Chlamydia titration For monitoring live organism shedding from swab samples, vaginal/cervical swabs were taken at the time points, as indicated. Each swab was soaked in 0.5 ml of sucrose-phosphate-glutamic acid (SPG) and vortexed with glass beads to release chlamydial organisms into supernatants. IFUs were titrated on HeLa cell monolayers in duplicates. The infected cultures were processed for immunofluorescence assays. Briefly, inclusions were counted in five random fields per coverslip under a fluorescence microscope. For coverslips with less than one IFU per view field, IFUs on entire coverslips were counted. Coverslips showing obvious cytotoxicity of HeLa cells were excluded. The total number of IFUs was calculated based on the mean IFUs per view, the ratio of the view area to that of the well, the dilution factor and inoculation volumes. Flow cytometry, immunoblotting, cDNA synthesis and quantitative RT-PCR (qRT-PCR) These methods were as we previously described (7, 10, 22). Antibodies (Supplemental Table 1) and oligonucleotides (Supplemental Table 2) used had been validated to be specific. Flow cytometry data were analyzed using the FlowJo® software (Tree Star). For immunoblotting, signals were quantified by Image J (NIH). For qRT-PCR, the ΔΔCt method was used to determine levels of transcripts and data were normalized to levels of CD79B (human) or Cd79b (mouse), which encodes the BCR Igβ chain that is constitutively expressed in B cells and, to a lesser degree, plasma cells. Statistical analysis Most statistical analyses were performed by one-tailed type 1 (for pairwise comparison) or type 2 student t-test. Survival data were analyzed by the Mantel-Cox log-rank test. Chlamydia infection data were analyzed by the Wilcoxon rank sum test. P values less than 0.05 were considered significant. More Materials and Methods in Supplemental Materials Results Rab7 is highly expressed in human and mouse lupus B cells Rab7 is upregulated in normal B cells by CSR-inducing stimuli (10), prompting us to analyze its expression in B cells from SLE patients and lupus mice. Peripheral blood mononuclear cells (PBMCs) from SLE patients expressed higher levels of RAB7A and AICDA transcripts than healthy subjects (Fig. 1A). Likewise, B cells, including PNAhi B cells, from lupus-prone MRL/Faslpr/lpr and C57/Sle1Sle2Sle3 mice displayed elevated Rab7 and Aicda transcripts as compared to those from age-matched C57 counterparts (Fig. 1B, 1C). A similar difference was observed in Rab7 and AID protein levels in lymphoid organs (Fig. 1D). Thus, lupus B cells dysregulate Rab7 expression. Targeting of Rab7 by CID 1067700 inhibits NF-κB activation and AID induction in B cells In view of treating lupus-prone mice with Rab7 inhibitor CID 1067700, we first tested the ability of this small molecule to inhibit CSR in vitro. CID 1067700 treatment of mouse B cells stimulated by CD154 or LPS plus IL-4 impaired Rab7 activation, as shown by reduced Rab7-GTP levels in RILP pull-down assays but normal Rab7 expression (Fig. 2A, 2B). This led to reduced AID expression and CSR to IgG1, despite normal B cell proliferation and germline IH-CH transcription, both of which are required for CSR (Fig. 2B–2D). Consistent with the requirement of AID for CSR to all Ig isotypes, CID 1067700 inhibited induction of CSR to IgG, IgA and IgE in human and mouse B cells (Fig. 2E–2H). CID 1067700 binds Rab7 with a high affinity (EC50: 10-20 nM), but was also shown to bind several small GTPases with much lower affinity (i.e., with EC50 being at least 80 nM) and could affect functions of these GTPases in cell lines (16, 24). In our hands, it directly bound purified recombinant Rab7 (Fig. 3A), likely through two binding sites on the surface of Rab7, as suggested by computational docking of CID 1067700 onto the crystal structure of Rab7, with one site overlapping with the GTP binding site and the other located on the opposite face (Fig. 3B). In stimulated primary B cells, which upregulate Rab7 expression, CID 1067700 specifically targeted Rab7, as its inhibitory activity was abrogated when Rab7 was ablated in Igh+/Cγ1-creRab7fl/fl conditional knockout B cells (Fig. 3C) – these knockout B cells showed reduced CSR (10). Conversely, CSR in CID 1067700-treated B cells was rescued by enforced expression of Rab7, but not Rab2a, a putative off-target of CID 1067700 (Fig. 3D). It was also increased by AID, which could rescue CSR in Rab7 knockout B cells (10). Like Rab7 deficiency, inhibition of Rab7 activity by CID 1067700 resulted in defective canonical NF-κB activation and normal non-canonical NF-κB activation (Fig. 2B). Expression of IKKβCA, a constitutively active mutant of IKKβ that can lead to canonical NF-κB hyperactivation (19) restored CSR in CID 1067700-treated B cells (Fig. 3E), indicating that the major defect in these cells is the impairment of canonical NF-κB activation. Thus, specific Rab7 inhibition in B cells hampers NF-κB activation, AID expression and CSR induction without affecting B cell proliferation. CID 1067700 blunts pathogenic autoantibody responses in lupus mice The Rab7 upregulation in lupus B cells, together with the role of Rab7 in NF-κB activation, AID expression and CSR, all of which are also elevated in lupus B cells and implicated in lupus pathogenesis (25, 26), suggested that Rab7 inhibition would hamper NF-κB-dependent pathogenic autoantibody responses in vivo, e.g., in MRL/Faslpr/lpr mice. These mice developed malar rash and dorsal skin lesions, lymphadenopathy, splenomegaly and nephritis, leading to mortality as early as at 11-week of age and an average life-span of about 20 weeks, with more than half of mice moribund between 17- and 23-week of age (Fig. 4A, 4B). When treated with CID 1067700 for only 10 weeks (starting at 10-week, one week before appearance of the first dead mouse in the untreated cohort), 14 of 16 MRL/Faslpr/lpr mice survived past the treatment period and half of mice were alive at 52 weeks, compared to only one in the untreated cohort (Fig. 4A). Treated mice had normal body weight, rare skin lesions and reduced lymphadenopathy and splenomegaly at 17 weeks (Fig. 4B–4D). They had some interstitial infiltration by monocytes (likely myeloid cells and T cells), but were largely free of glomerulonephritis, showing little proteinuria and decreased “crescent” formation and mesangial matrix expansion in glomeruli (Fig. 4E, 4F and not shown). Lupus nephritis is tightly associated with and/or caused by deposition of immunocomplexes (ICs) that contain class-switched IgG autoantibodies in the kidney. IgG-IC deposition was reduced in MRL/Faslpr/lpr mice treated with CID 1067700 (Fig. 5A). Serum levels of IgG anti-nuclear antibodies (ANAs) were also decreased in treated MRL/Faslpr/lpr and C57/Sle1Sle2Sle3 mice (Fig. 5B). As shown by time-course analyses of autoantibodies that bound double-strand DNA (anti–dsDNA), the ratio of pathogenic IgG classes, e.g., IgG1, IgG2a and IgG2b, over the non-pathogenic IgM class (IgG/IgM, depicted as RIgG) remained low in treated mice despite variations of IgG and IgM titers in different mice in the same cohort (Fig. 5C and data not shown). So did RIgG for total IgG, in contrast to the steadily increased anti–dsDNA and total RIgG, until peaking, in untreated mice (Fig. 5C). Thus, inhibition of Rab7 suppresses IgG autoantibody responses and prevents development of disease symptoms in lupus mice, leading to significantly increased life-span. CID 1067700 targets B cells and specifically impairs the CSR machinery in vivo The reduction of RIgG suggested that the CSR process was inhibited in CID 1067700-treated MRL/Faslpr/lpr mice. Indeed, these mice showed unchanged number of B cells but significantly decreased IgG+ B cells, concomitant with a slightly increased proportion of un-switched IgM+ B cells (Fig. 6A). Decreased CSR, as also shown by reduced post-recombination Iμ-Cγ transcripts (the molecular indicators of completed CSR), was associated with lower expression of CSR machinery genes, such as Aicda and 14-3-3γ, which is important for the targeting of AID to IgH switch (S) regions and is induced in an NF-κB-dependent manner (22, 27, 28); Rab7 expression was also slightly reduced (Fig. 6B). Expression of the A20 (Tnfaip3) gene, which dampens NF-κB activation and controls autoimmunity (29), or germline Iμ-Cμ transcription, however, was not affected; neither was expression of Prdm1 (encoding Blimp-1), Xbp1, which is essential for the plasma cell secretion function, or genes (Vps34, LC3 and Becn1) implicated in autophagy, which regulates plasma cell differentiation/functions (30, 31). Treated B cells were normal in proliferation, CD21/CD35 expression, induction of the CD80 activation markers and expression of CD40 and MHC II, which are important for B cells to interact with CD4+ T helper cells (Fig. 6C), thereby emphasizing the inherent nature of the CSR machinery defect. Rab7 has been suggested to be expressed in myeloid cells, in addition to B cells (32), prompting us to analyze the impact of Rab7 inhibition on CD11b+ macrophages and CD11c+ dendritic cells (DCs). The proportions of these cells, their survival in vitro and induced expression of cytokines that could influence B cells and autoantibody responses, such as BAFF and type I interferon (IFNα and IFNβ), were not affected by CID 1067700 treatment in MRL/Faslpr/lpr mice, with the only exception of reduced Il1b expression in CD11c+ DCs (Fig. 7A–7C). Reflecting the low expression of Rab7 in T cells and a modest role of this molecule in T cell functions (33), CID 067700 treatment did not change the proportions of total CD4+ T cells and those producing IFN-γ, which is critical for CSR to IgG2a and important in systemic autoimmunity (34–36), in MRL/Faslpr/lpr mice or survival of CD4+ lupus T cells in vitro (Fig. 7C, 7D). This, together with the ability of CID 1067700-treated C57 mice to clear Chlamydia infection, which is dependent on T cells, but not B cells or plasma cells (37, 38), showed that this small molecule does not have a major impact on T cells (Fig. 7E). Thus, CID 1067700 selectively targets B cells and impairs the CSR machinery in lupus-prone mice without altering proliferation/activation of B cells. Rab7 deficiency or Rab7 inhibition reduces ASCs Consistent with the surge of IgG+ ASCs during lupus flare, most cells expressing the plasma cell marker CD138 in MRL/Faslpr/lpr mice with active lupus (e.g., at 17-week) were IgG+, as shown by intracellular staining (Fig. 8A). Upon Rab7 inhibition, IgG+ plasma cells were greatly decreased (due to defective CSR) with a concomitant increase in the proportion of IgM+ plasma cells. However, the number of ASCs that produced IgM autoantibodies (e.g., anti–dsDNA) remained unchanged (Fig. 8B). This together with the lower number of IgG1+ and IgG2a+ ASCs indicated an overall reduction in ASCs upon treatment with CID 1067700. This Rab7 inhibitor had similar effect on ASCs in C57 mice, as indicated by the reduced antigen-specific IgG1+ ASCs and unchanged IgM+ ASCs in the spleen of mice injected with a T-dependent (NP-CGG) or T-independent (NP-LPS) antigen, which resulted in virtual abrogation of circulating antigen-specific IgGs (Fig. 8C). Likewise, Rab7 knockout in B cells and their plasma cell progenies, as occurring in Igh+/Cγ1-creRab7fl/fl mice, reduced total and antigen-specific IgG1+ ASCs (our previous findings, (10)) with no increase in IgM+ ASCs (Fig. 8D). ASCs in the bone marrow, in which long-lived plasma cells take residence (39), were even more sensitive to Rab7 inhibition or knockout, as both IgG+ and IgM+ ASCs were decreased (Fig. 8C, 8D). Finally, IgG+ ASCs preformed in vivo were vulnerable to CID 1067700 treatment, as they failed to produce antibodies ex vivo (Fig. 8E). Thus, abrogation of Rab7 expression or inhibition of its activity in normal and lupus-prone mice affects ASC production/maintenance, thereby compounding the defect in CSR to suppress class-switched antibody and autoantibody responses. Rab7 inhibition impairs plasma cell survival Despite their impact on ASCs (consisting of plasmablasts and plasma cells), neither CID 1067700 nor Rab7 knockout affected in vitro generation of CD138hi cells upon induction of B cells by all tested stimuli, such as LPS plus IL-4, TGF-β, BCR crosslinking and retinoic acid (RA), which led to more than 70% of live cells being CD138hi (Fig. 9A and not shown). This suggested that Rab7 was not involved in plasma cell generation and prompted us to analyze the effect of CID 1067700 on plasma cell survival. In MRL/Faslpr/lpr mice, CID 1067700 treatment resulted in increased death and reduced number of CD138hi cells (Fig. 9B, 9C). Even residual live CD138hi cells expressed lower levels of Cxcr4, Il6r and Vla4, all of which encode surface receptors that mediate plasma cell survival – by contrast, expression of Bcma and Taci, which are important for APRIL-dependent survival, was not affected (Fig. 9D). Expression of intracellular factors important for plasma cell survival (such as Irf4, Mcl1, and Hdac11) as well as the Atg5 autophagy gene was also affected. Expression of Prdm1 or Xbp1, which are important to maintain the plasma cell secretory function but dispensable for their survival (40), however, was not affected. To further confirm that Rab7 inhibition directly affects plasma cell survival, we treated CD138hi cells generated in vitro with CID 1067700. These cells displayed high levels of early and late apoptosis, as compared to the much slower loss of viability in their untreated counterparts (Fig. 9E). Enforced NF-κB activation through IKKβCA expression prevented CD138hi cells from being killed by CID 1067700 (Fig. 9F), showing that Rab7 plays a role in the plasma cell survival in vivo and in vitro, in part by mediating NF-κB-activation. Discussion Stemming from our previous findings (10), data reported here have further outlined an important role of Rab7 in antibody and autoantibody responses, owing to its functions in modulating Aicda expression, B cell class-switching and plasma cell survival. In spite of the partial impairments of these processes by Rab7 inhibition or knockout, their combined outcome was the virtual abrogation of class-switched specific antibodies in normal mice and paucity of pathogenic autoantibodies in lupus-prone mice. As we showed, small molecule Rab7 inhibition hampered CSR induced by all primary stimuli, including the one (CD154) produced by T cells, which mediate many autoimmune conditions in part by dysregulating B cells, and a ligand (R-848) of TLR7, an endosomal TLR that strongly promotes lupus pathogenesis and does so in a B cell-intrinsic manner (41). Thus, Rab7, which is an endosome-tethered protein, participates in B lineage cell differentiation processes initiated by many immune receptors irrespective of their initial locations, e.g., CD40 and TLR4 on the surface and TLR7 in the endosome. Surface immune receptors, however, need to be internalized for Rab7 to mediate CSR and plasma cell survival, likely in a manner dependent on Rab7+ endosomes. As indicated by our results showing IKKβCA-mediated rescue of CSR and plasma cell survival in CID 1067700-treated cells, the B cell- and plasma cell-intrinsic roles of Rab7 are, at least in part, through activation of NF-κB, a transcription factor central to B lineage cell functions (42, 43). Rab7-dependent NF-κB activation, however, is dispensable for B cell development and differentiation into plasma cells, as suggested by unchanged numbers of CD19+ B cells in CID 1067700-treated mice in vivo and generation of plasma cells from CID 1067700-treated B cells or Rab7 knockout B cells in vitro. It would also be absent during early activation of naïve B cells upon CD40 or TLR engagement, as these cells express low levels of Rab7 and contain sparse intracellular membrane structures. Only after Rab7 upregulation, likely together with expansion of the intracellular membrane network, would Rab7 nucleate the assembly of putative “intracellular signalosomes” on different membrane types (e.g., endosomes in B cells and autophagosomes in plasma cells) – through mechanisms that are not yet clear – for sustained NF-κB activation. This would be required for efficient AID induction in CSR, which takes at least 48 h to unfold, and continuous expression of genes in plasma cells to maintain survival. By contrast, Rab7-independent early NF-κB activation would occur through different signalosomes, e.g., those on the plasma membrane in B cells with cell surface receptor engagement. It also plays a role in the Rab7 gene transcription, possibly through several κB sites in the promoter region (unpublished data). Rab7 is inherently the preferred target of CID 1067700, as shown by its much lower EC50 as compared to the few small GTPase off-targets of this compound (16). As suggested by our docking analyses, CID 1067700 could potentially bind at two distinct sites on the Rab7 surface. Existence of a candidate binding site that is different from the GTP-binding site (the site that also exists in potential off-targets of CID 1067700) would, at least in part, explain the Rab7-specific inhibitory activity of CID 1067700. This together with upregulated Rab7 expression in B cells (and likely in plasma cells), including in PNAhi germinal center B cells in immunized mice, would underpin the specific inhibition of Rab7 by CID 1067700, as more molecules would be available for targeting. Likewise, the dysregulated Rab7 expression would make lupus B cells preferred targets of CID 1067700. Such dysregulation may be the consequence of a putative positive-feedback loop involving NF-κB hyperactivation (25, 26). It could also be exacerbated by female hormones, as suggested by enhancement of Rab7 expression by estrogens (our unpublished data). This together with estrogen upregulation of AID expression through the HoxC4 homeodomain transcription factor would contribute to the female bias of autoantibody-mediated systemic autoimmunity (11, 44, 45). In the subset of human lupus B cells that displayed moderate Rab7 expression, as shown in our studies, NF-κB hyperactivation may be through alternative mechanisms, such as A20 downregulation (29). Rab7 inhibition reduced expression of several genes instrumental to plasma cell survival, such as Cxcr4, Irf4 and Mcl1 (40, 46, 47). Expression of CXCR4 (a G protein-coupled receptor) has been shown to be boosted by NF-κB, consistent with the identification of several κB sites in the Cxcr4 gene locus (48), consistent with a role of Rab7 in NF-κB activation and rescue of survival of CID 1067700-treated plasma cells with enforced expression of IKKβCA. The presence of several Irf4-binding sites in the Cxcr4 promoter (49) also suggests that Cxcr4 transcription is regulated by Irf4, whose expression would be mediated by Rab7 (shown here) and promoted by NF-κB (50). Expression of Mcl1 in plasma cells is in general lower than that in B cells (51), perhaps explaining higher sensitivity of plasma cells to death upon Rab7 inhibition. Rab7 deficiency or inhibition in mice may also impair the autophagy-dependent antibody secretion, which together with decreased plasma cell survival would lead to marked reduction in ASCs. Like anti–dsDNA-specific ASCs in the spleen, kidney resident plasma cells that secret pathogenic nephrophilic antibodies (52–54) may be reduced, leading to significantly reduced kidney damage. As a small molecule, CID 1067700 injected i. p. was “metabolized” in vivo and required weekly administration to maintain detectable levels in the circulation and lymphoid organs (our unpublished observations). Despite this, Rab7 inhibition by CID 1067700 could prevent development of lupus pathology in MRL/Faslpr/lpr mice and extend the lifespan by more than 32 weeks, after a treatment period of only 10 weeks, suggesting that long-lasting changes had occurred in treated B cells – treated plasma cells would die. These changes likely include alterations in the epigenome in lupus B cells with extended survival (due to the Fas/lpr mutation) as well as changes in the BCR repertoire and/or the memory autoreactive B cell compartment. Putative epigenetic changes might occur through downregulation of NF-κB-dependent expression of histone modifying enzymes and/or re-balance of exaggerated Rab7-dependent autophagy, which modulates the microRNA machinery (15, 55). B cell-intrinsic mechanisms may be further complemented by changes in regulatory immune elements, such as T regulatory cells and DCs (56), possibly including altered type-I IFN gene signatures in plasmacytoid DCs (pDCs), as short-term pDC ablation could also elicit long-term prevention of nephritis in the BXSB lupus mouse mode (57). Such DC changes, however, would be indirect, as CID 1067700 did not decrease cytokine production by DCs in vitro. High Rab7 expression levels in DCs, however, do suggest that Rab7 regulate certain DC functions, perhaps antigen/autoantigen presentation through autophagy induction (58) and MyD88 signaling, which mediates skin lesions (59). By contrast, Rab7 is largely dispensable for T cell survival and functions (e.g., in clearing primary Chlamydia infection, as shown here). Addressing specific roles of Rab7 in these and other immune compartments requires generation of different conditional knockout mice. In contrast to the non-discriminating B cell-depletion lupus therapeutics approaches, CID 1067700 or its future derivatives with improved pharmacodynamics and pharmacokinetics properties would maintain IgM+ B cells and protective IgMs. It is conceivable that a Rab7 inhibitor can be combined with anti–BAFF or anti–CD20 mAb towards better therapeutic effects (60–62). Such an inhibitor may also be combined with a histone deacetylase (HDAC) inhibitor (HDI) to abrogate plasma cells by affecting plasma cell survival and generation, respectively – synthetic and naturally occurring (e.g., butyrate, a metabolite of gut microbiota) HDIs blunt antibody and autoantibody responses by inhibiting CSR/SHM and B cell differentiation into plasma cells, but not plasma cell survival (3, 21). The translational implications of Rab7-dependent NF-κB activation extend beyond lupus. Rab7 may promote B cell lymphomagenesis mediated by NF-κB hyperactivation (63–65), as suggested by the frequent DNA insertions/deletions and chromosomal translocations in/around the human RAB7A locus on the chromosome 3 (3q21) in hematologic malignancies (66–68). It may also play a role in NF-κB-mediated survival of multiple myelomas (69), as it does in maintaining the survival of plasma cells. Addressing these possibilities would greatly expand our studies towards the understanding of the role of Rab7 in immune regulation. Supplementary Material 1 This work was supported by National Institutes of Health (NIH) grants AI 079705 and AI 105813 (to P. C.), AI 124172 (to Z. X.), AI 104476 (to D. N. I.) and AI 047997 (to G. Z.). P. C. was also partially supported by the Alliance for Lupus Research (ALR) Target Identification in Lupus Grant ALR 295955 and the Zachry Foundation Distinguished Chair, D. N. I. by a Voelcker Fund Young Investigator Award, H. Z. by an Arthritis National Research Foundation research grant and R. W. by the Xiangya School of Medicine, Central South University of China. We thank Crystal Lafleur for technical help, Tian Shen for help in plasma cell induction, Egest J. Pone and Christie-Lynn Mortales for ELISPOT, Tamara McRae for PAS kidney histology analysis, Dr. Dmytro Kovalskyy for docking analyses, Dr. Benjamin J. Daniel and Karla M. Gorena in the UTHSCSA Flow Core Facility for help with cell sorting, and the UTHSCSA NMR and Analytical Ultracentrifugation Core Facility (supported in part by the NIH P30 CA054174 to the Cancer Therapy and Research Center of UTHSCSA) for NMR analysis. We thank Dr. Aimee L. Edinger for the RILP-expressing construct, Dr. Nu Zhang for help with cytokine intracellular staining and Dr. Xiaodong Li for help with myeloid cell differentiation assays. Abbreviations used AID activation-induced cytidine deaminase ASC antibody-secreting cell CSR class switch DNA recombination Rab7 Ras-related in brain7 SLE systemic lupus erythematosus FIGURE 1 Rab7 is highly expressed in human and mouse lupus B cells. (A) qRT-PCR analysis of levels of RAB7A and AICDA transcripts in PBMCs isolated from lupus patients or healthy subjects (n = 9). Expression of RAB7B, which is unrelated to RAB7A and is not evolutionarily conserved, was comparable in these samples (data not shown). (B) qRT-PCR analysis of Rab7 and Aicda transcripts in spleen B cells isolated from 10-week old MRL/Faslpr/lpr mice and 36-week old C57/Sle1/Sle2/Sle3 mice. Data are expressed as ratios of values to those in B cells from age-matched C57 mice (mean and s.d. of data from three independent experiments). (C) qRT-PCR analysis of Rab7 and Aicda transcripts in sorted spleen CD19+PNAhi cells (>95% pure, which displayed higher Rab7 expression than CD19+PNAlo cells, not shown) from 10-week MRL/Faslpr/lpr mice and NP-CGG-immunized C57 mice. Representative of two independent experiments (mean and s.e.m. of triplicate samples). (D) Immunoblotting analysis of Rab7 and AID in spleen and lymph node cells isolated from 10-week old MRL/Faslpr/lpr and C57 mice as well as CH12 B cells. FIGURE 2 CID 1067700 inhibits Rab7 activity, NF-κB activation, AID expression and CSR in B cells. (A) GST-RILP pull-down analysis of Rab7-GTP (bottom panels) in B cells stimulated with LPS plus IL-4 for 48 h in the presence of nil or CID 1067700. Expression of total Rab7 was also analyzed (top panels). (B) Immunoblotting of Rab7, phosphorylated and total p65 in the canonical NF-κB pathway, p52 in the non-canonical NF-κB activation pathway, AID and β-actin in stimulated B cells treated with nil or CID 1067700. Numbers on the right side indicate non-normalized values of quantified signals. (C) qRT-PCR analysis of levels of IgH germline Iμ-Cμ and Iγ1-Cγ1 transcripts, Aicda, post-recombination Iμ-Cγ1 and circle Iγ1-Cμ transcript in B cells stimulated with CD154 or LPS plus IL-4 and treated with different doses of CID 1067700. Data are expressed as ratios of values in CID 1067700-treated B cells to those in untreated cells (mean and s.e.m. of data from triplicate samples). (D) Flow cytometry analysis of proliferation and CSR to IgG1 in CFSE-labeled B cells after stimulation in the presence of nil or CID 1067700 (top panels) and depiction of the proportion of switched IgG1+ cells in B cells that had completed each cell division (bottom panels). (E–H) CSR to different Ig isotypes in human or mouse B cells stimulated by different combinations of a T-dependent or T-independent primary CSR-inducing stimulus and a secondary inducing stimulus, as indicated, in the presence of nil or CID 1067700. (A, B, E–H), representative of three independent experiments. FIGURE 3 CID 1067700 inhibit CSR in B cells by targeting Rab7. (A) NMR spectroscopy analysis of direct physical interaction of CID 1067700 with Rab7. Spectra of free Rab7 and Rab7:CID 1067700 were represented by red and green signals, respectively, with overlapping signals shown in black. Chemical shift perturbations and broadening observed for NMR signals (non-overlapping signals) of multiple Rab7 residues are indicative of direct physical interaction. (B) Docking analysis of two candidate CID 1067700 binding sites on the opposite faces of Rab7 (left panel). The more favorable docking score (Glide score –5.5) was obtained for CID 1067700 binding at the GTP-binding site of Rab7 (lower left panel) and the second binding site (Glide score –2.7) located on the opposite face of Rab7 when CID 1067700 adopts a different conformation (lower right panel). The interaction interface within the GTP-binding site was slightly different from that predicted by others (16). (C) Flow cytometry analysis of CSR to IgG1 in B cells from Igh+/Cγ1-creRab7+/fl and Igh+/Cγ1-creRab7fl/fl littermates after stimulation with LPS plus IL-4 in the presence of nil or CID 1067700. Representative of three independent experiments. (D) CSR to IgG1 in B cells pre-stimulated with LPS in the presence of CID 1067700, transduced with pMIG or pMIG-Rab7 and then stimulated with LPS plus IL-4 for 72 h (left panels). The histogram (right panel) depicts CSR rescue by Rab7, Rab2a, a potential off-target of CID 1067700 with EC50 (170 nM) much higher than Rab7 (10-20 nM), or AID. Other potential off-targets (http://tinyurl.com/obgvd3v) are three small GTPases with EC50 comparable to Rab2a (i.e., H-Ras, 20-145 nM; Rac1, 80 nM; and Cdc42, 91-129 nM), much weaker targets (i.e., autophagy protein ATG4B, EC50 of 13 μM, and histone lysine methyltransferase G9a, EC50 of 39 μM), and growth factors/receptors (i.e., FGF22, GFER, VLA-4 and EGFR) that have not been independently verified as potential off-targets and, to our best knowledge, are not known to regulate CSR (mean and s.e.m. of three independent experiments). (E) CSR to IgG1 in B cells isolated from Rosa26+/fl-STOP-fl-Ikkβca mice and transduced with pMIG or pMIG-Cre (which encoded the Cre recombinase to delete the “STOP” cassette in the Rosa26 locus and, therefore, allow for expression of IKKβCA) and then stimulated with LPS plus IL-4 in the presence of nil or CID 1067700 (mean and s.e.m. of four experiments). FIGURE 4 Rab7 inhibition by CID 1067700 prevents disease development in lupus-prone mice. (A) Survival curve of 16 MRL/Faslpr/lpr mice treated with nil or CID 1067700. The survival of mice maintained in our facility was similar to that reported by The Jackson Laboratory and other groups. (B) Skin lesions in 26-week old nil-or CID 1067700-treated MRL/Faslpr/lpr mice. (C–F) Spleens and cervical lymph nodes (C), physiological metrics as indicated (D), indices of kidney damages (E) and PAS staining of kidney sections (F) in 17-week old MRL/Faslpr/lpr mice treated with nil or CID 1067700 (C and F, representative of three independent experiments; D and E, mean and s.d. of four pairs). FIGURE 5 Rab7 inhibition blocks generation of pathogenic autoantibodies. (A) IgG-IC formation in the kidney 17-week nil- or CID 1067700-treated MRL/Faslpr/lpr mice (representative of kidney sections in three independent experiments). (B) ANA levels in serum samples (diluted 40-fold) from MRL/Faslpr/lpr mice, as in (A), and 54-week old nil- or CID 1067700-treated C57/Sle1Sle2Sle3 mice. ANAs to different autoantigens, as shown by different patterns of fluorescence images at higher magnifications, were all reduced in CID 1067700-treated mice (representative of three independent experiments). More diluted serum samples, e.g., by 160-fold, from CID 1067700-treated mice showed virtually no signals (not shown). (C) Ratios of anti–dsDNA (top panels) and total (bottom panels) IgG1, IgG2a and IgG2b titers to titers of the IgM counterparts in nil- or CID 1067700-treated MRL/Faslpr/lpr mice at different ages, as indicated. Data were normalized to values in mice at the age of from 10-week (set as 1), right before the treatment started, and were depicted as RIgG1, RIgG2a and RIgG2b (mean and s.e.m. of four mice per group). IgG3 titers were generally low (not shown). FIGURE 6 Rab7 inhibition hampers CSR in lupus-prone mice. (A) Flow cytometry analysis of number of B cells (left panel) and proportion of IgM+, IgG1+ and IgG2a+ cells among B cells (right panels) in 17-week old nil- or CID 1067700-treated MRL/Faslpr/lpr mice (mean and s.d. of four pairs of mice) – the proportion of these B cells was increased due to the reduced splenomegaly (data not shown). Both nil- and CID 1067700-treated MRL/Faslpr/lpr mice had B lymphocytopenia, with average of 3.2 × 107 CD19+ B cells per spleen, as compared to about 5.5 × 107 B cells in age-matched C57 mice (data not shown). Moribund MRL/Faslpr/lpr mice were excluded from analyses due to their severe B lymphocytopenia. (B) Expression of different transcripts, as indicated, in CD19+ B cells sorted from MRL/Faslpr/lpr mice, as in (A). Data were depicted as the ratio of values in CID 1067700-treated mice to those in nil-treated mice (*, p < 0.05; **, p < 0.01). (C) Flow cytometry analysis of B cell expression of markers (left and middle panel sets), as indicated, and proliferation (right panel set) in MRL/Faslpr/lpr mice, as in (A). FIGURE 7 Rab7 inhibition does not affect myeloid cells or T cells. (A) Flow cytometry analysis of CD11b+ and CD11c+ cells in 17-week old nil- or CID 1067700-treated MRL/Faslpr/lpr mice. (B) Expression of cytokine-encoding genes, as indicated, in CD11b+ and CD11c+ cells generated from the bone marrow cells isolated from untreated 10-week old MRL/Faslpr/lpr mice and then stimulated by CpG ODN in vitro in the presence of nil or CID 1067700 (mean and s.d. of triplicate samples; *, p < 0.05). (C) Survival of CD11b+ and CD11c+ myeloid cells as well as CD4+ T cells isolated from untreated 10-week old MRL/Faslpr/lpr mice and cultured in vitro in the presence of nil or different doses of CID 1067700 for 24 h or 48 h (mean and s.d. of triplicate samples). (D) Flow cytometry analysis of proportion of CD4+ T cells and expression of IFN-γ in CD4+ T cells in nil- or CID 1067700-treated MRL/Faslpr/lpr mice, as in (A). Expression of IL-17 is low despite the important role of this cytokine in lupus (70). Consistent with reduced splenomegaly, the number of T cells was reduced, possibly as the secondary effect of reduced activity of autoreactive B cells. No CD19+ cells expressed IFN-γ or IL-17 (not shown). (E) Shedding of chlamydia in C57 mice treated with nil or CID 1067700 (mean and s.e.m. of 5 mice in each group). B cells are not required for the clearance of primary chlamydia infection. (A, C, D), representative of three independent experiments. FIGURE 8 Rab7 deficiency or inhibition impairs ASC production/functions in vivo and in vitro. (A) Levels of intracellular IgM, IgG1 and IgG2a in spleen CD138+ cells from 17-week old nil- or CID 1067700-treated MRL/Faslpr/lpr mice (representative of three independent experiments). (B) ELISPOT analysis of IgM+, IgG1+ and IgG2a+ ASCs in 17-week old nil- or CID 1067700-treated MRL/Faslpr/lpr mice (right panels, mean and s.e.m. of four pairs of mice). (C) ELISPOT analysis of IgM+ and IgG+ (IgG1 or IgG3, as indicated) NP-binding ASCs in the spleen and bone marrow, as well as titers of circulating NP-binding IgM and IgG1 or IgG3 Abs, in nil- or CID 1067700-treated C57 mice immunized with NP-CGG (top panels) or NP-LPS (bottom panels). (mean and s.e.m. of three pairs of mice). (D) ELISPOT analysis of IgM+ NP-binding and total ASCs in the spleen and bone marrow in Igh+/Cγ1-creRab7fl/fl mice and their Igh+/Cγ1-creRab7+/fl littermates immunized with NP-CGG (mean and s.e.m. of three pairs of mice). (E) ELISPOT analysis of IgG1+ ASCs isolated form the bone marrow of C57 mice and treated in vitro with nil or CID 1067700n for 24 h before incubation for 24 h in the presence of nil or CID 1067700, totaling 48 h treatment (mean and s.e.m. of three independent experiments). FIGURE 9 Rab7 deficiency or inhibition impairs the survival of plasma cells, but not their generation. (A) Generation of CD19−/loCD138hi plasma cells in vitro from Igh+/Cγ1-creRab7fl/fl B cells and their Igh+/Cγ1-creRab7+/fl B cell counterparts upon stimulation by LPS plus IL-4 or IL-4, TGF-β, anti–δ/dex and RA (left panels) as well as from C57 B cells undergoing same stimulation and treated with nil or CID 1067700 (right panels). (B) Flow cytometry analysis of total (TCR− CD19−/lo)CD138hi cells in 17-week old nil- or CID107700-treated MRL/Faslpr/lpr mice (top panels) and proportion of dead (7–AAD+) cells (bottom panels). The proportion of (TCR−CD19+CD138−) B cells was increased, but their number and viability were not changed (data not shown). Representative of three independent experiments. (C) Flow cytometry analysis of the number of live (TCR−CD19−/lo)CD138hi cells in MRL/Faslpr/lpr mice, as in (B, mean and s.d. of four pairs of mice). (D) Expression of different transcripts, as indicated, in live 7–AAD−CD138hi cells sorted from 17-week old nil- or CID 1067700-treated MRL/Faslpr/lpr mice. Like Cd35 and Cd44 (which do not promote plasma cell survival), Vla4 and Cxcr4 are also involved in plasma cell homing to the bone marrow. Expression of iNos, as suggested to promote plasma cell survival in normal mice, was not detectable. Decreased expression of Iμ-Cγ1 and Iμ-Cγ2a in plasma cells was expected, due to reduced CSR. Data were depicted as the ratio of values in CID 1067700-treated mice to those in nil-treated mice (mean and s.d. of four pairs of mice; *, p < 0.05). (E) Flow cytometry analysis of early (Annexin V+ 7–AAD−) and late (Annexin VHi 7–AAD+) apoptosis in pre-generated CD19−/loCD138hi plasma cells (after stimulation by LPS plus IL-4, IL-5, TGF-β, anti–δ/dex and RA for 66 h) after treatment with nil or CID 1067700 for 24 h, 48 h and 72 h (the histogram depicts the mean and s.e.m. of three independent experiments; death of residual CD19+ B cells in the same culture was not affected by CID 1067700). Flow cytometry data (top panels) are representative of three independent experiments. (F) Survival of CD19−/loCD138hi plasma cells pre-generated in the presence of pMIG-Cre retrovirus to express IKKβCA and then treated with nil or CID 1067700 for 72 h. Representative of three independent experiments. 1 Xu Z Zan H Pone EJ Mai T Casali P 2012 Immunoglobulin class-switch DNA recombination: induction, targeting and beyond Nat Rev Immunol 12 517 531 22728528 2 Tsokos GC 2011 Systemic lupus erythematosus N Engl J Med 365 2110 2121 22129255 3 Zan H Casali P 2015 Epigenetics of peripheral B-cell differentiation and the antibody response Front Immunol 6 631 26697022 4 Elkon K Casali P 2008 Nature and functions of autoantibodies Nat Clin Pract Rheumatol 4 491 498 18756274 5 Venkatesh J Yoshifuji H Kawabata D Chinnasamy P Stanevsky A Grimaldi CM Cohen-Solal J Diamond B 2011 Antigen is required for maturation and activation of pathogenic anti-DNA antibodies and systemic inflammation J Immunol 186 5304 5312 21444762 6 Tipton CM Fucile CF Darce J Chida A Ichikawa T Gregoretti I Schieferl S Hom J Jenks S Feldman RJ Mehr R Wei C Lee FE Cheung WC Rosenberg AF Sanz I 2015 Diversity, cellular origin and autoreactivity of antibody-secreting cell population expansions in acute systemic lupus erythematosus Nat Immunol 16 755 765 26006014 7 Pone EJ Zhang J Mai T White CA Li G Sakakura JK Patel PJ Al-Qahtani A Zan H Xu Z Casali P 2012 BCR-signalling synergizes with TLR-signalling for induction of AID and immunoglobulin classswitching through the non-canonical NF-kappaB pathway Nat Commun 3 767 22473011 8 Zan H Zhang J Ardeshna S Xu Z Park SR Casali P 2009 Lupus-prone MRL/faslpr/lpr mice display increased AID expression and extensive DNA lesions, comprising deletions and insertions, in the immunoglobulin locus: concurrent upregulation of somatic hypermutation and class switch DNA recombination Autoimmunity 42 89 103 19156553 9 Jiang C Foley J Clayton N Kissling G Jokinen M Herbert R Diaz M 2007 Abrogation of lupus nephritis in activation-induced deaminase-deficient MRL/lpr mice J Immunol 178 7422 7431 17513793 10 Pone EJ Lam T Lou Z Wang R Chen Y Liu D Edinger AL Xu Z Casali P 2015 B cell Rab7 mediates induction of activation-induced cytidine deaminase expression and class-switching in T-dependent and T-independent antibody responses J Immunol 194 3065 3078 25740947 11 Park SR Zan H Pal Z Zhang J Al-Qahtani A Pone EJ Xu Z Mai T Casali P 2009 HoxC4 binds to the promoter of the cytidine deaminase AID gene to induce AID expression, class-switch DNA recombination and somatic hypermutation Nat Immunol 10 540 550 19363484 12 Tran TH Nakata M Suzuki K Begum NA Shinkura R Fagarasan S Honjo T Nagaoka H 2010 B cell-specific and stimulation-responsive enhancers derepress Aicda by overcoming the effects of silencers Nat Immunol 11 148 154 19966806 13 Nutt SL Hodgkin PD Tarlinton DM Corcoran LM 2015 The generation of antibody-secreting plasma cells Nat Rev Immunol 15 160 171 25698678 14 Minnich M Tagoh H Bonelt P Axelsson E Fischer M Cebolla B Tarakhovsky A Nutt SL Jaritz M Busslinger M 2016 Multifunctional role of the transcription factor Blimp-1 in coordinating plasma cell differentiation Nat Immunol 17 331 343 26779602 15 Lou Z Casali P Xu Z 2015 Regulation of B cell differentiation by intracellular membrane-associated proteins and microRNAs: role in the antibody response Front Immunol 6 537 26579118 16 Agola JO Hong L Surviladze Z Ursu O Waller A Strouse JJ Simpson DS Schroeder CE Oprea TI Golden JE Aube J Buranda T Sklar LA Wandinger-Ness A 2012 A competitive nucleotide binding inhibitor: in vitro characterization of Rab7 GTPase inhibition ACS Chem Biol 7 1095 1108 22486388 17 Perry D Sang A Yin Y Zheng YY Morel L 2011 Murine models of systemic lupus erythematosus J Biomed Biotechnol 2011 271694 21403825 18 Mohan C 2015 The long (and sometimes endless) road to murine lupus genes J Immunol 195 4043 4046 26477046 19 Sasaki Y Derudder E Hobeika E Pelanda R Reth M Rajewsky K Schmidt-Supprian M 2006 Canonical NF-kappaB activity, dispensable for B cell development, replaces BAFF-receptor signals and promotes B cell proliferation upon activation Immunity 24 729 739 16782029 20 Zhang Q Huang Y Gong S Yang Z Sun X Schenken R Zhong G 2015 In vivo and ex vivo imaging reveals a long-lasting Chlamydial infection in the mouse gastrointestinal tract following genital tract inoculation Infect Immun 83 3568 3577 26099591 21 White CA Pone EJ Lam T Tat C Hayama KL Li G Zan H Casali P 2014 Histone deacetylase inhibitors upregulate B cell microRNAs that silence AID and Blimp-1 expression for epigenetic modulation of antibody and autoantibody responses J Immunol 193 5933 5950 25392531 22 Li G White CA Lam T Pone EJ Tran DC Hayama KL Zan H Xu Z Casali P 2013 Combinatorial H3K9acS10ph histone modification in IgH locus S regions targets 14-3-3 adaptors and AID to specify antibody class-switch DNA recombination Cell Rep 5 702 714 24209747 23 Romero Rosales K Peralta ER Guenther GG Wong SY Edinger AL 2009 Rab7 activation by growth factor withdrawal contributes to the induction of apoptosis Mol Biol Cell 20 2831 2840 19386765 24 Hong L Guo Y BasuRay S Agola JO Romero E Simpson DS Schroeder CE Simons P Waller A Garcia M Carter M Ursu O Gouveia K Golden JE Aube J Wandinger-Ness A Sklar LA 2015 A Pan-GTPase inhibitor as a molecular probe PLoS One 10 e0134317 26247207 25 Zhang W Shi Q Xu X Chen H Lin W Zhang F Zeng X Zhang X Ba D He W 2012 Aberrant CD40-induced NF-kappaB activation in human lupus B lymphocytes PLoS One 7 e41644 22952582 26 Zubair A Frieri M 2013 NF-kappaB and systemic lupus erythematosus: examining the link J Nephrol 26 953 959 23807646 27 Xu Z Fulop Z Wu G Pone EJ Zhang J Mai T Thomas LM Al-Qahtani A White CA Park SR Steinacker P Li Z Yates J 3rd Herron B Otto M Zan H Fu H Casali P 2010 14-3-3 adaptor proteins recruit AID to 5′-AGCT-3′-rich switch regions for class switch recombination Nat Struct Mol Biol 17 1124 1135 20729863 28 Mai T Pone EJ Li G Lam TS Moehlman J Xu Z Casali P 2013 Induction of activation-induced cytidine deaminase-targeting adaptor 14-3-3gamma is mediated by NF-kappaB-dependent recruitment of CFP1 to the 5′-CpG-3′-rich 14-3-3gamma promoter and is sustained by E2A J Immunol 191 1895 1906 23851690 29 Ma A Malynn BA 2012 A20: linking a complex regulator of ubiquitylation to immunity and human disease Nat Rev Immunol 12 774 785 23059429 30 Pengo N Scolari M Oliva L Milan E Mainoldi F Raimondi A Fagioli C Merlini A Mariani E Pasqualetto E Orfanelli U Ponzoni M Sitia R Casola S Cenci S 2013 Plasma cells require autophagy for sustainable immunoglobulin production Nat Immunol 14 298 305 23354484 31 Conway KL Kuballa P Khor B Zhang M Shi HN Virgin HW Xavier RJ 2013 ATG5 regulates plasma cell differentiation Autophagy 9 528 537 23327930 32 Shay T Kang J 2013 Immunological Genome Project and systems immunology Trends Immunol 34 602 609 23631936 33 Roy SG Stevens MW So L Edinger AL 2013 Reciprocal effects of rab7 deletion in activated and neglected T cells Autophagy 9 1009 1023 23615463 34 Theofilopoulos AN Koundouris S Kono DH Lawson BR 2001 The role of IFN-gamma in systemic lupus erythematosus: a challenge to the Th1/Th2 paradigm in autoimmunity Arthritis Res 3 136 141 11299053 35 Domeier PP Chodisetti SB Soni C Schell SL Elias MJ Wong EB Cooper TK Kitamura D Rahman ZS 2016 IFN-γ receptor and STAT1 signaling in B cells are central to spontaneous germinal center formation and autoimmunity J Exp Med 213 10.1084/jem.20151722 36 Jackson SW Jacobs HM Arkatkar T Dam EM Scharping NE Kolhatkar NS Hou B Buckner JH Rawlings DJ 2016 B cell IFN-γ receptor signaling promotes autoimmune germinal centers via cell-intrinsic induction of BCL-6 J Exp Med 213 10.1084/jem.20151724 37 Morrison RP Caldwell HD 2002 Immunity to murine chlamydial genital infection Infect Immun 70 2741 2751 12010958 38 Morrison SG Su H Caldwell HD Morrison RP 2000 Immunity to murine Chlamydia trachomatis genital tract reinfection involves B cells and CD4(+) T cells but not CD8(+) T cells Infect Immun 68 6979 6987 11083822 39 Chernova I Jones DD Wilmore JR Bortnick A Yucel M Hershberg U Allman D 2014 Lasting antibody responses are mediated by a combination of newly formed and established bone marrow plasma cells drawn from clonally distinct precursors J Immunol 193 4971 4979 25326027 40 Tellierm J Shi W Minnich M Laio Y Crawford S Smyth GK Kallies A Busslinger M Nutt SL 2016 Blimp-1 controls plasma cell function through the regulation of immunoglobulin secretion and the unfolded protein response Nat Immunol 17 323 330 26779600 41 Walsh ER Pisitkun P Voynova E Deane JA Scott BL Caspi RR Bolland S 2012 Dual signaling by innate and adaptive immune receptors is required for TLR7-induced B-cell-mediated autoimmunity Proc Natl Acad Sci USA 109 16276 16281 22988104 42 Baltimore D 2011 NF-kappaB is 25 Nat Immunol 12 683 685 21772275 43 Heise N De Silva NS Silva K Carette A Simonetti G Pasparakis M Klein U 2014 Germinal center B cell maintenance and differentiation are controlled by distinct NF-kappaB transcription factor subunits J Exp Med 211 2103 2118 25180063 44 Mai T Zan H Zhang J Hawkins JS Xu Z Casali P 2010 Estrogen receptors bind to and activate the HOXC4/HoxC4 promoter to potentiate HoxC4-mediated activation-induced cytosine deaminase induction, immunoglobulin class switch DNA recombination, and somatic hypermutation J Biol Chem 285 37797 37810 20855884 45 White CA Seth Hawkins J Pone EJ Yu ES Al-Qahtani A Mai T Zan H Casali P 2011 AID dysregulation in lupus-prone MRL/Fas(lpr/lpr) mice increases class switch DNA recombination and promotes interchromosomal c-Myc/IgH loci translocations: modulation by HoxC4 Autoimmunity 44 585 598 21585311 46 O’Connor BP Gleeson MW Noelle RJ Erickson LD 2003 The rise and fall of long-lived humoral immunity: terminal differentiation of plasma cells in health and disease Immunol Rev 194 61 76 12846808 47 Peperzak V Vikstrom I Walker J Glaser SP LePage M Coquery CM Erickson LD Fairfax K Mackay F Strasser A Nutt SL Tarlinton DM 2013 Mcl-1 is essential for the survival of plasma cells Nat Immunol 14 290 297 23377201 48 Helbig G Christopherson KW 2nd Bhat-Nakshatri P Kumar S Kishimoto H Miller KD Broxmeyer HE Nakshatri H 2003 NF-kappaB promotes breast cancer cell migration and metastasis by inducing the expression of the chemokine receptor CXCR4 J Biol Chem 278 21631 21638 12690099 49 Liu X Chen X Zhong B Wang A Wang X Chu F Nurieva RI Yan X Chen P van der Flier LG Nakatsukasa H Neelapu SS Chen W Clevers H Tian Q Qi H Wei L Dong C 2014 Transcription factor achaete-scute homologue 2 initiates follicular T-helper-cell development Nature 507 513 518 24463518 50 Grumont RJ Gerondakis S 2000 Rel induces interferon regulatory factor 4 (IRF-4) expression in lymphocytes: modulation of interferon-regulated gene expression by rel/nuclear factor kappaB J Exp Med 191 1281 1292 10770796 51 Luckey CJ Bhattacharya D Goldrath AW Weissman IL Benoist C Mathis D 2006 Memory T and memory B cells share a transcriptional program of self-renewal with long-term hematopoietic stem cells Proc Natl Acad Sci USA 103 3304 3309 16492737 52 Sekine H Watanabe H Gilkeson GS 2004 Enrichment of anti-glomerular antigen antibody-producing cells in the kidneys of MRL/MpJ-Fas(lpr) mice J Immunol 172 3913 3921 15004199 53 Starke C Frey S Wellmann U Urbonaviciute V Herrmann M Amann K Schett G Winkler T Voll RE 2011 High frequency of autoantibody-secreting cells and long-lived plasma cells within inflamed kidneys of NZB/W F1 lupus mice Eur J Immunol 41 2107 2112 21484784 54 Mohan C Putterman C 2015 Genetics and pathogenesis of systemic lupus erythematosus and lupus nephritis Nat Rev Nephrol 11 329 341 25825084 55 Li G Zan H Xu Z Casali P 2013 Epigenetics of the antibody response Trends Immunol 34 460 470 23643790 56 Son M Kim SJ Diamond B 2016 SLE-associated risk factors affect DC function Immunol Rev 269 100 117 26683148 57 Rowland SL Riggs JM Gilfillan S Bugatti M Vermi W Kolbeck R Unanue ER Sanjuan MA Colonna M 2014 Early, transient depletion of plasmacytoid dendritic cells ameliorates autoimmunity in a lupus model J Exp Med 211 1977 1991 25180065 58 Deretic V Saitoh T Akira S 2013 Autophagy in infection, inflammation and immunity Nat Rev Immunol 13 722 737 24064518 59 Teichmann LL Schenten D Medzhitov R Kashgarian M Shlomchik MJ 2013 Signals via the adaptor MyD88 in B cells and DCs make distinct and synergistic contributions to immune activation and tissue damage in lupus Immunity 38 528 540 23499488 60 La Cava A 2013 Targeting the BLyS-APRIL signaling pathway in SLE Clin Immunol 148 322 327 23269199 61 Hahn BH 2013 Belimumab for systemic lupus erythematosus N Engl J Med 368 1528 1535 23594005 62 Wang W Rangel-Moreno J Owen T Barnard J Nevarez S Ichikawa HT Anolik JH 2014 Long-term B cell depletion in murine lupus eliminates autoantibody-secreting cells and is associated with alterations in the kidney plasma cell niche J Immunol 192 3011 3020 24574498 63 Calado DP Zhang B Srinivasan L Sasaki Y Seagal J Unitt C Rodig S Kutok J Tarakhovsky A Schmidt-Supprian M Rajewsky K 2010 Constitutive canonical NF-kappaB activation cooperates with disruption of BLIMP1 in the pathogenesis of activated B cell-like diffuse large cell lymphoma Cancer Cell 18 580 589 21156282 64 Staudt LM 2010 Oncogenic activation of NF-kappaB Cold Spring Harb Perspect Biol 2 a000109 000110.001101/cshperspect.a000109 20516126 65 Zhang B Calado DP Wang Z Frohler S Kochert K Qian Y Koralov SB Schmidt-Supprian M Sasaki Y Unitt C Rodig S Chen W Dalla-Favera R Alt FW Pasqualucci L Rajewsky K 2015 An oncogenic role for alternative NF-kappaB signaling in DLBCL revealed upon deregulated BCL6 expression Cell Rep 11 715 726 25921526 66 Kashuba VI Gizatullin RZ Protopopov AI Allikmets R Korolev S Li J Boldog F Tory K Zabarovska V Marcsek Z Sumegi J Klein G Zabarovsky ER Kisselev L 1997 NotI linking/jumping clones of human chromosome 3: mapping of the TFRC, RAB7 and HAUSP genes to regions rearranged in leukemia and deleted in solid tumors FEBS Lett 419 181 185 9428630 67 Mitelman F Mertens F Johansson B 1997 A breakpoint map of recurrent chromosomal rearrangements in human neoplasia Nat Genet 15 Spec No 417 474 9140409 68 Wieser R Schreiner U Pirc-Danoewinata H Aytekin M Schmidt HH Rieder H Fonatsch C 2001 Interphase fluorescence in situ hybridization assay for the detection of 3q21 rearrangements in myeloid malignancies Genes Chrom Cancer 32 373 380 11746978 69 Ni H Ergin M Huang Q Qin JZ Amin HM Martinez RL Saeed S Barton K Alkan S 2001 Analysis of expression of nuclear factor kappa B (NF-kappa B) in multiple myeloma: downregulation of NF-kappa B induces apoptosis Br J Haematol 115 279 286 11703322 70 Amarilyo G Lourenco EV Shi FD La Cava A 2014 IL-17 promotes murine lupus J Immunol 193 540 543 24920843

PMC005xxxxxx/PMC5113289.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8902617 1394 J Mol Endocrinol J. Mol. Endocrinol. Journal of molecular endocrinology 0952-5041 1479-6813 26242202 5113289 10.1530/JME-15-0124 NIHMS828823 Article Endothelial glucocorticoid receptor promoter methylation according to dexamethasone sensitivity Mata-Greenwood Eugenia Jackson P Naomi Pearce William J Zhang Lubo Divisions of Pharmacology and Physiology, Department of Basic Sciences, School of Medicine, Center for Perinatal Biology, Medical Center, Loma Linda University, Room A572, 11234 Anderson Street, Loma Linda, CA 92350, USA Correspondence should be addressed to E Mata-Greenwood ematagreenwood@llu.edu 9 11 2016 04 8 2015 10 2015 17 11 2016 55 2 133146 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. We have previously shown that in vitro sensitivity to dexamethasone (DEX) stimulation in human endothelial cells is positively regulated by the glucocorticoid receptor (NR3C1, GR). The present study determined the role of differential GR transcriptional regulation in glucocorticoid sensitivity. We studied 25 human umbilical vein endothelial cells (HUVECs) that had been previously characterized as DEX-sensitive (n = 15), or resistant (n = 10). Real-time PCR analysis of GR 5′UTR mRNA isoforms showed that all HUVECs expressed isoforms 1B, 1C, 1D, 1F, and 1H, and isoforms 1B and 1C were predominantly expressed. DEX-resistant cells expressed higher basal levels of the 5′UTR mRNA isoforms 1C and 1D, but lower levels of the 5′UTR mRNA isoform 1F than DEX-sensitive cells. DEX treatment significantly decreased GRα and GR-1C mRNA isoform expression in DEX-resistant cells only. Reporter luciferase assays indicated that differential GR mRNA isoform expression was not due to differential promoter usage between DEX-sensitive and DEX-resistant cells. Analysis of promoter methylation, however, showed that DEX-sensitive cells have higher methylation levels of promoter 1D and lower methylation levels of promoter 1F than DEX-resistant cells. Treatment with 5-aza-2-deoxycytidine abolished the differential 5′UTR mRNA isoform expression between DEX-sensitive and DEX-resistant cells. Finally, both GRα overexpression and 5-aza-2-deoxycytidine treatment eliminated the differences between sensitivity groups to DEX-mediated downregulation of endothelial nitric oxide synthase (NOS3), and upregulation of plasminogen activator inhibitor 1 (SERPINE1). In sum, human endothelial GR 5′UTR mRNA expression is regulated by promoter methylation with DEX-sensitive and DEX-resistant cells having different GR promoter methylation patterns. glucocorticoids endothelium methylation 5′UTR mRNA isoform Introduction Glucocorticoids are therapeutic agents used for reducing inflammation via targeting the glucocorticoid receptor (NR3C1, GR) of immune cells (Auphan et al. 1995, Barnes 1998). However, glucocorticoids also target the GR present in various other tissues, causing unwanted side effects. In the cardiovascular system, glucocorticoids induce short-term side effects that include hypertension, dyslipidemia and thrombosis. Chronic synthetic glucocorticoid-therapy has been associated with endothelial dysfunction and increased risk of cardiovascular events such as myocardial infarction (Yang & Zhang 2004, Walker 2007, Hadoke et al. 2009, Huang & Glass 2010, Kadmiel & Cidlowski 2013). These adverse events are mediated in part by glucocorticoid-dependent down-regulation of endothelial nitric oxide synthase (NOS3) and upregulation of plasminogen activator inhibitor 1 (SERPINE1) (Wallerath et al. 2004, Tamura et al. 2015). Of importance, human studies have revealed significant human variability in response to both endogenous (cortisol) and synthetic glucocorticoids (Ito et al. 2006, Kino 2007), but the mechanisms remain undetermined. Glucocorticoids mediate their biological effects by the ubiquitously expressed GR (Bamberger et al. 1996, Adcock et al. 2004, Heitzer et al. 2007). Alternative splicing produces 3′UTR mRNA isoforms GRα, GRβ and GRP. GRα protein is the biologically relevant isoform, capable of binding ligand that regulates transcription and stability of target genes. GRβ is localized in the nucleus and has a dominant negative effect on GRα through the formation of GRα/GRβ heterodimers. GRP has been reported to have both synergistic and antagonistic effects on GRα (Kassel & Herrlich 2007, Biddie et al. 2012). Previously, we have shown that human endothelial sensitivity to dexamethasone (DEX) varied according to the levels of GRα protein degradation (Mata-Greenwood et al. 2013). Human umbilical vein endothelial cells (HUVECs) that were resistant to in vitro stimulation by DEX had an increased expression of the E3 ubiquitin ligase gene BCL2-athanogene 1 (BAG1) and subsequent GR protein ubiquitination and proteasomal degradation. It was found that the basal protein levels of GRα correlated with the endothelial expression of key genes, such as NOS3 and SERPINE1 (Mata-Greenwood et al. 2013). However, the transcriptional regulation of GR expression in human endothelial cells and its role in endothelial glucocorticoid sensitivity have not been well defined, and warrant further exploration. GR mRNA expression is regulated by complex mechanisms that include transcriptional and epigenetic modification of GR promoters (Presul et al. 2007, Turner et al. 2010, Cao-Lei et al. 2011). The GR gene (NR3C1) contains at least nine 5′UTR first exons that are spliced to the common exon 2. The alternative first exons are divided into two promoter regions: a distal region located more than 30 kb upstream that contains exons 1A and 1I, and a proximal region located 5 kb upstream of the translation start site that contains exons 1B, 1C, 1D, 1E, 1F, 1H, and 1J. Each first exon is regulated by its own promoter that binds specific transcription factor complexes (Bockmuhl et al. 2011, Cao-Lei et al. 2011). Therefore, the transcription of the GR gene is regulated by complex promoter regions containing multiple transcription start sites. Furthermore, both proximal and distal GR promoters are embedded in a CpG island that is mostly methylation-free and therefore can be targeted for methylation. Recent human and rodent studies have shown tissue and disease specific expression of untranslated first exons, due in part to changes in GR promoter methylation (Turner et al. 2006, 2008, 2010, Bockmuhl et al. 2011). These studies have shown that the various GR promoters allow for tissue-specific mechanisms to control and adapt GR expression levels according to various developmental and pathophysiological stages. Previous studies have shown that rodent heart and aortic tissue express significant levels of GR transcripts (Presul et al. 2007, Xiong & Zhang 2013), but the transcriptional regulation of GR in specific cardiovascular cells (i.e., endothelial, smooth muscle, cardiomyocytes) remains unknown. Therefore, the main objective of the present study is to uncover the mechanisms that regulate the expression of GR mRNA isoforms in human endothelial cells that have been previously characterized as DEX-sensitive or DEX-resistant. We find that DNA methylation plays an important role in the fine-tuning determination of GR mRNA isoform expression in human endothelial cells. Our study provides novel insights into the upstream mechanisms that alter GR expression in the cardiovascular system and suggest that future use of epigenetic markers will aid in identifying glucocorticoid responsive individuals. Materials and methods Subjects and cell culture HUVECs isolated from 25 healthy term pregnancies were selected from a previous cohort study of 42 subjects (Mata-Greenwood et al. 2013) to further characterize the transcriptional regulation of GR in human endothelial cells. These HUVECs were previously characterized for endothelial cell purity and in vitro response to DEX (Mata-Greenwood et al. 2013). The study subject characteristics are summarized in Supplementary Table 1, see section on supplementary data given at the end of this article. The study was approved by the IRBs of the Loma Linda University and the University of California at San Diego. HUVECs were cultured using ECM media (ScienCell, Carlsbad, CA, USA). All assays were performed between passages 4 and 6. To study inter-individual differences in response to glucocorticoids, we stimulated confluent and quiescent HUVECs with DEX using DMSO as a solvent (final concentration of 0.05%). This synthetic glucocorticoid was chosen for its stability in the presence of cellular 11-β-dehydrogenase, and for its specificity for GRα binding with respect to other nuclear receptors, including the mineralocorticoid receptor (Wenting-van Wijk et al. 1999). To starve HUVECs, we used M199 (Sigma–Aldrich) supplemented with 0.95 mM HEPES, 0.1% BSA, 1% antibiotics and 1% fetal bovine serum (FBS). The acute monocytic leukemia cell line THP-1 from Sigma–Aldrich was used as a positive control as it is known to express all GR mRNA isoforms (Steer et al. 2000). THP-1 cells were grown in DMEM with 1% antibiotics and 10% FBS. RNA extraction and real-time PCR Total RNA was extracted with TriZOL-RNEasy kits (Invitrogen), quantified, and stored at −80 °C until analysis. The total RNA (1 µg) was reverse transcribed and real time PCR was performed for each sample in triplicate as previously described (Goyal et al. 2014). Taqman assays were obtained from Life Technologies for GR mRNA isoforms 1B (Hs01005211_m1), 1C (Hs01010775_m1), 1D (Hs03666144_m1) and the housekeeping gene 18S (Hs99999901_s1), and these were run according to the manufacturer’s recommendations using a Probe Quantitect PCR mix (Qiagen). Designed primers for SYBR green PCR analysis of the remaining GR mRNA isoforms and the housekeeping gene 18S, together with PCR conditions and accession numbers, are shown in Supplementary Table 2, see section on supplementary data given at the end of this article. SYBR green PCR was performed using Quantitect SYBR green PCR kit (Qiagen) using a 2.0 Lightcycler amplifier (Roche). PCR products were purified, sequenced, and used to obtain standard curves for each mRNA isoform. Extrapolation of unknowns from the standard curve was performed using Prism 3 (GraphPad Software, San Diego, CA, USA), predicting unknowns from the standard curve Ct values. Data is presented as fg mRNA/ng 18S. Alternatively, the percentages of each GR mRNA isoform were estimated for each subject as mRNA isoform/sum of all mRNA isoforms × 100. SDS–PAGE and immunoblotting Western blotting was performed as previously described (Mata-Greenwood et al. 2010). Briefly, protein extracts (30 µg) were prepared in a cold lysis buffer, heat denatured in a Laemmli buffer, separated on SDS-PAGE, and transferred to PVDF membranes. Membranes were blocked in 5% non-fat dried milk in 0.05% TBST for 1 h, and then probed in primary rabbit anti-GR (Santa Cruz Biotechnologies), monoclonal anti-BAG1 (Santa Cruz Biotechnologies), monoclonal anti-HSP90 (BD Biosciences, San Jose, CA, USA), and rabbit anti-FKBP51 (Stressmarq, Victoria, BC, Canada), diluted in blocking buffer (1 µg/ml) overnight at 4 °C. After three 10 min washes with TBST, the membranes were incubated with secondary antibodies that were diluted at 1:2000. Bound antibodies were visualized using the chemiluminscence substrate (ThermoFisher Scientific, Carlsbad, CA, USA). The membranes were then probed with monoclonal anti-β-Actin (Ambion, Austin, TX, USA). Data is presented as protein levels relative to β-Actin levels. Cloning of GR promoters GR promoters B, C, D, F, and H were cloned into the firefly luciferase-pGL3 reporter vectors (Promega), as previously described (Mata-Greenwood et al. 2010). GR promoter regions were selected to include putative essential DNA elements such as initiator (Inr) and downstream promoters (DPE). Briefly, 500 ng of HUVEC DNA (from specific donors) was used to amplify GR promoter regions using primers that contained specific restriction enzyme sites (Supplementary Table 3, see section on supplementary data given at the end of this article), followed by ligation into the basic luciferase pGL3 vector. The non-methylated GR promoter inserts were confirmed by DNA sequencing. Vector clones containing WT GR promoters were used. Cell transfection and luciferase assays Transfection of plasmid DNA was performed with the aid of HD Xtreme transfection reagent (Roche) as previously described (Mata-Greenwood et al. 2013). Briefly, confluent cells were transfected with the GR promoter/pGL3-luciferase constructs, using a 1:1 complex of DNA:Xtreme reagent according to the manufacturer’s protocol. TK-Renilla luciferase vector (Promega Corp.) was used as the internal control. The transfection was carried out at 37 °C for 6 h. The cells were allowed to recover in a complete culture medium for 18–20 h, and then treated with starvation media with or without DEX (0.2 and 1 µM) for another 24 h, before adding passive lysis. Firefly and Renilla luciferase activities were measured using a dual-reporter assay kit (Promega) according to the manufacturer’s protocol. Relative luciferase values were calculated as a ratio of firefly/renilla luciferase activities. Each treatment was tested in quadruplicates and averages per sample were used to calculate group averages and standard errors. Methyl-DNA immunoprecipitation Genomic DNA was isolated from confluent and quiescent HUVECs using the Wizard genomic DNA isolation kit (Promega). Methyl-DNA immunoprecipitation (MeDIP) was performed with the aid of a commercial kit according to the manufacturer’s protocol (Active Motif, Carlsbad, CA, USA). In brief, 20 µg of DNA was sheared into ~ 200–400 bp fragments and immunoprecipitated with a 5-methylcytosine-specific antibody at 4 °C overnight. Immunoprecipitated DNA was bound to magnetic beads, washed and eluted. Input (sheared DNA) and immunoprecipitated DNA (1 µl) was used to amplify specific promoter regions from the GR gene using SYBR green PCR, as previously described. MeDIP real-time PCR primers are shown in Supplementary Table 4, see section on supplementary data given at the end of this article. Promoter methylation levels were estimated as immunoprecipitated DNA/total (input) DNA × 100. Bisulfite-converted DNA sequencing To determine the methylation status of GR promoters, DNA samples were treated with bisulfite using the epiTect Bisulfite kit (Qiagen), according to the manufacturer’s instructions. Converted DNA was amplified using primers directed to bisulfite modified genomic DNA (Supplementary Table 4). Amplification conditions were as follows: 94 °C for 30 s, 40 °C for 30 s, and 72 °C for 40 sX40 cycles. Resulting amplicons were cloned into pCR4-TOPO sequencing vectors (Invitrogen) for automated fluorescent sequencing. Data were analyzed using the CLC Software, and clones showing <90% conversion or identified as clonal were not included in further analysis. The methylation levels for each subject were estimated as the methylated clones/#total clones sequenced × 100. A minimum of ten clones per amplicon were analyzed by sequencing. Averages and standard errors were then calculated for each sensitivity group. Overexpression of GRα Mammalian expressing vectors for the promoter-less full length GR-1Cα (NM_000176.2) were a gift from Dr John A. Cidlowski (NIEHS, Research Triangle, NC, USA). GR over-expression was achieved by transfecting log-phase proliferating HUVECs with 1 µg DNA/six well using the Xtreme transfection reagent as previously described. After 18 h of recovery, HUVECs were treated with DEX (1 µM) or not for an additional 24 h before analysis. Overexpression of GRα was confirmed by real-time PCR and immunoblotting. Global demethylation assays To induce global demethylation, we utilized a 5 µM dose of 5′-aza-2′-deoxycitidine (AZA) in 50% confluent cells. After 24 h of AZA treatment, cells were treated with both AZA and DEX (1 µM) or solvent for another 24 h before harvesting total RNA. DNA demethylation was confirmed by GR promoter D methylation levels of < 1%. Procoagulant activity assay The procoagulant activity of HUVECs was tested by a one-step recalcification time test, also known as the activated partial thromboplastin time (aPTT) as previously described (Mata-Greenwood et al. 2013). As discussed, 70% confluent HUVECs were transfected with GRα-expressing vectors, and 18 h after were treated with or without DEX (1 µM) for an additional 24 h. Cells were then washed, trypsinized, and resuspended in PBS at a concentration of one million cells per ml PBS. A 1:1 solution was prepared with cells and fresh plasma (collected from one single donor and anticoagulated with 3.8% sodium citrate, 1:9, v/v) and incubated for 180 s at 37 °C. After the addition of preheated CaCl2, the time to fibrin strand generation was recorded by a BCS XP hemostatic analyzer (Siemens, Munich, Germany). Statistical analysis Data are presented as means±s.e.m. Each individual variable was analyzed via ANOVA followed by post-hoc least significant difference (LSD) analysis to determine differences between sensitivity groups and between treatment groups. Homogeneity of variances was confirmed (P>0.05) using the Levene’s Test of Equality of Error Variance. For data sets with unequal variances (P<0.05), the data were log transformed and all further statistics were performed on the transformed data. For graphical presentation, the anti-logs of the mean log values were calculated and plotted. In all cases, statistical power was >0.08. A P value of <0.05 was regarded as significant. All statistical analysis was performed using SPSS, version 22. Results Differential GR mRNA isoform expression in HUVECs according to DEX-sensitivity We have previously characterized a small cohort of 42 HUVECs obtained from healthy pregnancies for DEX-sensitivity. DEX-sensitive HUVECs expressed significantly higher levels of GRα protein than DEX-resistant HUVECs (Mata-Greenwood et al. 2013). Analysis of study group characteristics indicated that the pre-pregnancy maternal BMI and average systolic blood pressure were significantly higher in DEX-sensitive HUVECs compared to those of DEX-resistant cells (Supplementary Table 1), but the remaining parameters were not significantly different. To further understand the transcriptional regulation of GR in endothelial cells, and the mechanisms leading to the differences in GR expression between DEX-sensitive and DEX-resistant HUVECs, we analyzed the expression of 5′UTR mRNA isoform transcripts in 25 HUVECs (15 DEX-sensitive and 10 DEX-resistant). HUVECs expressed GR proximal mRNA isoforms 1B, 1C, 1D, 1F, and 1H, but did not express the distal isoforms 1A and 1I, or the proximal isoforms 1E and 1J (Fig. 1A and B). DEX-resistant HUVECs expressed significantly higher basal, but similar DEX-stimulated, GR mRNA isoform 1C levels (Fig. 1B); and lower percentages of mRNA isoform 1F levels (Fig. 1C) than DEX-sensitive cells. DEX treatment significantly decreased the expression of mRNA isoform 1C in DEX-resistant cells only (Fig. 1B). The percentages of GR 5′UTR mRNA isoforms in all HUVECs were ~67% (1C), ~30% (1B), ~2% (1F), ~1% (1H), and ~0.2% (1D) (Fig. 1). DEX treatment did not significantly alter the percentages of GR 5′UTR mRNA isoforms in either DEX-sensitive or DEX-resistant cells (Fig. 1C). We also analyzed the expression profile of GR 3′UTR mRNA isoforms GRα, GRβ and GRP. GRα represented ~ 96% of all 3′UTR mRNA isoforms, followed by GRP (4%), and GRβ expression was slight (<0.0006%) in all HUVECs (Fig. 1E). DEX-resistant HUVECs expressed significantly higher basal levels of GRα than DEX-sensitive HUVECs (Fig. 1D) that correlate with higher basal levels of the predominant 5′UTR mRNA isoform 1C (Fig. 1B). Importantly, DEX treatment decreased the expression of GRα (Fig. 1D) and increased the percentage of GRP (Fig. 1E) in DEX-resistant cells only. Lastly, there were no significant differences in the levels of total GR mRNA (measured by exons 2–3) between the sensitivity groups. However, DEX downregulated total GR mRNA in DEX-resistant cells only (Fig. 1D). Role of GRα overexpression on its own transcriptional regulation and endothelial cell phenotype To clarify the role of differential GR 5′UTR mRNA isoform levels between sensitivity groups on endothelial phenotype, we overexpressed GRα in both DEX-sensitive and DEX-resistant HUVECs, using a promoter-less GR-1Cα-expressing vector. We hypothesized that GRα overexpression would increase sensitivity to DEX in any HUVEC tested. To validate our methods, we first showed our protocol significantly increased GR-1Cα mRNA levels by 31-fold and 20-fold, and GRα protein levels by 3.5-fold and 2.4-fold in DEX-sensitive and DEX-resistant cells respectively (Fig. 2A and B). DEX-treatment significantly decreased the levels of overexpressed GRα mRNA and protein in both sensitivity groups (Fig. 2A), suggesting DEX-dependent post-transcriptional and post-translational mechanisms of GR regulation (Pratt et al. 2006, Turner et al. 2010). However, our overexpression system had a smaller effect on GR protein levels (two- to fourfold increase) compared to GR mRNA (20- to 31-fold increase) levels that is most likely due to the reported GR protein degradation by the proteasome (Wallace et al. 2010, Mata-Greenwood et al. 2013). Of interest, overexpression of similar amount of GR-1Cα vectors yielded significantly lower levels of both GRα mRNA and protein in DEX-resistant cells compared to DEX-sensitive cells (Fig. 2B). This finding suggests that DEX-resistant cells have stronger negative autoregulatory mechanisms on GR expression compared to DEX-sensitive cells. We then investigated the role of increased expression of GRα in endothelial phenotype of both DEX-sensitive and DEX-resistant cells (Fig. 2C, D and E). We chose to study NOS3 and SERPINE1 expression, as these genes are key endothelial genes regulated by glucocorticoids, and their expression changes also regulate endothelial phenotype (Lou et al. 2001, Muzaffar et al. 2005, Kimura et al. 2009). GRα overexpression correlated with significant decreases in NOS3 (Fig. 2C) and increases in SERPINE1 (Fig. 2D) basal mRNA expression in both DEX-sensitive and DEX-resistant cells. Overexpression of GRα also increased the basal procoagulant activity of HUVECs as shown by the aPTT assay (Fig. 2E). However, there were significant differences between the sensitivity groups in response to DEX. DEX-treatment led to NOS3 down-regulation, SERPINE1 upregulation and aPTT increases in GRα-overexpressing DEX-sensitive cells, but not in GRα-overexpressing DEX-resistant cells (Fig. 2D and E). Therefore, overexpression of GR-1Cα alone did not correct the lack of response to DEX in resistant cells. However, GR-1Cα overexpression did alter basal endothelial phenotype in both sensitivity groups suggesting important non-ligand dependent effects of GR in human endothelial cells. We also investigated the effect of GR-1Cα over-expression on its own expression. We found that GR-1Cα overexpression (alone or in combination with DEX treatment) decreased the expression of GR mRNA isoform 1B in DEX-resistant cells, and that of isoform 1H in both sensitivity groups (Fig. 2F). GR-1Cα overexpression did not alter DEX effects on GR mRNA isoform expression (Fig. 2F). Finally, we investigated the expression of GR chaperones FKBP51, BAG1 and HSP90 in HUVECs overexpressing GRα. We found that resistant cells that overexpress GRα have significantly higher expression of the GRα inhibitor FKBP51 and higher levels of HSP90 than sensitive cells (Supplementary Figure 1, see section on supplementary data given at the end of this article), and could be a reason for the lack of DEX response in GRα overexpressing resistant cells. Therefore, differences in DEX-sensitivity in our HUVECs can also be accounted by differential regulation of GR chaperones. In summary, these results demonstrate that exogenous upregulation of GRα mRNA/protein levels can significantly alter both basal and glucocorticoid-stimulated endothelial phenotype in HUVECs, but does not eliminate the differences in GR expression and function regulation between the sensitivity groups. Correlation of GR 5′UTR mRNA isoform expression with promoter usage To further characterize the differences observed in GR mRNA isoform expression, we cloned five of the proximal TATA-less GR promoters to the pGL3 luciferase reporter vector to study promoter activity in HUVECs (Fig. 3A). Clones were sequenced and analyzed for putative internal elements (Inr), DPE, and transcription factor binding sites using the MattInspector software (Supplementary Figure 2, see section on supplementary data given at the end of this article). The consensus Inr element sequence (YYANWYY) located within the first 5 bp of the transcriptional starting site (+1) was found in promoter 1C only. The consensus TATA-less DPE elements (DPE=RGWYV) were found in promoters 1B and 1C, while partial DPEs were found in the remaining GR promoters (Supplementary Figure 2). Putative promoter elements for transcription factors were found for each promoter: those reported to be highly expressed in the cardiovascular system are shown in green. Luciferase assays revealed significant differences in GR promoter usage in all HUVECs, with promoter 1C having the highest activity (55%), followed by promoter 1B (30%), promoter 1D (10%), promoter 1F (8%), and promoter 1H (3%) (Fig. 3B). There were no significant differences in basal promoter usage between the sensitivity groups to account for the observed differences in GR 5′UTR mRNA isoforms 1C, 1D, and 1F (Fig. 1B). There were only small significant differences in DEX-stimulated promoter usage; sensitive cells responded to DEX with a slight increase in promoter 1B and 1D usage compared to DEX-resistant cells (Fig. 3B). DEX-treatment also significantly decreased promoter 1F usage in both DEX-sensitive and DEX-resistant cells, with non-significant decreases in promoter 1C usage in both groups as well (Fig. 3B). Therefore, our promoter assays suggest that the observed differences in GR mRNA isoforms 1C, 1D, and 1F between sensitivity groups (Fig. 1B and C) are not the result of differential transcriptional activation of GR promoters. We analyzed the correlation of promoter usage with 5′UTR mRNA isoform expression by plotting the average % mRNA levels with average % promoter usage in all HUVECs at basal, low DEX (0.2 µM) and high DEX (1 µM) dose (Fig. 3C). We found a significant positive correlation between these two parameters, with only promoter 1D usage being disproportionally higher than the expression of the 1D mRNA isoform (Fig. 3C), suggesting epigenetic regulation of promoter 1D. Role of promoter methylation in GR 5′UTR mRNA isoform expression Because our reporter assays failed to reveal transcriptional mechanisms that would account for the observed differences between DEX-sensitive and DEX-resistant cells in GRα expression regulation (particularly of mRNA isoforms 1C, 1D, and 1F), we hypothesized that there could be differential methylation patterns of GR promoters. We first studied promoter methylation levels by immunoprecipitation of 5-methylcytosine (Fig. 4A). As expected, promoters 1B and 1C showed very low methylation levels of <4 and <1.7% respectively (Fig. 4A). Promoter 1D showed the highest methylation levels of ~20%, followed by promoter 1F (~14%), and promoter 1H (~10%). There were significant differences in promoter methylation levels between DEX-sensitive and DEX-resistant HUVECs (Fig. 4A). DEX-resistant cells had significantly lower methylation levels of promoter 1D and higher methylation levels of promoter 1F, compared to DEX-sensitive cells (Fig. 4A). The basal promoter methylation levels (i.e. averages for all cells: sensitive and resistant) correlated inversely with the expression of GR 5′UTR mRNA isoforms, with promoter 1C having the lowest methylation levels and higher mRNA isoform 1C levels, and promoter 1D having the highest methylation levels together with lower mRNA isoform 1D expression levels (Fig. 4B). These results explain why the expression of the 5′UTR mRNA isoform 1D is lower than that of isoform 1F, although the cloned promoter 1D-driven luciferase activity was higher than that of promoter 1F (Figs 1B and 3B), as cloned promoter-vectors do not have any cytosine methylations. To confirm our MeDIP results, we then investigated the specific methylation sites in promoters 1D and 1F via bisulfite sequencing. Promoters 1D and 1F were chosen because they exhibited the highest methylation levels and because of differences in the expression of mRNA isoforms 1D and 1F between sensitivity groups. Bisulfite sequence analysis confirmed the differences observed in our MeDIP assays (Supplementary Figure 3, see section on supplementary data given at the end of this article and Fig. 4A). We found five methylation sites in promoter 1D with higher methylation percentages in DEX-sensitive cells than DEX-resistant cells (Supplementary Figure 3). DEX-resistant cells showed significant promoter 1F methylation of five different cytosines in comparison with none found in DEX-sensitive cells. In sum, bisulfite analysis confirmed that DEX-sensitive cells have higher methylation levels for promoter 1D but lower methylation levels for promoter 1F, compared to DEX-resistant cells. It is important to note that the MeDIP assay and bisulfite sequencing showed methylation levels for promoter 1F of ~7 and 0%, respectively, in DEX-sensitive cells. This discrepancy is likely due to the presence of other methylation sites further upstream or downstream from the selected DNA segment analyzed via bisulfite sequencing. Finally, we utilized AZA to induce global demethylation in order to study the role of GR promoter methylation on its own expression and function. Analysis of the GR 5′UTR mRNA isoform expression demonstrated that AZA treatment resulted in increased expression of all GR 5′UTR mRNA isoforms in DEX-sensitive cells, and in all isoforms except isoform 1D in DEX-resistant cells (Fig. 4C and D). Importantly, AZA treatment eliminated the differences in GR 5′UTR mRNA isoforms 1C, 1D and 1F expression between DEX-sensitive and DEX-resistant cells (Fig. 4C). Analysis of AZA-induced fold increases in GR mRNA expression further highlighted differences between sensitivity groups. AZA-treatment produced higher (fold of control) increases in the expression of GR mRNA isoforms 1B and 1D in DEX-sensitive cells compared to DEX resistant cells, and a higher fold increase of isoforms 1F and 1H in DEX-resistant cells compared to DEX sensitive cells, and these data correlated with GR promoter methylation levels (Fig. 4A and D). Consistent with the effect of AZA on GR expression, we found that AZA pretreatment increased the response to DEX in the resistant group in terms of NOS3 and SERPINE1 gene expression regulation, and thereby eliminated the differential response to DEX between the sensitivity groups (Fig. 4E and F). Of note, AZA treatment significantly increased the basal expression of SERPINE1 on all cells. This result is likely due to the effect of AZA on SERPINE1 promoter demethylation and basal transcriptional upregulation. However, we did not study the effect of AZA on the promoter methylation status of other genes. Therefore, AZA effects on in vitro DEX response could be mediated by demethylation of other genes that are differentially methylated in DEX-sensitive and DEX-resistant cells. In sum, we found significant differences in GR promoter methylation that result in differential GR 5′UTR mRNA isoform expression, and these differences further characterize DEX-sensitive and DEX-resistant HUVECs. Discussion We uncovered novel mechanisms of GR expression regulation in human endothelial cells that include differential expression of 5′UTR mRNA isoforms due to promoter methylation We have also found significant differences in GR promoter methylation and 5′UTR mRNA isoform expression according to DEX-sensitivity. The results of our studies on GR mRNA and protein regulation in HUVECs are summarized in Table 1. First, we have determined that the primary 5′UTR mRNA isoform present in human endothelial cells is isoform 1C, followed by isoform 1B, with smaller contributions from isoforms 1F, 1H, and 1D. Previous studies have shown similar dominance of expression of isoform 1C in tissues such as the liver, heart, kidney, and lung (Presul et al. 2007, Turner et al. 2008, 2010). Our promoter analysis suggests that the presence of putative Inr elements and DPE in promoter 1C leads to transcriptional dominance of mRNA isoform 1C compared to other isoforms. These DNA elements are known to recruit RNA polymerase complexes in TATA-less promoters (Butler & Kadonaga 2002, Yang et al. 2007, Juven-Gershon & Kadonaga 2010). Interestingly, mRNA isoforms 1J, 1D, 1H, and 1F have been reported to be highly expressed in non-vascular tissues such as the hippocampus and peripheral mononuclear cells (Turner et al. 2008, Cao-Lei et al. 2013). In our studies, these exons represent < 10% of all endothelial GR transcripts. Therefore, our data indicate that regulation of isoforms 1C and 1B expression, with smaller contributions from mRNA isoforms 1D, 1F and 1H, will have a higher impact on total GR mRNA expression. Previously we reported that increased in vitro DEX sensitivity of HUVECs correlated positively with GRα protein levels (Mata-Greenwood et al. 2013). DEX-sensitive HUVECs showed lower GRα protein turnover through the proteasome system, and this was partly due to lower expression of the E3 ubiquitin ligase BAG1. In the present study, we have found that increased GRα protein levels observed in DEX-sensitive, compared to DEX-resistant, cells are not the result of increased GRα mRNA levels, as previously hypothesized. However, we did observe significant differences in GRα mRNA isoform expression among DEX-sensitive and DEX-resistant HUVECs. Resistant cells, but not sensitive cells, responded to DEX treatment with a significant down-regulation of the main GR 5′UTR mRNA isoform 1C and 3′UTR mRNA isoform α (Table 1). Because glucocorticoids downregulate GRα mRNA expression (Petersen et al. 2004, Ramamoorthy & Cidlowski 2013), we now hypothesize that DEX-resistant cells show significant DEX-dependent downregulation of GR mRNA expression in addition to increased GR protein degradation via the proteasome. Our GRα overexpression data supports this hypothesis; DEX treated-resistant cells had lower levels of overexpressed GRα mRNA and protein than DEX-treated-sensitive cells. Therefore, DEX-resistance in endothelial cells could arise from a strong negative autoregulation that includes both mRNA and protein expression decreases. Another interesting result was that GRα overexpression did not increase in vitro response to DEX (measured as changes in NOS3, SERPINE1, and aPTT) in resistant cells (Table 1). Furthermore, GRα overexpression significantly altered basal endothelial phenotype (NOS3, SERPINE1, and aPTT levels) in both sensitivity groups with a greater effect shown in resistant cells. We currently speculate these data to be caused by differences in GR:chaperone interactions between sensitive and resistant cells. In support of this hypothesis, we found that GRα overexpression upregulated the expression of GR inhibitors BAG1 and FKBP51. Further studies on GR chaperones are therefore needed to fully understand the differential in vitro response to DEX in HUVECs. In summary, these data suggest that posttranslational regulation of GRα protein, such as GRα proteasomal degradation and GR:chaperone interactions, and not transcriptional mechanisms, are key factors in determining basal and DEX-stimulated levels of biologically active GR and, thereby, in the biological response to glucocorticoids (Table 1). Perhaps the most novel result of this study is the inverse association between GR 5′UTR mRNA isoform expression and GR promoter methylation together with the finding of significant differences in promoter methylation between DEX-sensitive and DEX-resistant cells (Table 1). This study is the first to report that methylation plays an important role in GR mRNA expression in endothelial cells. DEX-sensitive HUVECs showed significantly higher methylation of promoter 1D, but lower methylation of promoters 1F and 1H. Methylation levels of promoter 1D and 1F correlated inversely with the expression of their corresponding 5′UTR mRNA isoforms 1D and 1F. Of importance was the finding that AZA treatment abrogated the differences in GR 5′UTR mRNA isoform expression between sensitivity groups, thereby showing that methylation differences in GR promoters are a principal reason for the observed differences in GR 5′UTR mRNA isoforms expression between DEX-sensitive and DEX-resistant cells. Unexpectedly, AZA treatment increased the expression of GR mRNA isoforms 1B and 1C, even though their proximal promoters were highly unmethylated. We hypothesize that AZA demethylation of upstream sequences, such as promoter 1D or GR enhancers, increases the expression of mRNA isoforms 1B and 1C. Our hypothesis is in agreement with previous reports on YY1, a promoter 1D binding transcription factor, and the mediated transcriptional activation of isoform1B (Cao-Lei et al. 2011). To our knowledge, this study is the first to show that the expression of GR mRNA isoforms 1B and 1C is also regulated by methylation. Numerous studies have shown a significant role of promoter 1F methylation in brain tissue, and its correlation with psychiatric diseases (Moser et al. 2007, Alt et al. 2010, Labonte et al. 2014, Van der Knaap et al. 2014). In rats, maternal care has shown to decrease methylation levels of promoter 1–7 in brain and liver tissue (Szyf et al. 2005, Witzmann et al. 2012), while hypoxia and other stressors increase it (Xiong & Zhang 2013, Gonzalez-Rodriguez et al. 2014). Recent human studies on promoter 1F methylation in umbilical cord mononuclear cells revealed increased promoter 1F methylation in gestational infants exposed to maternal stressors (Filiberto et al. 2011, Drake et al. 2012, Sinclair et al. 2012, Hompes et al. 2013). This finding is of interest because our DEX-sensitive cells originated from mothers with significantly higher pre-pregnancy BMI than those from the DEX-resistant group. Therefore, we hypothesize that maternal weight could be an important factor (stressor) in determining fetal endothelial glucocorticoid sensitivity. Altogether, our data suggest that differential methylation of GR promoters have significance in regulating GR mRNA isoform expression and are associated with glucocorticoid sensitivity. Together, our findings show that GRα gene expression in human endothelial cells is highly complex and that GR promoter methylation patterns differ between DEX-sensitive and DEX-resistant HUVECs. Although these differences in GR promoter methylation and 5′UTR mRNA expression do not account for the differences observed in GRα protein expression, function, and further alterations of endothelial phenotype, we hypothesize that they could be exploited to develop epigenetic markers of vascular glucocorticoid sensitivity. Importantly, global demethylation studies with AZA eliminated the differences between sensitivity groups in the in vitro response to DEX, indicating an important role of methylation patterns in establishing HUVEC sensitivity to glucocorticoids. Future research into the mechanisms that lead to differential methylation of GR proximal promoters, and the potential role of maternal obesity in programming fetal endothelial glucocorticoid sensitivity, are warranted. Supplementary Material 01 We thank Dr John A Cidlowski (NIEHS) for providing the full length human GR-1Cα expressing vectors, Dr Fuxia Xiong for helping in the design of cloning primers, and Dr Donna Thorpe for helping with the statistical analysis. Funding The present study was supported by the AHA SDG award #0630297N to Dr E M-G and by the School of Medicine, Loma Linda University. Figure 1 Basal and DEX-stimulated expression of GR mRNA transcripts in DEX-sensitive (S) and DEX-resistant (R) HUVECs. (A) Structure of the human glucocorticoid receptor gene (NR3C1) that includes 9 exons 1 and 3 stop signals that result in both 5′UTR and 3′UTR mRNA isoforms. (B, C, D and E) Confluent and quiescent HUVECs were treated with solvent, DEX 0.2, and DEX1 µM for 24 h and GR mRNA isoforms were quantified by real-time PCR as described under methods. (B) Basal and DEX-stimulated expression levels of GR 5′UTR mRNA isoforms (1B, 1C, 1D, 1F, and 1H) expressed as fg GR isoform/ng 18S RNA. (C) Expression of GR 5′UTR mRNA isoforms as percent of total GR mRNA. (D) Basal and DEX-stimulated expression of GR 3′UTR mRNA isoforms (GRα, GRP and GRβ) as fg GR isoform/ng 18S RNA; (E) Expression of GR 3′UTR mRNA isoforms as percent of total GR mRNA. THP1 monocytoid cells were used as a positive control to verify that HUVECs do not express isoforms 1A, 1E, 1I and 1J. Bars represent the average±s.e.m. (n = 15 DEX-sensitive and 10 DEX-resistant). *P<0.05 DEX-treated vs untreated; #P<0.05 DEX-resistant cells vs DEX-sensitive cells. Figure 2 Role of GRα overexpression in endothelial phenotype, expression autoregulation, and response to DEX. HUVECs were transfected with a promoter-less GR-1Cα expressing vector, then treated for 24 h with or without DEX (1 µM). Total RNA was extracted, reverse-transcribed and quantified by real-time PCR. (A and B) GRα overexpression in HUVECs upregulated both mRNA (A) and protein (B) levels in both DEX-sensitive (n = 5) and DEX-resistant (n = 5) HUVECs. (C, D and E) Effect of GRα overexpression on NOS3 downregulation (C), SERPINE1 upregulation (D) and activated partial thromboplastin time, aPTT (E). (F) Effect of GRα overexpression on GR 5′UTR mRNA isoform levels. Bars represent the average±s.e.m. (n = 5 DEX-sensitive and 5 DEX-resistant). *P<0.05 DEX-treated vs untreated; #P<0.05 DEX-resistant vs DEX-sensitive cells; †P<0.05 basal vs GRα overexpression. Figure 3 Basal and DEX-stimulated transcriptional activity of proximal GR promoters. (A) Structure of the proximal CpG island of the human GR gene. (B and C) HUVECs (12 DEX-sensitive and 8 DEX-resistant) were transfected with GR promoter-luciferase vectors and then treated with solvent, DEX 0.2, or DEX 1 µM for 24 h. Promoter activity was estimated by the levels of firefly luciferase activity, normalized to Renilla luciferase activity and expressed in fold activity of the empty pGL3 firefly basic reporter vector. Transfections were performed in quadruples. (B) Basal and DEX-dependent GR promoter usage in HUVECs. (C) Linear regression analysis of promoter usage (percent luciferase activity) against 5′UTR mRNA isoform expression (percent isoform levels). Averages for all HUVECs for each treatment group (DEX 0, 0.2 and 1 µM) were used. Bars represent the average±s.e.m.. *P<0.05 DEX-treated vs untreated; #P<0.05 DEX-resistant cells vs DEX-sensitive cells. Figure 4 Role of GR promoter methylation on GR expression and response to DEX. (A) Methylation of proximal promoters as determined by immunoprecipitation of 5-methylcytosine in DEX-sensitive (S) and DEX-resistant (R) cells (n=8/group). Results are shown as average percentage methylation. (B) The correlation between basal HUVEC GR promoter methylation and GR 5′UTR mRNA isoform expression was determined by linear regression analysis. Each data point represents the average levels for each sensitivity group. Promoter methylation percentages are shown as Iog10 values. (C and D) Global demethylation was achieved with 48 h of AZA treatment, and the expression of GR 5′UTR mRNA isoform was determined by real-time PCR (n = 5/sensitivity group). Results are shown as fg GR isoform/ng 18S RNA (C), and fold expression of untreated cells (D). (E and F) HUVECs were treated with AZA as described for (C) and (D) in the presence or absence of DEX 1µM, and then analyzed for NOS3 (E) and SERPINE1 (F) mRNA expression by real-time PCR. Bars represent the average±s.e.m. *P<0.05 DEX-treated vs untreated; #P<0.05 DEX-resistant cells vs DEX-sensitive cells; †P<0.05 non-AZA vs AZA treatment. Table 1 Summary of major findings on GR expression regulation in HUVECs Figure 1: GR mRNA isoform levels Resistant cells have higher expression of GR isoforms 1C and 1D, and lower expression of isoform 1F than sensitive cells in basal conditions. Dexamethasone-treatment downregulated 5′UTR mRNA isoform 1C and 3′UTR mRNA isoform α in resistant cells only. Figure 2: Effect of GR-1Cα Overexpression GR-1Cα overexpression in resistant cells did not improve in vitro response to dexamethasone as determined by NOS3 downregulation, SERPINE1 upregulation, and decreases in aPTT. GR-1Cα overexpression further downregulated the expression of 5′UTR mRNA isoforms 1B (in resistant cells only) and 1H in both sensitivity groups. Figure 3: GR promoter activity There were no significant differences in basal GR promoter (unmethylated) activity between sensitive and resistant cells. Dexamethasone treatment decreased promoter 1F activity in both sensitivity groups, decreased promoter 1B activity in resistant cells only and increased promoter 1D activity in sensitive cells only. Promoter activity differences between sensitive and resistant cells did not correlate with differences in GR 5′UTR mRNA levels. Figure 4: GR promoter methylation and AZA treatment Sensitive cells have higher promoter 1D methylation levels and lower promoter 1F and 1H methylation levels compared to resistant cells. Promoter methylation levels correlated with GR 5′UTR mRNA levels. AZA treatment upregulated the expression of all GR mRNA isoforms except isoform 1D in resistant cells. AZA treatment improved in vitro dexamethasone response in resistant cells as determined by NOS3 downregulation and SERPINE1 upregulation. Supplementary data This is linked to the online version of the paper at http://dx.doi.org/10.1530/JME-15-0124. Declaration of interest The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported. References Adcock IM Ito K Barnes PJ Glucocorticoids: effects on gene transcription Proceedings of the American Thoracic Society 2004 1 247 254 16113442 Alt SR Turner JD Klok M Meijer OC Lakke EAJF DeRijk RH Muller CP Differential expression of glucocorticoid receptor transcripts in major depressive disorder is not epigenetically programmed Psychoneuroendocrinology 2010 35 544 556 19782477 Auphan N DiDonato JA Helmberg A Karin M Immunosuppression by glucocorticoids: inhibition of NF-κ B activity through induction of I κ B synthesis Science 1995 270 232 233 7569969 Bamberger CM Schulte HM Chrousos GP Molecular determinants of glucocorticoid receptor function and tissue sensitivity of glucocorticoids Endocrine Reviews 1996 17 245 261 8771358 Barnes PJ Anti-inflammatory actions of glucocorticoids: molecular mechanisms Clinical Science 1998 94 557 572 9854452 Biddie SC Conway-Campbell BL Lightman SL Dynamic regulation of glucocorticoid signaling in health and disease Rheumatology 2012 51 403 412 21891790 Bockmuhl Y Murgatroyd CA Kuczynska A Adcock IM Almeida OF Spengler D Differential regulation and function of 5′-Untranslated GR-Exon 1 transcripts Molecular Endocrinology 2011 25 1100 1110 21527501 Butler J Kadonaga JT The RNA polymerase II core promoter: a key component in the regulation of gene expression Genes and Development 2002 16 2583 2592 12381658 Cao-Lei L Leija SC Kumsta R Wust S Meyer J Turner JD Muller CP Transcriptional control of the human glucocorticoid receptor: identification and analysis of alternative promoter regions Human Genetics 2011 129 533 543 21234764 Cao-Lei L Suwansirikul S Jutavijittum P Meriaux SB Turner JD Muller CP Glucocorticoid receptor gene expression and promoter CpG modifications throughout the human brain Journal of Psychiatric Research 2013 47 1597 1607 23948638 Drake AJ McPherson RC Godfrey KM Cooper C Lillycrop KA Hanson MA Meehan RR Seckl JR Reynolds RM An unbalanced maternal diet in pregnancy associates with offspring epigenetic changes in genes controlling glucocorticoid action and foetal growth Clinical Endocrinology 2012 77 808 815 22642564 Filiberto AC Maccani MA Koestler D Wilhelm-Benartzi C Avissar-Whiting M Banister CE Gagne LA Marsit CJ Birthweight is associated with DNA promoter methylation of the glucocorticoid receptor in human placenta Epigenetics 2011 6 566 572 21521940 Gonzalez-Rodriguez PJ Xiong F Li Y Zhou J Zhang L Fetal hypoxia increases vulnerability of hypoxic-ischemic brain injury in neonatal rats: role of glucocorticoid receptors Neurobiology of Disease 2014 65 172 179 24513088 Goyal R Zhang L Blood AB Baylink DJ Longo LD Oshiro B Mata-Greenwood E Characterization of an animal model of pregnancy-induced vitamin D deficiency due to metabolic dysregulation American Journal of Physiology. Endocrinology and Metabolism 2014 306 256 266 Hadoke PWF Iqbal J Walker BR Therapeutic manipulation of glucocorticoid metabolism in cardiovascular disease British Journal of Pharmacology 2009 156 689 712 19239478 Heitzer MD Wolf IM Sanchez ER Witchel SF DeFranco DB Glucocorticoid receptor physiology Reviews in Endocrine & Metabolic Disorders 2007 8 321 330 18049904 Hompes T Izzi B Gellens E Morreels M Fieuws S Pexsters A Schops G Dom M Van Bree R Freson K Investigating the influence of maternal cortisol and emotional state during pregnancy on the DNA methylation status of the glucocorticoid receptor gene (NR3C1) promoter region in cord blood Journal of Psychiatric Research 2013 47 880 891 23566423 Huang W Glass CK Nuclear receptors and inflammation control: molecular mechanisms and pathophysiological relevance Atherosclerosis, Thrombosis, and Vascular Biology 2010 30 1542 1549 Ito K Chung KF Adcock IM Update on glucocorticoid action and resistance Journal of Allergy and Clinical Immunology 2006 117 522 543 16522450 Juven-Gershon T Kadonaga JT Regulation of gene expression via the core promoter and the basal transcriptional machinery Developmental Biology 2010 339 225 229 19682982 Kadmiel M Cidlowski JA Glucocorticoid receptor signaling in health and disease Trends in Pharmacological Sciences 2013 34 518 530 23953592 Kassel O Herrlich P Crosstalk between the glucocorticoid receptor and other transcription factors: molecular aspects Molecular and Cellular Endocrinology 2007 275 13 29 17689856 Kimura H Li X Torii K Okada T Kamiyama K Mikami D Takahashi N Yoshida H Dexamethasone enhances basal and TNF-α stimulated production of PAI-1 via the glucocorticoid receptor regardless of 11β-hydroxysteroid dehydrogenase 2 status in human proximal renal tubular cells Nephrology, Dialysis, Transplantation 2009 24 1759 1765 Kino T Tissue glucocorticoid sensitivity: beyond stochastic regulation on the diverse actions of glucocorticoids Hormone and Metabolic Research 2007 39 420 424 17578758 Labonte B Zaoulay N Yerko V Turecki G Brunet A Epigenetic modulation of glucocorticoid receptors in posttraumatic stress disorder Translational Psychiatry 2014 4 e368 24594779 Lou Y Wen C Li M Adams DJ Wang M Yang F Morris BJ Whitworth JA Decreased renal expression of nitric oxide synthase isoforms in adrenocorticotropin-induced and corticosterone-induced hypertension Hypertension 2001 37 1164 1170 11304519 Mata-Greenwood E Liao WX Wang W Zheng J Chen DB Activation of AP-1 transcription factors differentiates FGF2 and vascular endothelial growth factor regulation of endothelial nitric oxide synthase expression in placental artery endothelial cells Journal of Biological Chemistry 2010 285 17348 17358 20371606 Mata-Greenwood E Stewart JM Steinhorn RH Pearce WJ Role of BAG1 on differential sensitivity of human endothelial cells to glucocorticoids Atherosclerosis, Thrombosis, and Vascular Biology 2013 33 1046 1057 Moser D Molitor A Kumsta R Tatschner T Riederer P Meyer J The glucocorticoid receptor gene exon 1-F promoter is not methylated at the NGFI-A binding site in human hippocampus World Journal of Biological Psychiatry 2007 8 262 268 17853270 Muzaffar S Shukla N Angelini GD Jeremy JY Prednisolone augments superoxide formation in porcine pulmonary artery endothelial cells through differential effects on the expression of nitric oxide synthase and NADPH oxidase British Journal of Pharmacology 2005 145 688 697 15852033 Petersen KB Geng CD Vedeckis WV Three mechanisms are involved in glucocorticoid receptor autoregulation in a human T-lymphoblast cell line Biochemistry 2004 43 10851 10858 15323545 Pratt WB Morishima Y Murphy M Harrell M Chaperoning of glucocorticoid receptors Handbook of Experimental Pharmacology 2006 172 111 138 Presul E Schmidt S Kofler R Helmberg A Identification, tissue expression, and glucocorticoid responsiveness of alternative first exons of the human glucocorticoid receptor Journal of Molecular Endocrinology 2007 38 79 90 17242171 Ramamoorthy S Cidlowski JA Ligand-induced repression of the glucocorticoid receptor gene is mediated by an NCoR1 repression complex formed by long-range chromatin interactions with intragenic glucocorticoid response elements Molecular and Cellular Biology 2013 33 1711 1722 23428870 Sinclair D Fullerton JM Webster MJ Weickert CS Glucocorticoid receptor 1B and 1C mRNA transcript alterations in schizophrenia and bipolar disorder, and their possible regulation by GR gene variants PLoS ONE 2012 7 e31720 22427805 Steer JH Kroeger KM Abraham LJ Joyce DA Glucocorticoids suppress tumor necrosis factor-α expression by human monocytic THP-1 cells by suppressing transactivation through adjacent NF-kB and c-Jun-activating transcription factor-2 binding sites in the promoter Journal of Biological Chemistry 2000 275 18432 18440 10748079 Szyf M Weaver ICG Champagne FA Diorio J Meaney MJ Maternal programming of steroid receptor expression and phenotype through DNA methylation in the rat Frontiers in Neuroendocrinology 2005 26 139 162 16303171 Tamura Y Kawao N Yano M Okada K Okumoto K Chiba Y Matsuo O Kaji H Role of plasminogen activator inhibitor-1 in glucocorticoid-induced diabetes and osteopenia in mice Diabetes 2015 64 2194 2206 25552599 Turner JD Schote AB Macedo JA Pelascini LPL Muller CP Tissue specific glucocorticoid receptor expression, a role for alternative first exon usage? Biochemical Pharmacology 2006 72 1529 1537 16930562 Turner JD Pelascini LPL Macedo JA Muller CP Highly individual methylation patterns of alternative glucocorticoid receptor promoters suggest individualized epigenetic regulatory mechanisms Nucleic Acid Research 2008 36 7207 7218 Turner JD Alt SR Cao L Vernocchi S Trifonova S Battello N Muller CP Transcriptional control of the glucocorticoid receptor: CpG islands, epigenetics and more Biochemical Pharmacology 2010 80 1860 1868 20599772 Van der Knaap LF Riese H Hudziak JJ Verbiest MMPJ Verhulst FC Oldehinkel AJ van Oort FVA Glucocorticoid receptor gene (NR3C1) methylation following stressful events between birth and adolescence. The TRAILS study Translational Psychiatry 2014 4 e381 24713862 Walker BR Glucocorticoids and Cardiovascular Disease European Journal of Endocrinology 2007 157 545 559 17984234 Wallace AD Cao Y Chandramouleeswaran S Cidlowski JA Lysine 419 targets human glucocorticoid receptor for proteasomal degradation Steroids 2010 75 1016 1023 20619282 Wallerath T Gödecke A Molojawi A Li H Schrader J Förstermann U Dexamethasone lacks effect on blood pressure in mice with a disrupted endothelial NO synthase gene Nitric Oxide 2004 10 36 41 15050533 Wenting-van Wijk MJG Blankenstein MA Lafeber FPJG Bijlsma JWJ Relation of plasma dexamethasone to clinical response Clinical and Experimental Rheumatology 1999 17 305 312 10410263 Witzmann SR Turner JD Meriaux SB Meijer OC Muller CP Epigenetic regulation of the glucocorticoid receptor promoter 1(7) in adult rats Epigenetics 2012 7 1290 1301 23023726 Xiong F Zhang L Role of the hypothalamic-pituitary-adrenal axis in developmental programming of health and disease Frontiers in Neuroendocrinology 2013 34 27 46 23200813 Yang S Zhang L Glucocorticoids and vascular reactivity Current Vascular Pharmacology 2004 2 1 12 15320828 Yang C Bolotin E Jiang T Sladek FM Martinez E Prevalence of the Initiator over the TATA box in human and yeast genes and identification of DNA motifs enriched in human TATA-less core promoters Gene 2007 389 52 65 17123746

PMC005xxxxxx/PMC5113727.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101302316 33345 Cell Host Microbe Cell Host Microbe Cell host & microbe 1931-3128 1934-6069 27693306 5113727 10.1016/j.chom.2016.09.002 NIHMS816601 Article Diverse intestinal bacteria contain putative zwitterionic capsular polysaccharides with anti-inflammatory properties Neff C. Preston 1* Rhodes Matthew E. 1* Arnolds Kathleen L. 1 Collins Colm B. 2 Donnelly Jody 1 Nusbacher Nichole 1 Jedlicka Paul 3 Schneider Jennifer M. 1 McCarter Martin D. 4 Shaffer Michael 1 Mazmanian Sarkis K. 5 Palmer Brent E. 1⌘ Lozupone Catherine A. 1⌘ 1 Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO 80045 USA 2 Department of Pediatrics, University of Colorado School of Medicine, Aurora, CO 80045 USA 3 Department of Pathology, University of Colorado Anschutz Medical Campus, Aurora, CO 80045 USA 4 Department of Surgery, University of Colorado Anschutz Medical Campus, Aurora, CO 80045 USA 5 Division of Biology & Biological Engineering, California Institute of Technology, Pasadena, CA 91125, USA ⌘ Corresponding Author: Lead Contact: Dr. Catherine Lozupone (Catherine.Lozupone@ucdenver.edu; Phone 303-724-7942; Fax 303-724-7212).; Dr. Brent Palmer (brent.palmer@ucdenver.edu; Phone 303-724-7203; Fax 303-724-7212) * Authors contributed equally 19 9 2016 29 9 2016 12 10 2016 12 10 2017 20 4 535547 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Zwitterionic capsular polysaccharides (ZPS) are bacterial products that modulate T cells, including inducing anti-inflammatory IL-10-secreting T regulatory cells (Tregs). However, only a few diverse bacteria are known to modulate the host immune system via ZPS. We present a genomic screen for bacteria encoding ZPS molecules. We identify diverse host-associated bacteria, including commensals and pathogens with known anti-inflammatory properties, with the capacity to produce ZPSs. Human mononuclear cells stimulated with lysates from putative ZPS-producing bacteria induce significantly greater IL-10 production and higher proportions of Tregs than lysates from non ZPS-encoding relatives or a commensal strain of Bacteroides cellulosilyticus in which a putative ZPS biosynthetic operon was genetically disrupted. Similarly, wild-type B. cellulosilyticus DSM 14838, but not a close relative lacking a putative ZPS, attenuated experimental colitis in mice. Collectively, this screen identifies bacterial strains that may use ZPS to interact with the host as well as those with potential probiotic properties. eTOC Identifying anti-inflammatory host-associated bacteria can reveal strategies to combat inflammatory disease. Neff et al. conduct a genomic screen to identify bacteria with the capability to produce capsular zwitterionic polysaccharides (ZPS). These ZPS-producing bacteria induce increased levels of anti-inflammatory IL-10 and Tregs and prevent experimental colitis in mice. Introduction The trillions of microorganisms that colonize the human body (the microbiota) control many aspects of both innate and adaptive immune responses (Hooper and Macpherson 2010; Molloy, Bouladoux, and Belkaid 2012) and a healthy microbiota plays a crucial role in maintaining immune homeostasis. Accordingly, dysbiosis of the gut microbiota is associated with many diseases characterized by chronic gut inflammation, including Inflammatory Bowel Diseases (Kverka et al. 2011) and HIV infection (Lozupone et al. 2014). Host-associated bacteria that influence T cell populations are of key interest, since T cells play a central role in controlling host immune status. One key class of molecules known to influence T cell populations in the context of both health and disease are capsular zwitterionic polysaccharides (ZPS)(Surana and Kasper 2012). Unlike most naturally occurring polysaccharides composed of negatively charged sugar molecules, ZPSs contain positive and negative repeating charges and can activate CD4+ T cells in complex ways (Avci and Kasper 2010; Tzianabos et al. 1993). The alternating charges are crucial, as chemical modification eliminates the ability of ZPS to modulate immune responses (Tzianabos et al. 1993). Diverse bacteria produce immune-modulatory ZPSs, including SP1 of Streptococcus pneumonia (Velez et al. 2009), CP8 of Staphylococcus aureus (O’Riordan and Lee 2004) and O-chain antigen of Morganella morganii (Young et al. 2011). The best-studied ZPS is Polysaccharide A (PSA) of the intestinal bacterium Bacteroides fragilis NCTC 9343. Certain ZPSs are involved in disease, having the ability to induce abscess in a T cell dependent manner (Tzianabos et al. 1993). However, certain ZPSs also appear to play important and diverse roles in both maintaining and restoring intestinal immune homeostasis. For instance, studies in mice have shown that the ZPS of B. fragilis NCTC 9343 PSA, can drive a Th1-mediated response to help maintain Th1/Th2 balance in the gut (Mazmanian et al. 2005). PSA can also induce anti-inflammatory IL-10- secreting T regulatory cells (Tregs), to protect against both T-cell mediated and chemically induced colitis in mice (Mazmanian, Round, and Kasper 2008; Round and Mazmanian 2010). Finally, immune assays with human PBMC have shown that PSA can also induce CD4+ T cells with suppressive capacity in humans (Telesford et al. 2015; Kreisman and Cobb 2011). The bacteria known to modulate the host immune system using ZPS are limited, and include phylogenetically diverse organisms that make structurally similar ZPS. Specifically, B. fragilis NCTC 9343 and S. pneumoniae make similar ZPSs called PSA and SP1 respectively (Figure 1), in which the crucial positive charge is conferred by the same amino sugar, acetamido-amino-2,4,6-trideoxygalactose (AATGal). Most strains of B. fragilis contain an AATGal-ZPS but these vary in structure, with only ~25% of strains making PSA and with strain 638R making PSA2, a more structurally complex ZPS (Figure 1) that shares similar immunological properties to PSA and a biosynthesis locus on a homologous chromosomal region (Wang et al. 2000). The biological activities of varied ZPS from different bacterial strains are generally thought to be similar (Stephen, Groneck, and Kalka-Moll 2010), although the degree to which AATGal-ZPSs may vary in function is not well understood. As an example Sp1, PSA and PSA2 all induce abscess and SP1 and PSA both induce T cell activation in an APC dependent manner (Wang et al. 2000; Kalka-Moll et al. 2002). However, PSA and not SP1 induced CD4+39+FoxP3+ cells in a Dendritic Cell (DC)-dependent manner in stimulations with human cells derived from peripheral blood (Telesford et al. 2015), which has functional implications because CD39 expression was shown to be important for PSA-mediated protection against CNS inflammation in murine models (Wang et al. 2014). Further expansion of our knowledge of the diversity of ZPSs in nature may pave the way to discovery of ZPS with therapeutic potential. To further expand our knowledge of the breadth of AATGal-ZPSs and the bacteria that produce them, we screened complete and draft genomes for those containing orthologues to AATGal biosynthetic genes and for a subset of these genomes, characterized the gene content of Capsular Polysaccharide Biosynthesis (CPS) operons that contained them. Furthermore, we tested the immune-modulatory properties of AATGal-ZPS encoding bacteria using human blood and tissue assays and a mouse model of colitis. The availability of thousands of sequenced genomes from a continually increasing repertoire of cultured human gut inhabitants makes such genomic screens coupled with functional experiments a powerful approach for understanding the distribution of key biological properties across human gut bacteria. Results Genomic screen for bacteria that encode AATGal-ZPS molecules The PSA operon of B. fragilis NCTC 9343 had been previously cloned and sequenced (Coyne et al. 2001). This operon contains transcriptional regulators upaY and upaZ, which are conserved across CPS loci in the Bacteroides (Krinos et al. 2001; Xu et al. 2003), 4 glycosyl transferases (GT) and other key genes for CPS biosynthesis such as a flippase and polymerase (Figure 1). The key genes for the biosynthesis and attachment of the AATGal amino sugar that gives PSA its positive charges, and thus zwitterionic property, are wcfR, which encodes a protein that synthesizes AATGal and wcfS, which encodes a GT that transfers this amino sugar to the polysaccharide backbone, in an adjacent position to wcfR (Figure 1) (Coyne et al. 2001). Gene homologues to wcfR and wcfS downstream of upaY and upaZ transcriptional regulators, but not the other genes in the PSA operon, were conserved in almost all of 50 strains of B. fragilis tested, indicating that AATGal-ZPS but not canonical PSA are conserved across B. fragilis (Coyne et al. 2001). We predicted which other bacteria may produce AATGal-ZPS by first performing a BLAST search for B. fragilis NCTC 9343 wcfR gene homologues in protein coding genes from 8065 genomes. To define a similarity threshold at which the BLAST hits were potentially orthologous to wcfR and thus produce a functionally equivalent molecule, we considered the fact that B. fragilis NCTC 9343 has a wcfR gene homologue within its genome (BLAST e-value = 3e−48) that encodes a protein with a different function, the aminotransferase DegT (ref|wp_010992163). Plotting the e-value distribution of hits to B. fragilis wcfR produced a bimodal distribution, with the DegT in the lower distribution (Figure S1A). We hypothesized BLAST hits in the upper distribution (< 1e−90 in our search) may be orthologues to wcfR. We next determined whether wcfR orthologues had other essential elements of AATGal-ZPS nearby in the genome, including a homologue to the wcfS gene and the upaY and/or upaZ transcriptional regulators (using a threshold of e-value < 1e−5). Diverse taxa contain diverse putative AATGal-ZPS operons Of the 8065 genomes screened, 517 had BLAST hits to the B. fragilis wcfR gene with an e-value < 1e−90, including taxonomically diverse bacteria in the Bacteroidales, Erysipelotrichales, Clostridiales, and Bacillales orders (Figure 2). In at least 409 of these 517 genomes, wcfR homologues had a wcfS homologue adjacent or in very close proximity (within 3 genes). Most genomes from bacteria in the Bacteroidales order contained homologues to the upaY and/or upaZ transcriptional regulators upstream, but wcfR homologue containing bacteria from other bacteria lineages did not. Consistent with previous reports that B. fragilis and S. pneumoniae produce AATGal-ZPS with related genes (Coyne et al. 2001), among our predicted producers of AATGal-ZPS molecules were 217 strains of S. pneumoniae (Figures 1,2). Our screen also identified the PSA2 operon of B. fragilis 638R, which has a related but more complex structure compared to PSA, with five monosaccharides rather than four (Figure 1, S3)(Wang et al. 2000). Genomes usually only contained one AATGal-ZPS operon but in rare cases could contain multiple, including B. cellulosilyticus strains DSM 14838 and CL02T12C19 and B. fragilis strains HMW 615 and YCH46 (Figure S3, S4). Predicted producers of AATGal-ZPS were scattered throughout the 16S rRNA phylogenetic tree of the Bacteroides genus and phylogenetically interspersed with species that did not contain predicted AATGal-ZPS operons (Figure 3A), indicating a pattern of multiple gains and losses. A similar phylogenetically interspersed pattern was also observed for predicted AATGal-ZPS and non- AATGal-ZPS producers in the Erysipelotrichales (Figure 3B) and the Clostridiales (data not shown) lineages. For AATGal-ZPS encoding bacteria in the Bacteroidales order, we predicted the operon gene content based on the genes located between the upaY/upaZ transcriptional regulators and the wcfR/wcfS gene homologues and also using the Database of Prokaryotic Operons (Door2) application (Mao et al. 2009). To relate the operons to each other based on their gene content, we binned genes into families and clustered the operons that had similar collections of genes families. Interestingly, and consistent with previous reports (Coyne et al. 2001), only 4 of the 13 surveyed strains of B. fragilis had a canonical PSA operon, B. fragilis strain 638R had a unique operon consistent with its production of PSA2, and 6 strains shared an operon for a previously uncharacterized AATGal-ZPS (Figure S3). Consistent with PSA and PSA2 of B. fragilis having related but unique structures, the content of sugar biosynthesis and GT genes of the operons suggest that the AATGal-ZPSs in the Bacteroidales include variable polysaccharide backbones that generally have 4–5 sugar repeating subunits (Figure S2; Table S1). Establishing an AATGal-ZPS immune phenotype in human blood and tissue To establish assays in human blood for determining whether bacteria with putative AATGal-ZPS operons have anti-inflammatory properties, we first performed 3-day stimulations of peripheral blood mononuclear cells (PBMC) from healthy adults with lysate of wild-type (WT) B. fragilis NCTC 9343, and B. fragilis with the PSA operon knocked-out (B. fragilis ΔPSA). After three days of stimulation, the cell culture supernatant was subjected to IL-10 ELISA and the cell cultures to flow cytometry to assess immune cell populations (see methods). B. fragilis WT lysate induced significantly higher IL-10 levels (p<0.0001, Figure 4A) and lower levels of the inflammatory cytokines IL-6 and TNF-α (p=0.01, p=0.02, respectively, Figure S5A) than B. fragilis ΔPSA. We also used a 24 hour Intracellular Cytokine Staining (ICCS) assay to evaluate whether any IL-10 was being produced by CD4+ T cells in these mixed cell populations. WT B. fragilis stimulations produced a significantly higher fraction of CD4+ IL-10+ cells than B. fragilis ΔPSA (p=0.0015, Figure 4B) but the vast majority of these CD4+IL-10+ cells were CD25−FoxP3− (p<0.0001, Figure 4C). We also found a significant decrease in Treg induction by B. fragilis ΔPSA compared to WT when we used CD127 and CTLA4 as additional markers to FoxP3 and CD25 (Figures 4E). Consistent with reports that CD25 and FoxP3 alone are inadequate human Treg markers (Sakaguchi et al. 2010; Gavin et al. 2006), stimulations with WT B. fragilis did not result in a significantly greater proportion of CD4+CD25+FoxP3+ cells compared to B. fragilis ΔPSA (Figure 4D). To establish an assay for determining whether AATGal-ZPS containing bacteria can stimulate naïve T cells in an APC-dependent manner as has been previously described for B. fragilis PSA, we stimulated isolated CD14+ monocytes with bacterial lysates prior to being co-cultured with purified naïve T cells. Even though no memory or effector T cells were present, WT B. fragilis induced higher levels of IL-10 and Tregs compared to B. fragilis ΔPSA (p = 0.026, p = 0.021, Figure 4F and 4G). Since fewer cells were used for this assay (two hundred thousand naive T cells) compared to whole PBMC (two million total cells), lower levels of IL-10 were observed. Since immune cell subsets can behave differently depending on body compartment (Hu and Pasare, 2013), we also conducted assays using lamina propria mononuclear cells (LPMC) from resected gut tissue supplied to us following surgery (note that surgery was usually conducted to remove tumors. Sections of healthy tissue were used in the assays). WT B. fragilis induced more IL-10 and more CD25+FoxP3+CD127−CTLA-4+ Tregs than B. fragilis ΔPSA in LPMC (p = 0.0005, p=0.008 Figure 4H, and 4I, respectively). Taken together, our results show that an increased proportion of CD25+FoxP3+CD127−CTLA-4+ Tregs and increased IL-10 in supernatants from 3-day stimulations of PBMC and LPMC are both an indicator of PSA-like activity in bacteria. Assaying phylogenetically interspersed ZPS encoding and non-encoding bacteria for PSA-like immunomodulatory activity We selected 6 putative AATGal-ZPS-encoding and 7 non-encoding strains that were phylogenetically interspersed on a 16S rRNA tree of the Bacteroidales and Erysipelotrichales orders for immune assays (Figure 3). Using phylogenetically interspersed bacteria helps to reduce the effects of other genetic differences between strains that may be confounding. As an example, although any individual bacterial isolate will usually differ in many genes from even a very close phylogenetic relative, all of the putative ZPS-operon containing bacteria selected here would not be expected to share any genes to the exclusion of the phylogenetically interspersed non-ZPS encoding bacteria by chance. Finding anti-inflammatory properties established for PSA to be associated with our genomic predictions of AATGal ZPS-operon presence, would thus strongly implicate the ZPS-operons as an underlying reason. The selected bacterial strains also contained wcfR orthologues that were scattered throughout the wcfR tree (Figure 2), which allows us to test whether our selected e-value threshold is diagnostic of ZPS-like activity. Finally, they contained AATGal-ZPS operons with diverse gene content (Figure S2). Bacterial strains were grown to early stationary phase in liquid broth media, and a qPCR assay validated that the ZPS-encoding bacteria tested were expressing the wcfR gene under these growth conditions (Figure S1B). Stimulation of human PBMC with lysates of the predicted ZPS-producers elicited significantly higher median levels of IL-10 and CD25+FoxP3+CD127−CTLA-4+ Tregs in both the Bacteroidales and the Erysipelotrichales (Figure 5 A, B). Individual predicted ZPS-producing bacteria were also each compared to its closest assayed non ZPS-producing relative with a paired T-test and in almost all cases the predicted ZPS-producer induced significantly more IL-10 and Tregs, except for a couple of case that were borderline significant (Table 1). Our results for stimulations in LPMC were less clear. Stimulations with LPMC showed significantly greater median IL-10 production in ZPS-producers than non-producers in the Bacteroidales (p=0.039) but not in the Erysipelotrichales (p=0.21, Figure 5C). Median Treg stimulation was not significantly higher in ZPS-producers versus non in LPMC (Figure 5D). Paired T-tests between predicted ZPS-producing bacteria and their closest assayed non ZPS-producing relative were only sometimes significant (Table 1). However, the closest non-ZPS producing relatives of the ZPS-producers had typically only a 16S rRNA % ID of ~92% ID (Table 1), indicating that differences in other genes in the genomes may be obscuring the effects of the ZPS. When we compared B. cellulosilyticus and B. intestinalis, the most highly related strains with opposite designations, statistical significance was reached for IL-10 production and Treg stimulation in both PBMC (p=0.0026, p=0.047, Figures S6A and S6B, respectively) and LPMC (p=0.0019, p=0.028, Figures S6C and S6D, respectively). ZPS-knockout strains of Bacteroides cellulosilyticus To further confirm that the increased IL-10 and Treg levels in these bacteria were conferred by putative ZPS operons, we disrupted a wcfR gene in B. cellulosilyticus DSM 14838 via targeted insertional mutagenesis using the pKNOCK-bla-ermGb vector as described by (Alexeyev 1999). B. cellulosilyticus DSM 14838 has two putative AATGal-ZPS operons in its genome (Figure S2), one was encoded on Scaffold 5 (ZPS1) and one was encoded on Scaffold 9 (ZPS2). The operon encoding ZPS1 was simpler than the operon encoding ZPS2, and showed greater similarity in gene content to PSA, sharing homologues to the wcfQ and wcfP GTs and to the wzx3 flippase (Figure S4). Sequencing of the cDNA produced from mRNA of WT B. cellulosilyticus DSM 14838 grown to late log phase showed that both copies of the gene were expressed by the cell population. By PCR confirmation of both genomic DNA and mRNA, we confirmed that the wcfR gene of ZPS1 and not ZPS2 was disrupted. Stimulation of PBMC with B. cellulosilyticus ΔZPS1 resulted in the production of significantly less IL-10 and Tregs compared to WT bacteria (Figure 5E and 4F). As with B. fragilis and its PSA knockout counterpart, when isolated APCs were first stimulated with B. cellulosilyticus or B. cellulosilyticus ΔZPS1 and then co-cultured with purified naïve CD4+ T cells, the knockout strain induced less IL-10 and Tregs compared to wild type (Figure 5G and H), indicating that this AATGal-ZPS of B. cellulosilyticus can also induce Tregs from naïve T cells in an APC-mediated fashion. ZPS producing B. cellulosilyticus DSM 14838 protects against colitis in a TNBS model To further test whether our genomic screens for immune-modulatory ZPS operons are predictive of the anti-inflammatory effects described for B. fragilis PSA, we tested the effect of gavaging mice with the putative ZPS producer B. cellulosilyticus DSM 14838 in a Trinitrobenzenesulfonic acid (TNBS) induced colitis model. In two separate trials, female C57/Bl6 mice aged 8 – 12 weeks were gavaged with 5.0 × 108 cells of bacteria in 200 μL of culture media or culture media without bacteria, once a week for 3 weeks. On the day of the third treatment, mice were challenged with TNBS. Gavage of mice with B. cellulosilyticus led to significantly decreased weight loss induced by TNBS enema relative to vehicle control, measured by two-way repeated measures ANOVA (p = 0.014, Figure 6C). Vehicle-gavaged mice displayed a significant reduction in initial body weight by day 5 post-enema (95% ± 2.3; n = 6; p < 0.01) while those receiving B. cellulosilyticus were protected from weight loss and actually gained weight overall (102.3% ± 0.6; n = 6). In contrast, mice gavaged with B. intestinalis, a close relative of B. cellulosilyticus (97.4% identity over their aligned 16S rRNA genes) that does not encode a putative ZPS operon displayed no such protection from TNBS-induced weight loss. Gavage of mice with B. cellulosilyticus also led to decreased histological evidence of inflammation (Figure 6A) as assessed in blinded manner by a trained pathologist (Figure 6B). The vehicle group of TNBS treated animals displayed histological evidence of inflammation (2.28 ± 0.26) that was significantly higher than those receiving B. cellulosilyticus (1.40 ±0.27; p = 0.02), while those receiving B. intestinalis displayed worse inflammation than vehicle controls, though the difference was not statistically significant (3.30 ± 0.55; p = 0.051). The decrease in inflammation coincided with an increase in CD4+ FoxP3+ regulatory T cells in the colonic lamina propria of the vehicle group (5.3% ± 0.77; n = 4) compared to mice gavaged with B. cellulosilyticus (7.9% ± 0.82; n = 4; p = 0.04) (Figure 6D and E). In contrast those receiving B. intestinalis displayed no significant difference in Treg frequency from vehicle controls. DISCUSSION Here we show that we can predict bacteria with anti-inflammatory properties in in vitro assays in humans based on the presence of predicted AATGal-ZPS operons in their genomes. Furthermore, the one putative AATGal-ZPS producer tested, B. cellulosilyticus DSM 14838, showed a robust protection of colitis in mice, indicating that this bacterial strain has potential probiotic applications. B. cellulosilyticus appears to be a benign member of the Bacteroides, especially when compared to B. fragilis. It was originally isolated from healthy human feces for its ability to degrade cellulose (Robert et al. 2007) and is a gut symbiont that is notable for its extensive glycobiome (McNulty et al. 2013). Whether it is the 2 AATGal-ZPSs encoded by B. cellulosilyticus DSM 14838 that confers this anti-colitic property remains to be elucidated. The significant reduction of Treg and IL-10 in a B. cellulosilyticus strain in which one AATGal-ZPS had been genetically disrupted supports a role for this ZPS in anti-inflammatory properties of B. cellulosilyticus DSM 14838. A wide variety of bacteria from diverse phylogenetic lineages appear to produce AATGal-ZPSs that may modulate the immune system. We focused our analyses on bacteria within the Bacteroidales order and one isolate in the Erysipelotrichales order; Similar follow-up with a larger diversity of predicted AATGal-ZPS producing bacteria would also be useful for confirming the significance of our predictions. Our analysis of AATGal-ZPS operon gene content of bacteria in the Bacteroidales order suggests that we are identifying a diversity of ZPSs that includes PSA2 of B. fragilis strain 638R. Our immune assay results suggest that diverse AATGal-ZPS containing bacteria induce IL-10 and Tregs in stimulations of mixed immune cell populations from both PMBC and LPMC. Furthermore, assays with our B. cellulosilyticus ΔZPS1 strain indicates that for this putative AATGal-ZPS, Treg induction occurs at least in part via the APC-dependent stimulation of naïve T cells. Further work to more deeply define mechanisms of immune-modulation in these ZPSs and whether they differ between the various ZPS sub-types is needed, including experiments with AATGal-ZPSs purified from different bacteria, a greater diversity of genetic knock-out strains, human assays with more defined immune cell populations, and more validation in mouse models. Analysis of a wider diversity of AATGal-ZPSs may elucidate ones with therapeutic potential. Even for the best-studied AATGal-ZPS, PSA of B. fragilis, the mechanisms of immune-modulation, especially in humans, are far from being completely understood. Previous work establishing anti-inflammatory effects of B. fragilis/PSA in mice and humans have shown that PSA induces Foxp3+ Treg cells that express IL-10 upon antigen presentation on DCs (Round and Mazmanian, 2010)(Telesford et al. 2015), a result which we confirm here for both B. fragilis PSA and an AATGal-ZPS in B. cellulosilyticus DSM 14838. However, we also found that PSA induced IL-10 production by CD4+ T cells in mixed cell populations was mediated mostly by FoxP3- cells in humans, which is consistent with the results of Kreisman and Cobb despite considerable differences in assay design (Kreisman and Cobb 2011). It has also been shown previously that PSA can directly induce IL-10 from FoxP3−/CD4+ T cells via TLRs in mice (Round et al. 2011). The bacterial mediation of CD4+ T cell IL-10 production by Tr1 cells in humans is not unique to B. fragilis; the probiotic Bifidobacterium breve also induces colonic IL-10 producing Tr1 cells that do not express FoxP3 (Jeon et al. 2012). This work also evaluated the effects of PSA on immune cells in the gut compartment, and stimulations performed with mixed cell populations from LPMC from human gut tissues showed comparable results to those performed with PBMC when comparing B. fragilis to B. fragilis ΔPSA. This is an important finding since immune cells in different body compartments have the potential to respond very differently to environmental stimuli (Hu and Pasare 2013), and indicates that immune phenotypes in PBMC can be relevant to the gut compartment. However, for the broad comparison of predicted ZPS-producers to non ZPS-producers, the results for LPMC were not as strong as for PBMC with only increased IL-10 production in the Bacteroidales being modestly significant and no significant differences in Treg induction. The weaker phenotype in LPMC may be related to inherent challenges when working with these tissues. There was more variability in the responses across samples, which may be due to factors that are difficult to control for such as the presence of different bacterial populations already on the mucosal samples. Also, inherent differences in the bacteria that go beyond presence/absence of AATGal-ZPSs (e.g. production of other factors by these microbes that also influence immune phenotypes) may indicate why we had the power to observe a difference in B. fragilis versus knockout in LPMC but not in the broader comparisons of putative ZPS-producers versus non-producers. Further experiments with additional ZPS knock-out strains and purified ZPS will allow for a more robust analysis of whether diverse AATGal-ZPSs have an anti-inflammatory phenotype in human LPMC. Our techniques identify commensal bacteria already known to have anti-inflammatory properties Some AATGal-ZPS encoding bacteria have been shown to have anti-inflammatory properties in other studies. For instance, the crude lysate of P. distasonis, and particularly its membranous fraction, attenuated DSS-induced murine colitis while preventing increases in several pro-inflammatory cytokines and promoting significantly more CD4+CD25+FoxP3+ cells in mesenteric lymph nodes (Kverka et al. 2011). P. distasonis has also been previously shown to induce Tregs through a T cell receptor that was shared with B. uniformis, another species that we predict to produce an AATGal-ZPS (Lathrop et al. 2011). However, abolishment of function upon digestion with protease K indicated that this antigen is a protein and not a polysaccharide, signifying that P. distasonis and B. uniformis may share additional factors for stimulating Tregs. The predicted ZPS producers also included close relatives of 2 bacteria in a 17-strain consortium previously shown to induce Tregs in mice and protect against colitis and allergic diarrhea in mouse models (Atarashi et al. 2013) (Clostridium ramosum, and Clostridum 7_3_45FAA which is highly related to Clostridium symbiosum (Figure 2). Use of a cognate antigen-driven suppressor assay determined that specific bacterial antigens likely contributed to expansion of Tregs in this model (Atarashi et al. 2013). Our literature search also identified papers describing the convergence of surface structure properties of B. fragilis with distantly related ZPS-producers. For instance, Clostridium spiroforme, a predicted ZPS-producer in the Erysipelotrichales, had previously been reported to have a polysaccharide capsule similar to that of B. fragilis that mediates haemagglutination in both species (Baldassarri et al. 1989). Interestingly, the electron micrographs of C. spiroforme reported in (Baldassarri et al. 1989) show outer-membrane vesicles (OMVs) that are strikingly similar to those of B. fragilis (Shen et al. 2012) (Figure S7). OMVs of B. fragilis carry PSA and can induce immunomodulatory effects and prevent experimental colitis (Shen et al. 2012). Our techniques identify human pathogens The predicted producers of AATGal- ZPSs also include pathogenic bacteria, such as Clostridium botulinum (botulism), Clostridium perfringens (gas gangrene;bacteremias;wound infection), and Brachyspira hyodysenteriae (swine dysentery) and bacterial species that may opportunistically infect immunocompromised hosts such as C. ramosum (van der Vorm et al. 1999) and C. symbiosum (Elsayed and Zhang 2004). The presence of the AATGal-ZPS operon orthologues in pathogenic bacteria is not surprising given that ZPS producing bacteria including B. fragilis, S. pneumoniae, and S. aureus, often are commensal bacteria that can cause disease when they have escaped their normal habitats (Surana and Kasper 2012). B. fragilis is considered to be the most virulent of all of the Bacteroides species because of frequent isolation from a variety of clinical specimens including intra-abdominal abscess, blood, wound, perirectal, pelvic and other sites (Polk and Kasper 1977), and abscess formation requires PSA (Lindberg et al. 1982; Mazmanian and Kasper 2006; Onderdonk et al. 1977). S. aureus induces skin abscess in a manner that is also dependent on its ZPS (Weidenmaier, McLoughlin, and Lee 2010), and S. pneumoniae strains that produce Sp1 are the most common cause of bacterial pneumonia (Stephen, Groneck, and Kalka-Moll 2010). Although ZPSs can also in certain contexts be pro-inflammatory, e.g. as Th17 cell inducers (Mazmanian and Kasper 2006; Surana and Kasper 2012), Treg induction itself is known to be important for many pathogens as a mechanism for evasion of immune clearance and persistence in the host (Belkaid 2007; Mills 2004). Consistent with this notion, infection with putative ZPS-producer B. hyodysenteriae, which induces a severe colitis in pigs known as swine dysentery, induces a mucosal CD4+ T-cell response. This suggests that the induction of both pro-inflammatory and regulatory responses are important in the pathogenesis of this species (Hontecillas et al. 2005). CONCLUSIONS Taken together this work has established a greater diversity of AATGal-ZPSs and bacteria that carry them than has been previously appreciated. Much further study will be needed to understand how this expanded class of important immune-modulatory molecules mediate interactions with the host. EXPERIMENTAL PROCEDURES Putative ZPS operons were identified using a BLASTP search of the wcfR, wcfS, upaY, and upaZ genes of the PSA operon of B. fragilis NCTC 9343 against the entire collection of completed and draft genomes in the NCBI database on September of 2013. The proximity of wcfR, wcfS, upaY, and upaZ gene homologues on genomic contigs were determined using custom code. Operon gene membership was further defined using DOOR2 (Mao et al. 2014) and compared as detailed in the Supplemental Methods. Phylogenetic trees of the wcfR genes and the 16S rRNA genes from the surveyed bacteria were created as described in the Supplemental Methods. The bacterial isolates used in the immune assays were purchased from the ATCC or DSMZ and grown in rich media depending on their preferences as described in the Supplemental Methods. Bacteria were subjected to freeze/thaw or heat killing before being used in immune stimulations. Since immune stimulations with heat killed compared to freeze/thaw lysates resulted in no differences (data not shown), all reported stimulations were done with freeze/thaw lysates in order to avoid protein denaturing. The qPCR assays that were used to verify expression of wcfR in B. cellulosilyticus, B. fragilis WT, and B. uniformis are described in the Supplemental Methods and Figure S1B. The B. cellulosilyticusΔZPS1 strain was developed using the pKNOCK-bla-ermGb vector to disrupt wcfR via targeted insertional mutagenesis as previously described (Alexeyev 1999) and as detailed in the Supplemental Methods. To conduct the immune assays, human PBMCs were isolated by Ficol gradient centrifugation as previously described (Chain et al. 2013; Kassu et al. 2010; Neff et al. 2015) from the blood of 13 normal individuals. Informed consent was obtained and the study protocol was approved by the Colorado Multiple Institutional Review Board (COMIRB #14-1595). PBMCs were cultured with 10 μg freeze killed bacterial lysate for 3 days at 37°C. Six and 10-day stimulations were also performed and had similar results to those of 3-day stimulations (data not shown). Stimulations were performed in the presence of Streptomycin and Penicillin and in aerobic conditions and no bacterial growth was observed in the cell cultures. Unlike in the assays conducted by Kreisman and Cobb (Kreisman and Cobb 2011) in human PBMC, our stimulations were conducted in absence of exogenous IL-2; the IL-2 receptor CD25 is upregulated in the presence of IL-2, which could compromise our Treg staining. Cytokine secretion after 3-day stimulations with bacterial lysates was quantified in supernatant using ELISA Ready Set Go! (eBioscience), Tregs were enumerated using flow cytometry, and ICCS was used to determine IL-10+ cells, all as detailed in the Supplementary Methods. To determine whether induced IL-10 production and Tregs were due to restimulation of memory T cells, CD14+ monocytes were isolated by magnetic bead selection (Miltenyi Biotec), plated, stimulated with bacterial lysate for 4 hours and washed twice with PBS. Naïve T cells were negatively selected by magnetic beads (Miltenyi Biotec) and were added to the bacterial stimulated APCs. After three days supernatant was collected and subjected to IL-10 ELISA and cells were enumerated for Tregs. Intestinal lamina propria lymphocytes were isolated from resected gut tissue of jejunum, duodenum and colon, which were obtained from patients undergoing elective abdominal surgery. All patients signed a release to allow the unrestricted use of discarded tissues for research purposes and all protected patient information was de-identified to the laboratory investigators. The lamina propria lymphocytes were isolated from these tissues as detailed in the Supplemental methods. To test for protection in a murine TNBS colitis model, female C57/Bl6 mice aged 8 – 12 weeks were gavaged with bacteria or control PBS on days 0, 7, and 14 as described in the Supplemental Methods. Mice were cohoused prior to the intervention, and then housed in separate cages once the treatments began to prevent transmission of the introduced bacteria between mice. Mice were anesthetized, shaved and skin painted with 100μl of 1% TNBS in 100% EtOH on day 7 and received a rectal enema of 2.5% TNBS in 40% EtOH (5μl/g body weight) on day 14. Colonic tissue was prepared for histological assessment of disease as described in the Supplemental Materials or digested as previously described (Collins et al. 2013). CD4+CD25+FoxP3+ cells in colonic tissue were enumerated with flow cytometry as described in the Supplemental methods and statistical analysis was performed using GraphPad Prism. These experiments were approved by IACUC (B-104413(12)1E). Supplementary Material supplement Sources of Support We would like to thank Ashleigh Jones for her technical assistance with the mouse experiments. We would also like to express deep gratitude to Eric Martens for sharing with us the pKNOCK-bla-ermGb vector and for his guidance in conducting insertional transposon mutagenesis. This work was funded from R01 DK104047. Dr. Lozupone was also supported by K01 DK090285. Dr. Neff was supported by NIH T32 AI007405. Dr. Collins was supported by K01 DK099403. Figure 1 PSA operon of B. fragilis NCTC 9343 and chemical structures of B. fragilis NCTC 9343 PSA, B. fragilis 638R PSA2, and S. pneumoniae SP1. A. The PSA operon of B. fragilis and the conserved properties of B. fragilis AATGal-ZPS operons, as described by Coyne et al. (Coyne et al. 2001) B. Chemical structure of PSA of B. fragilis and related ZPSs. The shaded regions indicate the amino sugar AATGal that is synthesized by the wcfR gene and transferred by the wcfS gene to a polysaccharide backbone that varies in structure between these ZPSs. Genes shared across Bacteroidales AATGal-ZPSs, and examples of other AATGal-ZPS operons are shown in Figure S2, S3, and S4. Figure 2 A neighbor joining tree of BLAST hits to the wcfR gene of B. fragilis with an e-value <1e−90. Highly related sequences from different strains of the same species (or undesignated strains that are likely the same species) are collapsed into a wedge with the number of strains (genomes) within that wedge noted. Colors indicate the taxonomic order of the bacteria from which the wcfR sequence came. The wcfR genes of bacterial strains tested in immune assays are indicated with a star, showing that tested bacteria include those with wcfR homologues throughout this phylogeny. Figure S7 shows electron micrograph evidence of a convergence in surface structures in bacteria that are very distantly related phylogenetically but share highly similar wcfR gene homologues in their genomes. Figure 3 16S rRNA trees of all genomes that were screened in the orders Bacteroidales and Erysipelotrichales showing presence/absence of PSA-like operons. A. Strains surveyed in the Bacteroidales order and B. Strains surveyed in the Erysipelotrichales order. Species shown in red text had genes encoding both wcfR and wcfS and the genes were in adjacent or close positions within the operon. Those without a close homologue to wcfR are in black. Nodes in which all bacteria had the same color designation are collapsed into wedges with the number of strains noted in parentheses. The stars indicate predicted PSA-producing (in red) and non PSA-producing (in black) bacterial strains selected for immunologic testing. Figure 4 WT B. fragilis induces higher IL-10 production and proportion of Tregs compared to B. fragilis ΔPSA in human PBMC and LPMC. A. Levels of IL-10 in the supernatant of PBMC cultured with WT B. fragilis and B. fragilis ΔPSA for 3 days as determined by ELISA. B. Percent IL-10+ CD4+ T cells and C. CD4+IL-10+ T cells that are CD25+/−FoxP3+/− after one day stimulation with WT B. fragilis. D. Proportion of CD4+ T cells that are CD25+FoxP3+ in PBMC after 3 days of culture with WT B. fragilis or B. fragilis ΔPSA. E. Proportion CD4+ T cells that are CD25+FoxP3+CD127−CTLA4+ in PBMC after 3 days of culture with WT B. fragilis and B. fragilis ΔPSA. F. Levels of IL-10 in the supernatant and G. proportion of Tregs generated from purified naïve T cells mixed with bacteria stimulated purified CD14+ monocytes. H. Levels of IL-10 in the supernatant and I. proportion of CD4+ T cells that are CD25+FoxP3+CD127−CTLA4+ of LPMC after 3 days of culture with bacterial lysates. Statistical significance was calculated by paired T tests. Data for other cytokines (IL-6, TNF-α, IL-17, and IL-22) are in Figure S5A. Representative staining for CD25+FoxP3+CTLA4+CD127−Tregs in human PBMC and LPMC are in Figure S5B. Figure 5 (Left) Putative PSA producers in the Bacteroidales or Erysipelotrichales orders induce more IL-10 and Tregs compared to their non-PSA producing relatives in PBMC and LPMC. A. IL-10 levels and B. proportion of CD25+FoxP3+CTLA4+CD127−cells of CD4+ T cells in PBMC cultured with bacteria from the Bacteroidales (circles) or Erysipelotrichales order (squares) for 3 days. C. IL-10 levels and D. proportion CD25+FoxP3+CTLA4+CD127− cells of CD4+ T cells in LPMC cultured with bacteria from the Bacteroidales (circles) or Erysipelotrichales order (squares) for 3 days. Predicted PSA producers are in red and those that are not are in black. Statistical significance between medians of predicted and non-predicted PSA producers was calculated by nonparametric Mann-Whitney tests. (Right) WT B. cellulosilyticus induces higher IL-10 production and proportion of Tregs compared to a wcfR knockout in whole PBMC and purified naïve T cells. E. Levels of IL-10 and F. proportion CD4+ T cells that are CD25+FoxP3+CD127−CTLA4+ in PBMC after 3 days of culture with B. cellulosilyticus and B. cellulosilyticus ΔZPS1. G. Levels of IL-10 in the supernatant and H. proportion of Tregs generated from purified naïve T cells mixed with bacteria stimulated purified CD14+ monocytes. Statistical significance was calculated by paired T tests. Comparisons of stimulations with WT B. cellulosilyticus and the closely related B. intestinalis are in Figure S6. Figure 6 B. cellulosyliticus administration attenuates murine colitis in vivo. A. Representative micrographs of H&E stained colonic sections demonstrate that B. cellulosyliticus causes a decrease in intestinal inflammation relative to either vehicle controls or mice gavaged with B. intestinalis (black scale bar 200 μm). B. Histological inflammatory index scores show protection from inflammation in mice gavaged with B. cellulosilyticus relative to vehicle control or B. intestinalis. C. Weight loss curve data correlates with attenuated inflammation seen in histological scores. D. Representative contour plots showing expression of FoxP3+ T cells from the colonic lamina propria of each group, then enumerated E. * P < 0.05, ** P < 0.01. Table 1 Individual PSA-producing bacteria were compared with their closest non-PSA-producer relative for IL-10 and Treg induction in PBMC and LPMC. Statistical significance was calculated by paired T tests and the p-values for each comparison are noted in the table The 16S rRNA percent identity between pairs was calculated based on aligned sequenced in Arb (Kumar et al. 2005). PSA-producer Closest non PSA-producer 16S rRNA % Identity PBMC IL-10 PBMC Tregs LPMC IL-10 LPMC Tregs Bacteroides fragilis Bacteroides ovatus 94 0.007 0.041 0.304 0.025 Bacteroides cellulosilyticus Bacteroides intestinalis 97.4 0.003 0.047 0.002 0.028 Bacteroides uniformis Bacteroides stercoris 91.1 0.063 0.009 0.619 0.260 Bacteroides fluxus Bacteroides stercoris 93 0.015 0.034 0.184 0.495 Parabacteroides distasonis Prevotella stercorea 84.4 0.002 0.089 0.611 0.339 Clostridium spiroforme Catenibacterium mitsuoki 92.1 <0.0001 0.003 0.002 0.225 Highlights Genomic screen identifies bacteria with ability to produce zwitterionic polysaccharides. Predicted bacteria display anti-inflammatory immune modulation in human cell assays. Putative ZPS-producer Bacteroides cellulosilyticus protected mice from colitis. ZPS-producing bacteria may have therapeutic potential. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Author Contributions Conceptualization, C.A.L. and B.E.P.; Methodology, C.P.N., M.E.R., K.L.A., C.B.C, C.A.L., and B.E.P.; Software, M.E.R., M.S., and C.A.L.; Formal Analysis, C.P.N., M.E.R., K.L.A., C.B.C., B.E.P., and C.A.L.; Investigation, C.P.N., K.L.A., C.B.C., J.D., N.N., P.J., and J.M.S.; Resources, M.D.M. and S.K.M.; Data Curation, M.E.R., M.S., and C.P.N.; Writing – Original Draft, C.P.N., M.E.R., C.B.C., B.E.P., and C.A.L.; Writing – Review & Editing, K.L.A., M.S., S.K.M; Visualization, C.P.N., M.E.R., K.L.A., C.B.C., B.E.P., and C.A.L.; Supervision, B.E.P. and C.A.L., Project Administration, B.E.P. and C.A.L.; Funding Acquisition, C.A.L., B.E.P. and C.B.C. Alexeyev MF 1999 The pKNOCK series of broad-host-range mobilizable suicide vectors for gene knockout and targeted DNA insertion into the chromosome of gram-negative bacteria Biotechniques 26 824 6 28 10337469 Atarashi K Tanoue T Oshima K Suda W Nagano Y Nishikawa H Fukuda S Saito T Narushima S Hase K Kim S Fritz JV Wilmes P Ueha S Matsushima K Ohno H Olle B Sakaguchi S Taniguchi T Morita H Hattori M Honda K 2013 Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota Nature 500 232 6 23842501 Avci FY Kasper DL 2010 How bacterial carbohydrates influence the adaptive immune system Annu Rev Immunol 28 107 30 19968562 Baldassarri L Pantosti A Caprioli A Mastrantonio P Donelli G 1989 Haemagglutination and surface structures in strains of Clostridium spiroforme FEMS Microbiol Lett 51 1 4 2477303 Belkaid Y 2007 Regulatory T cells and infection: a dangerous necessity Nat Rev Immunol 7 875 88 17948021 Chain JL Martin AK Mack DG Maier LA Palmer BE Fontenot AP 2013 Impaired function of CTLA-4 in the lungs of patients with chronic beryllium disease contributes to persistent inflammation J Immunol 191 1648 56 23851684 Collins CB Aherne CM Ehrentraut SF Gerich ME McNamee EN McManus MC Lebsack MD Jedlicka P Azam T de Zoeten EF Dinarello CA Rivera-Nieves J 2013 Alpha-1-antitrypsin therapy ameliorates acute colitis and chronic murine ileitis Inflamm Bowel Dis 19 1964 73 23835442 Coyne MJ Tzianabos AO Mallory BC Carey VJ Kasper DL Comstock LE 2001 Polysaccharide biosynthesis locus required for virulence of Bacteroides fragilis Infect Immun 69 4342 50 11401972 Elsayed S Zhang K 2004 Bacteremia caused by Clostridium symbiosum J Clin Microbiol 42 4390 2 15365052 Gavin MA Torgerson TR Houston E DeRoos P Ho WY Stray-Pedersen A Ocheltree EL Greenberg PD Ochs HD Rudensky AY 2006 Single-cell analysis of normal and FOXP3-mutant human T cells: FOXP3 expression without regulatory T cell development Proc Natl Acad Sci U S A 103 6659 64 16617117 Hontecillas R Bassaganya-Riera J Wilson J Hutto DL Wannemuehler MJ 2005 CD4+ T-cell responses and distribution at the colonic mucosa during Brachyspira hyodysenteriae-induced colitis in pigs Immunology 115 127 35 15819705 Hooper LV Macpherson AJ 2010 Immune adaptations that maintain homeostasis with the intestinal microbiota Nat Rev Immunol 10 159 69 20182457 Hu W Pasare C 2013 Location, location, location: tissue-specific regulation of immune responses J Leukoc Biol 94 409 21 23825388 Jeon SG Kayama H Ueda Y Takahashi T Asahara T Tsuji H Tsuji NM Kiyono H Ma JS Kusu T Okumura R Hara H Yoshida H Yamamoto M Nomoto K Takeda K 2012 Probiotic Bifidobacterium breve induces IL-10-producing Tr1 cells in the colon PLoS Pathog 8 e1002714 22693446 Kalka-Moll WM Tzianabos AO Bryant PW Niemeyer M Ploegh HL Kasper DL 2002 Zwitterionic polysaccharides stimulate T cells by MHC class II-dependent interactions J Immunol 169 6149 53 12444118 Kassu A Marcus RA D’Souza MB Kelly-McKnight EA Golden-Mason L Akkina R Fontenot AP Wilson CC Palmer BE 2010 Regulation of virus-specific CD4+ T cell function by multiple costimulatory receptors during chronic HIV infection J Immunol 185 3007 18 20656923 Kreisman LS Cobb BA 2011 Glycoantigens induce human peripheral Tr1 cell differentiation with gut-homing specialization J Biol Chem 286 8810 8 21228275 Krinos CM Coyne MJ Weinacht KG Tzianabos AO Kasper DL Comstock LE 2001 Extensive surface diversity of a commensal microorganism by multiple DNA inversions Nature 414 555 8 11734857 Kumar Y Westram R Behrens S Fuchs B Glockner FO Amann R Meier H Ludwig W 2005 Graphical representation of ribosomal RNA probe accessibility data using ARB software package BMC Bioinformatics 6 61 15777482 Kverka M Zakostelska Z Klimesova K Sokol D Hudcovic T Hrncir T Rossmann P Mrazek J Kopecny J Verdu EF Tlaskalova-Hogenova H 2011 Oral administration of Parabacteroides distasonis antigens attenuates experimental murine colitis through modulation of immunity and microbiota composition Clin Exp Immunol 163 250 9 21087444 Lathrop SK Bloom SM Rao SM Nutsch K Lio CW Santacruz N Peterson DA Stappenbeck TS Hsieh CS 2011 Peripheral education of the immune system by colonic commensal microbiota Nature 478 250 4 21937990 Lindberg AA Weintraub A Kasper DL Lonngren J 1982 Virulence factors in infections with bacteroides fragilis: isolation and characterization of capsular polysaccharide and lipopolysaccharide Scand J Infect Dis Suppl 35 45 52 6188206 Lozupone CA Rhodes ME Neff CP Fontenot AP Campbell TB Palmer BE 2014 HIV-induced alteration in gut microbiota: driving factors, consequences, and effects of antiretroviral therapy Gut Microbes 5 562 70 25078714 Mao F Dam P Chou J Olman V Xu Y 2009 DOOR: a database for prokaryotic operons Nucleic Acids Res 37 D459 63 18988623 Mao X Ma Q Zhou C Chen X Zhang H Yang J Mao F Lai W Xu Y 2014 DOOR 2.0: presenting operons and their functions through dynamic and integrated views Nucleic Acids Res 42 D654 9 24214966 Mazmanian SK Kasper DL 2006 The love-hate relationship between bacterial polysaccharides and the host immune system Nat Rev Immunol 6 849 58 17024229 Mazmanian SK Liu CH Tzianabos AO Kasper DL 2005 An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system Cell 122 107 18 16009137 Mazmanian SK Round JL Kasper DL 2008 A microbial symbiosis factor prevents intestinal inflammatory disease Nature 453 620 5 18509436 McNulty NP Wu M Erickson AR Pan C Erickson BK Martens EC Pudlo NA Muegge BD Henrissat B Hettich RL Gordon JI 2013 Effects of diet on resource utilization by a model human gut microbiota containing Bacteroides cellulosilyticus WH2, a symbiont with an extensive glycobiome PLoS Biol 11 e1001637 23976882 Mills KH 2004 Regulatory T cells: friend or foe in immunity to infection? Nat Rev Immunol 4 841 55 15516964 Molloy MJ Bouladoux N Belkaid Y 2012 Intestinal microbiota: shaping local and systemic immune responses Semin Immunol 24 58 66 22178452 Neff CP Chain JL MaWhinney S Martin AK Linderman DJ Flores SC Campbell TB Palmer BE Fontenot AP 2015 Lymphocytic alveolitis is associated with the accumulation of functionally impaired HIV-specific T cells in the lung of antiretroviral therapy-naive subjects Am J Respir Crit Care Med 191 464 73 25536276 O’Riordan K Lee JC 2004 Staphylococcus aureus capsular polysaccharides Clin Microbiol Rev 17 218 34 14726462 Onderdonk AB Kasper DL Cisneros RL Bartlett JG 1977 The capsular polysaccharide of Bacteroides fragilis as a virulence factor: comparison of the pathogenic potential of encapsulated and unencapsulated strains J Infect Dis 136 82 9 886206 Polk BF Kasper DL 1977 Bacteroides fragilis subspecies in clinical isolates Ann Intern Med 86 569 71 322563 Robert C Chassard C Lawson PA Bernalier-Donadille A 2007 Bacteroides cellulosilyticus sp. nov., a cellulolytic bacterium from the human gut microbial community Int J Syst Evol Microbiol 57 1516 20 17625186 Round JL Lee SM Li J Tran G Jabri B Chatila TA Mazmanian SK 2011 The Toll-like receptor 2 pathway establishes colonization by a commensal of the human microbiota Science 332 974 7 21512004 Round JL Mazmanian SK 2010 Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota Proc Natl Acad Sci U S A 107 12204 9 20566854 Sakaguchi S Miyara M Costantino CM Hafler DA 2010 FOXP3+ regulatory T cells in the human immune system Nat Rev Immunol 10 490 500 20559327 Shen Y Giardino Torchia ML Lawson GW Karp CL Ashwell JD Mazmanian SK 2012 Outer membrane vesicles of a human commensal mediate immune regulation and disease protection Cell Host Microbe 12 509 20 22999859 Stephen TL Groneck L Kalka-Moll WM 2010 The modulation of adaptive immune responses by bacterial zwitterionic polysaccharides Int J Microbiol 2010 917075 21234388 Surana NK Kasper DL 2012 The yin yang of bacterial polysaccharides: lessons learned from B. fragilis PSA Immunol Rev 245 13 26 22168411 Telesford KM Yan W Ochoa-Reparaz J Pant A Kircher C Christy MA Begum-Haque S Kasper DL Kasper LH 2015 A commensal symbiotic factor derived from Bacteroides fragilis promotes human CD39(+)Foxp3(+) T cells and Treg function Gut Microbes 6 234 42 26230152 Tzianabos AO Onderdonk AB Rosner B Cisneros RL Kasper DL 1993 Structural features of polysaccharides that induce intra-abdominal abscesses Science 262 416 9 8211161 van der Vorm ER von Rosenstiel IA Spanjaard L Dankert J 1999 Gas gangrene in an immunocompromised girl due to a Clostridium ramosum infection Clin Infect Dis 28 923 4 10825071 Velez CD Lewis CJ Kasper DL Cobb BA 2009 Type I Streptococcus pneumoniae carbohydrate utilizes a nitric oxide and MHC II-dependent pathway for antigen presentation Immunology 127 73 82 18778282 Wang Y Kalka-Moll WM Roehrl MH Kasper DL 2000 Structural basis of the abscess-modulating polysaccharide A2 from Bacteroides fragilis Proc Natl Acad Sci U S A 97 13478 83 11106392 Wang Y Telesford KM Ochoa-Reparaz J Haque-Begum S Christy M Kasper EJ Wang L Wu Y Robson SC Kasper DL Kasper LH 2014 An intestinal commensal symbiosis factor controls neuroinflammation via TLR2-mediated CD39 signalling Nat Commun 5 4432 25043484 Weidenmaier C McLoughlin RM Lee JC 2010 The zwitterionic cell wall teichoic acid of Staphylococcus aureus provokes skin abscesses in mice by a novel CD4+ T-cell-dependent mechanism PLoS One 5 e13227 20949105 Xu J Bjursell MK Himrod J Deng S Carmichael LK Chiang HC Hooper LV Gordon JI 2003 A genomic view of the human-Bacteroides thetaiotaomicron symbiosis Science 299 2074 6 12663928 Young NM Kreisman LS Stupak J MacLean LL Cobb BA Richards JC 2011 Structural characterization and MHCII-dependent immunological properties of the zwitterionic O-chain antigen of Morganella morganii Glycobiology 21 1266 76 21321054

PMC005xxxxxx/PMC5113738.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9706581 34240 Curr Protoc Neurosci Curr Protoc Neurosci Current protocols in neuroscience 1934-8584 1934-8576 27696360 5113738 10.1002/cpns.16 NIHMS812620 Article Automatic dendritic spine quantification from confocal data with Neurolucida 360 Dickstein Dara L. 123 Dickstein Daniel R. 12 Janssen William G. M. 12 Hof Patrick R. 123 Glaser Jack 4 Rodriguez Alfredo 4 O’Connor Nate 4 Angstman Paul 4 Tappan Susan J. 4 1 Fishberg Department of Neuroscience, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA 2 Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA 3 Computational Neurobiology and Imaging Center, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA 4 MBF Bioscience, Williston, VT 05495, USA Corresponding author: Susan Tappan, susan@mbfbioscience.com 4 9 2016 3 10 2016 3 10 2016 03 10 2017 77 1.27.11.27.21 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Determining the density and morphology of dendritic spines is of high biological significance given the role of spines in synaptic plasticity and in neurodegenerative and neuropsychiatric disorders. Precise quantification of spines in three dimensions (3D) is essential for understanding the structural determinants of normal and pathological neuronal function. However, this quantification has been restricted to time- and labor-intensive methods such as electron microscopy and manual counting, which have limited throughput and are impractical for studies of large samples. While there have been some automated software packages that quantify spine number, they are limited in terms of their characterization of spine structure. This unit presents methods for objective dendritic spine morphometric analysis by providing image acquisition parameters needed to ensure optimal data series for proper spine detection, characterization, and quantification with Neurolucida 360. These protocols will be a valuable reference for scientists working towards quantifying and characterizing spines. dendritic spines neurons confocal microscopy automated quantification Introduction Because of their capacity for plasticity and their role as the location of excitatory synapses, accurately characterizing the structure of dendritic spines is of profound biological significance. Spine morphology determines the strength, stability and function of excitatory synaptic connections that subserve the neuronal networks underlying cognitive function. Developmental, aging-, and disease-related structural changes in neurons and dendritic spines, and their functional consequences, remain poorly understood. Therefore, there is value in imaging and quantifying specific spine populations, types, densities, and distribution along neuronal dendrites across brain regions or cortical layers. It is also important to assess entire neurons or dendritic trees as it permits to assess tree complexity and dendritic branching, both of which may be altered in disease states. Optimally both spines and neurons should be examined at high resolution, but frequently a compromise is necessary for throughput. The necessity to analyze data sets accurately, efficiently, and in true 3D has been a major bottleneck in deriving reliable relationships between altered neuronal function and changes in spine morphology. In this chapter, we provide protocols for both imaging and analysis scenarios, starting with imaging dendritic segments (Basic protocol 1), followed by the protocols for complete spine analysis (Basic protocol 2), and full neuron imaging and analysis (Basic protocol 3 and 4). BASIC PROTOCOL 1 TITLE: IMAGING OF FLUORESCENTLY LABELED DENDRITIC SEGMENTS High-resolution confocal microscopy is used to visualize spines on dendritic segments or complete neurons. Prior to performing microscopy, a number of labeling techniques can be employed which may include GFP-viral transfection, iontophoretic injection of a fluorescent dye (cell loading), DiOlistic labeling, or viral expression. We have routinely used iontophoretic injection of Lucifer Yellow in our laboratory; however, the Alexa Fluor dyes have also been used because of larger variety of choices (Boyan and Liu, 2014; Ding et al., 2009; Knafo et al., 2009a; Knafo et al., 2009b; Merino-Serrais et al., 2011; Wallace and Bear, 2004). For cell loading, it is imperative to fill a neuron sufficiently to the distal tips of all apical and basal dendrites or up to 10 minutes depending on the experimental conditions. It is also important not to overfill a neuron to avoid dye leaking out of the cell or dye-coupling with other neurons which can lead to multiple “ghost” neurons being filled. Once neurons are labeled with the fluorescent marker of choice, the tissue can be taken to the microscope. A confocal microscope with objective lenses that have a low numerical aperture for low resolution and a high numerical aperture for high resolution should be used. Imaging parameters for the confocal microscope should be selected for the particular fluorophore used, animal tissue, and confocal microscope. It is important to note that the same imaging parameters must be used for the entire study. Materials Charged or gelatin subbed glass slides (Fisherbrand Microscope slides; Fisher Scientific) Glass coverslips (number 1.5; Sigma-Aldrich, St. Louis,MO) Imaging spacers (SecureSeal, Electron Microscopy Sciences, Hatfield, PA) or nail polish to seal coverslips to the slide. VectaShield mounting medium (Vector Laboratories, Burlingame, CA), or any other preferred anti-fade fluorescence mounting media. Labeled tissue Confocal microscope equipped with low and high numerical aperture objective lenses. Immersion media; choose an immersion media that matches the specifications of the high NA objective that is on your system. Examples of immersion media and their refractive index (RI) are oil (RI = 1.53–1.54), glycerol (RI = 1.47) and water (RI = 1.34). This and the lens choice should also be based on what medium the sample is mounted in (aqueous, etc.) for optimal refractive index matching and aberration-reduced imaging. Setting up the confocal microscope Optimize the confocal laser scanning system for the particular fluorophore by matching the laser line to the excitation peak of the fluorophore (e.g., Argon 458 nm line for Lucifer Yellow). Match the dichroic beam-splitter in front of the laser to the laser used. Select an emission filter to maximize the amount of collected photons. We recommend a long-pass image filter unless multiple fluorophores are used or there is background photon emission due to autofluorescence. Choose appropriate image resolution (e.g., 512 × 512) and bit depth (16 or 8 bit). Most confocal microscopes can image with at least 12-bits. Filled neurons with large, bright soma and fine processes could benefit for more intensity resolution (assuming the sensor itself has the dynamic range). Taking high-resolution images of dendritic segments or sections of cells Sample the dendritic segment in an unbiased manner. Sampling should be random (for example, do not choose the best-looking spines). Choose a segment based on brain region, type of cells, and the type of inputs into the cell you wish to obtain. Measurements can be taken on entire neurons (see below), at specific distances from the soma, or by branch orders. In practice, it is best to choose segments that can be captured in a single field-of-view without exceeding the working distance of the highest resolution objective that will be used. In addition, we recommend avoiding bifurcations, sampling from primary dendrites (due to variation in spine density compared to higher order branches), and sampling the first and last 10% of the dendrite. If the neuron has been previously traced (using MBF Bioscience’s Neurolucida live on a wide-field microscope, or Neurolucida 360 on previously acquired images), use Neurolucida 360 Explorer to display the reconstructed trace with 3D Sholl rings placed at specific distances superimposed on the neuron (Figure 1A) to determine where to image along the dendritic arbor. Alternatively, Zen Zeiss software can be used to capture a low-resolution 3D image montage (e.g., with 10× objective) of the complete neuron and select the desired segments from this image by using the ruler tool and measuring at the desired distance, such as 50 or 100 µm from center of the soma, or desired branch order (Figure 1B). Note that this method selects branches on the basis of 2D alignment, instead of 3D distance. Image the dendritic segment. Regardless of how you choose to sample segment, locate the neuron at low magnification (10×). Once the segment of interest is located, switch to a high numerical aperture, high magnification objective (63× or 100×), locate the segment of interest again, and place it in the center of the field of view. We recommend imaging dendritic segments at high magnification (63× or 100× with an additional digital zoom) to obtain the fine structural detail of each spine. Set gain and offset to acquire the optimal image. It is very important to avoid saturated pixels. If the dendrite is saturated when imaging dendrites and spines, the size and shape may appear altered. This can lead to some spines (especially those above and below the dendrite) not being detected or not being properly characterized. Moreover, saturated pixels will result in incorrect deconvolution outcomes as described in the next section. Essentially, do not strive to get the perfect publication picture with minimized background noise. Instead, get an image with some background noise to facilitate successful post-processing deconvolution. Set the gain so that there are relatively few saturated pixels in the dendritic portion of the image and relatively zero-intensity pixels in the background of the image. A special color palette can be used on most microscopes to visualize saturated and zero intensity pixels in distinctive color (commonly red for saturated pixels and blue for pixels at zero; Figure 2). Select the averaging number, frame direction, and other parameters; they must be maintained throughout the entire experiment. We generally use a line or frame average of 4, and a pixel dwell time of approximately 6 µm per pixel. Reduced scanning speed produce cleaner images, which can lead to more accurate reconstructions. Select a pinhole size at “1 airy unit.” It is important to resist opening the pinhole to obtain a brighter signal with less noise. Resolution along the Z (or optical) axis decreases with increasing pinhole size. Set the X and Y parameters to 0.05 µm and the z-interval to 0.1 µm to allow the most accurate imaging (other laboratories may use different scaling to achieve a cubic voxel (Golden et al., 2013; Hao et al., 2006; Rocher et al., 2010)). We tend to spatially oversample in XYZ given the small size of spine heads and necks. Larger pixel sizes may miss these structures, and thus important information regarding the number of thin spines and length of neck will be underestimated. In addition, deconvolution can also benefit from the added information provided by oversampling. Set the RI correction value. This is an essential step to minimize the immersion RI mismatch of your sample to reduce the spherical aberration. The best-case scenario is immersion media RI = sample RI. Acquire the Z stack of the segment. Make sure to capture enough planes above and below the dendritic segment so that the entire segment for analysis is contained within the image stack. This will ensure complete resolution of the top and bottom of your sample and proper quantification and characterization of spine heads. A common practice is to acquire at least 0.5 µm above and below the sample. SUPPORT PROTOCOL TITLE: POST-PROCESSING DECONVOLUTION Deconvolution is a computational method that attempts to correct the optical distortion inherent in all light microscopic imaging systems (Holmes, 1992). This distortion, or point spread function (PSF), blurs all light passing through a microscope on its way to the detector and is heavily influenced by several system features, including the numerical aperture (NA) of the objective lens, and the refractive index of the objective lens immersion medium, specimen, and specimen mounting medium (Gibson and Lanni, 1992; Nasse and Woehl, 2010). Deconvolution of confocal image stacks results in images with an improved signal-to-noise characteristic and more easily resolved structures. Specifically, for dendritic spines, deconvolution allows better discretization of adjacent spines and well as more accurate measurements and classification (Rodriguez et al., 2008). Proper deconvolution is necessary for images with dendritic spines to reduce the optical Z-smearing of dendrites and spines in the ZY projection. Incomplete deconvolution can distort individual spines and cause problems when quantifying spine densities (as individual spines many not be discernible) and classifying spines into specific types (Rodriguez et al., 2008). Such errors can significantly impact the final data. Materials AutoQuant Image Deconvolution Software (Media Cybernetics; other software programs for deconvolution include those from Scientific Volume Imaging, Zeiss, FIJI, Matlab, etc., refer to their operating instructions for use). Minimal Computer Specification: 2.8 GHz Intel® quad-core 64-bit processor (Core i7 series) or better; RAM: 16 GB memory or higher; OS: Windows® 7 (64-bit); Graphics Card: 2 GB and OpenGL® 4.2 or higher (e.g., NVIDIA GeForce® GTX series, http://www.mediacy.com/index.aspx?page=AutoQuantX3_sys_req). Confocal images (check with the deconvolution software to ensure that native file formats from the microscope imaging software are compatible). Deconvolving images captured with a confocal microscope Open images in AutoQuant and make sure that they are correctly imported. If licensed deconvolution software is not available, one can deconvolve using some freeware software such as ImageJ (http://imagej.net/Parallel_Iterative_Deconvolution) Microscope settings (e.g., objective, N.A., voxel size) should be automatically imported. Ensure that these setting are correct and manually enter where necessary. Enter the emission wavelength. Save the output file as a 16-bit .tif file. Record the image scaling so that it can be entered into Neurolucida 360 for analysis. BASIC PROTOCOL 2 TITLE: DENDRITIC SPINE MODELING AND RECONSTRUCTION WITH NEUROLUCIDA 360 Morphometric analysis of image data containing dendritic branches with spines Neurolucida 360 from MBF Bioscience is a software platform for reconstructing neuronal morphology by tracing, editing, and visualizing image data from light microscopes in 3D. It is based, in part, on the laboratory version of NeuronStudio, originally developed at the Icahn School of Medicine at Mount Sinai by Susan Wearne and her colleagues (Rodriguez et al., 2003; Rodriguez et al., 2008; Rodriguez et al., 2006; Wearne et al., 2005). Key components and algorithms implemented in NeuronStudio for process and dendritic spine reconstruction have been further developed in Neurolucida 360. Built with three algorithms for user-guided and automatic tracing, Neurolucida 360 accurately models neurons visualized with multiple methodologies and imaging techniques. Further, when the algorithms are operated in user-guided mode, the researcher can switch algorithms on-the-fly to adjust for differing conditions along a single dendrite. Automatic dendritic spine detection models the protrusions from dendrites using a mesh to capture the surface and a 5-point segment to model the spine backbone. This results in a more accurate representation of the spine length and shape for better spine classification as well as a mechanism to modify the branch assignment when spines and branches are densely packed. The companion software for analytics, Neurolucida 360 Explorer, calculates a large number of metrics, including volume, length, plane angle, surface area, and includes a notation of whether the spine was classified and how it was classified (manual or automatic). Materials Neurolucida 360 v2.7 or later (MBF Bioscience). Neurolucida 360 Explorer Computer requirements: 2.8 GHz Intel® quad-core; RAM: 16 GB memory or higher (32 GB needed as image size increases); OS: Windows® 7, 8 or 10 (64-bit); Graphics Card: AMD Radeon series GPU with 2GB graphics memory or more (http://www.mbfbioscience.com/neurolucida360). ◦ Performance can vary quite drastically depending on changes in hardware (for example, solid state drives (SSDs) will speed up loading/saving quite dramatically) Image data with known scaling either embedded within the file, or written in your laboratory notebook. File formats accepted by Neurolucida 360 include: tiff, lsm (Zeiss), czi (Zeiss), oib/oif (Olympus), VSI (Olympus), lif (Leica), jpx, ids/ics (Nikon), bigTIFF (btf, ImageJ/FIJI). Protocol steps Refer to the video “Automatic dendritic spine modeling with Neurolucida 360” (DendriticSpines_with_Neurolucida360.mp4) for demonstration of this protocol. Load the image stack into Neurolucida 360 by dragging and dropping the file into the main window. Critically important: confirm the image scaling parameters and immersion media for proper scaling. Adjust the values if inaccurate, or provide values if image scaling is not embedded in the image (e.g., .tiff images). Zoom in to the region of interest using the mouse scroll wheel. Alternatively use the pivot point icon to center the image. Click the Tree icon and select the user-guided tracing mode. In the drop-down menu select the most suitable algorithm for the image data loaded. If desired, select the “pan to window center” checkbox to have the image re-center as you trace. Trace the backbone of the dendritic branch. Left-click to place the initial start point, move the mouse cursor along the branch to see the path (represented by open circles) and estimated thickness provided by the tracing algorithm. Click to have the algorithm trace between points (control+Z will undo the last point). The size of the open circles provides a preview of the thickness of the dendritic branch. Ensure that large spines do not over-influence the thickness by adjusting the path of the proposed trace. To end the trace, right-click (Figure 4). Record the tracing algorithm in the laboratory notebook, indicating which algorithm(s) was (were) used for tracing. Note: if automatic tracing is used, record all relevant parameters in addition to the tracing algorithm in the laboratory notebook. Once traced, inspect the tracing to ensure that the dendritic branch is accurately modeled for thickness in all 3 dimensions. Edit points as necessary (Figure 5). To edit, click the Edit button. Select the point mode to visualize the trace points of the branch. Each point can be moved individually by clicking the point with the left mouse button, or move points as a group (control+click to select multiple points) to adjust the position (control+Z to undo the last change). The dendritic diameter can be modified at each point by adjusting the thickness slider or by typing in a new value. Model dendritic spines. Click the Spine button to select the Spine detection mode. Detect all, and inspect the results. This will help parametrize the following settings (Figure 6): Outer range – this value defines the maximum distance from the dendritic surface that is used to search for spines. Minimum height – this value defines the minimum distance from the dendritic surface for a surface protrusion to be considered a spine. It helps prevent false positives due to surface irregularities. Reduce this value if the spine base is too large. Minimum count – this value is used to exclude individual objects that are too small to be considered a spine. This can be useful when trying to prevent image noise from being detected as spines. Alternative: Use Filter image noise, an image pre-processing filter that can be used with non-deconvolved images. Adjust parameters until the spine modeling algorithm detects the spines as desired. Alternatively, click a spine in the image to detect it. Alternatively, detect spines on a single branch of a complex dendritic arbor by selecting “use click to detect all spines on branch.” It is important to use the same parameters throughout the study; do not choose parameters arbitrarily since this will introduce bias and may affect the data and interpretation. Record the detection parameters in the laboratory notebook (Figure 6). Edit dendritic spines. Use the Edit mode to adjust the detection by splitting and merging spines. Each detected object is displayed in a different color. You may toggle the visibility of all reconstructed objects by using the +/− tool bar button. This allows you to see the underlying image data to determine if adjacent meshes need to be combined to make one spine, or if multiple spines are encompassed by one mesh that needs to be split (Figure 7). Advanced editing: When detecting spines on multiple branches, make sure to confirm that the spines are properly assigned to the correct branch. To view and change the branch assignment, select the Points button. The spine backbone is displayed as a series of 5 points. Move the point of attachment to the new branch by selecting and dragging it to the new location. The spine will be re-detected and assigned to the specified branch (Figure 8). Save the data file. Perform morphometric analyses. Open the data file in the Neurolucida 360 companion program, Neurolucida 360 Explorer. From Neurolucida 360 Explorer’s main menu, select Branched Structure Analyses from the Analysis menu. Click the Spines tab and select reports available with spines, spine details, and dendrites. The Dendrite Spine report includes the total number (and type, if classified), spine density per micrometer of dendritic length. The spine details report includes the many metrics, including total extent (the measure of the shortest-path distance from the dendritic surface to the furthest voxel of the spine) and spine backbone length (the length of the spine from the furthest included voxel along the backbone to the insertion onto the dendritic branch.), head diameter, head:neck ratio. Note: other analyses may be of interest, depending on your scientific question. ALTERNATE PROTOCOL 2 TITLE: SPINE CLASSIFICATION Accurate and effective dendritic spine classification remains a fundamental challenge for the neuroimaging research community. Dendritic spine morphology is thought to be crucial in synaptic plasticity and strength due to its compartmentalization of biochemical and electrical signals. A dendritic spine is a micron-sized protrusion, comprised of a spine head, where the excitatory synapse is located, and a spine neck that connects the spine to the dendritic shaft. Spines come in multiple shapes and sizes, with the more common subclasses being thin, mushroom and stubby (Harris et al., 1992; Jones and Powell, 1969; Nimchinsky et al., 2002; Peters and Kaiserman-Abramof, 1970; Spacek and Hartmann, 1983). Digital representation of spines using light microscopy has traditionally relied on manual counting from a computer screen and is prone to subjective errors. Despite the recent introduction of semi-automated tracing methods (e.g., NeuronStudio, Vaa3D, and Imaris), the problem of detecting and characterizing spine shapes automatically, in 3D, remains unsolved. Automatic classification with Neurolucida 360 greatly reduces human subjectivity and intra-operator variability. Dendritic spines from different brain areas, developmental stages, pathological conditions, or species may require different classification settings. Neurolucida 360 permits simple adjustment of the classification settings, however it is important that the classification settings are chosen on the basis of empirical research and remain unchanged throughout the study. After automatic detection with Neurolucida 360, dendritic spines can be automatically classified with a simple one-button operation into one of the following types: stubby, mushroom, thin, or filopodia (Figure 9). Alternate complex types (e.g., double-headed) can be assigned manually after detection. Each spine type is color-coded for easy visualization, and can be interactively re-classified to a different canonical or complex type. By default, spine classification is assigned according to parameters determined by Rodriguez et al. (2008), which were empirically determined for mouse hippocampal neurons as the consensus from multiple experienced researchers. Default settings for detection and classification should be considered starting points and not mandates. Both detection and parameters and classification settings will be influenced by imaging methodology as well as neuron characteristics (Benavides-Piccione et al., 2002; Harris et al., 1992; Nimchinsky et al., 2002; Rodriguez et al., 2008). Report quantification parameters in your manuscripts so that others can interpret, replicate, and build upon your work. When selecting parameters based on previously published peer-reviewed papers, note that papers that use manual spine measurements are typically measuring spine head sizes in the lateral XY dimension only, which may not be directly comparable to true 3D detection and measurement. Materials Neurolucida 360 v2.7 or later (MBF Bioscience) Neurolucida 360 Explorer (MBF Bioscience) Computer requirements: 2.8 GHz Intel® quad-core; RAM: 16 GB memory or higher (32 GB or more needed as image size increases); OS: Windows® 7, 8 or 10 (64-bit); Graphics Card: AMD Radeon series GPU with 2GB graphics memory or more (http://www.mbfbioscience.com/neurolucida360). ◦ Performance can vary drastically depending on changes in hardware (for example, solid state drives (SSDs) will dramatically speed up loading/saving image files). Data file from Neurolucida 360 with modeled dendritic spines. Protocol steps Step annotations If desired, classify the dendritic spines. Click the Classify button to instantaneously re-color the detected spines according to spine class. By default, the spines are colored red (thin), blue (mushroom), green (stubby), and yellow (filopodia) according to the metrics determined by Rodriquez et al., (2008). Alternatively: different metrics can be entered to define the spine classes according to values specific to the species or condition under study. Click Settings to enter the values appropriate for your experimental paradigm. Record the classification parameters in your laboratory notebook (Figure 10). Save the data file. Open the data file in Neurolucida 360 Explorer and select the spine details report from Branched Structure Analysis. Spine type and assigned type will be included in the spine details report. BASIC PROTOCOL 3 IMAGING COMPLETE NEURONS In some instances, complete neuronal reconstructions from high-resolution images (e.g., 1024×1024) are desired in order to analyze both neuronal complexity and spine density on entire neurons. Depending on the microscope and software capabilities, the entire neuron can be captured using an automated tiling function, with an appropriate amount of overlap between tiles (e.g., 10%), along with the Z-step distance. If there is no such tiling option, individual stacks need to be separately acquired at high resolution and magnification, and then stitched together using specific software programs such as Neurolucida 360 or Volume Integration and Alignment System (VIAS; http://research.mssm.edu/cnic/tools-vias.html). As in automated tiling performed by the acquisition software, there must be an appropriate amount of overlap (e.g., 10%) to accurately stitch the segments together as a post-acquisition operation. Find the neuron of interest on the microscope using the lowest magnification objective (e.g., 10×). An initial rapid image acquisition of the overall Z-depth and area is needed to set up for the high-resolution high magnification acquisition that follows this step. In order to determine both the area and maximum Z-depth using rapid acquisition methods, first set the system up as follows: change image resolution to a lowered value (e.g., 256×256), change pixel dwell time to its maximum speed (fastest pixel dwell time), and open the pinhole aperture to its maximum size (largest optical slice thickness). Using these conditions allows for rapid determination of the neurons area and Z-depth. Start image acquisition and adjust for gain/offset signal/background). Since the Airy Unit aperture setting is opened to its maximum, the image adjustments are used to optimize observation of the entire neurons. If the neurons dendritic branches extend beyond the field of view, change the digital magnification from Z=1 to Z=0.9. Continue this process until you have the entire neuron in the field of view (e.g., Z=0.8, etc). Start the Z-step function by clicking on the Z-stack window. Using the Z-step function, determine the upper and lower limits to include both the material, along with an additional Z buffer distance both above and below the tissue. It is important to capture all of the neurons terminal dendritic branches. Record both the Z-depth and the total area of the neuron. If needed, a low-resolution 3D capture can be taken at this time. Otherwise, set up for high-resolution full neuron capture. Do not move the XY position. Change to an oil-immersion objective (e.g., 63× oil/N.A. 1.4 or 100× oil/N.A. 1.4). Change the pinhole value to 1 Airy Unit and optimize for offset/gain. Set and optimize the parameters for high-resolution image capture, (e.g., 1024×1024 resolution, increased pixel dwell time, frame average of 2). Open the image information tab and record the x/y pixel resolution (e.g., 0.5×0.5×0.5 µm). Disengage the Z-step function, and open tiling by clicking open the tiling function window. Open the function and setup (if available) for on-line stitching. Determine the number of overlapping (8–10%) tiles needed to acquire the entire neuron using the area value from the low-resolution image. Engage the Z-step function. Using the Z-depth values from the low-resolution image, set the interval for voxel dimensions (e.g., 0.5×0.5×0.5 µm). Turn “online stitching” on. Set image overlap at 8–12% (Figure 11). Ensure both tiling and Z-stack functions are engaged, and that all image parameters are set. Image the neuron. Alternatively, image Z-stacks can be saved independently and stitched using off-line programs (Figure 12). BASIC PROTOCOL 4 (optional) TITLE: NEURON RECONSTRUCTION USING NEUROLUCIDA 360 Instructions for reconstructing the entire neuronal structure follow the basic protocol for reconstructing the single dendritic segment described previously, with a few additional steps to model the soma and confirm the origin of the trees at the base of the soma. Materials Neurolucida 360 v2.7 or later (MBF Bioscience) Computer requirements: 2.8 GHz Intel® quad-core; RAM: 16 GB memory or higher (32 GB or more needed as image size increases); OS: Windows® 7, 8 or 10 (64-bit); Graphics Card: AMD Radeon series GPU with 2GB graphics memory or more (http://www.mbfbioscience.com/neurolucida360). ◦ Performance can vary drastically depending on changes in hardware (for example, solid state drives (SSDs) will dramatically speed up loading/saving image files) Image data with known scaling either embedded within the file, or written in your laboratory notebook. File formats accepted by Neurolucida 360 include: tiff, lsm (Zeiss), czi (Zeiss), oib/oif (Olympus), VSI (Olympus), lif (Leica), jpx, ids/ics (Nikon), bigTIFF (btf, ImageJ/FIJI). The width, height, number of focal planes, and depth of each pixel all contribute to the size of image files. Compression algorithms reduce the impact of image dimensions on the storage system. Though compression has no direct impact on the memory required to navigate an image, a good rule of thumb we have found is to have two times the on-disk file size in available memory when using Neurolucida 360. Protocol steps Load the image stack into Neurolucida 360 by dragging and dropping the stack into the main window. Critically important: confirm the image scaling parameters and immersion media for proper scaling. Correct the values if inaccurate, or provide values if image scaling is not embedded in the image (e.g., tiff images). Zoom in to the region of interest using the mouse scroll wheel. Alternatively use the pivot point icon to center the image. To reconstruct the soma. Click the Soma button to enter the Soma mode. Adjust the circular cursor using Control+scroll wheel to create a discrete search region for the algorithm. Click the soma in the image to model it. If the soma is rendered as a cube, reduce the sensitivity, clear the soma, and re-detect. Trace the backbone of the dendritic branch. Click the Tree button. Select the user-guided tracing mode. In the drop-down menu, select the most suitable algorithm for the image data loaded. Select the “pan to window center” checkbox to have the image re-center as you trace. Click to place the initial start point, move the mouse cursor along the branch to see the path and estimated thickness provided by the tracing algorithm. Click to have the algorithm trace between points (control+Z will undo the last point). Ensure that large spines do not over-influence the thickness by adjusting the path of the proposed trace. To end the trace, right-click (Figure 4). Continue until all branches are traced. When desired, switch algorithms while tracing to match the algorithm to the image data presented. This will reduce the amount of editing needed. Directional kernels works well for punctuate label at distal branches, Rayburst Crawl and Voxel Scooping are helpful in areas of high complexity. Record the tracing algorithm in the laboratory notebook. Once traced, inspect the tracing to ensure that the dendritic branch is accurately modeled for thickness in all 3 dimensions. Edit points as necessary (Figure 5). Click the Edit button and select the point mode to visualize the trace points of the branch. You can move points individually by clicking a point with the left mouse button, or move points as a group (control+click to select multiple points) to adjust the position (control+Z to undo the last change). You can modify thickness at each point by using the thickness slider or typing in a new value. Confirm the origin of the dendritic trees. While it is important for branch analyses to have each tree begin at the soma, it is not required to trace in any particular direction. Return to the Edit panel and select the point mode to visualize the trace points of the branch. Draw a marquee around the soma small enough so that it does not contain full dendritic trees, but large enough to contain all the points closest to the soma. Once selected, the option to set all endings to “origins” becomes available only if some trees are initiated at a different location. Select the button to reset all trees to have their origin closest to the soma (Figure 13). Save the data file. Proceed to Basic Protocol 2 to model dendritic spines. COMMENTARY Background Information Dendritic spine morphology has become a central focus in research in the fields of learning and memory, aging, and neurodegenerative diseases (for review see (Dickstein et al., 2007; Dickstein et al., 2013; Hara et al., 2012) as well as other neuropsychiatric disorders such as autism (Durand et al., 2012; Hung et al., 2008; Hutsler and Zhang, 2010; Phillips and Pozzo-Miller, 2015), schizophrenia (Hayashi-Takagi et al., 2011; Ramos-Miguel et al., 2015), and addiction (Maze et al., 2010; Selvas et al., 2015) as they are an integral component of excitatory synapses. Excitatory synapses comprise the majority of connections in the central nervous system and play a vital role in learning, memory, and cognition. Abnormal development or regulation of these synapses has been implicated in many neurodevelopmental, psychiatric, and neurodegenerative disorders. Most excitatory synapses occur at specialized postsynaptic compartments known as dendritic spines, which are tiny protrusions from the dendrites of neurons (Carlisle and Kennedy, 2005; Ethell and Pasquale, 2005; Harris and Stevens, 1989; Nimchinsky et al., 2002). Based on their morphology, spines can be divided into “canonical” types (stubby, mushroom, thin) and complex types (cup-shaped, multi-headed or branched, and filopodia) (Ethell and Pasquale, 2005; Harris et al., 1992; Kasai et al., 2010; Nimchinsky et al., 2002). The variable structure of spines determines the strength, stability, and function of the synaptic connections that facilitate the neuronal networks in the brain. Understanding the dynamics of spine morphology will help researchers to address how the brain is able to process a continuous flow of sensory information and simultaneously store and consolidate memories, sometimes for a lifetime. However, precise quantification of spines parameters has been restricted to time and labor-intensive electron microscopy, which has limited throughput and is impractical for large-scale studies. Accordingly, there is growing interest in automating quantitative analysis of dendritic spine morphology at the light microscopic level. Traditionally, performing manual spine analysis has been, and in many situations still is, the approach of choice. This usually involves tracing dendrites and marking spines from confocal stacks as they appear in the XY plane with software packages such as Neurolucida (Brennan et al., 2009; Hao et al., 2006; Knafo et al., 2009a; Knafo et al., 2009b), Imaris (Vecellio et al., 2000), Metamorph (Wallace and Bear, 2004) and Arivis. Such counting is then followed by manual measurements of spine heads and necks using programs such as Photoshop (Hao et al., 2006). These methods, in addition to being very time-consuming, introduce much error based on observer bias and high inter-observer variability (Donohue and Ascoli, 2011). Moreover, underestimating the number of spines is a concern in these conditions since spines, which project on the Z plane of the dendrite, are often overlooked as they are obscured by the brighter dendritic shafts. Automatic algorithmic analysis of spine morphology at the light microscopic level provides for higher throughput by substantially increasing analysis speed, accuracy, and reproducibility compared to existing manual methods. In addition, it provides for observer independence and for quantitative analyses that are virtually impossible without automation (such as spine volume, surface, head diameter, neck diameter, etc.). NeuronStudio was our first semi-automated quantitative software based on the Rayburst Sampling algorithm. This software was used in multiple studies from our group and others on various animal models, such as mouse, rat and monkey, and in many brain areas including the prefrontal cortex, hippocampus, and nucleus accumbens (Bloss et al., 2011; Golden et al., 2013; Price et al., 2014; Radley et al., 2008; Rocher et al., 2010; Shansky et al., 2009; Steele et al., 2014). While programs such as Imaris (Bitplane), Amira, and NeuronStudio may make quantification easier, there remain issues regarding accuracy, in particular in defining the spine head volume and surface area (Donohue and Ascoli, 2011; Dumitriu et al., 2011). Here, we introduce a new computational approach for detection and shape analysis of dendritic spines that incorporate the algorithms of our previous software (NeuronStudio; http://research.mssm.edu/cnic/tools-ns.html) as well as numerous improvements to make the software more broadly usable. Neurolucida 360 can read multichannel, high-bit depth images, with file sizes that exceed 50 GB. The algorithms have been tested with various labeling techniques on neurons and spines from different mammalian species (mouse, rat, monkey) and brain regions. It is important to note that successful data acquisition always relies on the quality of the materials used, and of the labeling of the cells analyzed, independent of the software performance. It is essential that imaging be performed in optimal conditions, such as these detailed in this unit. We believe that the new quantitative software package, Neurolucida 360, provides the neuroscience research community with the ability to perform higher throughput automated 3D quantitative light microscopy spine analysis under standardized conditions to accelerate the characterization of dendritic spines with greater objectivity and reliability. Critical Parameters Insufficient labeling of cell – the saying “garbage in, garbage out” holds true here. The best imaging systems and most sophisticated algorithms cannot correct for inconsistent or improper methodology at the bench. Image scaling – For the most accurate estimates of quantitative measures (e.g., volume, extent, etc.) the correct image resolution and axial step size must be provided. Troubleshooting Problem Possible Cause Solution Poor images (low signal-to-noise) Poor/incomplete labeling of cells Only use tissue with strong fluorescent signal and complete arborization Z-smear Identified cell beyond working distance of objective Image only cells that lie within 80 µm of the tissue surface Low Z resolution Microscope type (e.g., confocal, spinning disk, 2 photon), pinhole size too large, objective is not sufficient to resolve the needed features Adjust imaging parameters to obtain the optimal image Dendrites appear distorted Air bubble in mounting media Remount the tissue Dendrite appears to be moving across the screen during confocal imaging Insufficient seal of coverslip If you used spacers when mounting the tissue, make sure there are enough spacers used for the thickness of the tissue. Each spacer is 120 µm If you used nail polish, either there is not enough to create a proper seal or it has not hardened enough. It is always best to wait a couple of days between mounting the tissue and imaging Make sure the objective does not compress the coverslip. This happens when the working distance is mismatched with the tissue Dendrite appears to be compressed while imaging The segment is too deep in the tissue (>80 µm) Image segments more parallel to the tissue section Dendrites appear to be flat once reconstructed Objective unable to fully adjust to the Z-step depth Image segments more parallel to the tissue section Software does not target the image Loading tiff images into Neurolucida 360 without correct scaling Correct size of voxels Dendrites appear beaded Poor perfusion/fixation leads to poorly loaded neurons Cells are not usable for analysis Cell loses fluores- cence intensity during imaging Poor perfusion/fixation leads to poorly loaded neurons Cells are not usable for analysis Overlapping den- dritic segments from adjacent cells/same cell in same imaging plane Certain techniques (e.g., viral expression using GFP, DiOlistic, cells loaded too closely) In certain cases this is unavoidable (e.g., GFP). If dendritic segments are separated in the Z-plane they can still be used. Anticipated Results After reconstruction in 3D with Neurolucida 360, a number of metrics are calculated. Using the companion software, Neurolucida 360 Explorer, spine analyses include number, total extent, plane angle, volume, surface area, contact area, XYZ coordinates, head diameter and length, neck length, head diameter to neck diameter ratio, as well as notations of attachment, type, and method for classification. Several analyses are also available for dendrites, including but not limited to: Sholl analysis, branch order, dendritic length, number of branch points, spine density, spine density by type, convex hull, polar histogram. These analyses can be exported directly to Microsoft Excel. The data file can also be exported in a format suitable for third-party 3D rendering software (e.g., Blender), or further utilized with programming software (such as MatLab) to perform additional computations. Time Considerations Image acquisition Acquisitions of dendritic segments or full neurons are the most time-consuming aspects of dendritic spine analysis. The time it takes to image dendritic segments can vary depending on the depth of the segment, the Z-step, scan time, and averaging. It is important to avoid shortcuts when acquiring confocal images to preserve the fine structural details of dendritic spines. Moreover, it is important to remain unbiased when choosing dendritic segments to image (e.g., avoid only imaging shallow dendritic segments to save time). The time to image a complete neuron will vary depending on the type of data to be collected from the cell. For dendritic spine density only, neurons can be imaged at a lower magnification. For more detailed information about spines (e.g., spine type, and estimates of surface area, volume, and neck length), higher magnifications are needed but this will increase the imaging duration as more image stacks need to be obtained and stitched together. A complete neuron at 100× can take up to 10 hours to image, depending on neuron size, complexity, and microscope configuration. Dendritic spine analysis Automatic dendritic spine analysis with Neurolucida 360 requires less than 1 minute for branch reconstruction. Parameterization of spine detection can take 3–10 minutes, with automatic detection occurring in less than 1 minute. Spine classification is instantaneous. Manual editing of branch and dendritic spines will vary depending on the complexity of the image data. For short dendritic segments similar to those described here, editing typically requires less than 5 minutes. Complete neuronal reconstruction The time required for automatic montaging of a complete neuron from discrete, overlapping image stacks will depend on the number of stacks. Typically, image montaging requires approximately 5 minutes with a powerful computer. Once the image has been stitched, the reconstruction of the entire neuronal structure will also depend on the complexity. Typically, user-guided reconstruction can take between 10 minutes and 3 hours. Supplementary Material supplemental video The development of Neurolucida 360 was funded by a NIMH Lab-to-Marketplace grant (R44 MH093011) to Paul J. Angstman and Patrick R. Hof. Videos were created by Pasang Sherpa. We appreciate the thoughtful edits by our colleague, Dr. Sandrine Dincki. Figure 1 Neuronal map for determining the dendritic segments to image. (A) 3D reconstruction of a CA1 pyramidal neuron with superimposed Sholl rings created in Neurolucida Explorer (MBF Bioscience). (B) Low magnification confocal image of a CA1 pyramidal neuron with superimposed concentric circles at measured distances from the center of the cell body using the Zeiss Zen 780 software. Note: the neurons in this figure are not the same. Figure 2 Example of optimally set gain and offset of dendritic segments Images were acquired on a Zeiss 780 laser scanning confocal microscope equipped with the Zen software. (A) Image demonstrates an open Zen operating window for high-resolution dendritic Z-stack imaging. Note the longitudinal dendritic segment with the presence of spines. Within this image, the gain (red) and offset (blue) are properly set to acquire the optimal finalized Z-stack image. The gain is set so there are virtually no saturated pixels (red) present prior to imaging. The offset (blue) appears somewhat mottled within a black background. This adjusted level of background noise is necessary to facilitate successful post-processing deconvolution. All other parameters (frame resolution, averaging number, frame direction, etc) are set prior to imaging. The standard ZEN operating window is open, along with some of the more advanced functions needed for high-resolution 3D imaging (e.g., Z-stacks, focus, etc). (B) A vertical dendritic representation with incorrect offset (blue) settings. While there is no over saturation of the gain (red), the offset (blue) is too high (not enough blue demonstrated). The effect on image reconstruction would be altered. Figure 3 Deconvolution of dendritic segments XY and ZY maximal projections of a typical image stack before (A) and after (B) deconvolution with AutoDeblur. Compared to the raw data (A), the deconvolved data exhibit good relative intensity equalization of spines and dendrites, and significantly reduced Z-axis “stretching” from optical smear, in the ZY projection (B). Adapted from (Rodriguez et al., 2008).XY and ZY maximal projections of a typical image stack before (A) and after (B) deconvolution with AutoDeblur. Compared to the raw data (A), the deconvolved data exhibit good relative intensity equalization of spines and dendrites, and significantly reduced Z-axis “stretching” from optical smear, in the ZY projection (B). (Adapted from (Rodriguez et al., 2008). Figure 4 Tracing the backbone of the dendritic segment Dendritic segment seen here is from a mouse pyramidal CA1 neuron filled with Lucifer Yellow. A cursor (red +) is moved along the dendritic segment to see the path and estimated thickness provided by the tracing algorithm. The open yellow circles provide a preview of dendritic branch thickness. It is important to confirm that large spines do not over-influence the thickness of the dendrite. Scale bar = 2 µm. Figure 5 Inspection of dendritic segment thickness (A) Neurolucida 360 showing inspection and correction of points (green) from the dendritic branch (yellow) that were drawn off center by the large spine (arrow). (B) To edit, view the branch in point mode, select the point, and move it to the correct, centered location on the dendritic branch. Scale bar = 0.5 µm. Figure 6 Merging dendritic spines Individual spine detections can be merged to create a single model for a dendritic spine. Inspecting the underlying image data (A) can show that a single spine was inaccurately modeled as two discrete objects (B). To correct, select each spine object in edit mode and select merge. The software remodels the spine to include all voxels previously split between the two objects as a single spine (C). Scale bar = 1 µm. Figure 7 Dendritic spine detection parameters The software has four detection parameters to set the conditions for modeling dendritic spines. The parameters, which will vary based on the imaging settings and experimental paradigm being tested, should be chosen based on empirical data, and remain constant during the study. Do not choose parameters arbitrarily since this will introduce bias and may affect the data and interpretation. Figure 8 Representation of dendritic spines in Neurolucida 360 The dendritic spine is modeled with a mesh to represent the surface and volume of the spine (A). The spine backbone (B) is represented with five points. The most distal point in the backbone indicates the furthest voxel from the dendritic surface, the centroid of the spine head (green) is the second point, and the last point represents where the spine connects with the dendrite. The shape of the dendritic spine is more accurately modeled, leading to better metrics and more complex spine classes. Spines can be re-assigned to nearby branches by dragging the last point from the original dendrite location to the desired location on the alternate branch. Figure 9 Spine classification using default parameters Dendritic spines are first modeled on the basis of a number of detection parameters, including distance from dendritic surface and apparent size. After detection, spines are colored to differentiate each modeled spine in close proximity (A). If desired, the detected spines can be classified using classification parameters as established by Rodriguez et al., 2008 (B) or through custom specifications. It is important that the same detection parameters and classification settings (if chosen) are used for all images in the experimental study. Scale bar = 2 µm. Figure 10 Setting spine classification parameters Different metrics can be entered to define the spine classes according to values specific to the species or condition under study. Parameters should not be changed during the study, and should be chosen on the basis of empirical evidence. Figure 11 Complete neuron image montage of a hippocampal pyramidal neuron acquired after whole-cell recording A single image was created using Neurolucida 360 to montage multiple 3D images of a mouse hippocampal pyramidal neuron labeled with biocytin. Multiple 2-channel z-series images were collected with a Zeiss LSM 710 confocal microscope, mounted on an AxioImager Z2, with a 20× Plan-apochromat objective, and a 25 mW multi-wavelength (458/488/514) argon laser and a 20 mW 561 nm diode DPSS laser (Alexa-549 displayed here). Dr. Piskorowski provided the image data from an experiment performed by Vincent Robert and Ludivine Therreau in accordance with European guidelines for the care and use of laboratory animals at the Université Paris Descartes. Note: this neuron was not imaged at a resolution high enough for concurrent spine analysis. Scale bar = 100 µm. Figure 12 Complete neuron reconstruction created with Neurolucida 360 Using Neurolucida 360, the pyramidal cell shown in Figure 11 was reconstructed using user-guided tracing and soma modeling. Scale bar = 100 µm. Figure 13 Setting the origin of branches as the closest point to the soma While it is important for branch analyses to have each tree begin at the soma, it is not required to trace in any particular direction. Confirm the root of each tree in edit mode, by drawing a marquee around the soma small enough so that it does not contain full dendritic trees, but large enough to contain all the points closest to the soma. Once selected, the option to set all endings to “origins” becomes available only if some trees are initiated at a different location. Select the button to reset all trees to have their origin closest to the soma. Significance Statement Understanding the role of dendritic spines is an important area of neuroscience research. We introduce a methodology for performing morphometric dendritic spine analysis from 3D confocal images of dendritic segments. The protocol describes the process of selecting segments for analysis, confocal image acquisition guidelines, deconvolution, and analysis with Neurolucida 360. Neurolucida 360 improves the reliability and accuracy of spine morphometrics while providing an objective means to rapidly analyze spines in 3D. Quantitative neuron and dendritic spine analyses could accelerate the understanding of the relationship between brain structure and function under physiological and pathological conditions and thereby improve the development of novel treatment strategies for complex CNS diseases. KEY REFERENCE Drs. Dickstein and Tappan hosted a webinar on this topic based on a previous version of Neurolucida 360. The recorded version of the webinar is available for viewing at this link: https://youtu.be/HczuQjeNcR4 INTERNET RESOURCES Software described in the protocol is available from the following companies. Free versions or free trials are available from each vendor. Zen Black image acquisition software (Zeiss) http://www.zeiss.com/microscopy/en_us/products/microscope-software/zen-lite.html AutoQuant image deconvolution software (Media Cybernetics) http://www.mediacy.com/index.aspx?page=AutoQuant Neurolucida 360 automated neuron reconstruction software (MBF Bioscience) http://www.mbfbioscience.com/neurolucida360 VIDEOS Automatic dendritic spine modeling with Neurolucida 360 (DendriticSpines_with_Neurolucida360.mp4) This video demonstrates all steps in Neurolucida 360 for modeling dendritic spines on dendritic segments from loading the image data to classification. LITERATURE CITED Benavides-Piccione R Ballesteros-Yanez I DeFelipe J Yuste R Cortical area and species differences in dendritic spine morphology J Neurocytol 2002 31 337 346 12815251 Bloss EB Janssen WG Ohm DT Yuk FJ Wadsworth S Saardi KM McEwen BS Morrison JH Evidence for reduced experience-dependent dendritic spine plasticity in the aging prefrontal cortex J Neurosci 2011 31 7831 7839 21613496 Boyan G Liu Y Dye coupling and immunostaining of astrocyte-like glia following intracellular injection of fluorochromes in brain slices of the grasshopper, Schistocerca gregaria Methods Mol Biol 2014 1082 99 113 24048929 Brennan AR Yuan P Dickstein DL Rocher AB Hof PR Manji H Arnsten AF Protein kinase C activity is associated with prefrontal cortical decline in aging Neurobiol Aging 2009 30 782 792 17919783 Carlisle HJ Kennedy MB Spine architecture and synaptic plasticity Trends Neurosci 2005 28 182 187 15808352 Dickstein DL Kabaso D Rocher AB Luebke JI Wearne SL Hof PR Changes in the structural complexity of the aged brain Aging Cell 2007 6 275 284 17465981 Dickstein DL Weaver CM Luebke JI Hof PR Dendritic spine changes associated with normal aging Neuroscience 2013 251 21 32 23069756 Ding JB Takasaki KT Sabatini BL Supraresolution imaging in brain slices using stimulated-emission depletion two-photon laser scanning microscopy Neuron 2009 63 429 437 19709626 Donohue DE Ascoli GA Automated reconstruction of neuronal morphology: an overview Brain Res Rev 2011 67 94 102 21118703 Dumitriu D Rodriguez A Morrison JH High-throughput, detailed, cell-specific neuroanatomy of dendritic spines using microinjection and confocal microscopy Nat Protoc 2011 6 1391 1411 21886104 Durand CM Perroy J Loll F Perrais D Fagni L Bourgeron T Montcouquiol M Sans N SHANK3 mutations identified in autism lead to modification of dendritic spine morphology via an actin-dependent mechanism Mol Psychiatry 2012 17 71 84 21606927 Ethell IM Pasquale EB Molecular mechanisms of dendritic spine development and remodeling Prog Neurobiol 2005 75 161 205 15882774 Gibson SF Lanni F Experimental test of an analytical model of aberration in an oil-immersion objective lens used in three-dimensional light microscopy J Opt Soc Am A 1992 9 154 166 1738047 Golden SA Christoffel DJ Heshmati M Hodes GE Magida J Davis K Cahill ME Dias C Ribeiro E Ables JL Kennedy PJ Robison AJ Gonzalez-Maeso J Neve RL Turecki G Ghose S Tamminga CA Russo SJ Epigenetic regulation of RAC1 induces synaptic remodeling in stress disorders and depression Nat Med 2013 19 337 344 23416703 Hao J Rapp PR Leffler AE Leffler SR Janssen WG Lou W McKay H Roberts JA Wearne SL Hof PR Morrison JH Estrogen alters spine number and morphology in prefrontal cortex of aged female rhesus monkeys J Neurosci 2006 26 2571 2578 16510735 Hara Y Rapp PR Morrison JH Neuronal and morphological bases of cognitive decline in aged rhesus monkeys Age (Dordr) 2012 34 1051 1073 21710198 Harris KM Jensen FE Tsao B Three-dimensional structure of dendritic spines and synapses in rat hippocampus (CA1) at postnatal day 15 and adult ages: implications for the maturation of synaptic physiology and long-term potentiation J Neurosci 1992 12 2685 2705 1613552 Harris KM Stevens JK Dendritic spines of CA 1 pyramidal cells in the rat hippocampus: serial electron microscopy with reference to their biophysical characteristics J Neurosci 1989 9 2982 2997 2769375 Hayashi-Takagi A Barker PB Sawa A Readdressing synaptic pruning theory for schizophrenia: Combination of brain imaging and cell biology Commun Integr Biol 2011 4 211 212 21655443 Holmes TJ Blind deconvolution of quantum-limited incoherent imagery: maximum-likelihood approach J Opt Soc Am A 1992 9 1052 1061 1634965 Hung AY Futai K Sala C Valtschanoff JG Ryu J Woodworth MA Kidd FL Sung CC Miyakawa T Bear MF Weinberg RJ Sheng M Smaller dendritic spines, weaker synaptic transmission, but enhanced spatial learning in mice lacking Shank1 J Neurosci 2008 28 1697 1708 18272690 Hutsler JJ Zhang H Increased dendritic spine densities on cortical projection neurons in autism spectrum disorders Brain Res 2010 1309 83 94 19896929 Jones EG Powell TP Morphological variations in the dendritic spines of the neocortex J Cell Sci 1969 5 509 529 5362339 Kasai H Fukuda M Watanabe S Hayashi-Takagi A Noguchi J Structural dynamics of dendritic spines in memory and cognition Trends Neurosci 2010 33 121 129 20138375 Knafo S Alonso-Nanclares L Gonzalez-Soriano J Merino-Serrais P Fernaud-Espinosa I Ferrer I DeFelipe J Widespread changes in dendritic spines in a model of Alzheimer's disease Cereb Cortex 2009a 19 586 592 18632740 Knafo S Venero C Merino-Serrais P Fernaud-Espinosa I Gonzalez-Soriano J Ferrer I Santpere G DeFelipe J Morphological alterations to neurons of the amygdala and impaired fear conditioning in a transgenic mouse model of Alzheimer's disease J Pathol 2009b 219 41 51 19449368 Maze I Covington HE 3rd Dietz DM LaPlant Q Renthal W Russo SJ Mechanic M Mouzon E Neve RL Haggarty SJ Ren Y Sampath SC Hurd YL Greengard P Tarakhovsky A Schaefer A Nestler EJ Essential role of the histone methyltransferase G9a in cocaine-induced plasticity Science 2010 327 213 216 20056891 Merino-Serrais P Knafo S Alonso-Nanclares L Fernaud-Espinosa I DeFelipe J Layer-specific alterations to CA1 dendritic spines in a mouse model of Alzheimer's disease Hippocampus 2011 21 1037 1044 20848609 Nasse MJ Woehl JC Realistic modeling of the illumination point spread function in confocal scanning optical microscopy J Opt Soc Am A Opt Image Sci Vis 2010 27 295 302 20126241 Nimchinsky EA Sabatini BL Svoboda K Structure and function of dendritic spines Annu Rev Physiol 2002 64 313 353 11826272 Peters A Kaiserman-Abramof IR The small pyramidal neuron of the rat cerebral cortex. The perikaryon, dendrites and spines Am J Anat 1970 127 321 355 4985058 Phillips M Pozzo-Miller L Dendritic spine dysgenesis in autism related disorders Neurosci Lett 2015 601 30 40 25578949 Price KA Varghese M Sowa A Yuk F Brautigam H Ehrlich ME Dickstein DL Altered synaptic structure in the hippocampus in a mouse model of Alzheimer's disease with soluble amyloid-beta oligomers and no plaque pathology Mol Neurodegener 2014 9 41 25312309 Radley JJ Rocher AB Rodriguez A Ehlenberger DB Dammann M McEwen BS Morrison JH Wearne SL Hof PR Repeated stress alters dendritic spine morphology in the rat medial prefrontal cortex J Comp Neurol 2008 507 1141 1150 18157834 Ramos-Miguel A Barr AM Honer WG Spines, synapses, and schizophrenia Biol Psychiatry 2015 78 741 743 26542741 Rocher AB Crimins JL Amatrudo JM Kinson MS Todd-Brown MA Lewis J Luebke JI Structural and functional changes in tau mutant mice neurons are not linked to the presence of NFTs Exp Neurol 2010 223 385 393 19665462 Rodriguez A Ehlenberger D Kelliher K Einstein M Henderson SC Morrison JH Hof PR Wearne SL Automated reconstruction of three-dimensional neuronal morphology from laser scanning microscopy images Methods 2003 30 94 105 12695107 Rodriguez A Ehlenberger DB Dickstein DL Hof PR Wearne SL Automated three-dimensional detection and shape classification of dendritic spines from fluorescence microscopy images PLoS One 2008 3 e1997 18431482 Rodriguez A Ehlenberger DB Hof PR Wearne SL Rayburst sampling, an algorithm for automated three-dimensional shape analysis from laser scanning microscopy images Nat Protoc 2006 1 2152 2161 17487207 Selvas A Coria SM Kastanauskaite A Fernaud-Espinosa I DeFelipe J Ambrosio E Miguens M Rat-strain dependent changes of dendritic and spine morphology in the hippocampus after cocaine self-administration Addict Biol 2015 Shansky RM Hamo C Hof PR McEwen BS Morrison JH Stress-induced dendritic remodeling in the prefrontal cortex is circuit specific Cereb Cortex 2009 19 2479 2484 19193712 Spacek J Hartmann M Three-dimensional analysis of dendritic spines. I. Quantitative observations related to dendritic spine and synaptic morphology in cerebral and cerebellar cortices Anat Embryol (Berl) 1983 167 289 310 6614508 Steele JW Brautigam H Short JA Sowa A Shi M Yadav A Weaver CM Westaway D Fraser PE St George-Hyslop PH Gandy S Hof PR Dickstein DL Early fear memory defects are associated with altered synaptic plasticity and molecular architecture in the TgCRND8 Alzheimer's disease mouse model J Comp Neurol 2014 522 2319 2335 24415002 Vecellio M Schwaller B Meyer M Hunziker W Celio MR Alterations in Purkinje cell spines of calbindin D-28 k and parvalbumin knock-out mice Eur J Neurosci 2000 12 945 954 10762324 Wallace W Bear MF A morphological correlate of synaptic scaling in visual cortex J Neurosci 2004 24 6928 6938 15295028 Wearne SL Rodriguez A Ehlenberger DB Rocher AB Henderson SC Hof PR New techniques for imaging, digitization and analysis of three-dimensional neural morphology on multiple scales Neuroscience 2005 136 661 680 16344143

PMC005xxxxxx/PMC5113819.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101284307 33047 Nat Protoc Nat Protoc Nature protocols 1754-2189 1750-2799 27560173 5113819 10.1038/nprot.2016.098 NIHMS802696 Article Generating kidney organoids from human pluripotent stem cells Takasato Minoru 12 Er Pei X 12 Chiu Han S 2 Little Melissa H 123 1 Murdoch Childrens Research Institute, Flemington Rd, Parkville, Melbourne, Victoria 3052, Australia. 2 Institute for Molecular Bioscience, The University of Queensland, St Lucia, Queensland 4072, Australia. 3 Department of Paediatrics, The University of Melbourne, Parkville, Victoria 3010, Australia. Correspondence should be addressed to MT (minoru.takasato@mcri.edu.au) or MHL (melissa.little@mcri.edu.au). 15 7 2016 18 8 2016 9 2016 01 3 2017 11 9 16811692 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The human kidney develops from four progenitor populations; nephron progenitors, ureteric epithelial progenitors, renal interstitial progenitors and endothelial progenitors; resulting in the formation of maximally 2 million nephrons. Until recently, methods differentiating human pluripotent stem cells (hPSCs) into either nephron progenitor or ureteric epithelial progenitor had been reported, consequently forming only nephrons or collecting ducts, respectively. Here, we detail a protocol that simultaneously induces all four progenitors to generate kidney organoids within which segmented nephrons are connected to collecting ducts and surrounded by renal interstitial cells and an endothelial network. As evidence of functional maturity, proximal tubules within organoids display megalin-mediated and cubilin-mediated endocytosis, and respond to a nephrotoxicant to undergo apoptosis. This protocol consists of 7 days of monolayer culture for intermediate mesoderm induction followed by 18 days of three-dimensional culture to facilitate self-organising renogenic events leading to organoid formation. Personnel experienced in culturing hPSCs are required to conduct this protocol. Kidney organoid Human pluripotent stem cell Directed differentiation INTRODUCTION Directed differentiation of human pluripotent stem cells (hPSCs), including human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs), is one of the most promising approaches to recreate organs for regenerative medicine. As human embryonic stem cells represent the epiblast stage of embryogenesis1, such directed differentiation is undertaken in a stepwise manner designed to recapitulate the developmental process from the epiblast to a specific tissue cell type. This methodology works more efficiently than single-step based methods2 and has been widely used to successfully induce various human tissues such as hematopoietic, cardiac, lung, pancreatic, hepatic, intestinal, cerebral and renal endpoints3–9. While a variety of cell types can be induced in vitro via stepwise differentiation protocols, three-dimensional structures are further required to recreate complex multicellular and functional organs. Generation of such three-dimensional mini-organs, called organoids10, from hPSCs has been reported in recent years, including organoids of brain, optic cup, stomach, intestine and liver11–15. In this protocol, we describe the methodology we have recently published for generating kidney organoids from hPSCs16. This expands on the brief step-by step protocol describing kidney organoid generation we previously uploaded to Protocol Exchange17. Development of the protocol The stepwise differentiation of hPSCs to kidney begins with the induction of the primitive streak which is the progenitor population for both endoderm and mesoderm. While the anterior primitive streak gives rise to the endoderm, the posterior primitive streak has potential to develop into the mesoderm, including the axial, paraxial, intermediate and lateral plate mesoderm18,19. The intermediate mesoderm differentiates to the ureteric epithelium and the metanephric mesenchyme, which are two key kidney progenitor populations subsequently undergoing a reciprocal interaction to form the kidney20. The ureteric epithelium develops into the collecting duct of the kidney and the ureter connecting the kidney with the bladder21. The metanephric mesenchyme gives rise to the cap mesenchyme which has been shown via lineage tracing to differentiate into all other epithelial cell types of the nephrons22. In addition to these two progenitors, endothelial and renal interstitium progenitors also arise from the intermediate mesoderm although it is not yet clear if these are subsets of the metanephric mesenchyme or distinct outcomes from intermediate mesoderm23. Primitive streak induction The first stage of differentiation in this protocol is induction of the posterior primitive streak. As previously investigated24,25, the posterior primitive streak can be differentiated from mouse embryonic stem cells by activating BMP, Nodal and canonical WNT signaling in two-dimensional culture methods. This method can also be successfully applied to hPSCs26. At this stage, we also culture cells under a monolayer culture condition to control anteroposterior cell fate of the primitive streak more precisely than in embryoid bodies,.Cell-autonomous effects and cell-cell interactions promote spontaneous differentiation within embryoid bodies whereas specific conditions of growth factors, concentration, and timing can be chosen in monolayer culture to produce more robust and uniform differentiation to a specific lineage. We previously demonstrated that the posterior primitive streak was induced by canonical WNT signalling or the combination of high and low doses of BMP4 and Activin A, respectively, in 2 days. In contrast, high Activin A with low BMP4 concentrations differentiated hPSCs into the anterior primitive streak9. Intermediate mesoderm induction The second stage is differentiation of posterior primitive streak cells into the intermediate mesoderm. Our previous study showed that hESC-derived posterior primitive streak spontaneously gives rise to the lateral plate mesoderm under APEL medium culture conditions9. As the intermediate mesoderm develops medial to the lateral plate mesoderm during embryogenesis, it is necessary to control the medial nature of the differentiation process using exogenous factors. Thus, again, we keep a monolayer culture condition to control M-L cell fate. There are only a few morphogens that have been proven to regulate M-L patterning in trunk mesoderm. These are BMP4 and FGF9. BMP4 is expressed in the lateral plate mesoderm and promotes lateral plate mesoderm development, whereas noggin (NOG)-mediated antagonism of BMP signaling is required for paraxial mesoderm while a low concentration of BMP4 induces the intermediate mesoderm27. FGF9 is specifically expressed in the intermediate mesoderm from the caudal through to the rostral trunk region28 and effectively directs the differentiation of hPSC-derived primitive streak to the intermediate mesoderm9. In our protocol, FGF9 is sufficient to specify the intermediate mesoderm without using NOG (Fig. 1). Kidney organoid The kidney functions as a three-dimensional organ, hence the culture conditions for differentiation needs to switch from monolayer to three-dimensional for the cells to form intact renal structures. While continued 2D culture may be adequate to induce specific target cell types, as we have previously demonstated9, the process of aggregation provides an increased cell density and volume within which cells are able to positionally reorganise with respect to each other. We chose day 7 of differentiation to transfer cells from a culture plate onto a transwell filter as an aggregate grown at an air-media interface. Day 7 represents the stage of intermediate mesoderm which is considered to include not only progenitors of ureteric epithelium and metanephric mesenchyme but also progenitors of renal interstitium and endothelium23. This methodology of aggregation culture has been previously optimized for ex vivo culture of mouse embryonic kidneys20 and reaggregations of mouse embryonic kidney cells29. A proper three-dimensional environment allows this kidney progenitor mixture to undergo self-renogenesis to form a kidney organoid. Alternative methods for generating kidney tissues Several studies in which kidney progenitors were induced from hPSCs have been reported. Xia et al. generated CK8-positive ureteric bud progenitor-like cells from hiPSCs and showed that those cells could integrate into ureteric epithelium in a re-aggregate with mouse embryonic kidney cells30. Other studies differentiated monolayer hPSCs into SIX2-positive metanephric mesenchyme that developed to renal tubules31–33. Taguchi and colleagues carefully investigated the developmental process of the metanephric mesenchyme commitment in mice and used that knowledge to obtain metanephric mesenchyme cells from mouse ESCs and human iPSCs based on an embryoid body culture method34. Their protocol focused on inducing the posterior intermediate mesoderm in order to obtain only the metanephric mesenchyme without collecting duct, renal interstitial and endothelial cells. These induced metanephric mesenchyme cells could differentiate into not only renal tubules but also glomeruli by being combined with mouse dorsal spinal cord, a source of WNT signals known to induce nephrogenesis from the metanephric mesenchyme. Their follow-up study showed that these nephrons could become vascularized by incorporating host blood vessels when transplanted under a mouse renal capsule35. Morizane et al. optimized the differentiation protocol using monolayer hPSCs to maximally induce the metanephric mesenchyme with 90% efficiency36. Hence, similarly to the above protocol, this did not induce collecting ducts and other non-epithelial renal cell types. The induced metanephric mesenchyme cells were also able to develop into nephron structures including renal tubules and glomeruli when stimulated by canonical WNT signaling using CHIR99021, a WNT agonist. Generated renal tubules revealed cell death in response to nephrotoxicants, cisplatin and gentamicin. Freedman et al. employed an approach in which they started the differentiation from hPSC-derived epiblast spheroids by sandwiching hPSCs between two layers of dilute Matrigel37. Epiblast spheroids underwent epithelial-to-mesenchymal transition (EMT) to form a monolayer, followed by a mesenchymal-to-epithelial transition (MET) when reaggregated, resulting in the formation of renal tubules, glomeruli and endothelial cells. While all these studies exclusively induced either the ureteric epithelium or metanephric mesenchyme and their derivative cell types, our unique methodology generates both cell types at the same time9. In addition, our optimized protocol enables the generation of kidney organoids that contain not only the collecting duct and nephrons but also renal interstitium and an endothelial network16. Within the organoids, individual nephrons segment into distal tubules, early loops of Henle, proximal tubules and glomeruli containing podocytes that elaborate foot processes and can undergo vascularization. Such segmented nephrons are connected with collecting ducts and surrounded by renal interstitium and an endothelial network. This protocol is clearly distinguished from others, as all anticipated cell types are simultaneously developed in the organoids. Applications and limitations of the protocol The differentiation of hPSCs using this protocol recapitulates the developmental process of human kidney organogenesis. Therefore, this protocol can be used as a platform for a variety of studies such as understanding human development, modeling renal disease and nephrotoxic drug screening. For instance, we have used this protocol to investigate mediolateral (M-L) and anteroposterior (A-P) patterning during trunk mesoderm development9,16,38. Also, as kidney organoids contain all components of the kidney, kidney organoids can be used for studying development of each distinct cell type present within the human kidney. Kidney organoids can be also utilized to test renal tubular damage in response to nephrotoxic drugs. We have confirmed proximal tubule specific cell death in kidney organoids by adding 5-20 μM cisplatin to the culture medium for 2 days16. This demonstrated the presence of proximal tubules which are sufficiently mature to appropriately respond to cisplatin, presumably as a result of the presence of basolateral organic cation transporter 2 (OCT2) and copper transporter 1 (CTR1) that mediate uptake of cisplatin into these tubular cells39,40. Hence, kidney organoids should be of value for the screening of pharmaceuticals. Modeling genetic renal disease using kidney organoids is another useful application. Kidney organoids using iPSCs that are generated from readily accessible somatic cells (e.g. skin fibroblasts or leucocytes) from inherited kidney disease patients may recapitulate features of these renal disorders in vitro37. Such disorders include autosomal dominant / recessive polycystic kidney disease (PKD), medullary cystic kidney (MCKD), nephronophthisis (NPHP), Alport syndrome and Bartter syndrome41. However, considering the likely variability in forming kidney organoids between iPSC lines, it will be desirable to compare mutant iPSC to a proper isogenic control iPSC line. This can be done using genome-editing tools like CRISPR/Cas9 system to artificially generate a specific genetic mutation in an iPSC line37. It may be further ideal to correct the mutation in patient-derived iPSCs, again using CRISPR/Cas9 technology, in order to generate an isogenic control iPSC line. This will only be possible in instances where the underlying mutation has been identified. Despite significant possibilities, the protocol has limitations and is likely to benefit from further protocol improvements. Kidney organoids at day 18 in three-dimensional culture, while transcriptionally similar to the first trimester human kidney16, have not matured to the level of the adult kidney. As consistent with a previous study of Cadherin expression in the developing mouse kidney42,43, we showed that more mature proximal tubules in kidney organoids expressed ECAD16. Cultured human proximal tubule cells also express ECAD44, however, some reports suggest that terminally differentiated proximal tubules in adult human kidney reveal no or very low ECAD expression45. This protocol has not generated kidney organoids that reach an adult stage of maturation. This is evident by a lack of expression of some mature renal tubule markers and capillary loops do not form in most glomeruli. Hence, it is highly likely that organoid-culture conditions may require further optimization to reach maturation. This immaturity may be problematic when studying disease modelling because most aforementioned inherited kidney diseases develop after birth, some not until adulthood. Also, it is important to note that the kidney organoid is not a kidney in miniature form from the anatomical and functional point of view. The mature kidney filters the blood to form a urinary filtrate before reclaiming the majority of this fluid to finally produce 1 to 1.5 litres of urine a day. During this process, the kidney also removes nitrogenous and toxic wastes, regulates electrolytes, maintains acid–base balance and regulates blood pressure, both via selective tubular reabsorbtion, secretion and hormone production. To recapitulate such diverse functions would require hPSC-derived kidney organoids not only to be reasonably big and comprised of mature cell types, but also develop a vascular access and a single exit path for urine, both of which are currently missing in current kidney organoids. Experimental Design hPSC culture for differentiation This protocol can be applied to both hESC and hiPSC. We have successfully applied the method described here to both hESCs (HES-3, commercially available) and hiPSCs (CRL1502 clone C32) that were authenticated and tested for mycoplasma infection46. Cells were maintained on mouse embryonic fibroblast (MEF) feeder layer and passaged using TrypLE Select. However, before the differentiation began, cells were adapted to feeder-free conditions with MEF-conditioned medium and Matrigel-coated culture dishes. This is to eliminate MEFs from the differentiation culture to avoid any unknown influence of MEFs to the differentiation. The differentiation begins with hPSC density at approximately 40-50% confluency (step 24), which can be obtained by plating about 15,000 cells / cm2 the day before (step 23). This plating cell number depends on the cell cycle speed and recovery rate, hence, should be adjusted for each hPSC line/clone. Intermediate mesoderm induction Although the intermediate mesoderm is the common origin of all known kidney progenitors, different regions in the intermediate mesoderm along with A-P axis contribute to distinct types of kidney progenitor. The anterior intermediate mesoderm (GATA3+) gives rise to the ureteric epithelium and the posterior intermediate mesoderm (HOXD11+) develops into the metanephric mesenchyme34,47,48. In an embryo, A-P patterning in the intermediate mesoderm is associated with primitive streak cell migration in which cells migrate from caudal to rostral during which they are exposed to canonical WNTs, followed by FGF9 and RA38. Hence, earlier migrating cells, which are exposed to WNTs for a shorter period of time, give rise to the anterior intermediate mesoderm, whereas cells migrating late are exposed to WNTs for longer and form the posterior intermediate mesoderm. In the protocol, hPSCs are treated with GSK-3 inhibitor, CHIR99021 to induce canonical WNT signaling (Fig. 1). This differentiates hPSCs into the posterior primitive streak that expresses T and MIXL1 with minimum expression of SOX17, a marker of the anterior primitive streak9. CHIR99021 is added to hPSC culture for 3 to 4 days, which is an intermediate duration designed to induce both the anterior and the posterior intermediate mesoderm at the same time (Fig. 1). In this protocol, we employ a CHIR99021 Conversely, a shorter (less than 3 days) or longer (more than 4 days) phase of CHIR99021 results in the predominant induction of the anterior or posterior intermediate mesoderm, respectively. The medium period of CHIR99021 may vary depending on hPSC clones and lines to use, therefore it needs to be optimized not to obtain one-sided intermediate mesoderm population of either GATA3+ or HOXD11+ at day 7 of the differentiation16. As generally seen in human hPSC differentiation, each hPSC line or clone has its own propensity to develop into a certain cell fate easier than to the other lineages49 and we experienced in the initial step of the differentiation that distinct clones respond to CHIR99021 differently. To generate kidney organoids reproducibly, it is recommended to test the protocol once with distinct hES/iPSC lines or distinct hiPSC clones even when derived from the same individual, and choose the most responsive lines or clones to use thereafter. Hence, an optimal CHIR99021 concentration and differentiation time period may be required for each line. e.g. a range of 2-5 days of CHIR99021 period first, then concentrations between 6 to 10 μM CHIR99021 if necessary. Kidney organoid formation The kidney progenitor aggregate includes the ureteric epithelium and the metanephric mesenchyme, hence their reciprocal interaction spontaneously initiates nephrogenesis. However, to maximize nephron formation, the protocol stimulates the organoids with a high concentration CHIR99021 for 1 hour. After the CHIR99021 pulse, the organoids should be supplemented with FGF9 to maximize the development of nephrons. We found 5 additional days of FGF9 supplementation was sufficient to obtain maximal nephrogenesis. Subsequently, the organoids are cultured in growth factor-free APEL medium50 for further self-organization (Fig. 1). The resulting organoids are regarded as successful based upon the presence of appropriately segmented nephrons and surrounding stromal and endothelial cell populations. MATERIALS REAGENT Human embryonic stem cells. We have used HES-3 (Wicell Research Institute Inc, cat. no. ES03) ! CAUTION Experiments using hPSCs must conform to all relevant governmental and institutional regulations relating to human ethics, biosafety and genetic modification This work was approved by the MCRI Institutional Biosafety Committee (212-2014PC2).It is recommended to check regularly to ensure cells are chromosomally stable and are not infected with mycoplasma. Human induced pluripotent stem cells. We have used CRL1502 clone C32 (gifted from Dr. Ernst Wolvetang, The University of Queensland, Australia) ! CAUTION Experiments using hPSCs must conform to all relevant governmental and institutional regulations relating to human ethics, biosafety and genetic modification. This work was approved by the Royal Childrens Hospital human ethics committee (HREC/14/QRBW/34; HREC/15/QRCH/126) and the MCRI Institutional Biosafety Committee (212-2014PC2). It is recommended to check regularly to ensure cells are chromosomally stable and are not infected with mycoplasma. 2-Mercaptoethanol (55 mM) (Thermo Fisher Scientific, cat. no. 21985-023) Antibiotic-Antimycotic (Thermo Fisher Scientific, cat. no. 15240-062) bFGF (Merck, cat. no. GF003-AF) BSA (Sigma Aldrich, cat. no. A3311-10G) CHIR99021 (R&D, cat. no. 4423/10) DMEM high glucose (Thermo Fisher Scientific, cat. no. 11995-073) DMEM/F-12 (Thermo Fisher Scientific, cat. no. 11320-082) DMSO (Sigma Aldrich, cat. no. D5879) DTT, 100 mM (Promega, cat. no. P1171) Dulbecco's phosphate-buffered saline (DPBS) (Thermo Fisher Scientific, cat. no. 14190-144) FGF9 (R&D, cat. no. 273-F9-025) Foetal Bovine Serum (FBS) (Interpath Services, cat. no. SFBSF) Gelatin (Sigma Aldrich, cat. no. G9391) GlutaMAX Supplement (Thermo Fisher Scientific, cat. no. 35050-061) Glycerol (Sigma Aldrich, cat. no. G9012-500ML) Heparin (Sigma Aldrich, cat. no. H4784-250MG) hESC-qualified Matrigel, LDEV-Free (Corning, cat. no. 354277) Knockout Serum Replacement (Thermo Fisher Scientific, cat. no. 10828-028) Non-Essential Amino Acids (Thermo Fisher Scientific, cat. no. 11140-050) Penicillin/Streptomycin (Thermo Fisher Scientific, cat. no. 15070-063) Recombinant human serum albumin, 10% (wt/vol) (Novozymes, 230-005) STEMdiff APEL Medium (Stem Cell Technologies, cat. no. 5210) TrypLE Select (Thermo Fisher Scientific, cat. no. 12563-029) Trypsin EDTA, 0.05% (Thermo Fisher Scientific, cat. no. 25300-054) Paraformaldehyde (Sigma-Aldrich, cat. no. 158127) Triton X-100 (TritonX) (Sigma-Aldrich, cat. no. T9284-500ML) Donkey serum (Merck Millipore, cat. no. S30-100ML) DAPI (6-Diamidino-2-Phenylindole Dihydrochloride) (Thermo Fisher Scientific, cat. no. D1306) Appropriate antibodies to detect proteins of interest (see Table 1 for primary antibodies we have successfully used). EQUIPMENT 175 cm2 tissue culture flask (Nunc, cat. no. 159910) 25 cm2 tissue culture flask (Nunc, cat. no. 156367) 6-well transwell cell culture plate, 0.4 μm pore polyester membrane (Corning, cat no. 3450) Benchtop centrifuge (Thermo Scientific, Heraeus Pico 17) Biological safety cabinet (LAF Technologies, TOP-SAFE 1.2) CO2 incubators (Thermo Scientific, Heracell 150i) Conical tubes, 15 ml (Corning, cat. no. 352096) Conical tubes, 50 ml (Corning, cat. no. 352070) Freezing container (Nalgene, Mr. Frosty) Glass-Bottom Dish, 35mm (Mattek, cat. no. P35G-0-14-C) Inverted contrasting tissue culture microscope (Nikon, TS100) Laser confocal microscope (Zeiss, LSM780) Media storage bottle, 500 ml (Corning, cat. no. 430282) Pipetboy (Integra Biosciences, PIPETBOY pro) Pipettes (Gilson, P10, P200 and P1000 PIPETMAN Classic) Serological pipettes, 10 ml (Corning, cat. no. 357551) Stericup 0.22 μm filter unit (Merck Millipore, cat. no. SCGPU05RE) Sterile filter pipette tips (1000, 200, wide bore 200 and 10 μl; Molecular BioProducts, cat. nos. 3581, 3551, 3531 and 3501, respectively) Sterile microcentrifuge tube, 1.5 ml (Eppendorf, cat. no. 3810X) Swing-out rotor centrifuge (Eppendorf, Centrifuge 5702) Syringe-driven 0.22 μm filter unit (Merck Millipore, cat. no. SLMP025SS) Ultrapure water generation system (Sartorius, arium pro) Water bath, 37 °C (Thermoline, TWB-24D) REAGENT SETUP Matrigel-coated 25 cm2 tissue culture flask To prepare Matrigel-coated 25 cm2 tissue culture flask, add 30 μl of hESC-qualified Matrigel into a 15 ml tube containing 3 ml of chilled DMEM/F-12. Mix it well and transfer it into a 25 cm2 tissue culture flask. Keep the flask at room temperature, 15 to 25 °C, for at least 30 min to allow Matrigel to coat the surface. Use the flask within a day. CRITICAL STEP Handle Matrigel on ice as it solidifies when warmed over 16 °C. P200 pipette tips should be also pre-chilled before use. bFGF (10 μg ml−1) Centrifuge the tube briefly before opening. Reconstitute to 10 μg ml−1 in a filtered solution of 0.5% (wt/vol) BSA, 1 mM DTT and 10% (vol/vol) glycerol in DPBS. Aliquot it into appropriate amounts and store them at −80°C for up to 6 months. bFGF can be stored at 4 °C for up to 2 weeks once it is thawed. FGF9 (100 μg ml−1) Centrifuge the tube briefly before opening. Reconstitute to 100 μg/ml in filtered DPBS containing 0.1% (wt/vol) human serum albumin. Aliquot it into appropriate amounts and store them at −80°C for up to 6 months. FGF9 can be stored at 4 °C for up to 2 weeks once it is thawed. Heparin solution (1 mg ml−1) Reconstitute to 1 mg ml−1 in ultrapure water and filter sterilize it through a polyethersulfone (PES) 0.22 μm syringe-driven filter unit. Heparin solution can be stored at 4 °C for more than 12 months. CHIR99021 (10 mM) Centrifuge the tube briefly before opening. Reconstitute 10 mg of CHIR99021 into 2.149 ml of DMSO to make 10 mM stock. Aliquot it into appropriate amounts and store them at −20°C. Gelatin solution (0.1%, wt/vol) For 500 ml of 0.1% gelatin solution, dissolve 0.5 g of gelatin in 500 ml of ultrapure water or DPBS and autoclave it. The solution can be stored at 4 °C for up to 3 months. Mouse Embryonic Fibroblast (MEF) culture medium For preparing 500 ml of MEF culture medium, combine 442.5 ml of DMEM high glucose, 50 ml of foetal bovine serum, 5 ml of GlutaMAX-1 and 2.5 ml of Penicillin/Streptomycin. Filter sterilize the medium through a polyethersulfone (PES) 0.22 μm vacuum-driven filter unit and store it at 4 °C for up to 2 weeks. hESC medium For preparing 500 ml of hESC medium, combine 386.5 ml of DMEM/F-12, 100 ml of Knockout serum replacement, 5 ml of GlutaMAX-1, 5 ml of Non-Essential Amino Acids, 2.5 ml of Penicillin/Streptomycin and 1 ml of 2-Mercaptoethanol (55 mM). Filter sterilize media through a polyethersulfone (PES) 0.22 μm vacuum-driven filter unit and store it at 4 °C for up to 2 weeks. To ensure bFGF is fresh, only supplement the medium with 10 ng ml−1 bFGF just before use for hPSC maintenance. MEF-conditioned hESC medium Seed 1 × 107 mitotically inactivated MEFs onto a 175 cm2 tissue culture flask containing 40 ml of MEF culture medium. Next day, change the culture medium to 40 ml of hESC medium without bFGF. After overnight culture, collect the conditioned medium into a 500 ml bottle and store it at 4 °C. Feed MEFs again with 40 ml of fresh hESC medium. Repeat this cycle another 6 times to pool 280 ml of the conditioned medium in total. Filter sterilize the medium through a polyethersulfone (PES) 0.22 μm vacuum-driven filter unit and aliquot it into 50 ml tubes. Store tubes at −20°C for up to 6 months. MEF-conditioned hESC medium can be stored at 4 °C for up to 2 weeks once it is thawed. To ensure bFGF is fresh, only supplement the medium with 10 ng ml−1 bFGF just before use for hPSC maintenance. APEL medium Once a bottle of APEL (100 ml volume) is opened, add 0.5-1 ml of Antibiotic-Antimycotic and store it at 4 °C for up to 2 weeks. Frozen stock of human pluripotent stem cells (hPSCs) For expanding hPSCs before cryopreserving them, cells are cultured in hESC medium supplemented with 10 ng ml−1 bFGF on mitotically inactivated MEF feeder layer by a single-cell culture method in which cells are passaged using TrypLE Select, as described in procedure steps 1-9. Once hPSCs reach the desired number, harvest cells using TrypLE Select, centrifuge (×400g for 3 min) and resuspend them in 10% DMSO / 90% FBS. Split cell-suspension into cryo-vials for 1 ml per vial. Typically, hPSCs of 100% confluent in a 75 cm2 tissue culture flask are split into 9 cryo-vials, so that each cryo-vial contains 1.5-2.0 × 106 cells. Freeze vials in a freezing container at −80 °C overnight and subsequently store them in liquid nitrogen. CRITICAL We have experienced a huge variability in differentiation success rate between experiments when cells were obtained from a continuously maintained cell culture pool. To minimize this variation, hPSCs should be thawed one by one from a cryopreserved stock for each experiment. 2% (wt/vol) paraformaldehyde (100 ml) Add 2 g of PFA and 10 μl of 10 N NaOH to 80 ml of DPBS at 60 °C and mix it until the powder dissolves. Adjust the final volume up to 100 ml by adding DPBS. Aliquot it into appropriate amounts and store them at −20 °C for up to one month. ! CAUTION Paraformaldehyde is toxic and must be handled inside a fume hood with masks and goggles. Blocking buffer (50 ml) Dissolve 5 ml of donkey serum and 30 μl of TritonX in 45 ml of PBST. Aliquot it into appropriate amounts. Blocking buffer can be stored at −20 °C for up to one year. PROCEDURE MEF feeder seeding TIMING 2 h 1. Coat a 25 cm2 tissue culture flask with 3ml of 0.1% gelatin solution. Incubate the flask in a 37 °C CO2 incubator for 1h. 2. Prepare 10 ml of prewarmed MEF culture medium in a 15 ml conical tube. 3. Thaw a frozen vial of mitotically inactivated mouse embryonic fibroblasts (MEFs) in 37 °C water bath until a small ice pellet remains. Transfer MEFs into a 15 ml conical tube containing prewarmed MEF culture medium in a drop wise manner and centrifuge at ×400g for 3 min. 4. Remove supernatant and resuspend MEFs in MEF culture medium. Count cell number using a hemocytometer. Make cell-suspension of 3 × 105 cells in 5 ml of MEF culture medium. Remove gelatin and seed cells onto a gelatin coated 25 cm2 tissue culture flask to obtain the density at 12 × 103 cells / cm2 and incubate it overnight in a 37 °C CO2 incubator. Human PSC thawing TIMING 30 min 5. Prepare 10 ml of prewarmed hESC medium in a 15 ml conical tube. 6. Thaw a frozen vial of hPSC containing 1.5-2.0 × 106 cells in 37 °C water bath until a small ice pellet remains. Transfer hPSCs into a 15 ml conical tube containing prewarmed DMEM/F-12 in a drop wise manner and centrifuge at ×400g for 3 min. 7. Remove supernatant and resuspend cells in 5 ml of hESC medium supplemented with 10 ng ml−1 bFGF. 8. Remove MEF culture medium from the flask containing mitotically inactivated MEFs and plate above hPSCs suspension. Incubate it overnight in a 37 °C CO2 incubator. Growing hPSCs on MEF feeder layer TIMING 4-5 d 9. Change 5 ml of hESC medium supplemented with 10 ng ml−1 bFGF daily for 4-5 days. CRITICAL STEP For optimal results, cells should be approximately 80-100% confluent after 4-5 days culture. If cells do not reach this confluency, allow them to grow for another day. ? TROUBLESHOOTING Growing hPSCs on Matrigel TIMING 2 d 10. Prepare Matrigel-coated 25 cm2 tissue culture flask. 11. To passage hPSCs onto Matrigel, wash hPSCs on MEF feeder layer in 25 cm2 tissue culture flask with 3 ml DPBS twice. Aspirate DPBS. 12. Add 1 ml of TrypLE Select to cells and incubate the flask at 37 °C for 3 min. 13. Pipette 11 ml of DMEM/F-12 to cells, mix and ensure cells have lifted off from the plastic surface. CRITICAL STEP Pipette cells no more than twice as hPSCs are very sensitive. 14. Collect 4 ml of cell suspension in a 15 ml tube to obtain a 1:3 split ratio and centrifuge it at ×400g for 3 min. Alternatively, a 1:2 split ratio can be chosen to achieve 70-100 % confluency at step 17. 15. Remove supernatant and add 5 ml of MEF-conditioned hESC medium supplemented with 10 ng ml−1 bFGF to cells. Mix it gently. 16. Aspirate Matrigel-containing DMEM/F-12 from a prepared Matrigel-coated 25 cm2 tissue culture flask and seed cells from step 15. Culture them in a 37 °C CO2 incubator for 2 days with daily changing MEF-conditioned hESC medium supplemented with 10 ng ml−1 bFGF. Plating hPSCs for differentiation TIMING 1 h 17. Wash hPSCs on Matrigel in 25 cm2 tissue culture flask with 3 ml of DPBS twice. Aspirate DPBS. 18. Add 1 ml of TrypLE Select to cells and incubate at 37 °C for 3 min. 19. Pipette 10 ml of DMEM/F-12 to cells, mix and ensure cells have lifted off from the plastic surface. CRITICAL STEP Do not pipette cells more than twice as hPSCs are very sensitive. 20. Collect cell suspension into 15 ml tube. Count cell number using a hemocytometer. 21. Calculate cell suspension volume to achieve 375,000 cells. 22. Aliquot cells into a 15 ml tube. Centrifuge the tube at ×400g for 3 min. 23. Remove supernatant and resuspend cells in 4 ml of MEF-conditioned hESC medium supplemented with 10 ng ml−1 bFGF. Seed cells onto a prepared Matrigel-coated 25 cm2 tissue culture flask to obtain the density at 15 × 103 cells / cm2. Culture them overnight in a 37 °C CO2 incubator. Inducing the intermediate mesoderm TIMING 7 d 24. Aspirate MEF-conditioned hESC medium from the 25 cm2 tissue culture flask. CRITICAL STEP Cells should reach to 40-50% confluent on this day. If not, adjust cell number of plating at Step 21. 25. Add 4 ml of APEL medium containing 8 μM CHIR99021 to hPSCs. 26. Culture them in a 37 °C CO2 incubator for 2-5 days, refreshing APEL medium containing 8 μM CHIR99021 every 2 days. CRITICAL STEP Duration of CHIR99021 determines the ratio of collecting duct/nephron in the organoid. Shorter periods of CHIR99021 phase induce more anterior intermediate mesoderm whereas a longer periods results in the more posterior part. To obtain both compartments, 3 or 4 days of CHIR99021 is recommended. ? TROUBLESHOOTING 27. After CHIR99021 phase, change culture medium to 8 ml of APEL medium supplemented with 200 ng ml−1 FGF9 and 1 μg ml−1 Heparin. CRITICAL STEP A shortage of FGF9 concentration and/or media volume causes an inefficient induction of the intermediate mesoderm. 28. Culture them in a 37 °C CO2 incubator, refreshing medium every 2 days until day 7 of the differentiation counting step 24 as day 0 this includes both CHIR99021 phase and FGF9 phase Forming aggregates for developing kidney organoids TIMING 11-18 d 29. Aspirate the culturing medium and wash with 3 ml of DPBS. Aspirate DPBS. 30. Add 1 ml of trypsin EDTA (0.05%) to cells and incubate them at 37 °C for 3 min. 31. Monitor under the microscope to make sure all cells have lifted from the surface after 2 min. If the cells are still attached to the surface, gently pipette the cells with trypsin and place back into the incubator for further 1 min. 32. Transfer cell suspension to a 15 ml tube containing 9 ml of MEF culture medium to neutralize trypsin. Centrifuge the tube at ×400g for 3 min. 33. Aspirate the supernatant and resuspend the cells with 3 ml of APEL medium. 34. Take out 10 μl of cell suspension and perform a cell count with a hemocytometer. 35. Each organoid will have roughly 5 × 105 cells. Aliquot the required amount of cell suspension into a 1.5 ml microcentrifuge tube. Centrifuge the tube at ×400g for 2 min to make a cell pellet. 36. Aliquot 1.2 ml of APEL containing 5 μM CHIR99021 into a 6-well transwell cell culture plate. The transwell filter attaches on the surface of the medium. 37. Pick a pellet up by using a P1000 or P200 wide bore tip. ? TROUBLESHOOTING 38. Carefully place pellets onto the 6-well transwell filter with minimal APEL medium carry over. Incubate pellets at 37 °C for 1 h. 39. After 1 h incubation, remove the medium of APEL containing 5 μM CHIR99021, and use 1.2 ml of APEL medium supplemented with 200 ng ml−1 FGF9 and 1 μg ml−1 Heparin. 40. Culture pellets for 5 days, refreshing FGF9 and Heparin-containing APEL medium every two days. ? TROUBLESHOOTING 41. After 5 days, change the culture medium to APEL medium without FGF9 and Heparin supplemented. 42. Culture the organoids for further 6 to 13 days, refreshing APEL medium every two days. If you wish to proceed to whole mount immunofluorescence, proceed to the next section. Whole mount immunofluorescence TIMING 3 d 43. After 18 days of organoid culture, if desired, fix the pellets in the transwell cell culture plate with 2% (wt/vol) paraformaldehyde at 4 °C for 20 minutes. 44. Remove the paraformaldehyde and wash three times with DPBS. PAUSE POINT Once fixed and washed, organoids can be stored at 4 °C for several months before immunofluorescence staining. 45. Aliquot around 150 μl of blocking buffer (10% donkey serum / 0.3% TritonX / DPBS) into the MatTek Glass Bottom Dishes. 46. Carefully cut the organoids off the filter and submerge the filter into the blocking buffer. 47. Block the organoids at room temperature for 2-3 h, gently keep shaking the dish using a rocker during incubation. 48. Prepare primary antibodies of choice in blocking buffer (0.3% TritonX / 10% Donkey serum / DPBS) with appropriate dilutions (Table 1). 49. Aspirate the blocking buffer off the MatTek dish, and aliquot 150 μl of blocking buffer containing primary antibodies into the MatTek dish. 50. Incubate the organoids with primary antibody solution at 4 °C over-night. CRITICAL STEP Make sure the organoid is completely submerged into the solution. Preferably, gently keep shaking the dish using a rocker during incubation. 51. Aspirate the primary antibody solution off the dish and wash with PBTX (0.3% TritonX / DPBS) for 6 times, 10 minutes each with gentle shaking. 52. Prepare secondary antibodies (1:400 dilution) of choice in PBTX. 53. Aspirate PBTX off the MatTek dish, and aliquot 150 μl of PBTX containing secondary antibodies into the MatTek dish. 54. Incubate organoids in secondary antibody solution at 4 °C over-night. 55. Remove secondary antibody solution and incubate with DAPI (1:1000 dilution) in DPBS for 3 h. 56. Wash organoids with DPBS for 10 min, 3 times. 57. Image organoids using a laser confocal microscope. ? TROUBLESHOOTING Troubleshooting advice can be found in Table 2. TIMING Steps 1-4, MEF feeder seeding: 2 h (Begin 9 days before hPSC differentiation at step 24) Steps 5-8, Human PSC thawing: 30 min Step 9, Growing hPSCs on MEF feeder layer: 4-5 d Steps 10-16, Growing hPSCs on Matrigel: 2 d Steps 17-23, Plating hPSCs for differentiation: 1 h Steps 24-28, Inducing the intermediate mesoderm: 7 d (Day 0 to 7) Steps 29-42, Forming aggregates for developing kidney organoids: 11-18 d (Day 7 to 25) Steps 43-57, Whole mount immunofluorescence: 3 d ANTICIPATED RESULTS This protocol produces kidney organoids from hPSCs with a high success rate. Success is defined as organoids containing segmented nephrons comprised of collecting duct, renal interstitium and endothelium. We have successfully generated organoids from both human ES cell lines (HES-3) and human iPSC lines (CRL1502 clone C32) and a number of patient-derived iPSC lines derived in house (unpublished). In addition, there is little variation between organoids within an experiment16. When the differentiation starts with CHIR99021 (Fig. 1 day 0), hPSCs on Matrigel should retain pluripotency as indicated by expression of stem cell markers such as OCT4 and NANOG, and with the anticipated morphology of a big nucleus, although defined borders of colonies are typically not seen at this stage (Fig. 2a,b). After 2 days of CHIR99021 exposure, colonies break apart into single cells with the morphology of spiky, typically triangular, shape (Fig. 2c). This is evidence of the epithelial-tomesenchymal transition (EMT) expected of hPSCs differentiating into the primitive streak. Once the 3-4 days of CHIR99021 phase ends (Fig. 2d) and the culture medium contains FGF9, cells start growing rapidly to reach confluence by day 7 of differentiation. The surface of the cell-layer at day 7 typically looks ‘hilly’ due to the contrast between thin and thick cell-layers (Fig. 2e). Occasionally cells become balls by day 7, which indicates the initial CHIR99021 duration or CHIR99021 concentration should be slightly shortened or reduced for an optimal result. If immunofluorescence on this day shows predominant population of either GATA3 or HOXD11-positive cells, adjust days of CHIR99021 to obtain both populations (Fig. 3a)16. The size of pellets comprised of 5 × 105 cells is approximately 2 mm in diameter (Fig. 2f). This pellet initiates nephrogenesis with developing renal vesicles in 3 days under the three-dimensional culture (organoid culture) conditions. The presence of renal vesicles can be confirmed under a tissue culture microscope (Fig. 2g,h). In the case of a failure of nephrogenesis, organoids look like a pancake without formation of any complex structures. After 11 days of organoid culture, if the differentiation has worked properly, obvious formation of glomeruli as well as tubular structures can be observed (Fig. 2i,j). The resulting kidney organoid will grow up to 6 mm in diameter by 18 days of organoid culture (Fig. 1 day 25). To evaluate tissues in kidney organoids, whole mount immunofluorescence can be performed (steps 44-58). Within the organoid, a number of different kidney cell types are visible (Table 1), including collecting ducts marked by PAX2, GATA3 and ECAD staining (Fig. 3b), loops of Henle marked by UMOD staining (Fig. 3c), proximal tubules marked by LTL and cubilin (CUBN) staining (Fig. 3d), the podocyte of the glomerulus marked by WT1 and NPHS1 staining (Fig. 3f) and basement membrane in glomeruli marked by LAM (Fig. 3g arrowheads), early mesangial cells marked by PDGFRA staining (Fig. 3h), distal tubule marked by ECAD staining in the absence of co-staining with either LTL or GATA3 (Fig. 3f), medullary interstitium marked by MEIS1 alone and cortical interstitium marked by MEIS1 and FOXD1 staining51 (Fig. 3e) and an endothelial network marked by CD31 (PECAM), KDR and SOX17 staining (Fig. 3h,i). Endothelial networks tend to run in between nephrons and also on the surface of the organoid (Fig. 3j). 4-color staining using NPHS1, LTL, ECAD and GATA3 antibodies/lectin enables to identify segmented nephrons in the kidney organoid (Fig. 3k). Typically, bigger sized kidney organoids develop nephrons densely in the peripheral region and have renal interstitium and endothelial network to grow in the middle (Fig. 3l). ACKNOWLEDGMENTS We are grateful to F. Froemling for experimental support and EJ. Wolvetang for providing iPSC lines. This research was supported by National Health and Medical Research Council (NHMRC) of Australia (APP1041277), the Australian Research Council (Stem Cells Australia, SRI110001002) and Organovo Inc. MHL is an NHMRC Senior Principal Research Fellow. We also acknowledge the Australian Cancer Research Foundation’s (ACRF) Cancer Biology Imaging Facility at The University of Queensland. Figure 1 Schematic diagram of the timeline for generating kidney organoids from hPSCs The protocol is based upon a step-wise differentiation protocol with sequential changing of culture media. hPSCs are cultured in MEF-CM (MEF-conditioned hESC medium). Differentiation begins in APEL/CHIR99021 (APEL medium supplemented with 8 μM CHIR99021) followed by APEL/FGF9/Heparin (APEL medium supplemented with 200 ng ml−1 FGF9 and 1 μg ml−1 heparin) and all growth factors are withdrawn in the final step. A pulse of CHIR99021 (5 μM) stimulates nephrogenesis in the organoids. Figure 2 Bright field images of hPSCs differentiation to kidney organoids Bright field images of the cells during differentiation. a, hPSC colonies on MEF feeder layer at day −3, before transfer onto Matrigel. b, hPSCs on Matrigel at day 0 before adding CHIR99021 to the medium. c, hPSCs differentiated into the posterior primitive streak cells with CHIR99021 induction showed a triangular shaped cell morphology. d, Monolayer culture at day 4 of differentiation e, Cells at day 7 of differentiation. Ideally, cultures nearly reach confluence at this time and form distinct areas of thick and thin layers across the surface. f, At day 7 of differentiation, cells were trypsinized and centrifuged to form an aggregate. This panel represents an aggregate at 30 min after transfer to transwell filter. g,h, A whole kidney organoid at day 10 of differentiation (3 days in organoid culture). A black square indicates the field of the right panel (h), which shows the formation of small renal vesicles. i,j, A whole kidney organoid at day 25 of differentiation (18 days in organoid culture). A black square indicates the field of the right panel (j). If the differentiation has been successful, glomerular structures (gl) can be recognized under microscope. Scale bars, 200 μm (a-e), 1 mm (f,g,i), 100 μm (h,j). Figure 3 Immunological characterization of structures within kidney organoids a, Staining for anterior intermediate mesoderm cells (GATA3) and posterior intermediate mesoderm cells (HOXD11) on day 7 of the differentiation (Step 29). b-l, Whole mount immunostaining of kidney organoids (Step 57). b, Co-staining for markers of collecting ducts (PAX2, GATA3 and ECAD). c, Staining for loop of Henle (UMOD and ECAD). d, Staining for proximal tubule markers (LTL and CUBN) shows evidence of an apical brush border (CUBN). e, Staining for cortical interstitium (MEIS1 and FOXD1, a white arrowhead) and medullary interstitium (MEIS1 only). FOXD1 also marks podocytes (WT1). f, Evidence for the formation of segmenting nephrons showing staining for early podocytes (WT1 and NPHS1), distal tubules (ECAD) and intervening unstained segments at day 18 of differentiation. An involute cluster of podocyes can be seen at the end of a forming nephron (center). g, At day 25 of differentiation, maturing glomeruli develop glomerular basement membrane (LAM, blank arrowheads) in the middle of podocytes (NPHS1). h, Staining for early mesangial cells (PDGFRA) invaginating into podocytes (NPHS1). i, Staining for the endothelial network (CD31 and SOX17). j, Staining of a representative whole mount kidney organoid illustrating the endothelial network (CD31) and glomerular podocytes (NPHS1). Cell nuclei stained by DAPI are shown in the inset panel. k, An example of a small whole kidney organoid (< 3 mm). l, An example of a larger kidney organoid (> 4 mm) Scale bars, 50 μm (a-i), 1 mm (j-l). Table 1 Antibodies for tissue characterization Cell type Antigen Host Supplier Cat. no. Dilution Posterior IM HOXD11 Mouse Santa Cruz Biotechnology sc-192 1:300 Anterior IM and collecting duct GATA3 Goat R&D Systems AF2605 1:300 Renal tubules and collecting duct PAX2 Rabbit Zymed Laboratories 71-6000 1:300 Maturing proximal, tubule, distal tubule and collecting duct ECAD Mouse BD Biosciences 610181 1:300 Loop of Henle UMOD Rabbit Biomedical Technologies BT-590 1:300 Proximal tubule LTL Biotin-conjugated Vector Laboratories B-1325 1:300 Proximal tubule CUBN Rabbit Santa Cruz Biotechnology sc-20607 1:150 Podocyte WT1 Rabbit Santa Cruz Biotechnology sc-192 1:100 Podocyte NPHS1 Sheep R&D Systems AF4269 1:300 Endothelium CD31 Mouse BD Pharmingen 555444 1:300 Endothelium KDR Rabbit Cell Signaling Technology 2479 1:300 Endothelium SOX17 Goat R&D Systems AF1924 1:300 Early mesangium PDGFRA Mouse BD Pharmingen 556001 1:200 Basement membrane LAM Rabbit Sigma-Aldrich L9393 1:300 Renal interstitium MEIS1/2 Mouse Activemotif ATM39795 1:300 Podocyte and renal interstitium FOXD1 Goat Santa Cruz Biotechnology sc-47585 1:200 Table 2 Troubleshooting table Step Problem Possible reason Solution 9 Cells do not recover or grow slowly after thawing Unhealthy cells at cryopreservation. Cryopreserve healthy cells. Alternatively, Rho kinase inhibitor (Y-27632) can be supplemented to hESC medium at 10 μM for 24 h post thawing. 26 Cells come off the plastic during CHIR99021 phase Cell density is too low Plate more cells at Step 23. 37 Pellets break up during a transfer Not enough centrifuging Perform two sequential centrifugations of ×400g for 2 min. Rotate the tube180° before the second spin at Step 35. Up to four sequential centrifugations can be performed. 40 No nephrogenesis happens Damaged cells due to centrifuge If centrifugations are repeated multiple times at Step 37, reduce the number of times or duration. Intermediate mesoderm induction fails Confirm FGF9 activity is intact. Use freshly thawed FGF9. Use more than 0.32 ml of FGF9/Heparin/APEL medium per cm2 at Step 27. Differentiation fails Confirm if posterior intermediate mesoderm markers, HOXD11, are positive at step 29 (Fig. 3a). If they are negative, modify culture conditions for your cells; e.g. initial cell density at Step 23, CHIR99021 concentration or CHIR99021 duration at Step 25, 26. 3 key references 1. Takasato M, Er PX, Chiu HS, et al (2015) Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis. Nature 526:564–8. doi: 10.1038/nature15695 2. Takasato M, Little MH (2015) The origin of the mammalian kidney: implications for recreating the kidney in vitro. Development 142:1937–1947. doi: 10.1242/dev.104802 3. Takasato M, Er PX, Becroft M, et al (2014) Directing human embryonic stem cell differentiation towards a renal lineage generates a self-organizing kidney. Nat Cell Biol 16:118–26. doi: 10.1038/ncb2894 Editorial summary This protocol describes stepwise differentiation of human pluripotent stem cells into 3D kidney organoids that contain segmented nephrons connected to collecting ducts, surrounded by renal interstitial cells and an endothelial network AUTHOR CONTRIBUTIONS MT and MHL wrote the manuscript. MT, PXE and HSC performed the experiments. COMPETING FINANCIAL INTERESTS The authors declare competing financial interests (see the HTML version of this article for details). MT and MHL are named inventors on a patent relating to this methodology. REFERENCES 1 Tesar PJ New cell lines from mouse epiblast share defining features with human embryonic stem cells Nature 2007 448 196 9 17597760 2 Takasato M Maier B Little MH Recreating kidney progenitors from pluripotent cells Pediatr. Nephrol 2014 29 543 52 24026757 3 Davis RP Targeting a GFP reporter gene to the MIXL1 locus of human embryonic stem cells identifies human primitive streak-like cells and enables isolation of primitive hematopoietic precursors Blood 2008 111 1876 1884 18032708 4 Yang L Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem- cell-derived population Nature 2008 453 524 8 18432194 5 Wang D Haviland DL Burns AR Zsigmond E Wetsel RA A pure population of lung alveolar epithelial type II cells derived from human embryonic stem cells Proc. Natl. Acad. Sci. U. S. A 2007 104 4449 54 17360544 6 Pagliuca FW Generation of functional human pancreatic β cells in vitro Cell 2014 159 428 39 25303535 7 McCracken KW Howell JC Wells JM Spence JR Generating human intestinala tissue from pluripotent stem cells in vitro Nat. Protoc 2011 6 1920 1928 22082986 8 Xia X Zhang S Differentiation of neuroepithelia from human embryonic stem cells Methods Mol. Biol 2009 549 51 8 19378195 9 Takasato M Directing human embryonic stem cell differentiation towards a renal lineage generates a self-organizing kidney Nat. Cell Biol 2014 16 118 26 24335651 10 Lancaster M. a. Knoblich J. a. Organogenesis in a dish: Modeling development and disease using organoid technologies Science (80-. ) 2014 345 1247125 1247125 11 Lancaster MA Cerebral organoids model human brain development and microcephaly Nature 2013 501 373 379 23995685 12 Nakano T Self-Formation of Optic Cups and Storable Stratified Neural Retina from Human ESCs Cell Stem Cell 2012 10 771 785 22704518 13 McCracken KW Modelling human development and disease in pluripotent stem-cell- derived gastric organoids Nature 2014 516 400 404 25363776 14 Spence JR Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro Nature 2011 470 105 109 21151107 15 Takebe T Vascularized and functional human liver from an iPSC-derived organ bud transplant Nature 2013 499 481 4 23823721 16 Takasato M Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis Nature 2015 526 564 8 26444236 17 Takasato M Er PX Chiu HS Little MH Generation of kidney organoids from human pluripotent stem cells Protoc. Exch 2015 10.1038/protex.2015.087 18 Iimura T Pourquié O Collinear activation of Hoxb genes during gastrulation is linked to mesoderm cell ingression Nature 2006 442 568 571 16760928 19 Sweetman D Wagstaff L Cooper O Weijer C Münsterberg A The migration of paraxial and lateral plate mesoderm cells emerging from the late primitive streak is controlled by different Wnt signals BMC Dev. Biol 2008 8 63 18541012 20 Saxen L Organogenesis of the Kidney 1987 Cambridge University Press 10.1017/CBO9780511565083 21 Mendelsohn C Using mouse models to understand normal and abnormal urogenital tract development Organogenesis 2009 5 306 14 19568352 22 Kobayashi A Six2 defines and regulates a multipotent self-renewing nephron progenitor population throughout mammalian kidney development Cell Stem Cell 2008 3 169 81 18682239 23 Mugford JW Sipilä P McMahon JA McMahon AP Osr1 expression demarcates a multi-potent population of intermediate mesoderm that undergoes progressive restriction to an Osr1-dependent nephron progenitor compartment within the mammalian kidney Dev. Biol 2008 324 88 98 18835385 24 Gadue P Huber TL Paddison PJ Keller GM Wnt and TGF-beta signaling are required for the induction of an in vitro model of primitive streak formation using embryonic stem cells Proc. Natl. Acad. Sci 2006 103 16806 16811 17077151 25 Tam PPL Loebel DAF Gene function in mouse embryogenesis: get set for gastrulation Nat. Rev. Genet 2007 8 368 81 17387317 26 Sumi T Tsuneyoshi N Nakatsuji N Suemori H Defining early lineage specification of human embryonic stem cells by the orchestrated balance of canonical Wnt/ -catenin, Activin/Nodal and BMP signaling Development 2008 135 2969 2979 18667462 27 James RG Schultheiss TM Bmp signaling promotes intermediate mesoderm gene expression in a dose-dependent, cell-autonomous and translation-dependent manner Dev. Biol 2005 288 113 125 16243309 28 Colvin JS Feldman B Nadeau JH Goldfarb M Ornitz DM Genomic organization and embryonic expression of the mouse fibroblast growth factor 9 gene Dev. Dyn 1999 216 72 88 10474167 29 Davies JA Unbekandt M Ineson J Lusis M Little MH Dissociation of embryonic kidney followed by re-aggregation as a method for chimeric analysis Methods Mol. Biol 2012 886 135 46 22639257 30 Xia Y Directed differentiation of human pluripotent cells to ureteric bud kidney progenitor-like cells Nat. Cell Biol 2013 15 1507 15 24240476 31 Lam AQ Rapid and Efficient Differentiation of Human Pluripotent Stem Cells into Intermediate Mesoderm That Forms Tubules Expressing Kidney Proximal Tubular Markers J. Am. Soc. Nephrol 2014 25 1211 1225 24357672 32 Kang M Han Y Differentiation of Human Pluripotent Stem Cells into Nephron Progenitor Cells in a Serum and Feeder Free System PLoS One 2014 9 e94888 24728509 33 Toyohara T Cell Therapy Using Human Induced Pluripotent Stem Cell-Derived Renal Progenitors Ameliorates Acute Kidney Injury in Mice Stem Cells Transl. Med 2015 4 980 992 26198166 34 Taguchi A Redefining the in vivo origin of metanephric nephron progenitors enables generation of complex kidney structures from pluripotent stem cells Cell Stem Cell 2014 14 53 67 24332837 35 Sharmin S Human Induced Pluripotent Stem Cell-Derived Podocytes Mature into Vascularized Glomeruli upon Experimental Transplantation J. Am. Soc. Nephrol 2015 1 14 10.1681/ASN.2015010096 36 Morizane R Nephron organoids derived from human pluripotent stem cells model kidney development and injury Nat. Biotechnol 2015 33 1193 1200 26458176 37 Freedman BS Modelling kidney disease with CRISPR-mutant kidney organoids derived from human pluripotent epiblast spheroids Nat. Commun 2015 6 8715 26493500 38 Takasato M Little MH The origin of the mammalian kidney: implications for recreating the kidney in vitro Development 2015 142 1937 1947 26015537 39 Ciarimboli G Cisplatin Nephrotoxicity Is Critically Mediated via the Human Organic Cation Transporter 2 Am. J. Pathol 2005 167 1477 1484 16314463 40 Ishida S Lee J Thiele DJ Herskowitz I Uptake of the anticancer drug cisplatin mediated by the copper transporter Ctr1 in yeast and mammals Proc. Natl. Acad. Sci. U. S. A 2002 99 14298 14302 12370430 41 Hildebrandt F Genetic kidney diseases Lancet 2010 375 1287 1295 20382325 42 Cho EA Differential expression and function of cadherin-6 during renal epithelium development Development 1998 125 803 812 9449663 43 Combes AN Davies JA Little MH Cell-cell interactions driving kidney morphogenesis Curr. Top. Dev. Biol 2015 112 467 508 25733149 44 Wieser M hTERT alone immortalizes epithelial cells of renal proximal tubules without changing their functional characteristics Am. J. Physiol. Renal Physiol 2008 295 F1365 75 18715936 45 Nouwen EJ Dauwe S van der Biest I De Broe ME Stage- and segment-specific expression of cell-adhesion molecules N-CAM, A-CAM, and L-CAM in the kidney Kidney Int 1993 44 147 58 8355456 46 Briggs JA Integration-free induced pluripotent stem cells model genetic and neural developmental features of down syndrome etiology Stem Cells 2013 31 467 78 23225669 47 Xu J Eya1 Interacts with Six2 and Myc to Regulate Expansion of the Nephron Progenitor Pool during Nephrogenesis Dev. Cell 2014 31 434 447 25458011 48 Mugford JW Sipilä P Kobayashi A Behringer RR McMahon AP Hoxd11 specifies a program of metanephric kidney development within the intermediate mesoderm of the mouse embryo Dev. Biol 2008 319 396 405 18485340 49 Osafune K Marked differences in differentiation propensity among human embryonic stem cell lines Nat. Biotechnol 2008 26 313 315 18278034 50 Ng ES Davis R Stanley EG Elefanty AG A protocol describing the use of a recombinant protein-based, animal product-free medium (APEL) for human embryonic stem cell differentiation as spin embryoid bodies Nat. Protoc 2008 3 768 76 18451785 51 Das A Stromal-epithelial crosstalk regulates kidney progenitor cell differentiation Nat. Cell Biol 2013 15 1035 44 23974041

PMC005xxxxxx/PMC5113833.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985117R 4816 J Immunol J. Immunol. Journal of immunology (Baltimore, Md. : 1950) 0022-1767 1550-6606 27798158 5113833 10.4049/jimmunol.1600836 NIHMS818721 Article Loss of neurokinin-1 receptor alters ocular surface homeostasis and promotes an early development of herpes stromal keratitis Gaddipati Subhash *§ Rao Pushpa § Jerome Andrew David § Burugula Bala Bharathi *§ Gerard Norma P ** Suvas Susmit *§¶# * Department of Ophthalmology, Wayne State University School of Medicine, Detroit, MI § Department of Anatomy & Cell Biology, Wayne State University School of Medicine, Detroit, MI ¶ Department of Immunology and Microbiology, Wayne State University School of Medicine, Detroit, MI ** Division of Respiratory Diseases, Boston’s Children hospital, Department of Medicine, Harvard Medical School, Boston, MA # Corresponding author: Dr. Susmit Suvas, 7223 Scott Hall, Departments of Ophthalmology, Anatomy & Cell Biology and Microbiology and Immunology, Wayne State University School of Medicine, Detroit, MI, Phone: 313-577-9820, Fax: 313-577-3125, ssuvas@med.wayne.edu 27 9 2016 17 10 2016 15 11 2016 15 11 2017 197 10 40214033 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Substance P neuropeptide and its receptor neurokinin-1 (NK1R) are reported to present on the ocular surface. In this study, mice lacking functional NK1R exhibited an excessive desquamation of apical corneal epithelial cells in association with an increased epithelial cell proliferation, increased epithelial cell density, but decreased epithelial cell size. The lack of NK1R also resulted in decreased density of corneal nerves, corneal epithelial dendritic cells, and a reduced volume of basal tears. Interestingly, massive accumulation of CD11c+CD11b+ conventional dendritic cells (cDCs) was noted in the bulbar conjunctiva and near the limbal area of corneas from NK1R−/− mice. After ocular HSV-1 infection, the number of cDCs and neutrophils infiltrating the infected corneas was significantly higher in NK1R−/− than C57BL/6J mice. This was associated with an increased viral load in infected corneas of NK1R−/− mice. As a result, the number of IFN-γ secreting virus specific CD4 T cells in the DLNs of NK1R−/− mice was much higher than infected C57BL/6J mice. An increased number of CD4 T cells and mature neutrophils (CD11b+Ly6ghigh) in the inflamed corneas of NK1R−/− mice was associated with an early development of severe HSK. Collectively, our results show that the altered corneal biology of uninfected NK1R−/− mice along with an enhanced immunological response after ocular HSV-1 infection cause an early development of HSK in NK1R−/− mice. Substance P inflammation cornea and herpes simplex virus-1 Introduction Neurokinin-1 receptor (NK1R) is the highest affinity receptor for substance P (SP), an eleven amino acid long neuropeptide (1, 2). NK1R is reported to express in corneal epithelial cells, and SP-NK1R interaction promotes the chemokine expression in primary cultures of human corneal epithelial cells (3, 4). Blocking of NK1R signaling, while using NK1R antagonists, is shown to ameliorate many pro-inflammatory conditions, including airway and ocular inflammation in animal models (5–8). Therefore, NK1R serves as a promising target to control chronic inflammation. Currently, NK1R antagonists are approved to prevent nausea and vomiting associated with cancer chemotherapy in clinics (9, 10). In addition to promoting inflammation, NK1R signaling is also reported to accelerate corneal epithelial wound healing in animal models (11, 12). However, the role of functional NK1R in regulating ocular surface homeostasis under steady-state condition is not known. Recurrent corneal HSV-1 infection can cause the development of herpes stromal keratitis (HSK), a chronic ocular inflammatory condition. If not controlled with steroids and anti-viral, HSK can cause the loss of vision. Mouse models have long been used to study the pathogenesis of HSK. Recently in a mouse model, we demonstrated that blocking NK1R signaling during the clinical period of the disease with spantide I (NK1R antagonist), ameliorated the severity of HSK (7). This suggested the pro-inflammatory nature of NK1R in an ongoing inflammatory condition However, it is not clear whether the lack of functional NK1R on the ocular surface under steady-state condition increases or decreases the susceptibility of eyes to develop HSK after corneal HSV-1 infection. To address the above question, we used mice lacking functional Nk1r gene (NK1R−/−) (13). Contrary to our expectation, an early development of severe HSK was noted in infected NK1R−/− mice when compared with infected C57BL/6J (B6) mice. While trying to understand the mechanism, we determined that in comparison to uninfected B6 mice, uninfected NK1R−/− mice exhibited excessive cell sloughing at the apical surface of the corneal epithelium in association with an increased epithelial cell proliferation, increased epithelial cell density, but decreased epithelial cell size. Additionally, a significant decrease in the number of resident corneal epithelial dendritic cells, but an increased accumulation of cDCs near limbal area was detected in naïve corneas of NK1R−/− mice than wild type B6 mice. Upon ocular HSV-1 infection, increased infiltration of cDCs and neutrophils was detected in the infected corneas from NK1R−/− mice. In addition, NK1R−/− mice corneas exhibited an increased viral titer at early time-points (days 2 and 4) post-infection. This was associated with an increased priming of virus specific IFN-γ secreting CD4 T cells in the DLNs of NK1R−/− mice. An increased number of CD4 T cells and mature neutrophils (CD11b+Ly6ghigh) in inflamed corneas of NK1R−/− mice was associated with an early development of severe HSK. Together, our results indicate the contribution of NK1R signaling in maintaining the homeostasis of the ocular surface under steady-state condition, and that the lack of functional NK1R increases the susceptibility of eyes to develop severe HSK upon ocular HSV-1 infection. Materials and Methods Mice and Ethics statement Eight to twelve week old male and female C57BL/6J mice were procured from the Jackson laboratory (Bar Harbor, ME). Breeding pairs of NK1R−/− mice were obtained from Dr. Norma P Gerard, and the mice were bred in an animal facility at Wayne State University School of Medicine (WSUSOM). Functional ablation of the NK1R gene in NK1R−/− mice was confirmed using tail biopsy and performing the PCR followed by electrophoresis (Supplementary Figure 1). Eight to twelve week old male and female B6 and NK1R−/− mice were used to carry out the experiments. All of the animals were housed in an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) approved specific pathogen free animal facility at Wayne State University School of Medicine (WSUSOM). The Institutional Animal Care and Use Committee (IACUC) of Wayne State University approved all of the experimental protocols and procedures. In addition, all of the experimental procedures were in complete agreement with the Association for Research in Vision and Ophthalmology resolution on the use of animals in research. Corneal HSV-1 infection and viral titration A virulent HSV-1 RE Tumpey strain used in the current study was propagated on a monolayer of Vero cells (American Type Culture Collection, Manassas, VA; CCL81) as described previously (14). To carry out an ocular HSV-1 infection, mice were first anesthetized by intra-peritoneal injection of Ketamine (33mg/Kg bodyweight) + Xylazine (20mg/Kg bodyweight) in 0.2 mL PBS. The corneas were scarified with trephine (Fine Science Tools, Foster city, CA) while twisting three to four times over the corneal surface. 1×104 plaque forming unit (p.f.u) of HSV-1 virus was then topically applied with a pipette to the eye in 3μL of 1x PBS followed by a gentle massage of the eyelids. The HSV-1 load in infected corneas of C57BL/6J and NK1R−/− groups of mice was determined on day 2, 4 and 5 post-infection as described previously (14). The same infected eye was not used to measure the viral load at different days post-infection. After collecting the eye-swabs, mice were euthanized on the indicated time-points post-infection and the corneas were processed for flow cytometry analysis. Clinical Scoring of HSK The eyes were examined on different day post-infection, while using a hand-held slit lamp biomicroscope (Kowa, Nagoya, Japan), to determine the extent of corneal opacity and angiogenesis. A standard scale for corneal opacity, ranging from 0–5, was used as described earlier (14). The neovascularization (NV) of the cornea was determined by measuring the centripetal growth of newly formed blood vessels in each quadrant of the cornea as described earlier (14). Lymph nodes and corneal cell preparation for flow cytometry HSV-1 infected B6 and NK1R−/− mice were euthanized on day 7 post-infection. A single cell suspension of the draining lymph nodes (DLNs) from individual mouse was prepared as described earlier (14). The infected eyes from both groups of mice were enucleated and collected in ice-cold RPMI medium with antibiotics. Under a dissecting microscope, a radial incision was made at the limbal region, and the cornea was separated from the underlying lens, iris, ciliary body, and scleral tissue using curved fine forceps (Miltex surgical instruments, PA). The individual cornea was suspended in 250 μL of RPMI, and 20 μL of liberase TL (2.5 mg/mL) was added followed by incubation at 37°C for 45 minutes on a tissue disruptor. At the end of the incubation period, the samples were triturated using 3 mL syringe plungers and passed through a 70 μm cell strainer. This was followed by pelleting down the cells at 315× g for 8 minutes in a refrigerated centrifuge. Cell surface staining Cell surface staining was carried out as described earlier (14). Briefly, the single cell suspension obtained from the individual cornea and DLN was washed with FACS buffer, followed by blocking of Fc receptors and incubation with fluorochrome conjugated antibodies. The following antibodies were used for the cell surface staining: Percp-Cy5.5 conjugated- anti-CD4 (RM4-5), FITC conjugated anti-Gr1 (RB6-8C5), Percp-Cy5.5 conjugated anti-CD11b (M1/70), PE-Cy7 conjugated anti-Ly6g (1A8), PE conjugated IAb (AF6-120.1), and Alexa 647 conjugated anti-CD11c (N418). All of the antibodies were purchased from BD biosciences and BioLegend, San Diego, CA. At the end of the cell surface staining, samples were acquired using a LSR II flow cytometer, and the data was analyzed using the FlowJo software (Ashland, OR, USA. V8.8.7). Corneal FITC painting assay HSV-1 infected C57BL/6 and NK1R−/− mice were anesthetized with isoflurane on day 2 post-infection. A 1% FITC solution was prepared by dissolving FITC isomer 1 (Sigma F7250) in 1:1 (vol/vol) acetone: dibutyl phthalate solution. The corneas of infected mice were painted under anesthesia with 3 μl of FITC solution. After 16h, the DLNs were collected and a single cell suspension was prepared while using a collagenase type II (Gibco 17101-015) enzyme solution. Cells were stained for IAb, CD11b, and CD11c molecules and the samples were acquired on a LSRFortessa flow cytometer. The data was analyzed using the FlowJo software. Intracellular cytokine and nuclear Ki-67 staining Single cell suspension of the lymph nodes was stimulated with 3 MOI (multiplicity of infection) of UV inactivated HSV-1 as described earlier (14). At the end of incubation period, cell surface staining was carried out, followed by permeabilization of the cells with Cytofix/Cytoperm (BD biosciences). PE-conjugated anti-IFN-γ (XMG1.2) antibody was used to stain IFN-γ expressing CD4 T cells. The cells were washed in perm wash (BD Biosciences) buffer with the final washing given in FACS buffer. For nuclear Ki-67 staining, cells were stained using a PE mouse anti-human Ki-67 kit (BD Pharmingen) as per manufacturer instruction. Samples were then acquired on the LSR II flow cytometer and the data was analyzed using the FlowJo software. Mouse cytokine protein array analysis Cytokine and chemokine levels in HSV-1 infected corneal tissues from B6 and NK1R−/− mice were measured using a Mouse Cytokine Array Panel A kit (R&D systems, Minneapolis, MN). Corneas dissected from naïve or infected eyes (day 4 and day 9 post-infection) of B6 and NK1R−/− mice were transferred to 1x PBS with protease inhibitor cocktail (PIC) and kept at −80°C. Sonication of the corneas was performed at 50% amplitude with a cycle of 15 second pulse, followed by 1 minute of resting on ice. A total of six cycles were given to each cornea. The tissue lysates (sonicated samples) obtained were centrifuged at 4°C at 15,000 rpm for 10 minutes. The supernatant was collected and the amount of protein in each sample was estimated using a BCA protein assay kit (Thermo Scientific). A total of four corneas from each group was pooled to obtain a minimum of 200 μg of protein for cytokine and chemokine array analysis as per the manufacturer instructions. Positive signals on the membranes were quantified using Image Studio Lite Version 4.0. Corneal whole mount staining Mice were euthanized and their eyes were enucleated with a pair of forceps. The eyeballs were immediately fixed in 4% paraformaldehyde for 30 min at room temperature. The corneal tissue was separated with a bard-parker #21 blade under a dissection microscope. Tissues were flattened with 6–8 partial cuts from the limbal to central cornea and stored in 30% sucrose solution overnight at 4°C. The next day, the corneas were permeabilized using 0.1% EDTA-0.01% Hyaluronidase type IV-S solution at 37°C for 90 min, and immediately followed by blocking with a 3% bovine serum albumin-1% Triton X100 solution for 90 min at room temperature. The corneal tissues were incubated with a primary antibody at an appropriate dilution in a humidified chamber (Supplementary table 1) for 16 hours in a cold room (4°C). The next day, the tissues were washed four times with 0.3% Triton X100-PBS solution (10 min each wash). If the primary antibody was unconjugated, a secondary antibody was added at an appropriate dilution as stated in Supplementary table 1 followed by overnight incubation in a cold room. Prior to acquisition, the corneal tissue was mounted with vectashield medium containing DAPI (H1200, Vector Laboratories, Burlingame, CA, USA). Images were acquired using a Leica TCS SP8 confocal microscope. An Image J version 1.49b software was used to quantitate the CD11c+ cells in the corneal wholemount from both groups of mice. The Image J “multi-point” tool was used for manual counting of the corneal nerve leash and limbal nerve trunk branch points. Cochet and Bonnet Esthesiometry Corneal sensitivity was measured using Cochet and Bonnet Esthesiometer attached with 12/100 mm nylon thread. This test was performed at 23–25°C room temperature by touching the central cornea of B6 or NK1R−/− mice. Numerical values were derived from a conversion table provided by the manufacturer (Luneau SAS). Phenol red thread tear test To measure the tear levels in the eyes of unmanipulated B6 and NK1R−/− mice, pH indicator phenol red treated cotton threads were obtained from Zone-Quick (ZQ-1010, Showa Yakuhin Kako, Japan). The mice were restrained by holding the lose skin at base of the neck without touching the facial skin or whiskers. The eyes were tested for tear level by placing the tip of the cotton thread on the bulbar conjunctiva of the lateral canthus for 30 seconds. The picture of the threads was taken and tear absorption was measured using a 10 cm scale. Bromodeoxyuridine (BrdU) incorporation assay For BrdU labeling of the corneal epithelium, 8 to 12-week old B6 and NK1R−/− mice were given a single intra-peritoneal injection of BrdU (10 mg BrdU/mL saline; 0.2 mL/mouse). The mice were euthanized on the next day after 18 hours, and their eyes were collected to snap freeze in OCT media. Longitudinal sections were cut at an 8μm thickness, and mounted on positively charged slides (Thermo Shandon Limited, UK). The sections were air dried overnight at room temperature and fixed in 4% buffered paraformaldehyde for 15 min followed by 3 washings with 1x PBS. For antigen retrieval, sections were immersed in 2N HCl in a covered Coplin jar for 30 minutes at room temperature. The acid was neutralized with 0.1M Borate buffer (pH-9.0) for 5 minutes with two buffer changes. The slides were blocked with 3% BSA in 1X PBS and incubated with a primary antibody (Abcam, MA, USA, BU1/75, ab6326) at a 1:100 dilutions in blocking buffer for 18 hrs at 4°C. The next day, the slides were washed three times and incubated with Alexa-488 conjugated anti-Rat antibody (Molecular Probes) at a 1:200 dilutions overnight at 4°C. The slides were washed and mounted with vectashield medium containing DAPI prior to acquisition. Statistical Analysis Statistical analysis was carried out using the Graph Pad Prism software (San Diego, CA). All p values were calculated using an unpaired, two-tailed Student’s t test. P < 0.05 was considered statistically significant. The results are displayed as the mean. Results Ocular HSV-1 infection causes an early development of corneal opacity and angiogenesis in NK1R−/− mice Previously, we determined that blocking NK1R signaling in HSV-1 infected inflamed corneas, during clinical period of disease, ameliorated the severity of HSK in a B6 mouse model (7). In this study, we analyzed whether the lack of functional NK1R on the ocular surface, under steady-state condition, also reduces the severity of HSK upon ocular HSV-1 infection. The corneas from both groups of mice were infected with HSV-1 as described in the methods section. Unexpectedly, our results showed significantly increased corneal opacity and angiogenesis on day 9 and 11 post-infection in HSV-1 infected corneas from NK1R−/− mice in comparison to the infected group of B6 mice (Fig. 1A, C and D). The incidence of corneal opacity and angiogenesis was also much higher in NK1R−/− mice when analyzed on day 9 and 11 post-infection (Fig. 1B). By day 14 and 16 post-infection, the extent of corneal opacity and angiogenesis was similar in eyes from both groups of mice. Collectively, our results show an early development of severe HSK in NK1R−/− mice after ocular HSV-1 infection. Excessive sloughing of apical corneal epithelial (ACE) cells in unmanipulated NK1R−/− mice To determine the underlying mechanism for our unexpected results, we microscopically evaluated the corneal surface of the eyes from uninfected B6 and NK1R−/− mice. No phenotypic differences in the appearance of the eyes were noted when comparing B6 and NK1R−/− mice (Fig. 2A). However, hematoxylin and eosin staining of paraffin-embedded corneal sections from NK1R−/− mice showed a much thinner apical layer of the corneal epithelium with occasional delaminating superficial epithelial cells in comparison to the corneas from B6 mice (Fig. 2B). The laser scanning confocal microscopy of whole mount corneas from B6 and NK1R−/− mice demonstrated significantly increased number of desquamating ACE cells (denoted by the * sign) in knockout mice (Fig. 2D). Excessive cell sloughing at the corneal apical surface could possibly stimulate compensatory cell proliferation in the corneal epithelium. Therefore, we next evaluated the corneal epithelial cell proliferation between B6 and NK1R−/− mice. Short-term Bromodeoxyuridine (BrdU) incorporation assay showed a significant increase in labeling of basal corneal epithelial cells in NK1R−/− compared to B6 mice (Fig. 2E). In addition, a significant increase in the number of hematoxylin stained nuclei was determined in the corneal epithelium of NK1R−/− than B6 mice (Fig. 2C). Increased epithelial cell density, but decreased epithelial cell size was also evident in the whole mount corneas from NK1R−/− mice (Fig. 2F). Together, our results demonstrated an excessive exfoliation and increased mitotic index of corneal epithelial cells in NK1R−/− mice. Loss of NK1R reduces the density of corneal epithelial DCs, but dramatically increases the number of cDCs in the bulbar conjunctiva and near the limbal area of uninfected cornea The normal corneal epithelium is populated with resident CD11c+ DCs, which are in close association with surrounding epithelial cells. The resident corneal DCs have been shown to promote the re-epithelialization of corneal wounds (15). An alteration in the homeostasis of the corneal epithelium might affect the homeostasis of corneal DCs. Therefore, we next compared the density of corneal DCs by carrying out the immunofluorescence staining of CD11c+ cells in the corneal whole mounts from unmanipulated B6 and NK1R−/− mice. As is shown in Fig. 3A, the resident corneal epithelial DCs present in the peripheral and central areas of the corneas from NK1R−/− mice were almost half in number when compared to DCs in the corneas from B6 mice. However, a more than two-fold increase in the number of CD11c+ cells stained with anti-CD11c (clone N418) antibody was evident in the bulbar conjunctiva and limbal region of the corneas from NK1R−/− mice when compared to the limbal area of the corneas from B6 mice (Fig. 3B and supplementary Figure 2). The majority of CD11c+ cells near the limbal area was also co-stained with CD11b molecule (Fig. 3C). Moreover, z-scans of the corneal wholemount near the limbal area showed that CD11b+CD11c+ cells were localized in both the corneal epithelium and anterior stroma, whereas CD11b+CD11c− macrophages were deep seated in the stroma of the corneas from both groups of mice (Supplementary movie S1). Together, our results showed an altered homeostasis of corneal resident epithelial DCs along with an increased accumulation of conventional dendritic cell type near the limbal area of the corneas from unmanipulated NK1R−/− mice. Reduced epithelial nerve density, stromal nerve trunk branching and basal level of tears in the corneas from NK1R−/− mice The corneal epithelium is a highly innervated tissue and the corneal nerves are reported in close association with corneal epithelial cells and dendritic cells (16, 17). We next ascertained whether the altered homeostasis of corneal epithelial and dendritic cells influences the density of corneal nerves. The corneal whole mount staining for class III β-tubulin, using Tuj-1 antibody, showed a significant decrease in the number of corneal subbasal nerve leashes in the peripheral region of the corneas from NK1R−/− mice (Fig. 4A). Additionally, the number of corneal stromal nerve trunk branch points near the limbal area was also significantly more reduced in NK1R−/− than B6 mice (Fig. 4A). The corneal nerve density was similar in the central corneal region when comparing both groups of mice (data not shown). Moreover, no significant differences (p>0.05) in corneal sensation were determined in the central corneal region of the eyes from both groups of mice, while using the Cochet-Bonnet Esthesiometer (Fig. 4B). Lastly, the phenol red thread tear test was carried out to evaluate the volume of unstimulated tears in the eyes from NK1R−/− and B6 mice. Our results showed a significant decrease (p<0.0001) in the volume of unstimulated tears measured in the eyes of NK1R−/− than B6 mice (Fig. 4C). Increased viral load in the corneas of NK1R−/− mice The intact apical surface of the corneal epithelium provides resistance to the establishment of HSV-1 infection in the mouse model. Our results clearly showed that in NK1R−/− mice, the apical layer of the corneal epithelium is much thinner and has more desquamating ACE cells. Therefore, viral load of HSV-1 on the corneal surface was compared at different time-points post-infection between B6 and NK1R−/− mice as described in the methods. Our results showed a moderate, but statistically significant increase in viral load in the corneas of NK1R−/− mice when compared with B6 mice on day 2 and 4 post-infection (Figure 5). No significant difference in viral load was determined between both groups of mice on day 5 post-infection. Pre-clinical and clinical phase of HSK document a differential level of cytokine/chemokines in infected corneas of NK1R−/− and B6 mice Development of HSK in a mouse model is categorized into a pre-clinical and a clinical phase of disease as reported earlier (18). During the pre-clinical phase (day 1 through 6 post-infection), active viral replication is reported with minimal corneal opacity and angiogenesis whereas, in the clinical stage of the disease (day 7 though 18 post-infection) severe corneal opacity and angiogenesis develop in response to chronic immunoinflammatory reactions in the inflamed corneas. We ascertained whether the presence or absence of functional NK1R affect the level of cytokines/chemokines in HSV-1 infected corneas during the pre-clinical and clinical phase of HSK. Infected corneas obtained from both groups of mice on day 4 (pre-clinical) and day 9 (clinical phase) post-infection were sonicated as described in the method section. Uninfected corneas from both groups of mice were also processed to determine the basal level of cytokines/chemokines. The amount of cytokines in the corneal lysates from both groups of mice was measured using a Mouse Cytokine Array assay kit as per the manufacturer’s instructions. Our results demonstrated that infected corneal lysates from NK1R−/− mice exhibited a marked increase in the amounts of neutrophil attracting chemokines CXCL1 (KC), CXCL2 (MIP-2) and CCL2 (MCP-1) in comparison to infected B6 mice, when measured on day 4 post-infection (Fig. 6). We also noted an increased amount of IL-1 receptor antagonist (IL-1Ra) in infected corneas of NK1R−/− than B6 mice on day 4 post-infection. On the other hand, during the clinical phase of HSK (day 9 post-infection), reduced levels of a number of cytokines/chemokines including the tissue inhibitor of metalloproteinase (TIMP-1), CCL2, and interleukin-1 receptor antagonist (IL-1ra) was measured in the corneal lysates of NK1R−/− than B6 mice (Fig. 6). No difference in the amounts of TIMP-1, CCL2, and IL-1ra protein was seen in uninfected corneal lysates from both groups of mice. Increased number of cDCs and neutrophils in infected corneas of NK1R−/− mice at an early time point after corneal HSV-1 infection In light of our observation demonstrating the massive accumulation of cDCs near the corneal limbal area of unmanipulated NK1R−/− mice (Fig. 3), we investigated the influx of cDCs in the peripheral, paracentral, and central region of HSV-1 infected corneas of B6 and NK1R−/− mice at an early time-point post-infection. As is shown in the corneal whole mount staining of CD11c+ cells, NK1R−/− mice exhibited an increased influx of CD11c+ cells in all three regions of infected corneas as stated above on day 2 post-infection (Fig. 7A). Moreover, the flow cytometry analysis of HSV-1 infected corneas on day 2 post-infection showed that almost all of CD11c+ cells co-expressed CD11b molecule, suggesting their cDCs phenotype (Fig. 7B). Our results also showed an increased influx of neutrophils in infected corneas of NK1R−/− than B6 mice on day 2 and 4 post-infection (Fig. 7C). Collectively, an increased immune cell influx was detected in infected corneas of NK1R−/− mice at early time-points post-infection. Increased numbers of CD11c+ dendritic cells observed in the DLNs of infected NK1R−/− mice did not migrate from the infected corneas The infiltrating DCs can pick up the viral antigens and migrate to the draining lymph nodes (DLNs) to prime virus specific T cells (19). Therefore, we next ascertained the number of CD11c+ cells in DLNs of HSV-1 infected B6 and NK1R−/− mice at early time-points post-infection. Our results showed a significant increase in the absolute number of CD11c+ cells in the DLNs of NK1R−/− than B6 mice on day 3 post ocular infection (Fig. 8A). No significant difference in the basal number of CD11c+ DCs was determined in the DLNs from both groups of uninfected mice. To address whether increased numbers of CD11c+ dendritic cells measured in the DLNs of infected NK1R−/− mice are the result of more CD11c+ cells migrating from the infected cornea to the DLN, we carried out FITC-painting of the infected corneas from both groups of mice. FITC painting of virus infected corneas was carried out on day 2 post-infection, and 16h later the number of IA-b highCD11C+CD11b+FITC+ cells were quantitated in the DLNs isolated from both groups of mice (Fig. 8B). Unexpectedly, our results showed a lesser number of FITC+CD11b+CD11c+ cells in the DLNs of NK1R−/− than C57BL/6 mice, when measured on day 3 post-infection. Taken together, our results showed that the increased number of CD11c+ DCs noted in the DLNs of infected NK1R−/− mice is not the outcome of an increased migration of cDCs from the infected corneas to DLNs. Increased expansion of CD4 T cells and IFN-γ secreting virus specific CD4 T cells in NK1R−/− mice Our results showed an increased viral load in infected corneas of NK1R−/− mice. Therefore, we next ascertained whether this results in an increased priming of CD4 T cells in the DLNs of NK1R−/− mice. The proliferation of CD4 T cells in the DLNs of infected B6 and NK1R−/− mice was compared by measuring the expression of Ki-67 nuclear antigen on day 3 and 5 post-infection using a flow cytometer. A significant increase in the proliferation of CD4 T cells in the DLNs of NK1R−/− mice was noted on day 3, but not on day 5 post-infection (Fig. 9A). To address HSV-1 antigen-specificity of CD4 T cells in the DLNs of infected B6 and NK1R−/− mice, a single cell suspension of the DLNs from both groups of infected mice was prepared on day 7 post-infection. As described in the methods section, the single cell suspension from both groups of mice was stimulated with UV-inactivated HSV-1 followed by intracellular cytokine staining to determine the number of IFN-γ secreting CD4 T cells. As is shown in Fig. 9B, about a two-fold increase in the number of IFN-γ secreting virus-specific CD4 T cell was detected in the DLNs of infected NK1R−/− mice, when compared to infected B6 mice. Increased influx of CD4 T cells and neutrophils into the inflamed corneas of infected NK1R−/− mice CD4 T cells and neutrophils play a pivotal role in the development of HSK (20, 21). Therefore, we next ascertained the influx of CD4 T cells in inflamed corneas of HSV-1 infected NK1R−/− and B6 mice. Single cell suspensions of individual corneas obtained from B6 and NK1R−/− mice on day 11 and 17 post-infection were stained for CD4 T cells, and the samples were acquired on a BD flow cytometer. Our results showed an increased frequency and absolute number of CD4 T cells in infected corneas of NK1R−/− mice on day 11 and 17 post-infection (Fig. 8). Similarly, an increased frequency and absolute number of CD11bhighLy6Ghigh mature neutrophils were seen in inflamed corneas of NK1R−/− mice on day 11 post-infection (Fig. 10). Discussion Our results clearly demonstrate an altered homeostasis of naive uninfected corneas in NK1R−/− mice. This involves lesser tear volume, reduced nerve density, reduced numbers of resident corneal epithelial DCs, increased accumulation of cDCs near the corneal limbal area, and an excessive exfoliation of apical corneal epithelial cells. As a result of this, the corneas of NK1R−/− mice are more susceptible to HSV-1 infection, as is evident from an increased viral load measured in the infected corneas at early time-points post-infection. An increased viral load was associated with the generation of an increased number of viral antigen specific CD4 T cells (as measured with intracellular cytokine assay) in NK1R−/− mice. Collectively, an altered corneal biology of uninfected NK1R−/− mice along with an enhanced immunological response after ocular HSV-1 infection cause an early development of HSK in NK1R−/− mice as depicted in the schematic of Figure 11. Our results showed an excessive exfoliation of apical corneal epithelial (ACE) cells in uninfected NK1R−/− mice. The exfoliation of ACE cells is regulated by tight junction proteins and adherens junction proteins. The ACE cells at the apical surface form tight junctions (TJs) and serve as a barrier to protect the ocular surface from microbial infections (22, 23). In addition, the adherens junction (AJ) proteins in the corneal epithelium are involved in stabilization of cell-cell adhesion (24). Thus, degradation, improper localization or reduced expression of membrane bound tight junction or adherens junction proteins in the corneal epithelium may increase the rate of exfoliation of ACE cells. NK1R signaling has been shown to enhance the expression of E-cadherin adhesion and ZO-1 tight junction protein in cultured human corneal epithelial cells (25, 26). This suggest that the lack of functional NK1R could cause a reduced expression or an improper localization of E-cadherin AJ and ZO-1 tight junction proteins in the corneal epithelium resulting in an excessive exfoliation of ACE cells, as determined in uninfected NK1R−/− mice. To maintain corneal epithelium homeostasis, excessive loss of ACE cells should be compensated by hyper-proliferation of basal epithelial cells, as determined in the corneas from NK1R−/− mice (Fig. 2). Additionally, rapid exfoliation of ACE cells may also impact the innervation of corneal nerves, as the latter has been reported to form complex structures with ACE cells (16). Due to the excessive loss of ACE cells, the ends of nerve fibers innervating the corneal epithelium may not make an intimate association with superficial corneal epithelial cells, and thereby retract from the corneal epithelium. Furthermore, the retraction of nerves fibers from the selective places in the corneal epithelium may also reduce the branching of stromal nerve trunks as depicted in our results. Afferent sensory nerves in the cornea are reported to play an important role in regulating tear secretion from the lacrimal gland through a neural reflex arc, which originates from the ocular surface (27). The reduced innervation of the corneas, as determined in NK1R−/− mice, might affect the neural reflex arc, and thereby reduce the secretion from the lacrimal gland, resulting in a lesser volume of basal tears as measured in the eyes of NK1R−/− mice. In addition, excessive exfoliation of ACE cells may also compromise tear film retention on the corneal surface, and promote corneal desiccation. This may also increase the severity of HSK upon ocular HSV-1 infection. In fact, a strong correlation between corneal desiccation, decreased density of corneal nerves, and pathogenesis of HSK are reported in human and mouse models of HSK (28–30). It is widely accepted that the corneal epithelium is populated with CD11c+ dendritic cells with long dendritic processes, which cover a significant area of the non-inflamed cornea (31). These DCs and their dendritic processes are in close association with corneal epithelial cells (15). However, the nature of the interaction between epithelial cells and resident DCs under steady state condition in corneal epithelium is not clear. In normal skin, E-cadherin adhesion protein mediate the interaction between epidermal keratinocytes and Langerhans cells (LCs), and this interaction is required for the retention of LCs in the skin (32). An aberrant expression of E-cadherin in human papillomavirus infected cervical tissue was associated with a significant reduction in Langerhans cells (33). Considering the role of SP-NK1R interaction in regulating the expression of E-cadherin in corneal epithelial cells (26), the lack of functional NK1R on the ocular surface may cause an aberrant expression of E-cadherin in the corneal epithelium, resulting in a decrease in the number of corneal epithelial DCs as noted in NK1R−/− mice. Moreover, signaling through NK1R has also been shown to promote the survival of bone-marrow derived DCs (34). Thus, the lack of NK1R on the ocular surface may result in the apoptosis of corneal epithelial resident DCs, and thereby decrease their number as determined in corneas of NK1R−/− mice. The massive accumulation of cDCs noted in the bulbar conjunctiva and near the corneal limbal area of NK1R−/− mice could be due to persistent low-grade inflammation in the conjunctival tissue of these mice. In support, we observed an increased number of Gr1+ and NKp46+ cells near the corneal limbal area of uninfected NK1R−/− than B6 mice (data not shown). Moreover, an increased vasodilation of feeder vessels was detected near the limbal area of NK1R−/− mice when compared with B6 mouse normal corneas (Supplementary Figure 3). Vasodilation of blood vessels along with an enhanced expression of selectins on vascular endothelial cells are reported to cause an increased infiltration of leukocytes from the peripheral blood into the skin tissue (35). At present, the possible cause of an increased vasodilation, as depicted in the feeder blood vessels near the corneal limbal area of NK1R−/− mice, is not clear. However, we suspect that it may be due to conjunctival inflammation in NK1R−/− mice. We are currently investigating this possibility. The severity of HSK in a mouse model is largely determined by viral load and the number of effector CD4 T cells and neutrophils in infected corneas. Increased viral load noted in NK1R−/− mice is possibly the outcome of two events. First, the easiness in establishing the primary HSV-1 infection, as the apical surface of NK1R−/− mice is disrupted due to excessive exfoliation of ACE cells. An increased viral load was associated with an increased influx of neutrophils as shown in our results on day 2 and day 4 post-infection (Figure 7). A recent study showed that infiltrating neutrophils in HSV-1 infected corneas can facilitate HSV-1 replication and dissemination (36). Thus, increased neutrophils in NK1R−/− mice may play a role in increasing the viral load at early time-points post-infection. The possible cause of an increased influx of neutrophils is the higher amounts of neutrophil attracting chemokines (CCL-2, CXCL1 and CXCL2) measured in infected corneas of NK1R−/− mice. An increased viral load may cause an increased generation of effector CD4 T cells as noted in the DLNs of NK1R−/− mice. A recent study showed that CD4 T cell priming in DLNs of HSV-1 infected mice could be due to the drainage of soluble viral antigens from infected corneas or the migration of cornea-infiltrating DCs loaded with viral antigens to the DLNs (19). Our results showed an increased number of cDCs in infected corneas and the DLNs of NK1R−/− than B6 mice. However, the results obtained from the corneal FITC-painting assay, to determine DC trafficking to DLNs, showed a lesser number of corneal cDCs migrating to the DLNs in NK1R−/− mice. DC migration is dependent upon their maturation (37), and NK1R signaling is reported to promote the maturation of DCs (38). The lack of functional NK1R on DCs in NK1R−/− mice may affect their maturation, and thereby reduce their migration from infected corneas to DLNs. Thus, the increased priming of CD4 T cells detected in DLNs of NK1R−/− mice is possibly the outcome of the drainage of soluble viral antigens from infected corneas. Together, our findings characterized a novel role of NK1R in maintaining the homeostasis of the ocular surface under a steady state condition. Moreover, the lack of NK1R increases the susceptibility of the eyes to ocular HSV-1 infection resulting in an increased incidence and early development of HSK. Several NK1R antagonists are in clinical trials for non-ocular conditions, including alcoholism and psychiatric conditions. It would be of interest to determine whether long-term treatment of NK1R antagonists, given for non-ocular conditions, increases the incidence of HSK or other microbe-induced ocular pathologies. Supplementary Material 1 We would like to thank Drs. Frank Giblin and Shravan Chintala for providing access to Zeiss Axio Imager Z2 fluorescence microscope facility. Many thanks to Ronald P. Barrett for help with SP8 confocal microscope. Funding Sources Supported by National Eye Institute Grant EY022417 awarded to Dr. Suvas, Core vision grant EY004068 awarded to Dr. Hazlett and Research to Prevent Blindness (RPB) grant awarded to Dr. Juzych. Figure 1 Lack of NK1R promotes an early development of HSK (A) Representative slit lamp pictures of B6, NK1R−/− mice eyes on day 11 post-ocular HSV-1 infection. (B) Frequency of eyes with severe opacity in both groups of mice. (C and D) Scatter plots showing corneal opacity and angiogenesis grading on day 9, 11, 14 and 16-post ocular infection. Data shown is the representative of three independent experiments (n = 5–10 mice per group). p values were calculated using two-tail student’s t-test (****p<0.0001, **p<0.01, *p<0.05 significant and ns p>0.05 non-significant). Figure 2 Lack of NK1R causes an excessive sloughing of apical corneal epithelial cells (A) Slit lamp bio-microscope images of naïve WT (top) and NK1R−/− (lower) showing corneal surface. (B) Representative H & E stained corneal tissue sections (8μ thick) from both groups at 20X magnification, boxed area shows the higher magnified view of that region. Apical corneal epithelial cell sloughing from the ocular surface of NK1R−/− mice indicated by arrow heads. Scale bar-60 μm. (C) Graphical representation for the number of epithelial cells quantified from H&E stained corneal sections of B6 and NK1R−/− mice. (D) Whole mount naïve corneas stained for ZO-1 (green) in B6 and NK1R−/− mice. Asterisks and scatter plot denote empty space resulting from the desquamation of ACE cells in outermost layer of the corneal epithelium. Scale bar-50 μm. (E) Representative corneal sections stained with anti BrdU Ab after 18 hours of BrdU pulsing in WT and NK1R−/− mice (BrdU positive cells indicated by arrows). Scale bar-50 μm. (F) Whole mount naïve corneas stained with E-cadherin (green) and counterstained with DAPI to demonstrate an increased epithelial cell density but decreased epithelial cell size at wing cell zone in NK1R−/− mouse cornea. Scale bar-25 μm. Images are representative of two independent experiments (n = 3–6 corneas per group). p values were calculated using two-tail student’s t-test (****p<0.0001, **p<0.01 significant). Figure 3 Lack of NK1R changes the homeostasis of corneal epithelial and limbal dendritic cells (A) Whole mount (montage images with 10X) corneas of naïve B6 and NK1R−/− mice stained with FITC conjugated anti-CD11c antibody (HL3 Clone). Scatter plot shows the total number of CD11c positive cells/each cornea from both the groups. (B) Low magnified limbal region of WT and NK1R−/− whole mount corneas stained with FITC conjugated anti-CD11c antibody (N418 clone, 5X). Dotted line denotes the limbal region. Scale bar-200 μm. Scatter plot shows the total number of CD11c positive cells at limbus/cornea from both the groups. (C) Co-localization of CD11c (N418) (Green) and CD11b (Red) in the limbal region of NK1R−/− mice corneal whole mounts. Nucleus was stained with DAPI (Blue) (20X). Scale bar-50 μm. Data shown is the representative of three independent experiments (n = 3–4 corneas per group). Cj-Conjunctiva, C Cornea-Central cornea. p values were calculated using two-tail student’s t-test. (**p<0.01 and *p<0.05 significant). Figure 4 Lack of NK1R reduces the corneal nerve density and basal levels of unstimulated tears (A) Confocal images of Tuj1 stained whole mount corneas from uninfected B6 and NK1R−/− mice. Images are montages images acquired with 20X objective lens. Scatter plots denote number of nerve leashes (bunches) and limbal stromal nerve trunk branch points from both groups of mice (n = 4 corneas per group). (B) Measurement of corneal sensitivity with Cochet and Bonnet Esthesiometer. (C) Image of phenol red threads used for measuring the tear quantity in B6 (top) and NK1R−/− (bottom) mice. The extent of color change from the thread edge denotes the abundance of tears. Scatter plot represents the volume (in mm) of tear absorption by phenol red coated cotton thread from both the groups. Data shown is the representative of two independent experiments (n= 10 corneas per group). p values were calculated using two-tail student’s t-test. (****p<0.0001, **p<0.01 and ns p>0.05). Figure 5 Increased viral load in corneas of NK1R−/− mice The viral load in individual corneas from both groups of mice was measured as plaque forming unit (p.f.u.) on day 2, 4, and 5 post-ocular infections. Each circle represents an individual eye from HSV-1 infected mice. Data shown is the sum of two independent experiments (n = 8–10 corneas per group). p values were calculated using two-tail student’s t-test. (*p<0.05 significant and p>0.05 non-significant). Figure 6 Differential amount of cytokine and chemokines in infected corneas of NK1R−/− and B6 mice during pre-clinical and clinical phase of HSK The cytokine and chemokine array blots of the corneal lysates prepared from uninfected, and HSV-1 infected corneas of B6 and NK1R−/− mice on day 4, and day 9 post-ocular infection. Bar diagram denote quantitation of the molecules involved in regulating the pathogenesis of HSK. A mouse cytokine array layout is provided to read the protein array blot images. Four corneal samples were pooled to carry out the assay. n = 8 corneas per group. Figure 7 Increased influx of cDCs and mature neutrophils in HSV-1 infected corneas of NK1R−/− than B6 mice (A) Corneal whole mounts stained with FITC conjugated anti-CD11c (clone N418) antibody from B6 and NK1R−/− mice on day 2 post-ocular HSV-1 infection. More magnified limbal, paracentral and central corneal images for both groups are included (n = 3–4 corneas per group). (B) Representative FACS plots denote that majority of infiltrating CD11c+ cells co-expressed CD11b molecule (n = 5 corneas per group). Graphs show frequency and absolute number of CD11b+CD11c+ cells in infected corneas on day 2 post-infection. (C) Representative FACS plots showing influx of leukocytes (CD45+) and neutrophils (CD11b+Ly6G+) in HSV-1 infected corneas on day 2 and 4 post-infection from both groups of mice. Graphs demonstrate the frequency and absolute number of leukocytes and neutrophils in infected corneas (n = 6 corneas per group). Data shown is the representative of two independent experiments. p values were calculated using two-tail student’s t-test. (*p<0.05, **p<0.01, ***p<0.001 significant and p>0.05 non-significant). Figure 8 Increased number of CD11c+ dendritic cells in the DLNs of HSV-1 infected NK1R−/− mice did not migrate from the infected corneas (A) FACS plots demonstrating the frequencies of CD11c+Gr1-ve DCs in naïve DLNs, and the DLNs obtained on day 3 post-infection, from both groups of mice. Bar diagrams denote the absolute number of DCs (n = 4–5 mice per group). (B) FACS plots showing the frequency of FITC+ DCs in the DLNs of infected mice, from both groups, on day 3 post-infection. The infected corneas of B6 and NK1R−/− mice were FITC-painted on day 2 post-infection and 16h later, the DLNs were removed. Bar diagrams show frequency and absolute number of FITC+ DCs in the DLNs of B6 and NK1R−/− mice on day 3 post-infection (n = 6 mice per group). Data shown is the representative of two independent experiments. (*p<0.05, **p<0.01, ****p<0.0001 significant and p>0.05 non-significant). Figure 9 Increased expansion and Th1 differentiation of CD4 T cells in DLNs of HSV-1 infected NK1R−/− mice (A) Representative FACS plots are showing the frequencies of Ki67 expressing CD4 T cells in the DLN of B6 and NK1R−/− mice on day 3 post-ocular HSV-1 infection. Scatter plots show frequency and absolute number of CD4+ Ki67+ cells in DLNs of uninfected and HSV-1 infected mice from both groups on day 3 and 5 post-infection (n = 3–6 mice group). (B) Representative FACS plots are denoting IFN-γ producing CD4 T cells after ex-vivo re-stimulation of lymph node cells from both groups of mice with UV inactivated HSV-1 on Day 7 post-infection. Scatter plots demonstrate frequency and absolute number of IFN-γ producing CD4 T cells from both groups of mice (n = 9–10 mice per group). p values were calculated using two-tail student’s t-test. (****<0.0001, *p<0.05 significant and p>0.05 non-significant) Figure 10 Increased influx of CD4 T cells and mature neutrophils (CD11bhighGr1high) in infected corneas of NK1R−/− mice (A) Representative FACS plots of CD4 T cells in infected corneas from both groups of mice on day 11 and 17 post-infection. Bar diagrams demonstrate the frequencies and absolute number of CD4 T cells in individual corneal tissue from both groups on day 11 and day 17 post-infection. Data shown is the representative of two independent experiments (n = 5–7 corneas per group). (B) Representative FACS plots of CD11bhighGr1high granulocytes in inflamed corneas on day 17 post-infection from both groups of mice. Bar diagrams denote frequency and absolute number of CD11bhigh cells in infected corneas from both the groups of mice on day 17 post-infection. (n = 7 corneas per group). p values were calculated using two-tail student’s t-test. (***p<0.001, **p<0.01, and *p<0.05 significant). Figure 11 Schematic of events in uninfected and HSV-1 infected corneas of NK1R−/− mice to cause an early development of HSK Cells in the apical layer of the corneal epithelium of uninfected NK1R−/− mice slough off prematurely due to an abnormal localization or expression of junctional complex proteins. This leads to an increased basal epithelial mitotic index, reduced corneal nerve density, and reduced volume of tears. Lack of NK1R survival signal is the most likely cause of the reduced number of resident corneal CD11c+ cells noted in NK1R−/− mice. Increased CD11b+CD11c+ cells found near the limbal area of uninfected NK1R−/− mice is possibly the outcome of low-grade inflammation developed due to lesser volume of tears. Altered ocular surface in NK1R−/− mice ease the establishment of corneal HSV-1 infection, and an increased influx of neutrophils in infected cornea further facilitate HSV-1 replication. Increased amount of viral antigens drained to the DLNs cause an increased expansion of HSV-1 specific effector CD4 T cells, which migrate to the infected cornea to cause an extensive tissue damage. Together, an altered ocular surface along with an enhanced immunological response after ocular HSV-1 infection cause an early development of severe HSK in NK1R−/− mice. 1 Steinhoff MS von Mentzer B Geppetti P Pothoulakis C Bunnett NW 2014 Tachykinins and their receptors: contributions to physiological control and the mechanisms of disease Physiol Rev 94 265 301 24382888 2 Douglas SD Leeman SE 2011 Neurokinin-1 receptor: functional significance in the immune system in reference to selected infections and inflammation Ann N Y Acad Sci 1217 83 95 21091716 3 Tran MT Lausch RN Oakes JE 2000 Substance P differentially stimulates IL-8 synthesis in human corneal epithelial cells Invest Ophthalmol Vis Sci 41 3871 3877 11053288 4 Watanabe M Nakayasu K Iwatsu M Kanai A 2002 Endogenous substance P in corneal epithelial cells and keratocytes Jpn J Ophthalmol 46 616 620 12543186 5 Bignami F Giacomini C Lorusso A Aramini A Rama P Ferrari G 2014 NK1 receptor antagonists as a new treatment for corneal neovascularization Invest Ophthalmol Vis Sci 55 6783 6794 25228541 6 Hazlett LD McClellan SA Barrett RP Liu J Zhang Y Lighvani S 2007 Spantide I decreases type I cytokines, enhances IL-10, and reduces corneal perforation in susceptible mice after Pseudomonas aeruginosa infection Invest Ophthalmol Vis Sci 48 797 807 17251480 7 Twardy BS Channappanavar R Suvas S 2011 Substance P in the corneal stroma regulates the severity of herpetic stromal keratitis lesions Invest Ophthalmol Vis Sci 52 8604 8613 21969295 8 Frode-Saleh TS Calixto JB Medeiros YS 1999 Analysis of the inflammatory response induced by substance P in the mouse pleural cavity Peptides 20 259 265 10422882 9 dos Santos LV Souza FH Brunetto AT Sasse AD da Silveira Nogueira Lima JP 2012 Neurokinin-1 receptor antagonists for chemotherapy-induced nausea and vomiting: a systematic review J Natl Cancer Inst 104 1280 1292 22911671 10 Navari RM Reinhardt RR Gralla RJ Kris MG Hesketh PJ Khojasteh A Kindler H Grote TH Pendergrass K Grunberg SM Carides AD Gertz BJ 1999 Reduction of cisplatin-induced emesis by a selective neurokinin-1-receptor antagonist. L-754,030 Antiemetic Trials Group N Engl J Med 340 190 195 9917226 11 Nakamura M Kawahara M Nakata K Nishida T 2003 Restoration of corneal epithelial barrier function and wound healing by substance P and IGF-1 in rats with capsaicin-induced neurotrophic keratopathy Invest Ophthalmol Vis Sci 44 2937 2940 12824234 12 Yang L Di G Qi X Qu M Wang Y Duan H Danielson P Xie L Zhou Q 2014 Substance P promotes diabetic corneal epithelial wound healing through molecular mechanisms mediated via the neurokinin-1 receptor Diabetes 63 4262 4274 25008176 13 Bozic CR Lu B Hopken UE Gerard C Gerard NP 1996 Neurogenic amplification of immune complex inflammation Science 273 1722 1725 8781237 14 Gaddipati S Estrada K Rao P Jerome AD Suvas S 2015 IL-2/anti-IL-2 antibody complex treatment inhibits the development but not the progression of herpetic stromal keratitis J Immunol 194 273 282 25411200 15 Gao N Yin J Yoon GS Mi QS Yu FS 2011 Dendritic cell-epithelium interplay is a determinant factor for corneal epithelial wound repair Am J Pathol 179 2243 2253 21924232 16 Kubilus JK Linsenmayer TF 2010 Developmental corneal innervation: interactions between nerves and specialized apical corneal epithelial cells Invest Ophthalmol Vis Sci 51 782 789 19741242 17 Leppin K Behrendt AK Reichard M Stachs O Guthoff RF Baltrusch S Eule JC Vollmar B 2014 Diabetes mellitus leads to accumulation of dendritic cells and nerve fiber damage of the subbasal nerve plexus in the cornea Invest Ophthalmol Vis Sci 55 3603 3615 24781935 18 Biswas PS Rouse BT 2005 Early events in HSV keratitis--setting the stage for a blinding disease Microbes Infect 7 799 810 15857807 19 Buela KA Hendricks RL 2015 Cornea-infiltrating and lymph node dendritic cells contribute to CD4+ T cell expansion after herpes simplex virus-1 ocular infection J Immunol 194 379 387 25422507 20 Doymaz MZ Rouse BT 1992 Herpetic stromal keratitis: an immunopathologic disease mediated by CD4+ T lymphocytes Invest Ophthalmol Vis Sci 33 2165 2173 1351475 21 Niemialtowski MG Rouse BT 1992 Predominance of Th1 cells in ocular tissues during herpetic stromal keratitis J Immunol 149 3035 3039 1357034 22 Ban Y Dota A Cooper LJ Fullwood NJ Nakamura T Tsuzuki M Mochida C Kinoshita S 2003 Tight junction-related protein expression and distribution in human corneal epithelium Exp Eye Res 76 663 669 12742348 23 Yi X Wang Y Yu FS 2000 Corneal epithelial tight junctions and their response to lipopolysaccharide challenge Invest Ophthalmol Vis Sci 41 4093 4100 11095601 24 Hartsock A Nelson WJ 2008 Adherens and tight junctions: structure, function and connections to the actin cytoskeleton Biochim Biophys Acta 1778 660 669 17854762 25 Ko JA Yanai R Nishida T 2009 Up-regulation of ZO-1 expression and barrier function in cultured human corneal epithelial cells by substance P FEBS Lett 583 2148 2153 19446555 26 Araki-Sasaki K Aizawa S Hiramoto M Nakamura M Iwase O Nakata K Sasaki Y Mano T Handa H Tano Y 2000 Substance P-induced cadherin expression and its signal transduction in a cloned human corneal epithelial cell line J Cell Physiol 182 189 195 10623882 27 Zoukhri D 2006 Effect of inflammation on lacrimal gland function Exp Eye Res 82 885 898 16309672 28 Hamrah P Cruzat A Dastjerdi MH Zheng L Shahatit BM Bayhan HA Dana R Pavan-Langston D 2010 Corneal sensation and subbasal nerve alterations in patients with herpes simplex keratitis: an in vivo confocal microscopy study Ophthalmology 117 1930 1936 20810171 29 Chucair-Elliott AJ Zheng M Carr DJ 2015 Degeneration and regeneration of corneal nerves in response to HSV-1 infection Invest Ophthalmol Vis Sci 56 1097 1107 25587055 30 Yun H Rowe AM Lathrop KL Harvey SA Hendricks RL 2014 Reversible nerve damage and corneal pathology in murine herpes simplex stromal keratitis J Virol 88 7870 7880 24789786 31 Knickelbein JE Watkins SC McMenamin PG Hendricks RL 2009 Stratification of Antigen-presenting Cells within the Normal Cornea Ophthalmol Eye Dis 1 45 54 20431695 32 Tang A Amagai M Granger LG Stanley JR Udey MC 1993 Adhesion of epidermal Langerhans cells to keratinocytes mediated by E-cadherin Nature 361 82 85 8421498 33 Leong CM Doorbar J Nindl I Yoon HS Hibma MH 2010 Deregulation of E-cadherin by human papillomavirus is not confined to high-risk, cancer-causing types Br J Dermatol 163 1253 1263 20698848 34 Janelsins BM Mathers AR Tkacheva OA Erdos G Shufesky WJ Morelli AE Larregina AT 2009 Proinflammatory tachykinins that signal through the neurokinin 1 receptor promote survival of dendritic cells and potent cellular immunity Blood 113 3017 3026 18987361 35 Ginhoux F Collin MP Bogunovic M Abel M Leboeuf M Helft J Ochando J Kissenpfennig A Malissen B Grisotto M Snoeck H Randolph G Merad M 2007 Blood-derived dermal langerin+ dendritic cells survey the skin in the steady state J Exp Med 204 3133 3146 18086862 36 Royer DJ Zheng M Conrady CD Carr DJ 2015 Granulocytes in Ocular HSV-1 Infection: Opposing Roles of Mast Cells and Neutrophils Invest Ophthalmol Vis Sci 56 3763 3775 26066745 37 Banchereau J Steinman RM 1998 Dendritic cells and the control of immunity Nature 392 245 252 9521319 38 Janelsins BM Sumpter TL Tkacheva OA Rojas-Canales DM Erdos G Mathers AR Shufesky WJ Storkus WJ Falo LD Jr Morelli AE Larregina AT 2013 Neurokinin-1 receptor agonists bias therapeutic dendritic cells to induce type 1 immunity by licensing host dendritic cells to produce IL-12 Blood 121 2923 2933 23365459

PMC005xxxxxx/PMC5114002.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101633964 42657 Austin J Pulm Respir Med Austin journal of pulmonary and respiratory medicine 27868109 5114002 NIHMS798832 Article Air Pollution and Lung Function Loss: The Importance of Metabolic Syndrome Zhang L 12 Crowley G 1 Haider SH 1 Zedan M 1 Kwon S 1 Nolan A 134* 1 Department of Medicine, Division of Pulmonary, Critical Care and Sleep Medicine, New York University, USA 2 Department of Respiratory Medicine, PLA, Army General Hospital, China 3 Department of Environmental Medicine, New York University, USA 4 Bureau of Health Services and Office of Medical Affairs, Fire Department of New York, USA * Corresponding author: Anna Nolan, Division of Pulmonary, Critical Care and Sleep, NYU School of Medicine, New Bellevue, 7N Room 24462 1st Avenue, New York, NY 10016, USA 30 6 2016 17 6 2016 2016 17 11 2016 3 2 1043This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Air pollution: an international issue A multinational study spanning over 20 years showed that approximately 5.5 million deaths were attributed to airpollution [1]. Even brief ambient Particulate Matter (PM) exposure significantly decreases FEV1 [2–6]. Development of airway injury following PM exposure is a major health concern worldwide and is associated with an increased risk for hospital admission due to respiratory and vascular diseases [7]. Furthermore, a reduction of pollutants has the potential to save five million lives per year [8]. Both diseases place tremendous burden on the world’s health resources. The costs and challenges of remedying the pollution burden may have a more profound effect on developing countries. As a pulmonologist in Beijing, pollution has always been a significant contributor to the disease burden of my patients. My experience in the Division of Pulmonary and Critical Care at New York University has raised my awareness about the relevance of air pollution and associated diseases. This brief overview will serve to highlight the challenges faced in dealing with this international issue of pollution and Metabolic Syndrome (MetSyn) which are both significant contributors to loss of lung function. Air pollution in China In 2010, approximately 1.2 million deaths may have been secondary to the negative health effects of ambient PM exposure. These deaths represent a 33% increase when compared to data published in 1990 [9]. In 2013, Beijing had a “haze event;” some estimates indicate PM2.5 levels reached a monthly average above 170 μg/m3, peaking at 677μg/m3. This haze event may have been responsible for an estimated 690 deaths, 5470 hospitalizations for respiratory illness, and an increase in hospitalizations for cardiovascular disease [10]. In my hometown of Beijing, the pollution remains severe. These high levels of PM originate from a number of sources; the burning of fossil fuels and biomass are significant contributors [11]. PM2.5 includes products of combustion and is considered on a mass basis to be more toxic due to its hazardous components which stay airborne longer and can more readily reach the lung [12,13]. Epidemiologic studies have associated PM2.5 pollution with adverse health effects, linking PM2.5 exposure to the development of vascular and pulmonary diseases [14–23]. PM exposure causes systemic inflammation and chronic lung disease Airflow obstruction from PM and smoke exposure is a heterogeneous process. PM exposure causes systemic inflammation, endothelial dysfunction, and subsequent end-organ damage [24–26]. The development of obstructive airways disease from PM induced inflammation is poorly understood [27]. In addition to the aforementioned short term effects of PM exposure, long-term exposure also impairs lung function [17,28–31]. A 3.4% change in forced vital capacity per 10 μg/m3 change in concentration of PM10 has been reported [28]. The group that I am working with has focused on the exposure that occurred after the destruction of the World Trade Center (WTC) complex on 9/11/2001. Similarly, many firefighters, rescue workers, and lower Manhattan residents who were exposed to WTC-PM and other toxins also experience a loss of lung function [32,33]. However, we know there to be heterogeneity in the development of lung disease even after a high intensity exposure; therefore, we and other investigators have tried to identify other cofactors such as MetSyn that contribute to disease development. MetSyn, the collection of risk factors including hypertension, dyslipidemia, insulin resistance, and abdominal obesity, is a key cofactor that, according to 2012 estimates, affects at least 34% of Americans [34–36]. Previous cross-sectional studies have suggested associations between impaired lung function and MetSyn [37–41]. Longitudinal assessment showed that lower baseline FEV1 was an independent predictor of development of MetSyn [42]. In New York City 2014, the average value of PM2.5 was 8.48 μg/m3 and the PM10 24-hour maximum concentration was 46.25 μg/m3. While these levels are well below what is found in my native city of Beijing, there may still be room for improvement. Nearly half of COPD patients exhibit MetSyn [43]. In air pollution studies, long-term PM exposure and coexisting MetSyn increase systemic inflammation [34]. In the WTC exposed FDNY cohort, dyslipidemia (defined as an elevated triglycerides and low high-density lipoprotein) was significantly predictive of developing a FEV1 less than the lower limit of normal [24]. The systemic inflammatory effects of lipids and subsequent end-organ effects are of great interest. In light of these findings, our work has focused on the effects of lipids and inflammation in the development of PM-induced lung injury. The Chinese economy has undergone significant growth in recent years. The lifestyle and dietary choices of the Chinese population have similarly undergone changes that have led to the estimated prevalence of obesity in the Chinese adult population to be 30% [44]. Childhood obesity (BMI≥ 95th percentile) is rising in China (from 1.2% to 10% for boys age 7–18 and from 1.1% to 5.2% for girls age 7–18 between 1985 and 2000, in Beijing) [45,46]. Unhealthy eating habits, however, are not the only factors contributing to the rise in obesity. As discussed in more detail in previous sections of this paper, PM exposure has been linked to a number of adverse health effects [14–23]. Recently, ambient levels of Beijing PM have been shown to induce systemic inflammation and metabolic dysfunction in an exposed rat population, resulting in significantly increased body weight (when compared to a population exposed to filtered air) [44]. If these findings are found to be generalizable to the human population, it would indicate ambient Beijing PM has led to metabolic dysfunction in the people of Beijing. The adverse effects from this exposure coupled with an increase in unhealthy lifestyle choices of a city-dwelling population would be a probable explanation of the high prevalence of MetSyn in China which has been cited to be as high as 21.3%, with urban populations being more likely to exhibit MetSyn (compared to rural populations, OR=1.27) [47]. Metabolic biomarkers of obstructive airway disease not only have prognostic utility, they can direct future research into mechanisms producing airflow obstruction and fuel future work into their downstream effects. Multiple biomarkers have been identified and studied in clinical trials [48]. Our studies have focused on the well-phenotyped World Trade Center (WTC)-exposed Fire Department of New York (FDNY) cohort. In this population, we have observed classic lipid and non-lipid vascular risk factors predict abnormal lung function in the WTC-PM exposed cohort [24,49–51]. This data fits into a larger set of studies demonstrating an increased risk of patients with dyslipidemia for developing COPD due to air pollution and smoking [52]. The mechanism of bioactive lipid induced pulmonary inflammation is poorly understood but is an area of significant interest [53,54]. Biologically active lipid metabolites may identify plausible pathways of disease that could be pharmacological targets. Given the high prevalence of metabolic syndrome in my home country of China and throughout the world, establishing a mechanistic link between lipid mediators of MetSyn and lung injury is crucially important. The adverse impact on quality of life and sizable cost of WTC-lung disease are both public health concerns. Validating metabolic contributors of PM associated lung disease in the Chinese population is necessary. Future work will focus on the metabolomics of these PM exposed populations. Finally, if metabolic biomarkers are validated predictors of PM associated lung disease, then targeted behavioral dietary modification may mitigate disease severity and improve their health and well-being. 1 GBD 2013 Risk Factors Collaborators Forouzanfar MH Alexander L Anderson HR Bachman VF Biryukov S Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013 Lancet 2015 386 10010 2287 2323 26364544 2 Hoek G Brunekreef B Acute effects of a winter air pollution episode on pulmonary function and respiratory symptoms of children Arch Environ Health 1993 48 328 335 8215597 3 Brunekreef B Kinney PL Ware JH Dockery D Speizer FE Spengler JD Sensitive subgroups and normal variation in pulmonary function response to air pollution episodes Environ Health Perspect 1991 90 189 193 2050060 4 Barnes PJ Chronic obstructive pulmonary disease: effects beyond the lungs PLoS Med 2010 7 e1000220 20305715 5 Peters JM Avol E Gauderman WJ Linn WS Navidi W London SJ A study of twelve Southern California communities with differing levels and types of air pollution. II. Effects on pulmonary function Am J Respir Crit Care Med 1999 159 768 775 10051249 6 Peters JM Avol E Navidi W London SJ Gauderman WJ Lurmann F A study of twelve Southern California communities with differing levels and types of air pollution. I. Prevalence of respiratory morbidity Am J Respir Crit Care Med 1999 159 760 767 10051248 7 Dominici F Peng RD Bell ML Pham L McDermott A Zeger SL Fine particulate air pollution and hospital admission for cardiovascular and respiratory diseases JAMA 2006 295 1127 1134 16522832 8 Gulland A Reduction in climate pollutants could save five million lives a year, WHO report says BMJ 2015 351 h5688 26504186 9 Liu S Zhou M Wang L Li Y Liu Y Liu J Burden of disease attributable to ambient particulate matter pollution in 1990 and 2010 in China Zhonghua Yu Fang Yi Xue Za Zhi 2015 49 327 333 26081541 10 Gao M Guttikunda SK Carmichael GR Wang Y Liu Z Stanier CO Health impacts and economic losses assessment of the 2013 severe haze event in Beijing area Sci Total Environ 2015 511 553 561 25585158 11 Rohde RA Muller RA Air Pollution in China: Mapping of Concentrations and Sources PLoS One 2015 10 e0135749 26291610 12 Oberdörster G Ferin J Lehnert BE Correlation between particle size, in vivo particle persistence, and lung injury Environ Health Perspect 1994 102 Suppl 5 173 179 13 Wilson WE Chow JC Claiborn C Fusheng W Engelbrecht J Watson JG Monitoring of particulate matter outdoors Chemosphere 2002 49 1009 1043 12492163 14 Blanc PD Eisner MD Earnest G Trupin L Balmes JR Yelin EH Further exploration of the links between occupational exposure and chronic obstructive pulmonary disease J Occup Environ Med 2009 51 804 810 19528835 15 Blanc PD Iribarren C Trupin L Earnest G Katz PP Balmes J Occupational exposures and the risk of COPD: dusty trades revisited Thorax 2009 64 6 12 18678700 16 Blanc PD Menezes AM Plana E Mannino DM Hallal PC Toren K Occupational exposures and COPD: an ecological analysis of international data Eur Respir J 2009 33 298 304 19010980 17 Brunekreef B Holgate ST Air pollution and health Lancet 2002 360 1233 1242 12401268 18 Dockery DW Pope CA 3rd Acute respiratory effects of particulate air pollution Annu Rev Public Health 1994 15 107 132 8054077 19 Dockery DW Pope CA 3rd Xu X Spengler JD Ware JH Fay ME An association between air pollution and mortality in six U.S. cities N Engl J Med 1993 329 1753 1759 8179653 20 Samet JM Dominici F Curriero FC Coursac I Zeger SL Fine particulate air pollution and mortality in 20 U.S. cities, 1987–1994 N Engl J Med 2000 343 1742 1749 11114312 21 Samet JM Dominici F Zeger SL Schwartz J Dockery DW The National Morbidity, Mortality, and Air Pollution Study. Part I: Methods and methodologic issues Res Rep Health Eff Inst 2000 5 14 discussion 75–84 22 Samet JM Zeger SL Dominici F Curriero F Coursac I Dockery DW The National Morbidity, Mortality, and Air Pollution Study. Part II: Morbidity and mortality from air pollution in the United States Res Rep Health Eff Inst 2000 94 5 70 11354823 23 Thurston GD Ito K Hayes CG Bates DV Lippmann M Respiratory hospital admissions and summertime haze air pollution in Toronto, Ontario: consideration of the role of acid aerosols Environ Res 1994 65 271 290 8187742 24 Naveed B Weiden MD Kwon S Gracely EJ Comfort AL Ferrier N Metabolic syndrome biomarkers predict lung function impairment: a nested case–control study Am J Respir Crit Care Med 2012 185 392 399 22095549 25 Gosker HR Schrauwen P Broekhuizen R Hesselink MK Moonen-Kornips E Ward KA Exercise training restores uncoupling protein-3 content in limb muscles of patients with chronic obstructive pulmonary disease Am J Physiol Endocrinol Metab 2006 290 E976 981 16352674 26 Gan WQ Man S Senthilselvan A Sin D Association between chronic obstructive pulmonary disease and systemic inflammation: a systematic review and a meta-analysis Thorax 2004 59 574 580 15223864 27 Barnes PJ Chronic obstructive pulmonary disease: effects beyond the lungs PLoS Med 2010 7 e1000220 20305715 28 Ackermann-Liebrich U Leuenberger P Schwartz J Schindler C Monn C Bolognini G Lung function and long term exposure to air pollutants in Switzerland. Study on Air Pollution and Lung Diseases in Adults (SAPALDIA) Team Am J Respir Crit Care Med 1997 155 122 129 9001300 29 Franco Suglia S Gryparis A Schwartz J Wright RJ Association between traffic-related black carbon exposure and lung function among urban women Environ Health Perspect 2008 116 1333 1337 18941574 30 Götschi T Heinrich J Sunyer J Künzli N Long-term effects of ambient air pollution on lung function: a review Epidemiology 2008 19 690 701 18703932 31 Kan H Heiss G Rose KM Whitsel E Lurmann F London SJ Traffic exposure and lung function in adults: the Atherosclerosis Risk in Communities study Thorax 2007 62 873 879 17442705 32 Weiden MD Kwon S Caraher E Berger KI Reibman J Rom WN Biomarkers of World Trade Center Particulate Matter Exposure: Physiology of Distal Airway and Blood Biomarkers that Predict FEV1 Decline Seminars in Respiratory and Critical Care Medicine 2015 36 323 333 26024341 33 Szema AM Savary KW Ying BL Lai K Post 9/11: high asthma rates among children in Chinatown, New York Allergy Asthma Proc 2009 30 605 611 19772715 34 Chen JC Schwartz J Metabolic syndrome and inflammatory responses to long-term particulate air pollutants Environ Health Perspect 2008 116 612 617 18470293 35 Grundy SM Cleeman JI Daniels SR Donato KA Eckel RH Franklin BA Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement Circulation 2005 112 2735 2752 16157765 36 Aguilar M Bhuket T Torres S Liu B Wong RJ Prevalence of the metabolic syndrome in the United States, 2003–2012 JAMA 2015 313 1973 1974 25988468 37 Lazarus R Sparrow D Weiss ST Baseline ventilatory function predicts the development of higher levels of fasting insulin and fasting insulin resistance index: the Normative Aging Study Eur Respir J 1998 12 641 645 9762793 38 Lawlor DA Ebrahim S Smith GD Associations of measures of lung function with insulin resistance and Type 2 diabetes: findings from the British Women’s Heart and Health Study Diabetologia 2004 47 195 203 14704837 39 Lin WY Yao CA Wang HC Huang KC Impaired lung function is associated with obesity and metabolic syndrome in adults Obesity (Silver Spring) 2006 14 1654 1661 17030977 40 Fimognari FL Pasqualetti P Moro L Franco A Piccirillo G Pastorelli R The association between metabolic syndrome and restrictive ventilatory dysfunction in older persons J Gerontol A Biol Sci Med Sci 2007 62 760 765 17634324 41 Leone N Courbon D Thomas F Bean K Jégo B Leynaert B Lung Function Impairment and Metabolic Syndrome: The Critical Role of Abdominal Obesity American Journal of Respiratory and Critical Care Medicine 2009 179 509 516 19136371 42 Hsiao FC Wu CZ Su SC Sun MT Hsieh CH Hung YJ Baseline forced expiratory volume in the first second as an independent predictor of development of the metabolic syndrome Metabolism 2010 59 848 853 20006363 43 Watz H Waschki B Kirsten A Müller KC Kretschmar G Meyer T The metabolic syndrome in patients with chronic bronchitis and COPD: frequency and associated consequences for systemic inflammation and physical inactivity Chest 2009 136 1039 1046 19542257 44 Wei Y Zhang JJ Li Z Gow A Chung KF Hu M Chronic exposure to air pollution particles increases the risk of obesity and metabolic syndrome: findings from a natural experiment in Beijing FASEB J 2016 30 2115 2122 26891735 45 Ji CY Report on childhood obesity in China (1)–body mass index reference for screening overweight and obesity in Chinese school-age children Biomed Environ Sci 2005 18 390 400 16544521 46 Ji CY The prevalence of childhood overweight/obesity and the epidemic changes in 1985–2000 for Chinese school-age children and adolescents Obes Rev 2008 9 Suppl 1 78 81 18307704 47 Xi B He D Hu Y Zhou D Prevalence of metabolic syndrome and its influencing factors among the Chinese adults: the China Health and Nutrition Survey in 2009 Prev Med 2013 57 867 871 24103567 48 Barnes PJ The cytokine network in asthma and chronic obstructive pulmonary disease J Clin Invest 2008 118 3546 3556 18982161 49 Nolan A Naveed B Comfort AL Ferrier N Hall CB Kwon S Inflammatory biomarkers predict airflow obstruction after exposure to World Trade Center dust Chest 2012 142 412 418 21998260 50 Weiden MD Naveed B Kwon S Cho SJ Comfort AL Prezant DJ Cardiovascular biomarkers predict susceptibility to lung injury in World Trade Center dust-exposed firefighters Eur Respir J 2013 41 1023 1030 22903969 51 Tsukiji J Cho SJ Echevarria GC Kwon S Joseph P Schenck EJ Lysophosphatidic acid and apolipoprotein A1 predict increased risk of developing World Trade Center-lung injury: a nested case-control study Biomarkers 2014 19 159 165 24548082 52 Tiengo A Fadini GP Avogaro A The metabolic syndrome, diabetes and lung dysfunction Diabetes Metab 2008 34 447 454 18829364 53 Bowler RP Jacobson S Cruickshank C Hughes GJ Siska C Ory DS Plasma sphingolipids associated with chronic obstructive pulmonary disease phenotypes Am J Respir Crit Care Med 2015 191 275 284 25494452 54 Mirzaie M Kheradmand F Bioactive Lipids in Emphysema. Decoding Fat to Reveal COPD Phenotypes Am J Respir Crit Care Med 2015 191 241 243 25635483

PMC005xxxxxx/PMC5114019.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2984819R 7707 Tetrahedron Lett Tetrahedron Lett. Tetrahedron letters 0040-4039 27867229 5114019 10.1016/j.tetlet.2012.03.073 NIHMS798100 Article Eucalyptals D and E, new cytotoxic phloroglucinols from the fruits of Eucalyptus globulus and assignment of absolute configuration Wang Ji ab†‡ Zhai Wen-Zhu ab†‡ Zou Yike c§ Zhu Jing-Jing a† Xiong Juan b‡ Zhao Yun a† Yang Guo-Xun b‡ Fan Hui a† Hamann Mark T. c§ Xia Gang a† Hu Jin-Feng ab*†‡ a Department of Natural Products for Chemical Genetic Research, Key Laboratory of Brain Functional Genomics, Ministry of Education, East China Normal University, No. 3663 North Zhongshan Rd., Shanghai 200062, PR China b Department of Natural Products Chemistry, School of Pharmacy, Fudan University, No. 826 Zhangheng Rd., Shanghai 201203, PR China c Department of Pharmacognosy, School of Pharmacy, The University of Mississippi, University, MS 38677, USA * Corresponding author. Tel./fax: +86 21 51980172. fhu@brain.ecnu.edu.cn, jfhu@fudan.edu.cn (J.-F. Hu) † Responsible for phytochemistry work. ‡ Responsible for cytotoxic assay. § Responsible for ECD calculation. 28 6 2016 25 3 2012 23 5 2012 17 11 2016 53 21 26542658 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Two new phloroglucinols, named eucalyptals D (1) and E (2), along with a related known compound (euglobal-In-3, 3) were isolated from the fruits of Eucalyptus globulus. Their structures were established on the basis of extensive spectroscopic studies, revealing that they share a common 3,5-diformyl-isopentyl phloroglucinol unit, but each is instead coupled to a different sesquiterpenoid skeleton (aromadendrene in 1, cadinene in 2, and a spirosesquiterpene in 3). Compound 1 possessed an unusual seven-membered D ring with an ether bridge between C-2 of the aromadendrene moiety and C-2′ of the aromatic unit. The absolute configuration of the isolates was defined by the comparison of experimental and calculated electronic circular dichroism (ECD) spectra. Compounds 1–3 exhibited significant in vitro cytotoxicities against a few human cancer cell lines (Huh-7, Jurkat, BGC-823, and KE-97) using the CellTiter-Glo™ luminescent cell viability assay method. Eucalyptus globulus Myrtaceae Phloroglucinols Eucalyptals Cytotoxicities Introduction The phloroglucinol-terpene adducts are a class of secondary metabolites with interesting bioactive structures unique to plants in the family Myrtaceae, especially Eucalyptus species.1–3 The blue gum plant Eucalyptus globulus Labill was autochthonous to Australia.4 It was recorded that this plant was first introduced to China in late 1890s.5 Now it is widely cultivated in southern and south-western China, especially in Yunnan and Jiangxi provinces. The dried mature fruits, earned a common Chinese name ‘Yi-Kou-Zhong’, have been used as an herbal medicine for the treatment of influenza, inflammation, and eczema.6 The fruits have been documented to yield cytotoxic 3,5-diformyl-isopentyl phloroglucinol-coupled sesquiterpenes.7,8 In an ongoing project towards the discovery of new antitumor agents from natural products,9,10 two (1,2) new and one (3) related known phloroglucinol-coupled sesquiterpenes (Fig. 1) were isolated from ‘Yi-Kou-Zhong’ collected from the noted Jing-gang Mountain. We herein report the characterization of the new compounds and their cytotoxic effects against four human cancer cell lines (Huh-7 hepatocarcinoma, Jurkat T cell lymphoblast, BGC-823 and KE-97 gastric carcinoma). Results and discussion Air-dried and powdered fruits of E. globulus were extracted with 95% ethanol at room temperature (rt), and the extract was concentrated in vacuo. The crude extract was suspended in H2O and then exhaustively extracted with EtOAc. Our phytochemical reinvestigation on the EtOAc fraction gave rise to three 3,5-diformyl-isopentyl phloroglucinol-coupled sesquiterpenes (1: 3.4 mg, 2: 1.7 mg, 3: 4.2 mg). The isolation was indeed guided by observing the typical signals of the formyl protons in the 1H NMR spectrum. Comparison of the spectroscopic data and physical properties with those reported in the literature, compound 3 was identified as euglobal-In-3, a phloroglucinol possessing an unusual spirosesquiterpene skeleton previously isolated from the leaves of Eucalyptus incrassate.11 The absolute configuration of 3 was deduced herein by means of the electronic circular dichroism (ECD) calculation for the first time (see below). Eucalyptal D (1)12 was obtained as a pale yellow gum. The molecular formula of 1 was determined to be C28H36O5 based on a prominent molecular ion peak at m/z 452.2562 [M]+ in its HR-EIMS. The IR and UV data of compound 1 were in good agreement with those of euglobal-In-3 (3),11 indicating that 1 also has a similarly substituted phloroglucinol moiety. The 1H NMR spectral data (in CDCl3, Table 1) of 1 showed two D2O-exchangable hydrogen-bonded phenolic hydroxyls [δ 13.35 (1H, s, 4′-OH), 13.22 (1H, s, 6′-OH)], two aldehyde protons [δ 10.09 (1H, s, H-7′), 10.20 (1H, s, H-8′)], one oxymethine [δ 5.36 (1H, br d, J = 4.5 Hz, H-2)], a methine proton [δ 3.19 (1H, br d, J = 11.0 Hz, H-9′)] adjacent to the aromatic ring and isobutyl side chain, four singlet methyls [δ 1.60 (vinylic Me-14), 1.18 (Me-15), 1.06 (Me-12), 0.99 (Me-13)], two doublet methyls [δ 0.95 (3H, d, J = 6.3 Hz, Me-13′), 0.77 (3H, d, J = 6.6 Hz, Me-12′)], and two typical methine protons resonating at δ 0.68 (1H, dd, J = 11.3, 9.9 Hz, H-6) and 0.49 (1H, ddd, J = 9.9, 9.5, 6.3 Hz, H-7) attributed to a gem-dimethyl cyclopropane ring. 13C and DEPT NMR spectra of 1 exhibited twenty-eight signals classified as six sp3 methyl, four sp3 methylene, six sp3 (one oxygenated at δ 86.5) and two sp2 (δ 193.3, 192.1) methine, two sp3, and eight sp2 (δ 169.5, 168.3, 167.2, 138.0, 135.5, 115.9, 108.2, 104.6) quaternary carbons (Table 1). These results demonstrated that 1 was a phloroglucinol-coupled sesquiterpene, and the 3,5-diformyl-isopentyl phloroglucinol unit of which was secured by 2D NMR experiments. An obvious isobutyl spin system was found in the COSY spectrum. In the HMBC spectrum, methine proton H-9′ was correlated with C-l′, C-2′, C-6′, and C-11′, while formyl proton H-7′ was correlated with C-2′, C-3′, and H-8′ was correlated with C-5′ and C-6′ (Fig. 2). Two separated spin systems of –OCHCH2– (H-2 and H2-3) and – CHCHCHCH2CH2– (H-5, H-6, H-7, H2-8, andH2-9) were found for the sesquiterpene unit in the COSY spectrum of 1. One bond proton–carbon connectivities and the long range proton–carbon couplings in this unit were then ascertained by HSQC and HMBC NMR experiments (Fig. 2), which led to the determination of an aromadendrene-type sesquiterpene moiety involving a 5-membered ring (A), 7-membered ring (B), and 3-membered ring (C). The molecular formula of 1 accounted for eleven degrees of unsaturation. Apart from six double-bond equivalents in the 3,5-diformyl-isopentyl phloroglucinol unit, and the tricyclic sesquiterpene unit bearing an unambiguous double-bond (olefinic carbons resonating at δ 138.0 and 135.5), the remaining one degree of unsaturation must be an additional ring connecting between the two units. The linkage positions could be thereafter easily elucidated by a key HMBC correlation between the oxymethine proton H-2 at δ 5.36 and C-2′ at δ 168.3. Meanwhile, H-9′ at δ 3.19 was further correlated with C-3 (δ 39.9), and C-15 (δ 22.4), H-5 at δ 2.08 (1H, d, J = 11.5 Hz) was correlated with C-9′ at δ 43.0 in the HMBC NMR spectrum. In fact, similar phloroglucinol–aromadendrene coupled compounds (e.g. macrocarpals A and B, euglobal V) were previously obtained from the fruits and leaves of E. globulus.8,13–16 However, such a seven-membered D ring constructed with an ether bridge between C-2 of the tricyclic aromadendrene moiety and C-2′ of the aromatic E ring in compound 1 has never been encountered. The relative stereochemistry of 1 was determined by the analysis of the proton coupling constants (Table 1) and the NOE correlations in the NOESY NMR spectrum (Fig. 3). A large coupling constant (11.5 Hz) between H-5 and H-6 indicated that both H-5 and H-6 took opposite axial positions. Clear NOE correlations were observed between H-5 and Me-12 at δ 1.06, between H-5 and H-9′, between H-6 at δ 0.68 and H-7 at δ 0.49, between H-6 and Me-15 at δ 1.18, between Me-15 and H-3a at δ 1.53, between H-3b at δ 2.32 and H2-10′ at δ 1.42/1.30, as well as between H2-3 and H-2. These data revealed that H-2, H-6, H-7, Me-15, and the isobutyl group adopted the same orientation, whereas H-5, H-9′ and the cyclopropane ring assumed the same orientation as depicted in Fig. 1. Eucalyptal E (2)12 was also obtained as a pale yellow gum. The molecular weight of 2 and its chemical formula of C28H38O7 were deduced from its HR-EIMS, which resulted in an [M-H2O]+ ion peak at m/z 468.2516. The IR and UV data of 2 closely resembled those of compounds 1 and 3,11 indicating that 2 also has a substituted phloroglucinol moiety. The 1H and 13C NMR data (Table 1) of 2 showed general features similar to those of compound 4 (eucalyptal B, Fig. 1), a known 3,5-diformyl-isopentyl phloroglucinol-coupled cadinene previously also isolated from the fruits of E. globulus in 2007.7 The 1H NMR spectral data of 2 exhibited two D2O-exchangable phenolic hydroxyls [δ 13.35 (1H, s, 4′-OH), 13.38 (1H, s, 6′-OH)], two aldehyde protons [δ 10.06 (1H, s, H-7′), 10.17 (1H, s, H-8′)], an olefinic proton resonating at δ 5.62 (1H, brs, H-2), one oxymethine [δ 4.55 (1H, d, J = 10.0 Hz, H-5α)], a methine proton [δ 2.66 (1H, t, J = 5.6 Hz, H-9′)] adjacent to the aromatic ring and isobutyl side chain, four tertiary methyl groups with singlets at δ 1.26 (3H, Me-12), 1.30 (3H, Me-13), 1.28 (3H, Me-14), and 0.88 (3H, Me-15), and two secondary methyl doublets at δ 0.95 (3H, d, J = 6.5 Hz, Me-12′) and 0.97 (3H, d, J = 6.5 Hz). Similar to 1, the 13C NMR and DEPT NMR spectra of 2 also gave 28 signals [six sp3 methyl, four sp3 methylene, five sp3 (one oxygenated at δ 77.3) and three sp2 (δ 191.7×2, 115.4) methine, three sp3 (two oxygenated at δ 74.9 and 70.4) and seven sp2 (δ 169.3, 167.9, 162.2, 138.5, 107.0, 104.0, 103.4) quaternary carbons] (Table 1). The obvious difference between 2 and 4 was in the cadinene moiety. A tertiary methyl group (geminal to a carbon bearing an oxygen) at δ 1.28 (3H, Me-14) was present in 2, instead of the Δ10(14) exomethylene group in 4.7 Moreover, a trisubstituted double bond at C-1 was present in 2. These structural features were confirmed by 2D NMR experiments, and the planar structure of 2 was determined. The relative stereochemistry of 2 was determined by a combination of analysis of the proton coupling constants (Table 1) and the NOE correlations in the NOESY NMR spectrum (Fig. 3). In the proposed structure of 2, a large coupling constant (10.0 Hz) between H-5 at δ 4.55 and H-6 at δ 2.62 indicated that both H-5 and H-6 took opposite axial positions. Clear NOE correlations were observed between H-5 and Me-12 at δ 1.26, between H-5 and H-3a at δ 2.51, between H-6 and H-7 at δ 2.01, between H-6 and Me-15 at δ 0.88, between Me-15 and H-3b at δ 2.05, between Me-15 and H-9′ at δ 2.66, as well as between Hax-9 at δ 1.83 (br d, J = 16.1, 11.3 Hz) and Me-14 at δ 1.28, but not between Me-14 and H-6 (otherwise they should have a strong NOE correlation). The above data mentioned that H-5, the 2-hydroxyisopropyl group at C-7, Me-14, and the isobutyl group at C-9′ took the same orientation, whereas H-6, H-7, 10-OH, Me-15, and H-9′ adopted the same orientation as illustrated in Figure 1. The overall absolute configuration of phloroglucinol-coupled sesquiterpenes has never been determined. To our knowledge, the absolute configuration at C-9′ of macrocarpals H, I, and J was once determined by a computational chemical method involving Monte Carlo (MC) and semiempirical molecular orbital calculations.17 The absolute stereochemistry at C-10 of a derivative of macrocarpals A and C was established by chemical transformation and modified Mosher’s method; however, the absolute configuration of entire structure of the macrocarpals A and C was still unknown.14 To date, the calculated experimental electronic circular dichroism (ECD) spectral data have been successfully utilized to determine the absolute configuration of complex naturally occurring compounds (e.g. sorbiterrin A,18 psiguadials,19 and discorhabdins20). To further determine the absolute configuration of compounds 1–3, a computational approach was applied, and the simulated ECD curves were then overlaid with the experimental CD spectra for comparison. Briefly, conformational analyses were carried out by the Amber force field method followed by the PM6 semi-empirical method in tandem to acquire pre-optimized structures. As a result, a group of low energy conformations of compounds 1–3 were generated. Each low energy conformation was further optimized by the density functional theory (DFT) method at the B3LYP/6-31g(d,p) level in the gas phase and applying the Polarizable Continuum Model (PCM) solvation model, respectively, and the optimized geometries were then followed by excited state calculations using the time-dependent density functional theory (TDDFT) method at the same theoretical level. Resulting from the simulated ECD curves overlaid with the experimental CD spectra of each compound (Fig. 4) (see Supplementary data), the absolute configuration of compounds 1–3 was thus finally assigned as (2R, 4R, 5R, 6R, 7R, 9′S)-eucalyptal D (1), (4R, 5S, 6R, 7R, 10R, 9′S)-eucalyptal E (2), and (1S, 4R, 5S, 10R, 9′S)-euglobal-In-3 (3), respectively. From a structural point of view, the above three isolates (1–3) have a common 3,5-diformyl-isopentyl phloroglucinol unit, but each is instead coupled to a different sesquiterpenoid framework. With regard to their biosynthetic pathway, these phloroglucinol-sesquiterpene adducts (1–3) are thought to be derived biogenetically from a phloroglucinol precursor (a) and a bicyclogermacrene (b) (Fig. 5). In fact, biogenetic pathways have been previously proposed for phloroglucinol-coupled aromadendrenes (e.g. macrocarpals A–C,14) cadinenes (e.g. eucalyptals A–C,7) and spirosesquiterpenes (e.g. euglobal-In-211) obtained from the plant E. globulus. Since compound 1 contained an unusual seven-membered D ring with an ether bridge between C-2 of the aromadendrene moiety and C-2′ of the aromatic unit, a possible biosynthetic pathway is therefore highlighted in Fig. 5. The key step is producing a stable intermediate (c) possessing an allyl cation with a p–π conjugated effect, followed by a carbocation-induced cyclization under acidic condition to form such a seven-membered D ring in compound 1. The cytotoxicities of compounds 1–3 against human BGC-823 and KE-97 gastric, Huh-7 hepatocarcinoma, and Jurkat T lymphoma cancer cell lines were tested by the CellTiter-Glo™ luminescent cell viability assay method.10 The results were summarized in Table 2. Supplementary Material 2 The work was supported by an NSFC Grant (No. 90713040), STCSM Grants (Nos. 06DZ19002, 07DZ22006, 11DZ1921203), and the Fundamental Research Funds for the Central Universities (No. 78210102). The ECD calculation was carried out in MH’s group, to whom the computational resources were provided by the Mississippi Center for Supercomputing Research (MCSR), the United States of America. Figure 1 The structures of phloroglucinol-coupled sesquiterpenes 1–4. Figure 2 Key COSY ( ) and HMBC (H→C) correlations of 1. Figure 3 Observed key NOE (H H) correlations of 1 and 2. Figure 4 Simulated ECD spectra of 1–3 (after Boltzmann averaging) compared with the experimental CD spectra. Figure 5 Hypothetical biosynthetic pathway of 1. Table 1 1H and 13C NMR data of 1 and 2 (in CDCl3, δ in ppm, J in Hz) No. 1a 2b δ H δ C δ H δ C 1 138.0 138.5 2 5.36 (br d, 4.5) 86.5 5.62 (br s) 115.4 3 1.53 (dd, 14.2, 4.5, H-3a) 2.32 (br d, 14.2, H-3b) 39.9 2.51 (br d, 18.0, H-3a) 2.05 (dd, 18.0, 4.3, H-3b) 34.6 4 43.7 43.1 5 2.08 (d, 11.5) 47.1 4.55 (d, 10.0) 77.3 6 0.68 (dd, 11.5, 9.9) 32.4 2.62 (br d, 10.0) 43.5 7 0.49 (ddd, 10.0, 9.9, 5.3) 24.9 2.01 (m) 29.6 8 1.41 (overlapped, H-8a) 1.58 (overlapped, H-8b) 21.3 2.23 (m) 1.65 (overlapped) 25.1 9 2.14 (overlapped, H-9a) 2.04 (overlapped, H-9b) 36.4 1.83 (br dd, 16.0, 10.5, Hax) 1.66 (overlapped, Heq) 31.1 10 135.5 70.4 11 18.6 74.9 12 1.06 (s, 3H) 15.6 1.26 (s, 3H) 30.0 13 0.99 (s, 3H) 28.1 1.30 (s, 3H) 29.6 14 1.60 (s, 3H) 22.0 1.28 (s, 3H) 22.3 15 1.18 (s, 3H) 22.4 0.88 (s, 3H) 20.9 1′ 115.9 107.0 2′ 168.3 162.2 3′ 108.2 104.0 4′ 167.2 167.9 5′ 104.6 103.4 6′ 169.5 169.3 7′ 10.09 (s) 193.3 10.06 (s) 191.7 8′ 10.20 (s) 192.1 10.17 (s) 191.7 9′ 3.19 (br d, 11.0) 43.0 2.66 (t, 5.6) 36.3 10′ 1.30 (overlapped, H-a) 1.42 (overlapped, H-b) 41.7 1.13 (m) 1.38 (m) 43.5 11′ 1.11 (m) 25.8 1.70 (m, overlapped) 27.6 12′ 0.95 (d, 6.3) 21.4 0.95 (d, 6.5) 23.0 13′ 0.77 (d, 6.6) 24.5 0.97 (d, 6.5) 23.1 4′-OH 13.35 (s) 13.35 (s) 6′-OH 13.22 (s) 13.38 (s) a Recorded at 500 and 125 MHz for 1H and 13C, respectively. b Recorded at 400 and 100 MHz for 1H and 13C, respectively. Table 2 Cytotoxicities of compounds 1–3 Compound IC50 (μM, mean ± SEM, n = 4) KE-97 Jurkat BGC-823 Huh-7 1 5.20 ± 1.16 7.16 ± 2.84 6.65 ± 1.36 11.73 ± 1.81 2 7.22 ± 2.06 10.71 ± 4.21 10.16 ± 0.89 24.57 ± 4.45 3 4.63 ± 0.86 10.50 ± 3.25 14.27 ± 0.78 10.77 ± 3.55 STa 0.13 ± 0.01 0.14 ± 0.03 0.17 ± 0.02 0.23 ± 0.08 a Staurosporine (ST): Positive control. Supplementary data Supplementary data (NMR, HR-EIMS, CD/UV curves, and experimental/calculated electronic circular dichroism (ECD) spectra of compounds 1–3) associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tetlet.2012.03.073. References and notes 1 Eyles A Davies NW Mohammed C J Chem Ecol 2003 29 881 898 12775149 2 Eschler BM Pass DM Willis R Foley WJ Biochem Syst Ecol 2000 28 813 824 10913843 3 Bharate SB Singh IP Bioorg Med Chem Lett 2011 21 4310 4315 21665468 4 Kawabata T Hasegawa T Nojiri Y Uchida C Tsubata T Kato H Takano F Ohta T Heterocycles 2011 83 631 636 5 Editorial Commission of China Flora Flora of China 53 Science Press Beijing 1984 28 6 Jiangsu New Medical College Dictionary of Chinese Materia Medica Shanghai Scientific & Technical Publishing House Shanghai 1985 1793 7 Yin S Xue JJ Fan CQ Miao ZH Ding J Yue JM Org Lett 2007 9 5549 5552 18020353 8 Yang XW Guo QM Wang Y Xu W Tian L Tian XJ Bioorg Med Chem Lett 2007 17 1107 1111 17118653 9 Li XW Weng L Gao X Zhao Y Pang F Liu JH Zhang HF Hu JF Bioorg Med Chem Lett 2011 21 366 372 21109433 10 Wu SB Bao QY Wang WX Zhao Y Xia G Zhao Z Zeng HQ Hu JF Planta Med 2011 77 922 928 21243584 11 Takasaki M Konoshima T Kozuka M Haruna M Ito K Nat Med 1997 51 486 490 12 The new phloroglucinol–sesquiterpene adducts 1 and 2 were named eucalyptals D and E, respectively. The trivial name was sequentially derived from the work of Yue and co-workers (Ref.7 herein). Eucalyptal D (1): Pale yellow gum; [α]D22 104.7 (c 0.34, CHCl3); UV (CH3OH) λmax (log ε) nm: 274 (4.38). IR (KBr) νmax (cm−1): 3424 (br), 2924, 1633, 1601, 1448, 1384, 1173, 1021; For 1H, 13C NMR data, see Table 1; EI-MS: m/z 452 [M]+; HR EI-MS: m/z 452.2562 [M]+ (C28H36O5; calcd 452.2563, Δ = −0.2 ppm). Eucalyptal E (2): Pale yellow gum; [α]D22 −135.3 (c 0.17, CHCl3); UV (CH3OH) λmax (log ε) nm: 276 (4.52). IR (KBr) νmax (cm−1): 3444 (br), 2925, 1632, 1445, 1384, 1045; For 1H, 13C NMR data, see Table 1; EI-MS: m/z 468 ([M-H2O]+); HR EI-MS: m/z 468.2516 [M–H2O]+ (C28H36O6; calcd 468.2512, Δ = 0.9 ppm). 13 Osawa K Yasuda H J Nat Prod 1996 59 823 827 8864235 14 Nishizawa M Emura M Kan Y Yamada H Ogawa K Hamanaka N Tetrahedron Lett 1992 33 2983 2986 15 Takasaki M Konoshima T Fujitani K Yoshida S Nishimura H Tokuda H Nishino H Iwashima A Kozuka M Chem Pharm Bull 1990 38 2737 2739 1963812 16 Amano T Komiya T Hori M Goto M J Chromatogr 1981 208 347 355 17 Osawa K Yasuda H Morita H Takeya K Itokawa H Chem Pharm Bull 1997 45 1216 1217 18 Chen L Zhu TJ Ding YQ Khan IA Gu QQ Li DH Tetrahedron Lett 2012 53 325 328 19 Shao M Wang Y Liu Z Zhang DM Cao HH Jiang RW Fan CL Zhang XQ Chen HR Yao XS Ye WC Org Lett 2010 12 5040 5043 20929258 20 Hu JF Fan H Xiong J Wu SB Chem Rev 2011 111 5465 5491 21688850

PMC005xxxxxx/PMC5114023.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0042124 4284 Int J Cancer Int. J. Cancer International journal of cancer 0020-7136 1097-0215 27038352 5114023 10.1002/ijc.30117 NIHMS828607 Article The association of soy food consumption with the risk of subtype of breast cancers defined by hormone receptor and HER2 status Baglia Michelle L Zheng Wei Li Honglan Yang Gong Gao Jing Gao Yu-Tang Shu Xiao-Ou Division of Epidemiology, Department of Medicine, Vanderbilt-Ingram Cancer Center, Vanderbilt University School of Medicine, Nashville, TN, 37203 (MLB, WZ, GY, XOS); Department of Epidemiology, Shanghai Cancer Institute, Shanghai, People’s Republic of China, 200032 (HL, JG, YTG) Corresponding author: Xiao-Ou Shu, M.D., Ph.D., Vanderbilt Epidemiology Center, Vanderbilt University Medical Center, 2525 West End Avenue, Suite 600 (IMPH), Nashville, TN 37203-1738, Phone: 615-936-0713, Fax: 615-936-8291, xiao-ou.shu@vanderbilt.edu 10 11 2016 05 5 2016 15 8 2016 15 8 2017 139 4 742748 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Soy-food intake has previously been associated with reduced breast cancer risk. Epidemiological evidence for subgroups of breast cancer, particularly by menopausal and hormone receptor status, is less consistent. To evaluate the role of hormone receptor and menopausal status on the association between soy-food intake and breast cancer risk, we measured usual soy-food intake in adolescence and adulthood via food frequency questionnaire in 70,578 Chinese women, aged 40-70 years, recruited to the Shanghai Women’s Health Study (1996-2000). After a median follow-up of 13.2 years (range:0.01-15.0), 1,034 incident breast cancer cases were identified. Using Cox models, we found that adult soy intake was inversely associated with breast cancer risk (hazard ratio-HR) for fifth versus first quintile soy protein intake=0.78; 95% confidence interval (CI):0.63-0.97). The association was predominantly seen in premenopausal women (HR=0.46; 95% CI:0.29-0.74). Analyses further stratified by hormone receptor status showed that adult soy intake was associated with significantly decreased risk of ER+/PR+ breast cancer in postmenopausal women (HR=0.72; 95% CI:0.53-0.96) and decreased risk of ER−/PR− breast cancer in premenopausal women (HR=0.46; 95% CI:0.22-0.97). The soy association did not vary by HER2 status. Furthermore, we found that high soy intake during adulthood and adolescence was associated with reduced premenopausal breast cancer risk (HR=0.53; 95% CI:0.32-0.88; comparing third versus first tertile) while high adulthood soy intake was associated with postmenopausal breast cancer only when adolescent intake was low (HR=0.63; 95% CI:0.43-0.91). Our study suggests that hormonal status, menopausal status, and time window of exposure are important factors influencing the soy-breast cancer association. soy breast cancer adolescence menopausal status hormone receptor status INTRODUCTION Breast cancer is the leading cancer in women worldwide 1. Low rates of breast cancer among women in Asian countries and its rapid increase following emigration to western cultures has led to lifestyle factors, particularly dietary factors, being postulated as an explanation for this pattern 2-4. Epidemiological studies have shown that soy food consumption may decrease the risk of breast cancer and these data were summarized in several recent meta-analyses/systematic reviews 5-8. However, the conclusions from these studies are mixed with one suggesting that the association may be limited to premenopausal women 5, one reporting a stronger effect in postmenopausal women 6, and two concluding no modifying effect of menopausal status 7, 8. The majority of previous studies investigating the association between soy food intake and breast cancer risk have been case-control studies which are subject to recall and selection biases. The few reports from cohort studies have also been inconsistent9, 10,11 and were limited by small numbers of breast cancer cases 9, 11 or did not include all sources of commonly consumed soy foods 11. Isoflavones, which have a chemical structure similar to 17β-estradiol, are the most abundant phytoestrogen in soy food and have been shown to compete with endogenous estrogen to bind estrogen receptors 12. Isoflavones have weak estrogenic potency and endogenous estrogen level may affect their action; isoflavones exert estrogenic-like effects in estrogen-deprived environments and anti-estrogenic effects when endogenous estrogen levels are high 13. Experimental studies have suggested that soy may be protective against hormone-related cancers14. Breast cancer is a heterogeneous disease and biological differences in subtypes depending on the expression of receptors, such as estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor-2 (HER2) have been well recognized. Several risk factors for subtypes of breast cancer also differ; compared to receptor-negative tumors, ER+/PR+ tumors are more strongly associated with some reproductive factors, such as parity, timing of births, and age at menopause 15, 16. Due to the ability of isoflavones to bind estrogen receptors, the association between soy intake and breast cancer risk may vary by hormone receptor status of the tumor, though few studies have assessed this association 8, 14. Of the 8 case-control studies we found in the literature that considered hormone receptor status as a potential effect modifier, 3 reported that the protective association between soy and breast cancer was consistent across all subtypes of ER/PR tumors 17-19. Two studies found a greater reduction in risk in those with ER+ tumors compared to ER− tumors 20 or ER+/PR+ tumors compared to other ER/PR status 21, while another study found reduced risk for ER+/PR+ and ER−/PR− tumors but not mixed subtypes 22. One study only observed a decreased risk of ER+/PR+ breast cancer with high adolescent intake 23. To our knowledge, only one study evaluated the association by HER2 status, and it reported null results for all types, i.e. ER+, HER2−, and ER+/PR+/HER2-tumors 24. We only identified one prospective study that investigated soy intake and breast cancer risk by ER and/or PR status of tumors and reported no effect modification among postmenopausal women25. This study, however, was limited by small sample size and did not report the results for premenopausal women by hormone receptor status25. We have previously reported the association of adolescent and adult intake of soy with breast cancer risk in the Shanghai Women’s Health Study (SWHS).10 In the present study, we updated the association between soy food intake and breast cancer risk with a longer follow-up time and provided new information on the association by ER, PR, and HER2 status of the breast cancer. MATERIALS AND METHODS Study Population This study utilized the resources generated from the population-based Shanghai Women’s Health Study (SWHS) for which the methodology has been previously described 26, 27. Briefly, 74,942 women aged 40 to 70 years were recruited to the study from seven urban communities in Shanghai, China between 1996 and 2000, with a participation rate of nearly 93%. Information on demographic and lifestyle factors, as well as participant characteristics including reproductive factors, menstrual history, and medical history, was collected using structured questionnaires through in-person interviews by trained interviewers. Using a standard protocol, anthropomorphic measures, including height, weight, and waist and hip circumferences, were taken at the baseline interview. Self-reported weight information was collected at the 3rd and 4th follow-up interviews. Clinical information and tumor characteristics, including hormone receptor status (ER, PR, and HER2), were extracted from medical charts. Written, informed consent was obtained from all participants and the institutional review boards at all participating institutions approved the study. Ascertainment of Breast Cancer Cases Active surveys were conducted every 2-3 years in the SWHS to collect information on occurrence of cancer and other chronic diseases with the following response rates: 1st in-person follow-up, 99.8%; second, 98.7%; third, 96.7%, and fourth, 92%. Annual record linkage to the population-based Shanghai Cancer Registry was used to identify cancer cases. Women with a first cancer diagnosis of breast cancer (ICD-9 code 174) were defined as cases for this study 28. Cancer diagnosis was verified through in-person visits and review of medical charts obtained from the diagnostic hospital. Dietary Assessment Habitual dietary intake was measured using a validated food frequency questionnaire (FFQ) at baseline and 2-3 years later at the 1st follow-up interview; the latter was completed by 92% of cohort members who were alive at the time of the 1st follow-up. The FFQ used in this study was a comprehensive dietary assessment and was designed to measure the consumption of soy foods commonly consumed in Shanghai, including soymilk, tofu, fresh soy beans, and other soy foods 27. Estimated energy and nutrient intakes, including soy and isoflavones, were calculated by summing products of food intake amount multiplying the nutrient content of the specific food item based on the Chinese Food Composition Tables 2002 29. Statistical Analysis Women were excluded for the present study if they reported a previous cancer diagnosis at baseline (n=1,598), extreme baseline total energy intakes (<500 or ≥3500 kcal/day, n=125), or prior or current hormone replacement therapy (HRT) use (n=2,585) resulting in a total sample size of 70,578 women. Adult habitual dietary intake of soy food was assessed from the baseline and 1st follow-up surveys. Analyses performed used two methods for estimating adult soy intake: (1) the soy protein and soy isoflavone intake based on the baseline survey data only and (2) an average soy protein and isoflavone intake calculated by averaging the soy intake levels from the baseline and 1st follow-up surveys. The latter method was not used for women who developed cancer, diabetes, myocardial infarction, or stroke between the baseline and follow-up surveys because the diagnosis of these conditions may have modified some women’s dietary habits. For these women, the baseline soy measure was used for all analyses of adult intake. Adolescent soy intake was assessed at the baseline interview; women were asked to report their intake of commonly consumed soy foods between the ages of 13 and 15. Baseline variables were evaluated for their association with breast cancer and soy protein intake using generalized linear models for continuous variables and chi-square tests for categorical variables. Soy intake was analyzed using quintiles when sample size allowed, where the lowest quintile served as the reference group, in order to assess the pattern of association with a greater range of intake. Using Cox proportional hazards regression models, the hazard ratios (HR) and 95% confidence intervals (CI) for the association between soy intake and breast cancer risk were estimated. We used age as the time scale for the analyses, where age at enrollment was defined as the entry time and age at diagnosis or censoring was defined as the exit time. Censoring was defined as date of death, last date of follow-up, or the latest date of record linkage. The covariates included in the model were age at enrollment, body mass index (BMI), age at first live birth, physical activity (yes/no), education, family history of breast cancer, season of recruitment, total energy intake, and menopausal status. Menopausal status, which was updated at follow-up interviews, was treated as a time-varying covariate in the analyses. Menopausal status was further evaluated as an effect modifier by stratifying results by menopausal status. Categorical soy intake variables were treated as ordinal variables to evaluate the linear trend. Further analyses stratified by hormonal receptor status were performed. For these analyses, soy intake variables were categorized into tertiles of intake for analysis in each stratified group. The joint effect of average adult and adolescent soy protein intake was evaluated using tertiles of intake. These results were additionally stratified by menopausal status. Results from analyses on soy protein and soy isoflavones were very similar, therefore only the former is included in the tables (see supplemental tables 2 and 3 for soy isoflavone analyses). All statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC). RESULTS Over a median follow-up time of 13.2 years, 1,034 incident breast cancer cases were identified. The average (standard deviation) age at diagnosis was 59.6 (9.3) years. Consistent with the literature on the established risk factors for breast cancer 30 and our previous study,10 breast cancer cases were more likely to have a higher education, higher income, younger age at menarche, older age at menopause, longer total years of menstruation, older age at first live birth, fewer number of live births, and higher BMI compared to non-cases (Supplemental Table 1). Women in the highest quintile of soy protein (median intake 16.4 grams per day) were more likely to have a higher education, to exercise regularly, to be older at menopause onset, to be younger at age of first live birth, to have a family history of breast cancer, to have a higher BMI, to have a higher waist-to-hip ratio, and to consume more overall calories as well as more meats and vegetables. These factors were included in the analysis as covariates. The rate of cigarette smoking was very low in this cohort and was not related to breast cancer risk. Additionally, soy food intake was not related to weight change during the follow-up period. Table 1 shows the association between averaged adult and adolescent soy protein intake and breast cancer risk overall and by menopausal status at diagnosis. These updated results are similar to those previously published.10 Overall, soy protein was associated with a modest decrease in risk of breast cancer. Compared to the lowest quintile of soy intake, the highest quintile of averaged soy protein intake during adulthood was associated with reduced breast cancer risk (HR=0.78; 95% confidence interval (CI): 0.63, 0.97; Ptrend=0.007); the association was predominantly seen in premenopausal women (HR=0.46; 95% CI: 0.29, 0.74; Ptrend=0.004). No significant association was observed among postmenopausal women. Additionally adjusting for adolescent soy intake did not materially change the observed associations (data not shown). Analyses based on baseline soy intake information showed very similar results (data not shown). Adolescent soy intake was not significantly associated with breast cancer risk among premenopausal women; although the HR for the fifth quintile compared to the first quintile was in the direction of reduced risk (HR=0.69; 95% CI: 0.45, 1.04; Ptrend=0.08). No association between adolescent soy intake and postmenopausal breast cancer risk was observed. When stratified by hormone receptor status, HRs for high adult soy intake were below 1.0 for most subtypes, though only one reached statistical significance (Table 2). Compared to women in the lowest tertile of adult average soy protein intake, there was a significantly decreased risk of ER+/PR+ breast cancer among all women (HR=0.75; 95% CI: 0.58, 0.98; Ptrend=0.03), and the association was only significant in postmenopausal women (HR=0.72; 95% CI: 0.53, 0.96; Ptrend=0.02). A significantly decreased risk of ER−/PR− breast cancer was observed for premenopausal women (HR=0.46; 95% CI: 0.22, 0.97; Ptrend=0.04) in the highest tertile of soy intake. Analyses using baseline adult soy intake showed very similar results (data not shown). HRs associated with higher adult soy intake for HER2+ and HER2− breast cancer were both below 1.0 but none of the point estimates were statistically significant. Among premenopausal women, only high soy protein intake in both adulthood and adolescence was significantly associated with the risk (HR=0.53; 95% CI: 0.32, 0.88) (Table 3). Among postmenopausal women, high soy protein intake during adulthood was associated with a significantly decreased risk of breast only when adolescent intake was low (HR=0.63; 95% CI: 0.43, 0.91). High adolescent intake alone or high adult and adolescent intake was not significantly associated with the risk among postmenopausal women. None of the tests for multiplicative interaction between adult and adolescent soy intake reached statistical significance. Among breast cancer patients, those who had high soy food intake during both adolescence and adulthood had a later age of cancer diagnosis (mean age=61.3, SD=8.2) than cases who had low level soy food intake during both periods (mean age=57.5, SD=9.1) (p=0.0001). High consumption during adulthood alone (mean age=59.6, SD=8.5) and during adolescence alone (median age=61.7, SD=11.0) were also related to delayed age at cancer diagnosis. DISCUSSION In this population-based cohort study, we found a decreased risk of breast cancer with high soy food intake among all women, particularly in premenopausal women, consistent with several previous reports, 5, 31, 32 including our own previous study.10 We found that the significant association with ER+/PR+ breast cancer was mainly seen among postmenopausal women, while the association with ER−/PR− breast cancer was predominantly confined to premenopausal women. Furthermore, among premenopausal women, high soy intake during both adult and adolescent periods was associated with the biggest reduction in risk; however, among postmenopausal women, high adult soy intake was only associated with a decreased risk of breast cancer when adolescent intake was low. We did not observe the soy and breast cancer risk association to vary by HER2 status. These results suggest that both adult and adolescent soy food intake may be associated with a reduction in risk of breast cancer, and that this association may vary by menopausal status and ER/PR status. Several systematic reviews and meta-analyses have been conducted to summarize the literature on the association between soy intake and breast cancer risks 5-8. One meta-analysis, which included 12 case-control studies and 6 cohort or nested case-control studies, concluded that the association was strongest in premenopausal women; however, many of the included studies were not designed to assess the association between breast cancer and soy intake and the studies differed in the level of control for confounding 5. Another meta-analysis restricted to prospective studies observed the strongest association in postmenopausal women; however, the studies included were highly heterogeneous which resulted in difficult interpretations of the results 6. A more recent meta-analysis additionally considered the region and type of study conducted and found an association between soy intake and breast cancer risk in Asian countries, but not Western countries, and found no modifying effect of menopausal status; however, there was heterogeneity among the included studies and varying dietary intake measures 7. In our study we observed that the association between soy intake and reduced breast cancer risk was predominantly seen in premenopausal women. This finding is supported by studies that have shown that genistein, an isoflavone found in soy foods, inhibits growth of breast cancer cells in culture when endogenous estrogen levels are high33. None of the meta-analyses evaluated the potential modifying effect of hormone receptor status although this has been evaluated in a few previous studies with inconsistent findings. The majority of the previous studies were case-control studies and had several limitations, such as including a small number of breast cancer cases available for analysis 18, 20, low soy intake in the population studied 17, 20, 23, 25, lack of a comprehensive soy intake assessment 18, 24, or recall bias 17-24. The large sample size in our study allowed for analyses stratified by both ER/PR status and menopausal status. Our findings that soy intake may be associated with a reduction in risk of different breast cancer subtypes in premenopausal and postmenopausal women offers a potential explanation for the inconsistent findings of previous studies. Soy food is a rich source of isoflavones, which are structurally similar to estrogens; studies have shown that isoflavones compete with endogenous estrogen for the estrogen receptor making it biologically plausible that isoflavones protect against breast cancer development 12. Other anti-cancer properties of soy food, such as anti-angiogenic effects, anti-proliferative effects, antioxidative DNA topoisomerase I and II inhibition, as well as tyrosine kinases, and proteases effects 12, 14, may also explain the inverse association between isoflavones and breast cancer risk, particularly for estrogen receptor negative breast cancer. The reduced risk of breast cancer associated with soy intake has primarily been observed in Asian populations, with studies in Western populations generally showing no association,6 which is likely due to the lower level of soy food consumption as the soy intake levels in Asian populations are considerably higher than that of Western populations. On the other hand, several studies have also suggested that early or lifetime soy exposure may explain this discrepancy 10, 34, 35. In our study, we found only adolescent and adult soy intake in conjunction were significantly associated with reduced breast cancer risk among premenopausal women, suggesting a relatively long period of exposure may be required for soy food consumption to exert its protective effect. Although our finding that only high adult soy food intake alone was associated with postmenopausal breast cancer risk appears to be contradictory to the long-term exposure hypothesis at first glance, it is likely that soyfood, like most cancer preventive agents, would only be able to prevent some but not all breast cancer in women. Under the long-term exposure hypothesis, these “soy responsive” breast cancers would be prevented when women had sufficient exposure, e.g., high adolescent plus high adulthood exposure for pre-menopausal women, and high adulthood exposure for postmenopausal women. Because “soy responsive” breast cancer would have been already prevented during the pre-menopausal period for women with both high adolescent and adulthood soyfood intake, no additional benefit would be observed among post-menopausal women. This may also explain the null association observed in western populations as soy food consumption has only become popular in recent years. It is noteworthy that we found that soy food consumption during adolescence and adulthood were both related to late age at cancer diagnosis among breast cancer patients. This observation adds to the evidence supporting a causal association between soy food intake and breast cancer risk. More research is warranted to understand the underlying biological mechanisms. Diets high in soy food may be associated with healthier dietary and lifestyle behaviors. These dietary and other associated lifestyle factors which are associated with soy intake and breast cancer risk, could contribute to the observed inverse association seen in our population. We have carefully adjusted for a wide range of dietary and other lifestyle variables in our study although residual confounding can’t be completely ruled out. Additional adjustment for dietary pattern in our study did not change the study results. For some subgroup analyses, particularly those by HER2 status and mixed ER/PR status, the statistical power of the study was low. We had to analyze the data using tertile categorization which prevented us from investigating more extreme soy food intake in these specific groups of women. Additionally, some measurement errors, with the ER, PR, and HER2 status information obtained from multiple hospitals and a long lag time for recalling adolescent soy food intake, may have biased our results towards the null. The strengths of our study include its prospective design, large sample size, and low lost to follow-up rate. Adult soy intake was assessed using a validated, culturally appropriate FFQ designed to capture usual soy intake in our population. Multiple dietary assessments (at baseline and at the 1st follow-up survey) improved our soy intake measurement. In conclusion, we found that soy food intake was inversely associated with breast cancer risk, particularly in premenopausal women. The association between soy food intake and breast cancer risk may be modified by menopausal status and hormone receptor status. Supplementary Material Supplementary Material ACKNOWLEDGEMENTS The authors wish to thank the participants and research staff of the Shanghai Women’s Health Study for their contribution to the study. Funding: This work was supported by the United States National Institutes of Health (R37 CA070867). Michelle Baglia is funded by a CTSA TL1 fellowship. Abbreviations ER estrogen receptor PR progesterone receptor HER2 human epidermal growth factor-2 BMI body mass index Table 1 Adult and Adolescent Soy Protein Intake and Breast Cancer Risk By Menopausal Status Soy Protein Intake Level (RRa,b (95%CI)) Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Ptrend ADULT INTAKE (median(g/day)) 3.5 6.0 8.2 10.9 16.0 Overall 1.00 (reference) 1.01 (0.83, 1.22) 1.00 (0.82, 1.21) 0.87 (0.71, 1.06) 0.78 (0.63, 0.97) 0.007 N(cases) 212 222 222 196 182 Premenopausalc (N(cases)=273) 1.00 (reference) 0.97 (0.69, 1.36) 0.86 (0.60, 1.24) 0.98 (0.68, 1.42) 0.46 (0.29, 0.74) 0.004 N(cases) 68 65 54 59 27 Postmenopausalc (N(cases)=761) 1.00 (reference) 1.03 (0.82, 1.30) 1.06 (0.84, 1.33) 0.83 (0.65, 1.06) 0.90 (0.71, 1.16) 0.15 N(cases) 144 157 168 137 155 ADOLESCENT INTAKE (median(g/day)) 1.8 3.9 6.2 9.2 15.6 Overall 1.00 (reference) 1.04 (0.86, 1.26) 1.12 (0.92, 1.35) 1.04 (0.86, 1.26) 0.95 (0.77, 1.16) 0.43 N(cases) 198 213 226 212 185 Premenopausalc 1.00 (reference) 0.77 (0.54, 1.11) 0.96 (0.68, 1.36) 0.74 (0.51, 1.07) 0.69 (0.45, 1.04) 0.08 N(cases) 63 55 69 51 35 Postmenopausalc 1.00 (reference) 1.17 (0.93, 1.47) 1.19 (0.94, 1.49) 1.18 (0.94, 1.49) 1.06 (0.84, 1.34) 1.00 N(cases) 135 158 157 161 150 a Cox proportional hazards models adjusted for age, body mass index, age at first live birth, physical activity, education, family history of breast cancer, season of recruitment, and menopause (time-varying) were used for analyses b Adult intakes additionally adjusted for total energy intake and juvenile intakes adjusted for total juvenile rice intake c Stratified based on menopausal status at breast cancer diagnosis Table 2 Adult and Adolescent Soy Protein Intake and Breast Cancer Risk By Menopausal Status and Receptor Status Adult Soy Protein Intake Level (RRa,b (95%CI)) Adolescent Soy Protein Intake Level (RRa,b (95%CI)) Tertile 1 Tertile 2 Tertile 3 Tertile 1 Tertile 2 Tertile 3 median=4.5g/day median=8.2g/day median=13.5g/day P trend median=2.6g/day median=6.2g/day median=12.5g/day P trend OVERALL ER+ (N(cases)=550) 1.00 (ref) 0.96 (0.78, 1.18) 0.86 (0.68, 1.07) 0.16 1.00 (ref) 1.28 (1.04, 1.57) 1.09 (0.87, 1.35) 0.76 ER− (N(cases)=288) 1.00 (ref) 1.25 (0.95, 1.66) 0.87 (0.63, 1.21) 0.32 1.00 (ref) 1.13 (0.86, 1.49) 0.98 (0.73, 1.32) 0.77 ER+/PR+ (N(cases)=409) 1.00 (ref) 0.91 (0.72, 1.15) 0.75 (0.58, 0.98) 0.03 1.00 (ref) 1.32 (1.03, 1.68) 1.16 (0.90, 1.48) 0.46 ER−/PR− (N(cases)=246) 1.00 (ref) 1.20 (0.89, 1.62) 0.83 (0.59, 1.18) 0.25 1.00 (ref) 1.12 (0.82, 1.52) 1.12 (0.82, 1.54) 0.52 ER+/PR− (N(cases)=124) 1.00 (ref) 1.18 (0.75, 1.85) 1.27 (0.78, 2.05) 0.35 1.00 (ref) 1.05 (0.69, 1.59) 0.81 (0.51, 1.28) 0.32 HER2+ (N(cases)=158) 1.00 (ref) 1.04 (0.72, 1.51) 0.79 (0.51, 1.21) 0.26 1.00 (ref) 1.23 (0.85, 1.77) 0.79 (0.52, 1.19) 0.17 HER2− (N(cases)=434) 1.00 (ref) 0.92 (0.73, 1.16) 0.83 (0.65, 1.07) 0.15 1.00 (ref) 1.11 (0.88, 1.40) 1.13 (0.89, 1.43) 0.37 PREMENOPAUSAL ER+ (N(cases)=135) 1.00 (ref) 1.00 (0.66, 1.50) 0.91 (0.57, 1.44) 0.67 1.00 (ref) 1.10 (0.74, 1.63) 0.86 (0.55, 1.35) 0.46 ER− (N(cases)=88) 1.00 (ref) 0.91 (0.56, 1.46) 0.60 (0.33, 1.11) 0.11 1.00 (ref) 1.31 (0.81, 2.11) 0.77 (0.43, 1.39) 0.30 ER+/PR+ (N(cases)=103) 1.00 (ref) 0.92 (0.58, 1.48) 0.91 (0.54, 1.52) 0.72 1.00 (ref) 1.01 (0.64, 1.61) 1.01 (0.62, 1.66) 0.97 ER−/PR− (N(cases)=68) 1.00 (ref) 0.85 (0.50, 1.44) 0.46 (0.22, 0.97) 0.04 1.00 (ref) 1.30 (0.73, 2.30) 1.09 (0.58, 2.05) 0.89 ER+/PR− (N(cases)=30) 1.00 (ref) 1.32 (0.56, 3.12) 1.04 (0.38, 2.83) 0.97 1.00 (ref) 1.48 (0.68, 3.25) 0.33 (0.09, 1.21) 0.08 HER2+ (N(cases)=49) 1.00 (ref) 1.07 (0.56, 2.05) 0.73 (0.33, 1.61) 0.44 1.00 (ref) 1.43 (0.75, 2.72) 0.64 (0.28, 1.47) 0.22 HER2− (N(cases)=122) 1.00 (ref) 0.90 (0.58, 1.38) 0.93 (0.58, 1.51) 0.78 1.00 (ref) 0.84 (0.55, 1.27) 0.84 (0.54, 1.32) 0.49 POSTMENOPAUSAL ER+ (N(cases)=415) 1.00 (ref) 0.95 (0.75, 1.21) 0.84 (0.65, 1.09) 0.19 1.00 (ref) 1.35 (1.06, 1.72) 1.16 (0.90, 1.49) 0.48 ER− (N(cases)=200) 1.00 (ref) 1.52 (1.07, 2.15) 1.05 (0.71, 1.57) 0.90 1.00 (ref) 1.05 (0.75, 1.48) 1.06 (0.75, 1.50) 0.76 ER+/PR+ (N(cases)=306) 1.00 (ref) 0.90 (0.69, 1.19) 0.72 (0.53, 0.96) 0.02 1.00 (ref) 1.45 (1.09, 1.92) 1.21 (0.90, 1.62) 0.44 ER−/PR− (N(cases)=178) 1.00 (ref) 1.44 (1.00, 2.08) 1.04 (0.69, 1.58) 0.91 1.00 (ref) 1.05 (0.73, 1.51) 1.13 (0.78, 1.62) 0.52 ER+/PR− (N(cases)=94) 1.00 (ref) 1.14 (0.66, 1.94) 1.35 (0.78, 2.33) 0.28 1.00 (ref) 0.90 (0.55, 1.48) 0.94 (0.57, 1.54) 0.84 HER2+ (N(cases)=109) 1.00 (ref) 1.02 (0.65, 1.61) 0.81 (0.49, 1.36) 0.40 1.00 (ref) 1.12 (0.72, 1.75) 0.83 (0.51, 1.34) 0.37 HER2− (N(cases)=312) 1.00 (ref) 0.94 (0.71, 1.23) 0.80 (0.60, 1.08) 0.14 1.00 (ref) 1.26 (0.95, 1.66) 1.26 (0.95, 1.68) 0.15 a Cox proportional hazards models adjusted for age, body mass index, age at first live birth, physical activity, education, family history of breast cancer, season of recruitment, and menopause (time-varying) were used for analyses b Adult intakes additionally adjusted for total energy intake and juvenile intakes adjusted for total juvenile rice intake c Stratified based on menopausal status at breast cancer diagnosis Table 3 Joint Effect of Adult and Adolescent Soy Protein Intake on Breast Cancer Risk Adult Average Soy Protein Intake (RRa (95%CI)) Juvenile Soy Protein Intake ≤6.32 6.33-10.43 ≥10.44 Premenopausalb (N(cases)=273)    ≤4.24 1.00 (reference) 0.62 (0.38, 1.00) 0.56 (0.31, 1.02)    4.25-8.61 0.86 (0.58, 1.28) 0.86 (0.57, 1.29) 0.78 (0.49, 1.25)    ≥8.62 0.56 (0.31, 1.00) 0.80 (0.51, 1.26) 0.53 (0.32, 0.88) Postmenopausalb (N(cases)=761)    ≤4.24 1.00 (reference) 0.96 (0.72, 1.29) 0.63 (0.43, 0.91)    4.25-8.61 1.07 (0.81, 1.43) 1.08 (0.82, 1.43) 0.94 (0.69, 1.26)    ≥8.62 0.94 (0.67, 1.32) 1.01 (0.76, 1.35) 0.98 (0.74, 1.29) a Cox proportional hazards models adjusted for age, body mass index, age at first live birth, physical activity, education, family history of breast cancer, season of recruitment, total adult energy, total juvenile rice intake, and menopause (time-varying) were used for analyses b Stratified based on menopausal status at breast cancer diagnosis Novelty and Impact: Epidemiological evidence is inconsistent and limited with regard to whether the soyfood and breast cancer risk association differ by hormone receptors and HER2 status. In a large cohort study, we found that soyfood intake was associated with both ER/PR positive and negative breast cancer risk but the association differed by menopausal status. No modification by HER2 status was observed. These results suggest that soyfood may influence breast cancer risk via multiple mechanisms. The authors have no conflicts of interest to disclose. WZ, XOS and YTG designed research, MLB analyzed data, MLB and XOS wrote the paper, WZ and GY provided critical review, HL, JG, and GY supervised field operation and cancer confirmation, MLB and XOS had primary responsibility for final content. All authors read and approved the final manuscript. REFERENCES 1 Ferlay J Shin HR Bray F Forman D Mathers C Parkin DM Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008 Int J Cancer 2010 127 2893 917 21351269 2 Parkin DM Muir CS Cancer Incidence in Five Continents. Comparability and quality of data IARC scientific publications 1992 45 173 1284606 3 Ziegler RG Hoover RN Pike MC Hildesheim A Nomura AM West DW Wu-Williams AH Kolonel LN Horn-Ross PL Rosenthal JF Hyer MB Migration patterns and breast cancer risk in Asian-American women J Natl Cancer Inst 1993 85 1819 27 8230262 4 Wu AH Ziegler RG Horn-Ross PL Nomura AM West DW Kolonel LN Rosenthal JF Hoover RN Pike MC Tofu and risk of breast cancer in Asian-Americans Cancer Epidemiol Biomarkers Prev 1996 5 901 6 8922298 5 Trock BJ Hilakivi-Clarke L Clarke R Meta-analysis of soy intake and breast cancer risk J Natl Cancer Inst 2006 98 459 71 16595782 6 Dong JY Qin LQ Soy isoflavones consumption and risk of breast cancer incidence or recurrence: a meta-analysis of prospective studies Breast Cancer Res Treat 2011 125 315 23 21113655 7 Chen M Rao Y Zheng Y Wei S Li Y Guo T Yin P Association between soy isoflavone intake and breast cancer risk for pre- and post-menopausal women: a meta-analysis of epidemiological studies PLoS One 2014 9 e89288 24586662 8 Fritz H Seely D Flower G Skidmore B Fernandes R Vadeboncoeur S Kennedy D Cooley K Wong R Sagar S Sabri E Fergusson D Soy, red clover, and isoflavones and breast cancer: a systematic review PLoS One 2013 8 e81968 24312387 9 Yamamoto S Sobue T Kobayashi M Sasaki S Tsugane S Soy, isoflavones, and breast cancer risk in Japan J Natl Cancer Inst 2003 95 906 13 12813174 10 Lee SA Shu XO Li H Yang G Cai H Wen W Ji BT Gao J Gao YT Zheng W Adolescent and adult soy food intake and breast cancer risk: results from the Shanghai Women’s Health Study Am J Clin Nutr 2009 89 1920 6 19403632 11 Nishio K Niwa Y Toyoshima H Tamakoshi K Kondo T Yatsuya H Yamamoto A Suzuki S Tokudome S Lin Y Wakai K Hamajima N Consumption of soy foods and the risk of breast cancer: findings from the Japan Collaborative Cohort (JACC) Study Cancer Causes Control 2007 18 801 8 17619154 12 Adlercreutz H Mazur W Phyto-oestrogens and Western diseases Annals of medicine 1997 29 95 120 9187225 13 Taylor CK Levy RM Elliott JC Burnett BP The effect of genistein aglycone on cancer and cancer risk: a review of in vitro, preclinical, and clinical studies Nutrition reviews 2009 67 398 415 19566600 14 Messina M McCaskill-Stevens W Lampe JW Addressing the soy and breast cancer relationship: review, commentary, and workshop proceedings J Natl Cancer Inst 2006 98 1275 84 16985246 15 Colditz GA Rosner BA Chen WY Holmes MD Hankinson SE Risk factors for breast cancer according to estrogen and progesterone receptor status J Natl Cancer Inst 2004 96 218 28 14759989 16 Huang WY Newman B Millikan RC Schell MJ Hulka BS Moorman PG Hormone-related factors and risk of breast cancer in relation to estrogen receptor and progesterone receptor status Am J Epidemiol 2000 151 703 14 10752798 17 Zhang C Ho SC Lin F Cheng S Fu J Chen Y Soy product and isoflavone intake and breast cancer risk defined by hormone receptor status Cancer science 2010 101 501 7 19860847 18 Cho YA Kim J Park KS Lim SY Shin A Sung MK Ro J Effect of dietary soy intake on breast cancer risk according to menopause and hormone receptor status Eur J Clin Nutr 2010 64 924 32 20571498 19 Iwasaki M Hamada GS Nishimoto IN Netto MM Motola J Jr. Laginha FM Kasuga Y Yokoyama S Onuma H Nishimura H Kusama R Kobayashi M Dietary isoflavone intake and breast cancer risk in case-control studies in Japanese, Japanese Brazilians, and non-Japanese Brazilians Breast Cancer Res Treat 2009 116 401 11 18777206 20 Touillaud MS Pillow PC Jakovljevic J Bondy ML Singletary SE Li D Chang S Effect of dietary intake of phytoestrogens on estrogen receptor status in premenopausal women with breast cancer Nutrition and cancer 2005 51 162 9 15860438 21 Dai Q Shu XO Jin F Potter JD Kushi LH Teas J Gao YT Zheng W Population-based case-control study of soyfood intake and breast cancer risk in Shanghai British journal of cancer 2001 85 372 8 11487268 22 Zhang M Yang H Holman CD Dietary intake of isoflavones and breast cancer risk by estrogen and progesterone receptor status Breast Cancer Res Treat 2009 118 553 63 19252980 23 Anderson LN Cotterchio M Boucher BA Kreiger N Phytoestrogen intake from foods, during adolescence and adulthood, and risk of breast cancer by estrogen and progesterone receptor tumor subgroup among Ontario women Int J Cancer 2013 132 1683 92 22907507 24 Suzuki T Matsuo K Tsunoda N Hirose K Hiraki A Kawase T Yamashita T Iwata H Tanaka H Tajima K Effect of soybean on breast cancer according to receptor status: a case-control study in Japan Int J Cancer 2008 123 1674 80 18623079 25 Wu AH Koh WP Wang R Lee HP Yu MC Soy intake and breast cancer risk in Singapore Chinese Health Study British journal of cancer 2008 99 196 200 18594543 26 Zheng W Chow WH Yang G Jin F Rothman N Blair A Li HL Wen W Ji BT Li Q Shu XO Gao YT The Shanghai Women’s Health Study: rationale, study design, and baseline characteristics Am J Epidemiol 2005 162 1123 31 16236996 27 Shu XO Yang G Jin F Liu D Kushi L Wen W Gao YT Zheng W Validity and reproducibility of the food frequency questionnaire used in the Shanghai Women’s Health Study Eur J Clin Nutr 2004 58 17 23 14679362 28 International Classification of Diseases, 9th revision National Center for Health Statistics 2008 29 Yang Y Wang GY Pan XC Chinese Food Composition Tables 2002 Beijing University Medical Press 2002 30 McPherson K Steel CM Dixon JM ABC of breast diseases. Breast cancer-epidemiology, risk factors, and genetics BMJ 2000 321 624 8 10977847 31 Lee HP Gourley L Duffy SW Esteve J Lee J Day NE Risk factors for breast cancer by age and menopausal status: a case-control study in Singapore Cancer Causes Control 1992 3 313 22 1617118 32 Hirose K Tajima K Hamajima N Inoue M Takezaki T Kuroishi T Yoshida M Tokudome S A large-scale, hospital-based case-control study of risk factors of breast cancer according to menopausal status Japanese journal of cancer research : Gann 1995 86 146 54 7730137 33 Lamartiniere CA Protection against breast cancer with genistein: a component of soy Am J Clin Nutr 2000 71 1705S 7S discussion 8S-9S 10837323 34 Wu AH Wan P Hankin J Tseng CC Yu MC Pike MC Adolescent and adult soy intake and risk of breast cancer in Asian-Americans Carcinogenesis 2002 23 1491 6 12189192 35 Shu XO Jin F Dai Q Wen W Potter JD Kushi LH Ruan Z Gao YT Zheng W Soyfood intake during adolescence and subsequent risk of breast cancer among Chinese women Cancer Epidemiol Biomarkers Prev 2001 10 483 8 11352858

PMC005xxxxxx/PMC5114024.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2984170R 7706 Tetrahedron Tetrahedron Tetrahedron 0040-4020 27867228 5114024 10.1016/j.tet.2012.04.025 NIHMS798098 Article Simplexolides A–E and plakorfuran A, six butyrate derived polyketides from the marine sponge Plakortis simplex Liu Xiang-Fang ab Shen Yang b* Yang Fan ac Hamann Mark T. c Jiao Wei-Hua a Zhang Hong-Jun a Chen Wan-Sheng a Lin Hou-Wen a* a Laboratory of Marine Drugs, Department of Pharmacy, Changzheng Hospital, Second Military Medical University, Shanghai 200003, People's Republic of China b Department of Pharmacy, Shanghai Children's Hospital, Shanghai Jiao Tong University, 1400 West Beijing Road, Shanghai 200040, People's Republic of China c Department of Pharmacognosy and the National Center for Natural Products Research (NCNPR), School of Pharmacy, The University of Mississippi, University, MS 38677, USA * Corresponding authors. Tel.: +86 21 62792098 (Y.S.); tel./fax: +86 21 65585154 (H.-W.L.); shenyang@medmail.com.cn (Y. Shen), franklin67@126.com (H.-W. Lin) 28 6 2016 14 4 2012 17 6 2012 17 11 2016 68 24 46354640 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Six new polyketides, simplexolides A–E (1–5) and a furan ester, plakorfuran A (6), together with four known furanylidenic methyl esters (7–10) were isolated from the marine sponge Plakortis simplex. Compounds 1–5 feature a tetrahydrofuran ring opened seco-plakortone skeleton. These new structures, including relative configurations, were determined on the basis of extensive analysis of spectroscopic data. The absolute configurations of 1–6 were established by the modified Mosher's method, and the CD exciton chirality method. However, configurations of the remote stereocenters at C-8 in compounds 1–5 were not determined. Antifungal, cytotoxicity, antileismanial, and antimalarial activities of these poly-ketides were evaluated. Plakortis simplex Simplexolide Mosher's method Antifungal Cytotoxicity Antimalarial Absolute configuration 1. Introduction Sponges of the genera Plakortis and Plakinastrella have been widely investigated for their biologically active polyketides.1–8 Apart from a large number of cyclic peroxides they also contain a group of γ-lactones, including bicyclic peroxylactones (plakortolides),9,10 bicyclic furanolactones (plakortones),8,11 N-alkylated lactones (amphiasterins),12 α,β-unsaturated lactones (butenolides),13 and some seco derivatives (seco-plakortolides).8 Plakortones A–F, featuring bicyclic furanolactones, are a class of ethyl branched (butyrate derived) polyketides from the genus Plakortis. They show interesting biological activities that have been the focus of extensive chemical and pharmaceutical studies. Plakortones A–D are cardiac sacroplasmic reticulum Ca2+-pumping ATPase activators, being active at micromolar concentrations;11 plakortones B–F exhibited moderate cytotoxicity against the murine fibro sarcoma cell line WEHI 164;14 plakortone G, a closely related analogof the γ-lactone isolated by Stierle and Faulkner,15 without a tetrahydrofuran moiety, showed antimalarial activity against both the D6 and W2 clones of Plasmodium falciparum.16 The absolute configurations of these plakortones were not confirmed during the initial isolation. The total syntheses of plakortones B, D, and E helped to establish their absolute configurations later.17–19 However, the absolute configurations of the other four plakortones A, C, F, and G still remain unknown. Recently, plakortones L, N, and P were isolated from the sponge of Plakinastrella clathrata.8 They have bicyclic furanolactone fragments like plakortones A–F, but with methyl rather than ethyl substituents. During the course of our continuing search for new drug leads from marine sponges collected off the Xisha Islands in the South China Sea, besides the previously reported two unusual polyketides simplextones A and B,20 we have recently isolated and identified six new polyketides, simplexolides A–E (1–5) and plakorfuran A (6), together with four known furanylidenic methyl esters (2Z,6R,8R,9E) [3-ethyl-5-(2-ethyl-hex-3-enyl)-6-ethyl-5H-furan-2-ylidene]-acetic acid methyl ester (7),15 methyl (2Z,6R,8S)-4,6-diethyl-3,6-epoxy-8-methyldeca-2,4-dienoate (8),21 methyl (2Z,6R,8S)-3,6-epoxy-4,6,8-triethyldodeca-2,4-dienoate (9),21 and (2Z,6R,8R,9E)[3-ethyl-5-(2-ethyl-hex-3-enyl)-6-methyl-5H-furan-2-ylidene]-acetic acid methyl ester (10)22 from the marine sponge Plakortis simplex. The primary difference between compounds 1–5 and the previously isolated plakortones A–F from Plakortis spp. is the opening of the tetrahydrofuran ring in 1–5. Herein, we describe the isolation, structure elucidation, and initial biological evaluation of compounds 1–6. 2. Results and discussion A sample of the sponge P. simplex, collected off the Xisha Islands in the South China Sea, was exhaustively extracted with MeOH. The extract was suspended in water and successively extracted with n-hexane, CH2Cl2, EtOAc, and n-BuOH. The CH2Cl2-soluble extract was subjected to repeated column chromatography followed by reversed-phase preparative HPLC to afford compounds 1–10 as colorless oils. The molecular formula of compound 1 was assigned as C17H30O3 by its HRESIMS (m/z 305.2094, [M+Na]+) and NMR data (Tables 1 and 2), displaying three degrees of unsaturation. The IR absorptions indicated the presence of hydroxyl (3451 cm−1), double bond (1655 cm−1), and ester carbonyl (1763 cm−1) groups. Analysis of the 1H NMR data (Table 1) and HSQC spectrum revealed the presence of four methyls, seven methylenes, two sp3 methines, one of which was oxygenated, one sp2 methine, one oxygenated sp3 quaternary carbon, and two sp2 quaternary carbons. By interpretation of 1H–1H COSY correlations, it was possible to establish five partial structures of consecutive proton systems: H-2/H-3, H-7/H-8/Me-17, H-11/Me-12, H-13/Me-14, and H-15/Me-16 (Fig. 1). The HMBC correlations from Me-14 (δH 1.01) to C-4 (δC 93.3), from H-13a (δH 1.64) to C-3 (δC 72.4) and C-5 (δC 120.1), and from H-3 (δH 4.22) to C-1 (δC 174.9) indicated the attachment of an ethyl group at C-4. Moreover, the HMBC correlations from Me-16 (δH 1.03) to C-6 (δC 93.3) and from H-15 (δH 2.13) to C-5 (δC 120.1), C-6 (δC 148.6), and C-7 (δC 44.9) demonstrated the linkage of C-5, C-7, and C-15 via C-6. The HMBC correlations from Me-17 (δH 0.85) to C-9 (δC 36.5), from H-9a (δH 1.11) to C-11 (δC 22.9), and from Me-12 (δH 0.89) to C-10 (δC 29.2) defined the structure of the C-1 to C-12 portion of the molecule. The C-1 ester carbonyl was confirmed to form a γ-butyrolactone with the oxygenated C-4, which was severely downfield shifted at δC 93.3, to consume the remaining one degree of unsaturation. The relative configuration of 1 was established on the basis of NOESY data (Fig. 1). The crucial NOE correlations between H-3 (δH 4.22) and H-13a (δH 1.64) suggested that these protons were oriented on the same face of the γ-butyrolactone moiety. The E-geometry of the Δ5,6 double bond was deduced from a NOESY correlation between H-5 (δH 5.09) and H-7a (δH 1.87). The absolute configuration of C-3 was determined by applying the modified Mosher's method to the secondary hydroxyl group.23,24 Compound 1 was reacted with (R)-(−)- and (S)-(+)- α -methoxy- α -tri-fluoromethylphenylacetic acid (MTPA) chlorides to give MTPA esters 1a and 1b, respectively. A consistent distribution of positive and negative Δδ values around C-3 allowed the assignment of S-configuration for C-3 (Fig. 2). On the basis of the previously determined relative configuration, a 3S,4S configuration was assigned to simplexolide A (1). However, configuration of the remote stereocenter at C-8 was not determined. Simplexolide B (2) was clearly an isomer of compound 1 on the basis of the identical molecular formula of C17H30O3 obtained by HRESIMS (m/z 305.2095 [M+Na]+). Comparison of the NMR data between 2 and 1 (Table 1) suggested that they had the same planar structures except for the configuration of the double bond at C-5–C-6, which was assigned as Z geometry on the basis of a NOESY correlation between H-5 and H-15 (Fig. 1). The NOE correlation between H-3 and H-5 indicated the same orientation of these two protons. The absolute configuration at C-3 was determined to be S by the modified Mosher's method (Fig. 2), implying that 1 and 2 were epimeric at C-4. On the basis of above-mentioned analysis, a 3S,4R configuration was assigned to simplexolide B (2). The molecular formula of compound 3 was established as C18H32O3 on the basis of HRESIMS (m/z 319.2247, [M+Na]+) and NMR data. The 1H and 13C NMR data indicated that 3 is a homolog of 2. The 1H NMR spectrum of 3 was almost identical to that of 2 except that a methyl group was replaced by an ethyl group at C-8 (Table 2). The geometry of the Δ5,6 double bond was assigned as E based on the NOESY correlation between H-5 and H-7. The crucial NOE correlation between H-3 and H-5 indicated that the relative configuration was assumed to be the same as 2. The CD spectrum of 3 displayed a positive Cotton effect at 196 nm, which was similar to that of 2 (Fig. 3), suggesting that the absolute configuration of 3 was the same as that of 2. Thus, a 3S,4R configuration was assigned to simplexolide C (3). Compound 4 exhibited a quasi-molecular ion peak at m/z 319.2251 ([M+Na]+, calcd 319.2249), consistent with the molecular formula of C18H32O3. The 1H and 13C NMR data of 4 (Tables 1 and 2) were similar to those of 3. The geometry of the double bond at C-5 was assigned as Z, supported by the NOE correlation between H-5 and H-15. The CD spectrum of 4 revealed a pronounced positive Cotton effect with maxima observed at 196 nm (Fig. 3), and this closely matched the CD spectrum of 3. Therefore, the absolute configuration of simplexolide D (4) was assigned to be 3S,4R, consistent with those of 3. The positive HRESIMS spectrum of simplexolide E (5) exhibited a pseudomolecular ion peak at m/z 305.2091 [M+Na]+, consistent with the molecular formula of C17H30O3, implying three degrees of unsaturation. Analysis of its NMR spectroscopic data revealed nearly identical structural features to those of 2, except for the geometry of the double bond at C-5, which was assigned as E on the basis of an NOE correlation between H-5 and H-7. The CD spectra of 5 showed a positive Cotton effect at 196 nm (Fig. 3), similar to that of 2, suggesting that the absolute configuration of 5 was 3S,4R. The HRESIMS of compound 6 exhibited a pseudomolecular ion peak at m/z 345.2040 [M+Na]+ and established a molecular formula of C19H30O4, indicating five degrees of unsaturation. The 13C NMR (Table 2) displayed 19 carbon signals, which were identified by the assistance of the DEPT spectrum as 5 methyls, 5 methylenes, 3 sp2 methines, 1 oxygenated sp3 methine, and 5 quaternary carbons. The carbon resonances at δC 84.0 (C-6) and 171.4 (C-3) are identical as those for a furano α,β-unsaturated ester.14,15 Furthermore, an ester carbonyl was recognized as being present in 6 from its 13C NMR signal at C 166.8 (qC, C-1) and strong IR absorptions at 1752 cm−1. By interpretation of COSY correlations (Fig. 1), it was possible to establish the proton connections between H-15 and H-16, H-17 and H-18, H-9 and H-10, H-13 and H-14, and between H-11 and H-12. The connectivities of these partial structures were further established by the HMBC correlations (Fig. 1). The conjunction of C-10 and C-11 was elucidated on the basis of the HMBC correlation from H3-12 to C-10. Moreover, the HMBC correlations from H3-18 and H-10 to C-8 indicated the attachment between C-9 and C-17 via C-8. The HMBC correlations observed from H3-16 to C-6, and from H-15 and H-17 to C-7 indicated the connection of C-8 and C-15 through C-6 and C-7. The HMBC correlations of H3-19/C-1, H3-14 and H-2/C-4, H-5/C-3 and C-13, and H-15/C-5 further confirmed the existence of an unsaturated furan ester. With this assignment secured, the final oxymethine at C-10 had to be substituted with a hydroxyl group to satisfy the molecular formula. Finally, the E-geometry of the Δ8,9 double bond was deduced from a NOESY correlation between H-7 and H-9. The crucial NOE correlation between H-2 and H-13 indicated that the geometry of the Δ2,3 double bond is assumed to be Z (Fig. 1). This completed the assignment of the planar structure of plakorfuran A (6). The absolute configuration at C-10 was determined to be R by applying a modified Mosher's ester method to the secondary hydroxyl group (Fig. 2). The absolute configuration at C-6 was determined by applying the CD exciton chirality method.25 The CD spectrum of 6 revealed a negative Cotton effect at λmax 285 nm (Δε −5.49) and a positive Cotton effect at λmax 205 nm (Δε 5.85) due to the transition interaction between two different chromophores of the unsaturated furan ester and the Δ8,9 double bond (Fig. 4), indicating a negative chirality for 6, thus concluding the absolute configuration as 6R,10R. Simplexolides A–E (1–5) feature a previously unknown tetrahydrofuran opened plakortone skeleton. A possible biogenetic pathway is proposed as shown in Scheme 1. Acetate, propionate, and butyrate units are required to assemble the polyketide skeleton.26 The double bond at Δ3,4 is oxidized to an epoxide. With acyl carrier protein (ACP) domain releasing, the carboxylic acid cyclizes onto the opening epoxide togenerate a γ-lactone with the insertion of a hydroxyl.27 In this lactonization step, a pair of stereoisomers are formed at C-4 due to the nucleophilic attack of the hydroperoxy group onto C-4 of the epoxide group from both of the side. Compounds 1, 2, and 5–10 were tested for antifungal, cytotoxicity, antileismanial, and antimalarial activities (Table 3). Compounds 2, 5, and 7–10 showed weak to moderate antifungal activity against the fungi Cryptococcus neoformans. The cytotoxic activity of compounds 1 and 2 against five human cancer cell lines, HCT-116 (colon cancer), HeLa (cervical cancer), SW480 (colon cancer), QGY-7703 (hepatocarcinoma), and A549 (lung carcinoma) was also assayed. The results showed that compound 1 exhibited much stronger cytotoxicity against the five cancer cell lines, which was due to the stereochemistry influenced. Compounds 2 and 10 showed moderate antileismanial activity againt Leishmania donovani, while compounds 7–9 were a little weaker, and compounds 1, 5 and 6 were inactive. Furthermore, the antimalarial activity against chloroquine sensitive (D6, Sierra Leone) and resistant (W2, Indo China) strains of P. falciparum was tested. In this assay, only compound 10 displayed antimalarial activity against D6 and W2 with IC50 values of 2.0 and 2.0 μg/mL, respectively, and both with SI [IC50(fibroblast)/IC50(parasite)] of 2.4, while the other compounds were inactive. 3. Experimental section 3.1. General experimental procedures Optical rotations were determined with a Perkin–Elmer 341 polarimeter equipped with a 1 mm cell. The CD spectra were obtained with a JASCO J-715 spectropolarimeter. IR spectra were recorded on a Bruker Vector 22 spectrometer using KBr pellets. The NMR experiments were conducted on Bruker AVANCE-600 and Bruker AMX-500 MHz instruments in CDCl3 with TMS as an internal standard. HRESIMS and ESIMS were obtained on a Q-Tof micro YA019 mass spectrometer. Reversed-phase HPLC was performed using Sunfire C18 (5 μm) and YMC-Pack Pro C18 RS (5 μm) columns with a Waters 1525/2998 liquid chromatograph. Column chromatography (CC) was carried out on silica gel 60 (200–300 mesh; Yantai, China), and Sephadex LH-20 (Pharmacia). TLC was carried out using HSGF 254 plates and visualized by spraying with anisaldehyde/H2SO4 reagent. Samples were weighed on a Shushi analytical balance. 3.2. Animal material The sponge specimen was collected around Yongxing Island and seven connected islets in the South China Sea in June 2007, and were identified by Prof. Jin-He Li (Institute of Oceanology, Chinese Academy of Sciences, China). A voucher sample (No. B-3) was deposited in the Laboratory of Marine Drugs, Department of Pharmacy, Changzheng Hospital, Second Military Medical University, China. 3.3. Extraction and isolation The air-dried and powdered sponge (2.0 kg, dry weight) was extracted with MeOH, and the crude extract was concentrated under reduced pressure at 45 °C to yield 500 g of residue. The residue was then extracted successively with n-hexane, CH2Cl2, EtOAc, and n-BuOH. The CH2Cl2 extract (41 g) was separated by vacuum liquid chromatography (VLC) on silica gel using CH2Cl2/MeOH as the eluent to give three fractions (Fr. A–C). The lower polarity part fraction A (CH2Cl2/MeOH 25:1) was subjected to VLC eluting with petroleum ether/EtOAc (PE) to give four subfractions (Fr. A1–A4). Fraction A3 was purified by reversed-phase preparative HPLC (Sunfire C18, 5 μm, 10×250 mm) to yield 31.7 mg of compound 1 (CH3OH/H2O 95:5, 2.0 mL/min, UV detection at 210 nm, tR=17.0 min), 22.1 mg of compound 6 (CH3CN/H2O 50:50, 2.0 mL/min, UV detection at 282 nm, tR=17.2 min), a mixture of compounds 3 and 4 (CH3CN/H2O 70:30, 2.0 mL/min, UV detection at 200 nm, tR=29.0 min), and a mixture of compounds 2 and 5 (CH3CN/H2O 70:30, 2.0 mL/min, UV detection at 200 nm, tR=24.4 min). The mixture of compounds 3 and 4 and the mixture of compounds 2 and 5 were further purified by reversed-phase preparative HPLC (YMC-Pack Pro C18 RS, 5 μm, 10×250 mm) to yield 21.3 mg of compound 3 (CH3CN/H2O 85:15, 2.0 mL/min, UV detection at 200 nm, tR=18.4 min), 1.8 mg of compound 5 (CH3CN/H2O 85:15, 2.0 mL/min, UV detection at 200 nm tR=17.4 min), 8.2 mg of compound 2 (CH3CN/H2O 85:15, 2.0 mL/min, UV detection at 200 nm, tR=66.9 min), and 52.1 mg of compound 5 (H3CN/H2O 85:15, 2.0 mL/min, UV detection at 200 nm, tR=62.6 min). Similarly, the four known compounds 7 (22.0 mg), 8 (15.1 mg), 9 (8.3 mg), and 10 (2.2 mg) were obtained from fraction A1. 3.3.1. Simplexolide A(1) Colorless oil; [α]D23+1 (c 0.215, MeOH); 1R (KBr) νmax 3451, 2960, 2927, 2873, 1763, 1655, 1464, 1412, 1378, 1343, 1290, 1248, 1210, 1182, 1119, 1103, 1090, 1071, 1014, 980, 958, 926, 885, 800 cm−1; 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3,150 MHz) data, see Tables 1 and 2; HRESIMS m/z 305.2094 [M+Na]+ (calcd for C17H30O3Na, 305.2093). CD spectrum (c 1.1 mg/mL, CH3CN), 193 nm (Δε -0.65), 222 nm (Δε 0.84). 3.3.2. Simplexolide B (2) Colorless oil; [α]D23+2 (c 0.120, MeOH); IR (KBr) νmax 3439, 2962, 2930, 2874, 1758, 1655, 1460, 1400, 1379, 1341, 1280, 1250, 1208, 1165, 1111, 1063, 1023, 972, 954, 916, 879, 799 cm−1; 1H NMR (CDCl3, 500 MHz) and 13C NMR (CDCl3, 125 MHz) data, see Tables 1 and 2; HRESIMS m/z 305.2095 [M+Na]+ (calcd for C17H30O3Na, 305.2093). CD spectrum (c 0.5 mg/mL, CH3CN), 196 nm (Δε+5.16). 3.3.3. Simplexolide C (3) Colorless oil; [α]D23−9 (c 0.090, MeOH); IR (KBr) νmax 3444, 2962, 2928, 2874, 1758, 1655, 1464, 1399, 1379, 1341, 1287, 1251, 1203, 1164, 1113, 1099, 1022, 975, 951, 915, 879, 800 cm−1; 1H NMR (CDCl3, 500 MHz) and 13C NMR (CDCl3, 125 MHz) data, see Tables 1 and 2; HRESIMS m/z 319.2247 [M+Na]+ (calcd for C18H32O3Na, 319.2249). CD spectrum (c 0.5 mg/mL, CH3CN), 196 nm (Δε +3.43). 3.3.4. Simplexolide D (4) Colorless oil; [α]D23+14 (c 0.130, MeOH); IR (KBr) νmax 3448, 2962, 2930, 2874, 1759, 1655, 1460, 1400, 1379, 1340, 1280, 1250, 1205, 1163, 1111, 1036, 1024, 975, 955, 914, 798cm−1; 1H NMR (CDCl3, 500 MHz) and 13C NMR (CDCl3, 125 MHz) data, see Tables 1 and 2; HRESIMS m/z 319.2251 [M+Na]+ (calcd for C18H32O3Na, 319.2249). CD spectrum (c 0.9 mg/mL, CH3CN), 196 nm (Δε +3.99). 3.3.5. Simplexolide E (5) Colorless oil; [α]D23−18 (c 0.170, MeOH); IR (KBr) νmax 3446, 2961, 2927, 2874, 1758, 1655, 1465, 1400, 1379, 1342, 1288, 1252, 1202, 1166, 1112, 1063, 1021, 974, 951, 916, 879, 800 cm−1; 1H NMR (CDCl3, 500 MHz) and 13C NMR (CDCl3, 125 MHz) data, see Tables 1 and 2; HRESIMS m/z 305.2091 [M+Na]+ (calcd for C17H30O3Na, 305.2093). CD spectrum (c 0.7 mg/mL, CH3CN), 196 nm (Δε + 2.57). 3.3.6. Plakorfuran A (6) Colorless oil; [α]D23−96 (c 0.085, MeOH); 1H NMR(CDCl3, 500 MHz) and 13C NMR(CDCl3,125 MHz) data, see Tables 1 and 2; HRESIMS m/z 345.2040 [M+Na]+ (C19H30O4Na, calcd 345.2042). CD spectrum (c 0.85 mg/mL, CH3CN), 205 nm ((Δε 5.85), 239 nm ((Δε 1.70), 285 nm ((Δε −5.49). 3.4. Preparation of MTPA esters 1a and 1b Simplexolide A (1; 1.0 mg and 0.8 mg, respectively) was reacted with R-(−)- or S-(+)-MTPACl (15 μL) in freshly distilled dry pyridine (500 μL) and stirred under N2 at room temperature for 18 h, respectively, and the solvent was removed in vacuo. The products were purified by mini-column chromatography on silica gel (200 mesh, petroleum ether/EtOAc, 1:1) to afford S-(−)- and R-(+)-MTPA esters 1a and 1b, respectively. 3.5. Preparation of MTPA esters 2a and 2b Simplexolide C (3; 1 mg each) was similarly processed to give S-(−)- and R-(+)-MTPA esters 3a and 3b, respectively. 3.6. Preparation of MTPA esters 6a and 6b Plakorfuran A (6; 1 mg each) was similarly processed to give S-(−)- and R-(+)-MTPA esters 6a and 6b, respectively. 3.7. Biological tests Antifungal assay against C. neoformans was performed as described by Ikhlas A. Khan et al.28 Amphotericin B was used as the positive control. Cytotoxicity was determined against human cancer cell lines HCT-116 (colon cancer), HeLa (cervical cancer), SW480 (colon cancer), QGY-7703 (hepatocarcinoma), and A549 (lung carcinoma) using the MTT assay method. Camptothecin was used as the positive control. The experimental details of this assay were carried out according to a previously described procedure.29 Antimalarial activity was determined in vitro against chloroquine sensitive (D6, Sierra Leone) and resistant (W2, Indo China) strains of P. falciparum by measuring plasmodial LDH activity.30 Chloroquine was used as the positive control. In vitro antileishmanial activity was tested on a culture of L. donovani promastigotes. In a 96-well microplate assay, test compounds were diluted to an appropriate concentration and added to the Leishmania promastigotes culture (2×106 cell/mL). The plates were incubated at 26 °C for 72 h and growth of Leishmania promastigotes was determined by Alamar blue assay.31 Pentamidine and Amphotericin B were used as the standard antileishmanial agents. The growth inhibition curve was used to compute the IC50 value for each compound. Supplementary Material 2 This work was supported by the National Natural Science Foundation of China (No. 81072573), the Major Program of Modernization of Chinese Medicine (STCSM, 09dZ1975800), the NIH, NIAID, Division of AIDS (No. AI 27094), and the USDA Agricultural Research Service Specific Cooperative Agreement (No. 58-6408-2-0009). The authors thank Dr. Melissa R. Jacob and Ms. Marsha A. Wright for antifungal assay assistance, and Dr. Shabana Khan for antimalarial assay assistance. Fig. 1 COSY ( ), key HMBC (→), and selected NOE ( ) correlations of 1,2, and 6. Fig. 2 ΔδS–R values (ppm) for the MTPA derivatives of 1, 2, and 6 in CDCl3. Fig. 3 CD curves of compounds 2–5. Fig. 4 Experimental CD spectrum of 6. Bold lines denote the electric transition dipole of the chromophores for 6. Scheme 1 Plausible biogenetic pathway for simplexolides A–E (1–5). Table 1 1H NMR data of compounds 1–6 in CDCl3 No. 1a 2b 3b 4b 5b 6b 2 2.59, dd (18.0, 1.2) 2.42, d (17.5) 2.43, d (17.5) 2.42, d (17.5) 2.46, d (17.5) 4.80, s 2.85, dd (18.0, 6.0) 2.80, dd (17.5, 5.5) 2.83, dd (17.5, 5.5) 2.82, dd (17.5, 5.5) 2.83, dd (17.5, 5.5) 3 4.22, d (5.4) 4.30, d (5.0) 4.32, t (4.0) 4.34, t (4.0) 4.31, d (5.5) 5 5.09, s 5.14, s 5.04 s 5.15 s 5.03, s 6.22, s 7 1.87, dd (13.8, 8.4) 2.11, dd (14.0, 9.5) 1.95, m 2.15, dd (14.5, 7.5) 1.70, dd (13.5, 9.0) 2.53 d (14.0) 2.17, m 2.21, dd (13.5, 7.0) 2.23, dd (14.5, 7.5) 2.13, dd (13.5, 5.5) 2.36 d (14.0) 8 1.62, m 1.65, m 1.21, m 1.45, m 1.58, m 9 1.11, m 1.14, m 1.41, m 1.17, m 1.10, m 5.17, d (9.0) 1.23, m 1.28, m 1.27, m 1.27, m 10 1.32, m 1.30, m 1.23, m 1.28, m 1.27, m 4.24, dt (9.0, 6.5) 11 1.30, m 1.27, m 1.24, m 1.28, m 1.27, m 1.55, m 1.39, m 12 0.89, t (7.2) 0.90, t (7.5) 0.84, t (7.5) 0.99, t (7.5) 0.89, t (7.5) 0.86, t (7.5) 13 1.64, dq (7.2, 7.2) 1.89, m 1.88, m 1.92, m 1.89, dq (15.0, 7.5) 2.15, m 1.72, dq (7.2, 7.2) 1.99, m 2.02, m 2.02, m 2.03, dq (15.0, 7.5) 14 1.01, t (7.2) 0.95, t (7.5) 0.99, t (7.5) 0.98, t (7.5) 1.01, t (7.5) 1.15, t (7.5) 15 2.13, m 2.04, m 2.10, dq (7.5, 7.5) 2.03, m 2.03, dq (14.0, 7.0) 1.88, m 2.27, dq (7.5, 7.5) 2.33, dq (14.0, 7.0) 1.76, m 16 1.03, t (7.2) 0.99, t (7.5) 1.00, t (7.5) 1.00, t (7.5) 0.98, t (7.5) 0.81, t (7.5) 17 0.85. d (6.6) 0.84. d (6.5) 1.25, m 1.20, m 0.81. d (6.5) 2.09, m 18 0.89, t (7.5) 0.85, t (7.5) 0.97, t (7.5) 19 3.68, s a Recorded at 600 MHz. b Recorded at 500 MHz. Table 2 13 C NMR data of compounds 1–6 in CDCl3 No. 1a 2b 3b 4b 5b 6b 1 174.9 qC 175.7 qC 175.1 qC 175.0 qC 175.6 qC 166.8 qC 2 37.6 CH2 38.8 CH2 38.8 CH2 38.8 CH2 38.9 CH2 84.0 CH 3 72.4 CH 74.4 CH 74.6 CH 74.6 CH 74.5 CH 171.4 qC 4 93.3 qC 92.8 qC 92.3 qC 92.2 qC 92.7 qC 140.3 qC 5 120.1 CH 123.4 CH 124.0 CH 123.5 CH 124.0 CH 139.4 CH 6 148.6 qC 147.3 qC 147.3 qC 147.7 qC 147.0 qC 97.4 qC 7 44.9 CH2 37.5 CH2 41.4 CH2 35.1 CH2 44.7 CH2 43.6 CH2 8 30.8 CH 30.6 CH 32.6 CH 37.0 CH 30.9 CH 139.0 qC 9 36.5 CH2 37.1 CH2 36.8 CH2 33.1 CH2 36.8 CH2 132.9 CH 10 29.2CH2 29.2 CH2 28.8 CH2 29.2 CH2 29.3 CH2 69.6 CH 11 22.9 CH2 22.9 CH2 23.1 CH2 23.1 CH2 23.0 CH2 30.4 CH2 12 14.0 CH3 14.1 CH3 14.1 CH3 14.1 CH3 14.1 CH3 9.7 CH3 13 31.8 CH2 27.4 CH2 27.2 CH2 27.4 CH2 27.1 CH2 18.5 CH2 14 8.1 CH3 8.3 CH3 8.4 CH3 8.4 CH3 8.4 CH3 12.1 CH3 15 23.9 CH2 29.8 CH2 23.4 CH2 30.0 CH2 23.5 CH2 30.9 CH2 16 12.9 CH3 13.0 CH3 12.8 CH3 13.1 CH3 12.8 CH3 8.1 CH3 17 19.6 CH3 19.1 CH3 25.5 CH2 26.1 CH2 19.5 CH3 24.5 CH2 18 10.6 CH3 11.6 CH3 13.6 CH3 19 50.6 CH3 a Recorded at 150 MHz. b Recorded at 125 MHz. Table 3 Biological activities of 1, 2 and 5–10 Compound Antifungal IC50 (μg/ml)/MIC (μg/ml) Cytotoxicity IC50 (μg/ml) Antileismanial IC50 (μg/ml) Antimalarial IC50 (μg/ml) HCT-116 HeLa SW480 QGY-7703 A549 D6 W2 1 – 9.23 27.43 17.08 26.53 16.79 >40 >50 >50 2 10.49/20 4.34 6.51 11.28 17.98 10.77 13.82 >50 >50 5 3.66 >50 – – – – >40 >50 >50 6 – >50 – – – – >40 >50 >50 7 20.00 – – – – – 38.56 >50 >50 8 15.30 – – – – – 31.24 >50 >50 9 20.00 – – – – – 20.20 >50 >50 10 11.72 – – – – – 7.11 2.0 2.0 Amphotericin B 0.317 – – – – – 0.34 – – Camptothecin – 3.22 2.43 8.26 1.41 0.81 – – – Pentamidine – – – – – – 1.62 – – Chloroquine – – – – – – – 0.06 0.83 Supplementary data: Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.tet.2012.04.025. These data include MOL files and InChiKeys of the most important compounds described in this article. References and notes 1 Higgs MD Faulkner DJ J Org Chem 1978 43 3454 3457 2 Braekman JC Daloze D De Groote S Fernandes JB Van Soest RWM J Nat Prod 1998 61 1038 1042 9722495 3 Hu JF Gao HF Kelly M Hamann MT Tetrahedron 2001 57 9379 9383 4 Kossuga MH Nascimento AM Reimao JQ Tempone AG Taniwaki NN Veloso K Ferreira AG Cavalcanti BC Pessoa C Moraes MO Mayer AMS Hajdu E Berlinck RGS J Nat Prod 2008 71 334 339 18177008 5 Feng Y Davis RA Sykes M Avery VM Camp D Quinn RJ J Nat Prod 2010 73 716 719 20235550 6 Rahm F Hayes P Kitching W Heterocycles 2004 64 523 575 7 Qureshi A Salva J Harper MK Faulkner DJ J Nat Prod 1998 61 1539 1542 9868160 8 Yong KWL De Voss JJ Hooper JNA Garson MJ J Nat Prod 2011 74 194 207 21261297 9 Rudi A Afanii R Gravalos LG Aknin M Gaydou E Vacelet J Kashman Y J Nat Prod 2003 66 682 685 12762807 10 Perry TL Dickerson A Khan AA Kondru RK Beratan DN Wipf P Kelly M Hamann MT Tetrahedron 2001 57 1483 1487 11 Patil AD Freyer AJ Bean MF Carte BK Westley JW Johnson RK Lahouratate P Tetrahedron 1996 52 377 394 12 Zampella A Giannini C Debitus C D'Auria MV Tetrahedron 2001 57 257 263 13 De Guzman FS Schmitz FJ J Nat Prod 1990 53 926 931 14 Cafieri F Fattorusso E Taglialatela-Scafati O Ianaro A Tetrahedron 1999 55 13831 13840 15 Stierle DB Faulkner DJ J Org Chem 1980 45 3396 3401 16 Gochfeld DJ Hamann MT J Nat Prod 2001 64 1477 1479 11720540 17 Xie XG Wu XW Lee HK Peng XS Wong HN Chem —Eur J 2010 16 6933 6941 20432418 18 Hayes PY Kitching W J Am Chem Soc 2002 124 9718 9719 12175225 19 Akiyama M Isoda Y Nishimoto M Narazaki M Oka H Kuboki A Ohira S Tetrahedron Lett 2006 47 2287 2290 20 Liu XF Song YL Zhang HJ Yang F Yu HB Jiao WH Piao SJ Chen WS Lin HW Org Lett 2011 13 3154 3157 21604759 21 Compagnone RS Pina IC Rangel HR Dagger F Suarez AI Venkata Rami Reddy M John Faulkner D Tetrahedron 1998 54 3057 3068 22 Epifanio R Pinheiro L Alves N J Braz Chem Soc 2005 16 1367 1371 23 Ohtani I Kusumi T Kashman Y Kakisawa H J Am Chem Soc 1991 113 4092 4096 24 Hoye TR Jeffrey CS Shao F Nat Protoc 2007 2 2451 2458 17947986 25 Cai G Bozhkova N Odingo J Berova N Nakanishi K J Am Chem Soc 1993 115 7192 7198 26 Huang XH van Soest R Roberge M Andersen R J Org Lett 2003 6 75 78 27 Richard GB Mary JG James S J Chem Soc, Chem Commun 1980 1165 1167 28 Ma G Khan SI Jacob MR Tekwani BL Li Z Pasco DS Walker LA Khan IA Antimicrob Agents Chemother 2004 48 4450 4452 15504880 29 Carmichael J De Graff WG Gazdar AF Minna JD Mitchell JB Cancer Res 1987 47 936 942 3802100 30 Muhammad I Bedir E Khan SI Tekwani BL Khan IA Takamatsu S Pelletier J Walker LA J Nat Prod 2004 67 772 777 15165136 31 Mikus J Steverding D Parasitol Int 2000 48 265 269 11227767

PMC005xxxxxx/PMC5114165.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101282001 33781 Brain Struct Funct Brain Struct Funct Brain structure & function 1863-2653 1863-2661 26983815 5114165 10.1007/s00429-016-1208-y NIHMS770542 Article Exposure to a diet high in fat attenuates dendritic spine density in the medial prefrontal cortex Dingess Paige M. 2 Darling Rebecca A. 2 Dolence E. Kurt 1 Culver Bruce W. 1 Brown Travis E. 12* 1 School of Pharmacy, University of Wyoming, Laramie, WY 82071 2 Neuroscience Program, University of Wyoming, Laramie, WY 82071 * To whom correspondence should be addressed: Travis E. Brown, PhD, University of Wyoming, 1000 E. University Ave., Dept. 3375, Laramie, WY 82071, Phone: 307-766-6129, Fax: 307-766-2953, tbrown53@uwyo.edu 2 11 2016 17 3 2016 3 2017 01 3 2018 222 2 10771085 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. A key factor in the development of obesity is the overconsumption of food calorically high in fat. Overconsumption of food high in fat not only promotes weight gain but elicits changes in reward processing. No studies to date have examined whether consumption of a high-fat (HF) diet alters structural plasticity in brain areas critical for reward processing, which may account for persistent changes in behavior and psychological function by reorganizing synaptic connectivity. To test whether dietary fat may induce structural plasticity we placed rats on one of three dietary conditions: ad libitum standard chow (SC), ad libitum 60% HF (HF-AL), or calorically matched 60% HF (HF-CM) for 3 weeks and then quantified dendritic spine density and type on basal and apical dendrites of pyramidal cells in layer V of the medial prefrontal cortex (mPFC) and medium spiny neurons (MSNs) of the nucleus accumbens. Our results demonstrate a significant reduction in the density of thin spines on the apical and basal segments of dendrites within the infralimbic, but not prelimbic, mPFC. high-fat prefrontal cortex plasticity dendritic spines Introduction At present, more than one third of the U.S. adult population is classified as obese, making obesity a substantial and relevant health concern [1]. Obesity contributes to and/or exacerbates a host of health problems, including heart disease [2], diabetes [2], stroke [3] and certain types of cancers [4], all of which could be mitigated by reducing the prevalence of obesity [5]. The development of obesity is likely a consequence of diverse factors that broadly include genetics [6], lifestyle [7] inactivity [8], pregnancy [9], unhealthy diet [10], socioeconomic status [11], and environmental influences [12]. Although there are numerous causes contributing to the obesity epidemic what is ubiquitous among most theories is that excessive consumption of foods calorically high in fat results in weight gain and obesity [13]. A hallmark of obesity is continued unhealthy eating behaviors despite knowledge of negative social and physiological consequences [14]. It has been hypothesized that exposure to dietary high-fat (HF) results in a decrease in natural reward sensitivity and therefore contributes to compensatory hedonic overeating behaviors (i.e., eating in the absence of an energetic demand) [15, 16]. In the rodent model, overconsumption of diets high in fat has consistently and efficiently been shown to induce obesity and induce a number of neurobiological changes within the mesolimbic dopamine reward system [17, 18]. For example, obese rats have a reduction in striatal dopamine D2 receptors (D2Rs) and knockdown of D2Rs potentiates compulsive food seeking to HF food, suggesting a deficit in brain reward processing [19]. Additionally, rats fed a diet consisting of Crisco, cheddar cheese and peanut butter for 15 weeks have lower extracellular dopamine (DA) levels within the nucleus accumbens (NAc) compared to standard chow fed animals [20]. Consistent with adaptations in reward processing, rats susceptible to diet-induced obesity are more prone to food-cue triggered motivation and show increased sensitivity to cocaine, suggesting basal differences in the function of the mesolimbic circuits [21, 22]. In addition to rodent studies, clinical research has shown evidence of dysregulated D2R signaling within obese human populations. Specifically, Wang et al., (2001) showed that striatal D2R availability is attenuated in obese individuals and that this decrease is negatively correlated with body mass index (BMI) [23]. It has also been demonstrated that the down regulation of striatal D2Rs in morbidly obese patients is linked with a decrease in brain glucose metabolism, a measure of brain function, within the prefrontal cortex (PFC) [24]. As demonstrated, previous reports of reward circuit modulation have focused on changes in dopaminergic plasticity, particularly in regards to DA and D2R levels but little is known about the structural plasticity that may underlie those changes. One way to examine structural change is to quantify dendritic spine density and spine type. Dendritic spines are considered to be the primary postsynaptic structures where excitatory synaptic inputs, primarily glutamatergic, are integrated [25]. Activity within the PFC is shaped by several factors. Dopamine in particular has been implicated in mediating primary PFC cognitive functions, including memory and reward [26], and the interaction between DA and glutamate has been studied as a major mechanism underlying neuronal excitability [27, 28] and may subserve plasticity. The influence of DA on glutamate function in the cortex is partially dependent upon what DA receptor is expressed on the neuron [29]. In the PFC, it has been demonstrated that D2 and D1 receptors decrease and increase pyramidal cell excitability, respectively [30]. Despite the well-established relationship between DA and glutamate and the relatively large amount of literature examining DA plasticity in obesity, very little research has aimed to elucidate changes in excitatory transmission. Functionally, increases in dendritic spine density and shape are hypothesized to be the direct result of long-term potentiation (LTP) [31, 32], particularly on dendritic segments where spine density is already low [33], and can shrink in size following long-term depression (LTD) [34]. As such, changes in spine density and shape are thought to reflect synaptic activity and neuronal remodeling. Dendritic spine plasticity has been extensively studied in response to pathological and non-pathological paradigms. For example, postmortem studies in human patients diagnosed with mental retardation [35] or schizophrenia [36] show a marked decrease in spine density in the CA1 region of the hippocampus and on pyramidal neurons of the PFC, respectively. Alternatively, an increase in spine density has been observed in the rat hippocampal dentate gyrus following consolidation of water maze training, a spatial learning task [37]. Within the reward circuit, an increase in spine density has been demonstrated in NAc shell (NAc-Sh) and mPFC following exposure to the psychostimulants amphetamine and cocaine [38]. Stress too has been reported to have a positive effect on spine density in the NAc-Sh, prelimbic PFC (PL-PFC) and orbitofrontal cortex (OFC) [39]. In contrast, morphine [40] and sucrose [41] elicit the opposite effect on spine density within the NAc-Sh and OFC, respectively. The aim of our study was to examine whether exposure to dietary HF would induce changes in dendritic spine density within the PFC and NAc similarly to other appetitive stimuli. Rats were maintained on either ad libitum standard chow (SC), ad libitum 60% HF (HF-AL) or 60% HF calorically matched (HF-CM) diets for 3 weeks. Following 3 weeks of feeding on SC, HF-CM, or HF-AL diets, dendritic spine density and type were analyzed within the PL-PFC, infralimbic prefrontal cortex (IL-PFC), the NAc-Sh and NAc core (NAc-C) using DiI staining to visualize the spines. Methods Animals Eighteen adult (PND 60) male Sprague-Dawley rats obtained from our breeding colony and housed in clear plastic cages in a temperature-controlled room with a 12:12-h light-dark cycle with lights on at 0700 were utilized for experimental procedures. Rats were 285-397 g at the beginning of the experiment. All experimental procedures and protocols for animal studies were approved by the University of Wyoming Institutional Animal Care and Use Committee in accordance with international guidelines on the ethical use of animals. Maintenance diets Rats were exposed to one of three dietary conditions in their home cages for three weeks: standard chow (SC, n=6), 60% high-fat (HF-AL, n=6), or 60% calorically-matched high-fat (HF-CM, n=6). The nutritional content of the SC diet included ∼29% protein, ∼58% carbohydrate, and ∼13% fat by kilocalorie with a total fuel value of 3.36 kcal/g (Laboratory Rodent Diet 5001, St. Louis, MO). The nutritional content of the HF diet included ∼20% protein, ∼20% carbohydrate, and ∼60% fat by kilocalorie with a gross fuel value of 5.24 kcal/g (Research Diets, New Brunswick, NJ). Animals in the HF-CM group were also fed the 60% HF diet but were food restricted such that their daily caloric intake, and therefore weight gain, did not significantly differ from the SC controls at the time of sacrifice (Figure 1B, Table 1). Each day, food consumption was measured and averaged for each dietary condition. The total amount of food consumed by the SC group was then multiplied by the caloric value of the diet (3.36 kcal/g) in order to obtain an average caloric intake. The average caloric intake was then divided by the caloric value of the HF diet (5.24 kcal/g) to estimate the amount of food to be given to the HF-CM group. This method ensured that the caloric intake of both the SC and HF-CM groups were the same. To validate that this procedure produced similar weight gain in the SC and HF-SM groups, all animals were weighed every other day. Furthermore, following sacrifice, epididymal fat was dissected and weighed as a proxy for overall fat composition. This gonadal fat deposit is among the largest adipose deposits in the rat and is therefore often used in diet studies to determine whether the dietary manipulation induced fat accumulation, in addition to changes in body weight [42]. Quantification of dendritic spine density After dietary exposure, rats were euthanized via cardiac perfusion (200 mL, 0.9% saline followed by 300 mL, 1.5% paraformaldehyde (in 0.1M phosphate buffer (PBS)). After washing in PBS, brains were coronally sectioned into 200 μm slices with a Leica VT1200S vibratome (Leica, Buffalo Grove, IL) and briefly collected in PBS. Slices were fixed in 4% paraformaldehyde in PBS for 20 minutes, incubated with Vybrant-DiI cell-labeling solution (1:200, Invitrogen, USA) for 1 hour at room temperature and placed in PBS at 4°C for 48 hours to allow dye diffusion within membranes. Finally, slices were mounted on glass slides with Vectashield (Vector, Burlingame, CA) and imaged using a Zeiss confocal microscope. For cortical pyramidal cells, dendritic spines were quantified on the terminal tips of third or fourth order basal dendrites and second or third order apical dendrites in layer V of the PL-PLC and IL-PFC. For medium spiny neurons, spines on third order or greater terminal tips were quantified in the NAc-Sh and NAc-C [43]. In all regions of interest, visible spines located within the terminal 10 μm were manually counted using 40 × magnification. In addition to total density, spine type was also analyzed following parameters previously described [44]. Briefly, spines were identified as thin type if the ratio of head diameter to neck diameter was greater than 1.1 and maximum head diameter was less than 0.4 μm. Spines were identified as mushroom type if the ratio of head to neck diameter was greater than 1.1 and the maximum head diameter was greater than 0.4 μm. Spines with a head to neck ratio of less than 1.1 were classified as stubby and those that bifurcated above the connection between spine and dendrite were classified as cup shaped. Filopodium were identified as long, thin protrusions lacking a head. Within each animal 4-8 dendrites were selected from distinct cells for analysis in each brain region and counts from these dendrites were then averaged. Dendrites were only selected if staining of the cell was complete enough that the branch order and differentiation of spine types could be determined. In total, 32-40 dendrites were analyzed per group for each brain region. Slides were coded so that the person responsible for cell selection and analysis was blind to the experimental condition. Statistical analyses were performed by averaging dendritic spine counts across animals per brain region. Group differences were assessed using two-way ANOVA and any post hoc comparisons were completed with Tukey's multiple comparisons test (p<0.05). Results To determine whether dietary manipulation may alter structural plasticity within the reward circuitry of the brain we quantified changes in dendritic spine density within the PFC and NAc, brain regions known to play a vital role in reward processing [45, 46]. There was a significant dietary effect on weight gain (F(2, 15) = 14.40, p<0.01), a significant day effect (F(10, 150) = 352, p<0.01), and significant dietary and day interaction (F(20, 150) = 9.50, p<0.01). Post-hoc analysis revealed that rats in the HF-AL group showed significant weight gain compared to HF-CM and SC fed rats days 9-21 (Figure 1B). In addition to gaining significant weight, rats maintained on the ad libitum high-fat diet showed a significant increase in estimated body fat composition measured by the ratio of epididymal fat to body weight (SC: 1.18 ± 0.09, HF-CM: 1.6 ± 0.08, HF-AL: 2.49 ± 0.25, Figure 1C). Basal and apical spine density was quantified from pyramidal cells within the PFC-PL and the PFC-IL regions of the prefrontal cortex (Figures 2-3, Table 2) and spines were quantified from medium spiny neurons (MSNs) in the shell and core of the NAc (Figure 4, Table 2). To our surprise the dietary exposure had no significant effect on the total spine density or spine type of MSNs within the NAc-C (Table 2). There was also no dietary effect on the spine density or spine type of pyramidal cells within the PL-PFC (Table 2). However, there was a significant dietary effect within the IL-PFC on the total spine density for both apical and basal dendrites (Apical: SC: 11.43 ± 0.09, HF-CM: 9.89 ± 0.16, HF-AL: 9.91 ± 0.22; F(2, 10) =57.34, p<0.01; Basal: SC: 12.00 ± 0.16, HF-CM: 10.08 ± 0.12, HF-AL: 10.05 ± 0.14, F(2, 10) = 74.22, p<0.01). An analysis of the spine type revealed that this reduction was due largely to an attenuation of thin spines in the IL-PFC. There was a significant dietary and spine type interaction (F(8, 150) = 14.82, p<0.01). Post-hoc analysis revealed that rats in both the HF-AL and HF-CM groups showed a significant attenuation in thin spines (Apical: SC: 7.24 ± 0.20, HF-CM: 5.39 ± 0.19, HF-AL: 5.50 ± 0.09; Basal: SC: 7.37 ± 0.25, HF-CM: 5.33 ± 0.28, HF-AL: 5.64 ± 0.11, Figure 3B). Interestingly, we also observed an increase in stubby type spines on apical dendrites in the HF-CM group compared to SC (SC: 1.16 ± 0.14, HF-CM: 1.69 ± 0.14) and a decrease in mushroom type spines on apical dendrites in the HF-CM group compared to SC (SC: 2.75 ± 0.14, HF-CM: 2.28 ± 0.12). Discussion Our study looked to expand upon previous research that has shown neurobiological changes as a result of exposure to dietary HF. To the best of our knowledge, no laboratories have examined whether dietary HF induces changes in dendritic spine density within the reward circuit. In the present study, rats exposed to dietary HF independent of weight gain had attenuation in spine density on both apical and basal dendrites of the IL-PFC compared to SC controls (Figure 3). This reduction in spine density can be mostly attributed to a loss of thin spines. This is not surprising given that thin spines are the most numerous comparatively and form and disappear more rapidly in the context of varying levels of synaptic activity, suggesting that they are more susceptible to plasticity [47]. Thin spines have been widely characterized under a multitude of circumstances. For example, a loss of thin spines has been observed in the monkey prefrontal cortex following aging [48] while an increase of thin spines in this region has been demonstrated subsequent of seven days forced abstinence from cocaine self-administration in rats [49]. The loss of thin spines in response to HF exposure may underlie previous reports that HF consumption impairs learning and memory, evidenced by performance on both hippocampal and frontal-dependent tasks [50]. While there were also significant changes between SC and HF-CM groups in mushroom and stubby spines in the IL-PFC, it is unlikely a dietary effect since the HF-AL group did not also show this trend. It is however possible that this finding may be attributed to food restriction given the nature of the feeding regimen in the HF-CM group, but future studies are required to elucidate this. It is further hypothesized that a lack of change in spine type or density in the nucleus accumbens may be due to a lack of ability to discriminate between D1 and D2 dopamine receptor-containing neurons, given their differential change in spine density following treatments such as withdrawal from cocaine [51]. Quantification of spine density is a common approach used to examine experience-induced modifications to synaptic organization [35] [36] [37]. Although it is an indirect measure, providing information about changes to the postsynaptic cell exclusively, electron microscopy (EM) studies have demonstrated that changes in spine density correlate with sites of synaptic contact [52, 53]. Specifically, time-lapse two-photon imaging studies within the hippocampus shows an increased spine density following induction of long-term potentiation (LTP), a phenomenon known to increase synaptic strength [25]. Similarly, following long-term depression (LTD), or loss of synaptic efficacy, shrinkage in spine size is observed [34] suggesting that changes in spine dynamics is correlated with synaptic activity, or lack thereof. To examine a potential neurobiological consequence of the HF diet, we quantified dendritic spine density in the PFC (IL and PL) and NAc (core and shell), key reward areas long characterized as being involved in reward seeking and reward reinforcement, respectively [38, 54, 55]. The PL-PFC has been known to initiate reward-seeking, as found in cocaine studies demonstrating that inhibition of this region blocks several forms of reinstated drug seeking [56-58]. The significance of the IL-PFC is presently less established but is suggested to play an opposing role in reward related behavior. Specifically, it has been demonstrated that the IL-PFC is responsible for the inhibition of cocaine seeking in extinguished rats, as measured by an increase in cocaine reinstatement following inactivation of this region [59]. Similarly, inactivation of the IL-PFC immediately following extinction impairs the maintenance of consolidation of extinction after cocaine self-administration [60]. The above findings suggest that while the PL-PFC facilitates reward seeking, the IL-PFC is an integral component of inhibitory control [61]. Our results demonstrate that exposure to HF decreases dendritic spine density and type in the IL-PFC within both the HF-AL and HF-CM groups compared to SC controls, while no differences between groups were observed in the PL-PFC or NAc. This is the first report of morphological changes following exposure to a HF diet. Given the proposed role of the IL-PFC in inhibitory control of drug reward seeking, we hypothesize that the decrease in spine density may underlie challenges in dieting behavior after exposure to fatty foods. This finding is supported by other reports of hypofunctionality in the PFC, notably the decrease in glucose metabolism in obese patients [14]. Future studies will expand upon these findings to study the downstream consequence of altering excitatory input and output from the IL-PFC in controlling feeding behaviors. The authors would like to thank Dr. Zhaojie Zhang, the director of the Neuroscience microscopy facility at the University of Wyoming, for his help and guidance imaging the spine data. We would also like to thank Kevin Schlidt and Morgan Deters for their assistance with animal care. We are also grateful for the support contributed by NIGMS grant P30 GM103398, and the College of Health Sciences Seed Grant from the University of Wyoming. Fig 1 Ad Libitum exposure to a high-fat diet for 3 weeks increases weight gain and body-fat composition (A) Experimental timeline. (B) Percent weight gain throughout dietary exposure. (C) Body-fat composition measured as the ratio of epididymal fat to body weight. Values represent the mean ± S.E.M. * indicates significant increase in HF-AL from SC and HF-CM fed rats. Fig 2 Exposure to a high-fat diet for 3 weeks has no effect on spine density within the prelimbic region of the prefrontal cortex (A) Representative prelimbic PFC apical (bottom) and basal (top) DiI-stained dendrites from each dietary condition. (B) Quantification of spine type from the basal (left) and apical (right) dendrites of pyramidal cells in the PL-PFC. Values represent the mean ± S.E.M. Fig 3 Exposure to a high-fat diet for 3 weeks decreases spine density within the infralimbic prefrontal cortex (A) Representative IL-PFC apical (bottom) and basal (top) DiI-stained dendrites from each dietary condition. (B) Quantification of spine type from the basal (left) and apical (right) dendrites of pyramidal cells in the IL-PFC. Values represent the mean ± S.E.M. Fig 4 Exposure to a high-fat diet for 3 weeks has no effect on spine density within the nucleus accumbens (A) Representative nucleus accumbens core (top) and shell (bottom) DiI-stained dendrites from each dietary condition. (B) Quantification of spine type from medium spiny neurons of the core (left) and shell (right). Values represent the mean ± S.E.M Table 1 Dietary compositions and percent weight gain for rats exposed to SC, HF-AL, and HF-CM dietary conditions for 3 weeks % Calories % Weight Δ Protein Carbohydrates Fat 3 wk Standard Chow 28.5 58.0 13.5 124.1 ± 0.96 High-Fat Calorically Matched 20.0 20.0 60.0 124.4 ± 2.16 High-Fat Ad Libitum 20.0 20.0 60.0 139.7 ± 3.65* * indicates significant weight gain from SC and HF-CM fed rats. Values represent the mean ± S.E.M Table 2 Summary of dendritic spine quantification A Prelimbic PFC Basal Apical SC HF-CM HF-AL SC HF-CM HF-AL Filopodium 0.27 ± 0.11 0.29 ± 0.09 0.23 ± 0.09 Filopodium 0.28 ± 0.08 0.36 ± 0.09 0.22 ± 0.08 Thin 7.30 ± 0.13 7.32 ± 0.10 7.29 ± 0.16 Thin 7.21 ± 0.12 7.06 ± 0.20 7.27 ± 0.16 Stubby 1.74 ± 0.08 1.56 ± 0.15 1.79 ± 0.10 Stubby 1.40 ± 0.03 1.68 ± 0.16 1.60 ± 0.06 Mushroom 2.54 ± 0.14 2.64 ± 0.16 2.59 ± 0.14 Mushroom 2.53 ± 0.15 2.44 ± 0.11 2.43 ± 0.14 Cup Shaped 0.22 ± 0.06 0.16 ± 0.04 0.16 ± 0.03 Cup Shaped 0.19 ± 0.05 0.08 ± 0.03 0.09 ± 0.04 Total 12.07 ± 0.19 12.01 ± 0.22 12.07 ± 0.23 Total 11.62 ± 0.24 11.61 ± 0.24 11.62 ± 0.11 B Infralimbic PFC Basal Apical SC HF-CM HF-AL SC HF-CM HF-AL Filopodium 0.54 ± 0.12 0.40 ± 0.12 0.31 ± 0.08 Filopodium 0.31 ± 0.12 0.39 ± 0.07 0.28 ± 0.05 Thin 7.37 ± 0.25 5.33 ± 0 28 * 5.64 ± 0.11 * Thin 7.24 ± 0.20 5.39 ± 0.19 * 5.50 ± 0.09 * Stubby 1.67 ± 0.13 1.58 ± 0.12 1.45 ± 0.07 Stubby 1.16 ± 0.14 1.69 ±0.14 $ 1.50 ± 0.20 Mushroom 2.20 ± 0.33 2.57 ± 0.10 2.48 ± 0.07 Mushroom 2.75 ± 0.14 2.28 ± 0.12 # 2.56 ± 0.19 Cup Shaped 0.24 ± 0.07 0.21 ± 0.06 0.16 ± 0.04 Cup Shaped 0.05 ± 0.03 0.14 ± 0.05 0.09 ± 0.04 Total 12.01 ± 0.16 10.08 ± 0.12 10.05 ± .14 Total 11.43 ± 0.09 9.89 ± 0.16 9.91 ± 0.22 C Nucleus Accumbens Core Shell SC HF-CM HF-AL SC HF-CM HF-AL Filopodium 0.41 ± 0.05 0.37 ± 0.12 0.23 ± 0.02 Filopodium 0.32 ± 0.09 0.30 ± 0.12 0.24 ± 0.08 Thin 9.45 ± 0.16 9.41 ± 0.25 9.4 ± 0.07 Thin 9.44 ± 0.13 9.36 ± 0.10 9.37 ± 0.17 Stubby 1.68 ± 0.06 1.91 ± 0.09 1.91 ± 0.15 Stubby 1.53 ± 0.14 1.67 ± 0.23 1.72 ± 0.09 Mushroom 2.67 ± 0.06 2.60 ± 0.08 2.55 ± 0.19 Mushroom 2.65 ± 0.12 2.52 ± 0.11 2.53 ± 0.19 Cup Shaped 0.29 ± 0.04 0.13 ± 0.05 0.19 ± 0.05 Cup Shaped 0.15 ± 0.07 0.16 ± 0.11 0.12 ± 0.05 Total 14.50 ± 0.08 14.42 ± 0.21 14.41 ± 0.24 Total 14.08 ± 0.18 13.99 ± 0.20 13.96 ± 0.23 * indicates significant decrease in HF-AL and HF-CM from SC fed rats. $ indicates significant increase in HF-CM from SC fed rats. # indicates significant decrease in HF-CM from SC fed rats. Values represent the mean ± S.E.M 1 Ogden CL Prevalence of overweight and obesity in the United States, 1999-2004 JAMA 2006 295 13 1549 55 16595758 2 Wild SH Byrne CD ABC of obesity. Risk factors for diabetes and coronary heart disease BMJ 2006 333 7576 1009 11 17095784 3 Kurth T Body mass index and the risk of stroke in men Arch Intern Med 2002 162 22 2557 62 12456227 4 Calle EE Obesity and cancer BMJ 2007 335 7630 1107 8 17986715 5 Finkelstein EA Annual medical spending attributable to obesity: payer-and service-specific estimates Health Aff (Millwood) 2009 28 5 w822 31 19635784 6 Yamada Y Genetic factors for obesity Int J Mol Med 2006 18 5 843 51 17016614 7 Shigeta H Lifestyle, obesity, and insulin resistance Diabetes Care 2001 24 3 608 8 Spence D Inactivity and obesity BMJ 2011 343 d5093 21831949 9 Yogev Y Catalano PM Pregnancy and obesity Obstet Gynecol Clin North Am 2009 36 2 285 300 viii 19501314 10 Hensrud DD Diet and obesity Curr Opin Gastroenterol 2004 20 2 119 24 15703632 11 Sobal J Stunkard AJ Socioeconomic status and obesity: a review of the literature Psychol Bull 1989 105 2 260 75 2648443 12 Hebebrand J Hinney A Environmental and genetic risk factors in obesity Child Adolesc Psychiatr Clin N Am 2009 18 1 83 94 19014859 13 Bray GA Paeratakul S Popkin BM Dietary fat and obesity: a review of animal, clinical and epidemiological studies Physiol Behav 2004 83 4 549 55 15621059 14 Volkow ND Overlapping neuronal circuits in addiction and obesity: evidence of systems pathology Philos Trans R Soc Lond B Biol Sci 2008 363 1507 3191 200 18640912 15 Cordeira JW Brain-derived neurotrophic factor regulates hedonic feeding by acting on the mesolimbic dopamine system J Neurosci 2010 30 7 2533 41 20164338 16 Geiger BM Evidence for defective mesolimbic dopamine exocytosis in obesity-prone rats FASEB J 2008 22 8 2740 6 18477764 17 Masek J Fabry P High-fat diet and the development of obesity in albino rats Experientia 1959 15 444 5 14421993 18 Schemmel R Mickelsen O Gill JL Dietary obesity in rats: Body weight and body fat accretion in seven strains of rats J Nutr 1970 100 9 1041 8 5456549 19 Johnson PM Kenny PJ Dopamine D2 receptors in addiction-like reward dysfunction and compulsive eating in obese rats Nat Neurosci 2010 13 5 635 41 20348917 20 Geiger BM Deficits of mesolimbic dopamine neurotransmission in rat dietary obesity Neuroscience 2009 159 4 1193 9 19409204 21 Robinson MJ Individual Differences in Cue-Induced Motivation and Striatal Systems in Rats Susceptible to Diet-Induced Obesity Neuropsychopharmacology 2015 40 9 2113 23 25761571 22 Vollbrecht PJ Pre-existing differences in motivation for food and sensitivity to cocaine-induced locomotion in obesity-prone rats Physiol Behav 2015 152 Pt A 151 160 26423787 23 Wang GJ Brain dopamine and obesity Lancet 2001 357 9253 354 7 11210998 24 Volkow ND Low dopamine striatal D2 receptors are associated with prefrontal metabolism in obese subjects: possible contributing factors Neuroimage 2008 42 4 1537 43 18598772 25 Engert F Bonhoeffer T Dendritic spine changes associated with hippocampal long-term synaptic plasticity Nature 1999 399 6731 66 70 10331391 26 Schultz W Getting formal with dopamine and reward Neuron 2002 36 2 241 63 12383780 27 Ferrario CR Withdrawal from cocaine self-administration alters NMDA receptor-mediated Ca2+ entry in nucleus accumbens dendritic spines PLoS One 2012 7 8 e40898 22870207 28 Baldwin AE Sadeghian K Kelley AE Appetitive instrumental learning requires coincident activation of NMDA and dopamine D1 receptors within the medial prefrontal cortex J Neurosci 2002 22 3 1063 71 11826135 29 Nicola SM Surmeier J Malenka RC Dopaminergic modulation of neuronal excitability in the striatum and nucleus accumbens Annu Rev Neurosci 2000 23 185 215 10845063 30 Tseng KY O'Donnell P Dopamine-glutamate interactions controlling prefrontal cortical pyramidal cell excitability involve multiple signaling mechanisms J Neurosci 2004 24 22 5131 9 15175382 31 Tominaga-Yoshino K Repetitive activation of protein kinase A induces slow and persistent potentiation associated with synaptogenesis in cultured hippocampus Neurosci Res 2002 44 4 357 67 12445624 32 Tominaga-Yoshino K Repetitive induction of late-phase LTP produces long-lasting synaptic enhancement accompanied by synaptogenesis in cultured hippocampal slices Hippocampus 2008 18 3 281 93 18058822 33 Oe Y Dendritic spine dynamics in synaptogenesis after repeated LTP inductions: dependence on pre-existing spine density Sci Rep 2013 3 1957 23739837 34 Zhou Q Homma KJ Poo MM Shrinkage of dendritic spines associated with long-term depression of hippocampal synapses Neuron 2004 44 5 749 57 15572107 35 Irwin SA Galvez R Greenough WT Dendritic spine structural anomalies in fragile-X mental retardation syndrome Cereb Cortex 2000 10 10 1038 44 11007554 36 Glantz LA Lewis DA Decreased dendritic spine density on prefrontal cortical pyramidal neurons in schizophrenia Arch Gen Psychiatry 2000 57 1 65 73 10632234 37 O'Malley A Transient spine density increases in the mid-molecular layer of hippocampal dentate gyrus accompany consolidation of a spatial learning task in the rodent Neuroscience 2000 99 2 229 32 10938428 38 Robinson TE Kolb B Alterations in the morphology of dendrites and dendritic spines in the nucleus accumbens and prefrontal cortex following repeated treatment with amphetamine or cocaine Eur J Neurosci 1999 11 5 1598 604 10215912 39 Muhammad A Kolb B Maternal separation altered behavior and neuronal spine density without influencing amphetamine sensitization Behav Brain Res 2011 223 1 7 16 21515311 40 Diana M Spiga S Acquas E Persistent and reversible morphine withdrawal-induced morphological changes in the nucleus accumbens Ann N Y Acad Sci 2006 1074 446 57 17105943 41 Crombag HS Opposite effects of amphetamine self-administration experience on dendritic spines in the medial and orbital prefrontal cortex Cereb Cortex 2005 15 3 341 8 15269111 42 Bjorndal B Different adipose depots: their role in the development of metabolic syndrome and mitochondrial response to hypolipidemic agents J Obes 2011 2011 490650 21403826 43 Ferrario CR Neural and behavioral plasticity associated with the transition from controlled to escalated cocaine use Biol Psychiatry 2005 58 9 751 9 16098484 44 Bloss EB Evidence for reduced experience-dependent dendritic spine plasticity in the aging prefrontal cortex J Neurosci 2011 31 21 7831 9 21613496 45 Nestler EJ The neurobiology of cocaine addiction Sci Pract Perspect 2005 3 1 4 10 18552739 46 Kalivas PW Volkow ND The neural basis of addiction: a pathology of motivation and choice Am J Psychiatry 2005 162 8 1403 13 16055761 47 Holtmaat A Experience-dependent and cell-type-specific spine growth in the neocortex Nature 2006 441 7096 979 83 16791195 48 Dumitriu D Selective changes in thin spine density and morphology in monkey prefrontal cortex correlate with aging-related cognitive impairment J Neurosci 2010 30 22 7507 15 20519525 49 Rasakham K Synapse density and dendritic complexity are reduced in the prefrontal cortex following seven days of forced abstinence from cocaine self-administration PLoS One 2014 9 7 e102524 25072653 50 Greenwood CE Winocur G Learning and memory impairment in rats fed a high saturated fat diet Behav Neural Biol 1990 53 1 74 87 2302144 51 Lee KW Cocaine-induced dendritic spine formation in D1 and D2 dopamine receptor-containing medium spiny neurons in nucleus accumbens Proc Natl Acad Sci U S A 2006 103 9 3399 404 16492766 52 Gray EG Electron microscopy of synaptic contacts on dendrite spines of the cerebral cortex Nature 1959 183 4675 1592 3 13666826 53 Kleim JA Synaptogenesis and Fos expression in the motor cortex of the adult rat after motor skill learning J Neurosci 1996 16 14 4529 35 8699262 54 Cornish JL Duffy P Kalivas PW A role for nucleus accumbens glutamate transmission in the relapse to cocaine-seeking behavior Neuroscience 1999 93 4 1359 67 10501460 55 Dong Y Cocaine-induced plasticity of intrinsic membrane properties in prefrontal cortex pyramidal neurons: adaptations in potassium currents J Neurosci 2005 25 4 936 40 15673674 56 McFarland K Kalivas PW The circuitry mediating cocaine-induced reinstatement of drug-seeking behavior J Neurosci 2001 21 21 8655 63 11606653 57 Capriles N A role for the prefrontal cortex in stress- and cocaine-induced reinstatement of cocaine seeking in rats Psychopharmacology (Berl) 2003 168 1-2 66 74 12442201 58 McLaughlin J See RE Selective inactivation of the dorsomedial prefrontal cortex and the basolateral amygdala attenuates conditioned-cued reinstatement of extinguished cocaine-seeking behavior in rats Psychopharmacology (Berl) 2003 168 1-2 57 65 12845418 59 Peters J LaLumiere RT Kalivas PW Infralimbic prefrontal cortex is responsible for inhibiting cocaine seeking in extinguished rats J Neurosci 2008 28 23 6046 53 18524910 60 LaLumiere RT Niehoff KE Kalivas PW The infralimbic cortex regulates the consolidation of extinction after cocaine self-administration Learn Mem 2010 17 4 168 75 20332188 61 Bechara A Decision making, impulse control and loss of willpower to resist drugs: a neurocognitive perspective Nat Neurosci 2005 8 11 1458 63 16251988

PMC005xxxxxx/PMC5114175.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0370121 6712 Prog Neurobiol Prog. Neurobiol. Progress in neurobiology 0301-0082 1873-5118 26721620 5114175 10.1016/j.pneurobio.2015.12.002 NIHMS757198 Article Challenges in the development of therapeutics for narcolepsy Black Sarah Wurts a Yamanaka Akihiro b Kilduff Thomas S. a* a Center for Neuroscience, Biosciences Division, SRI International, Menlo Park, CA 94025, USA b Research Institute of Environmental Medicine, Nagoya University, Nagoya 464-8601, Japan * Corresponding author at: Center for Neuroscience, Biosciences Division, SRI International, 333 Ravenswood Avenue, Menlo Park, CA 94025, USA. Tel.: +1 650 859 5509. thomas.kilduff@sri.com (T.S. Kilduff). 3 11 2016 23 12 2015 5 2017 01 5 2018 152 89113 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Narcolepsy is a neurological disorder that afflicts 1 in 2000 individuals and is characterized by excessive daytime sleepiness and cataplexy—a sudden loss of muscle tone triggered by positive emotions. Features of narcolepsy include dysregulation of arousal state boundaries as well as autonomic and metabolic disturbances. Disruption of neurotransmission through the hypocretin/orexin (Hcrt) system, usually by degeneration of the HCRT-producing neurons in the posterior hypothalamus, results in narcolepsy. The cause of Hcrt neurodegeneration is unknown but thought to be related to autoimmune processes. Current treatments for narcolepsy are symptomatic, including wake-promoting therapeutics that increase presynaptic dopamine release and anticataplectic agents that activate monoaminergic neurotransmission. Sodium oxybate is the only medication approved by the US Food and Drug Administration that alleviates both sleep/wake disturbances and cataplexy. Development of therapeutics for narcolepsy has been challenged by historical misunderstanding of the disease, its many disparate symptoms and, until recently, its unknown etiology. Animal models have been essential to elucidating the neuropathology underlying narcolepsy. These models have also aided understanding the neurobiology of the Hcrt system, mechanisms of cataplexy, and the pharmacology of narcolepsy medications. Transgenic rodent models will be critical in the development of novel therapeutics for the treatment of narcolepsy, particularly efforts directed to overcome challenges in the development of hypocretin replacement therapy. Narcolepsy Cataplexy Orexin Hypocretin Neurodegeneration Animal models 1. Overview of narcolepsy 1.1. History The neurological disorder narcolepsy began with descriptions of patients who experienced attacks of muscle weakness with retained consciousness after excitation (Westphal, 1877) and the frequent, urgent need to sleep (Gélineau, 1880; Schenck et al., 2007). The term “narcolepsy” (literally “seized by somnolence”) was coined by Gelineau and was described over one hundred years ago as a syndrome with many disparate features that seemed to defy reconciliation with a unified mechanism of etiology. Overwhelming sleepiness, emotionally-triggered muscle paralysis, onset of sleep attacks possibly associated with head trauma, sleep attacks during physical activity, nighttime sleeplessness, hallucinations, mental sluggishness and obesity were all described in these early reports. The attacks of paralysis were considered distinct from epilepsy because they were not accompanied by loss of consciousness, sensation was maintained, and neither tonic convulsions nor clonic movement were observed. These episodes were first characterized by Löwenfeld (Löwenfeld, 1902) but eventually termed “cataplexy” (literally “struck down as if stupefied”) by Henneberg and Adie in the early 20th century (Guilleminault et al., 2007) to refer to sudden, emotionally-triggered bilateral loss of muscle tone. 1.2. Clinical features 1.2.1. The narcolepsy tetrad The symptoms of narcolepsy were first organized into the classic tetrad to aid diagnosis (Yoss and Daly, 1957) and include: Excessive daytime sleepiness (EDS): sudden or persistent need to sleep during the day, independent of the amount or quality of previous nighttime sleep; cataplexy: sudden loss of muscle tone, usually triggered by positive, rather than negative, emotional stimuli (Anic-Labat et al., 1999); hypnogogic hallucinations: unreal, vivid auditory or visual perceptions at sleep onset; sleep paralysis: temporary inability to move while falling asleep or awakening. Of these symptoms, only cataplexy is unique to narcolepsy—the others can occur in people without narcolepsy, for example, in instances of severe sleep deprivation (Guilleminault and Cao, 2011). In addition, not all of these symptoms are present in all narcoleptic individuals. Cataplexy, while pathognomonic of narcolepsy, only occurs in 60–70% of narcoleptic patients (Bassetti and Aldrich, 1996) and, consequently, a distinction is made between diagnosis of narcoleptic patients with and without cataplexy (see Section 1.2.2). Beyond the classic tetrad, other features are now well acknowledged to contribute to the symptomatology of narcolepsy. Many of these features have been discovered through advances in basic sleep research over the last 50 years. 1.2.2. Non-tetrad symptoms: arousal state instability The discovery of rapid eye movement (REM) sleep as a distinct sleep state characterized by activated EEG, muscle atonia, bursts of eye movements, and vivid dreaming (Aserinsky and Kleitman, 1953; Dement and Kleitman, 1957; Jouvet et al., 1959) heralded a new understanding of the narcolepsy syndrome. Subsequent research revealed that patients with narcolepsy entered REM sleep sooner than the typical 90 min interval after sleep onset that is observed in non-narcoleptic controls (Rechtschaffen et al., 1963). Because these sleep-onset REM periods (SOREMPs) persisted during daytime naps (Dement et al., 1966), narcolepsy came to be viewed as a disorder of REM sleep timing. Intrusions of inappropriately timed dream imagery and muscle atonia of REM sleep were also thought to underlie the hypnogogic hallucinations and sleep paralysis experienced by patients with narcolepsy. Cataplexy came to be viewed as an initiation of REM sleep atonia during wakefulness—a hypothesis corroborated by the finding that only patients who experienced cataplexy also exhibited SOREMPs (Dement et al., 1966). These observations led to the implementation of the Multiple Sleep Latency Test (MSLT) to detect SOREMPs, and to objectively measure daytime sleepiness as an aid in the diagnosis of narcolepsy (Mitler et al., 1979; Richardson et al., 1978). In the MSLT, patients are provided a series of 4–6 nap opportunities at 2 h intervals that begin 2 h after the onset of wakefulness following a night of standard polysomnography. During the nap opportunities, mean sleep latency within 8 min and REM onset within 15 min (a SOREMP) in at least 2 naps was considered pathologic (Mathis and Hess, 2007). Recently, the diagnostic criteria for narcolepsy have changed such that a SOREMP during polysomnographic monitoring on the night prior to the MSLT can count as one of the two or more SOREMPS required for diagnosis (Eichler, 2014). Thus, three of the tetrad symptoms (cataplexy, hypnogogic hallucinations and sleep paralysis) appeared to be linked to a REM sleep abnormality in narcolepsy, but EDS did not fit this paradigm. As the timing of sleep and wakefulness across the 24 h day/night cycle became recognized as due to interaction between the circadian timekeeping system and a sleep homeostatic mechanism (Borbely et al., 1984; Daan et al., 1984; Edgar et al., 1993), the pathological sleepiness in narcolepsy was also examined in these terms. Patients with narcolepsy not only have difficulty staying awake during the day, they also sleep poorly at night (Roth et al., 2013). Could the EDS of narcolepsy be a compensatory response to lost night time sleep? Slow wave activity during NREM sleep (EEG spectral power in the 0.75–4.5 Hz frequency band), an index of homeostatic sleep drive, was shown in narcoleptics to increase after experimentally-induced sleep deprivation and to decrease during recovery sleep, which revealed that homeostatic regulation was operative (Tafti et al., 1992). Despite shorter latencies from bedtime to sleep onset, patients with narcolepsy indicate that they experience insomnia more frequently than controls (Parkes et al., 1998). These self-reports have been supported by objective polysomnography in which frequent nocturnal awakenings and reduced REM sleep latency are thought to contribute to fragmented nighttime sleep in narcolepsy (Harsh et al., 2000). Despite the large sample size in this study, however, the relationship between disrupted nighttime sleep and sleepiness the following day was found to be weak and thus does not explain the EDS of narcolepsy (Harsh et al., 2000). In addition, the defined, but attenuated, distribution of sleep, wake, and sleepiness across the 24 h day/night cycle in patients with narcolepsy has been interpreted to indicate an intact circadian timekeeping system (Broughton et al., 1988; Dantz et al., 1994). Thus, neither dysfunctional sleep homeostatic mechanisms nor abnormal circa-dian processes fully explain the pattern of fragmented arousal states. The observations that SOREMPs, the extent of nocturnal awakening, and EDS strongly correlate with cataplexy (Harsh et al., 2000) foreshadowed the current view that narcolepsy is best described as a disorder of arousal state instability. It has now been firmly established that the loss of arousal state boundary control, first described by Broughton et al. (1986), arises from the loss of the hypothalamic neuropeptide hypocretin-1 (HCRT1), also known as orexin (see Section 2.2 and Section 2.4). The 3rd Edition of the International Classification of Sleep Disorders, (Eichler, 2014) formalized the distinction of narcolepsy with loss of HCRT1 (which usually presents with cataplexy) as “narcolepsy type 1”, and without HCRT1 loss or cataplexy as “narcolepsy type 2.” Narcolepsy type 1 was intended to capture patients who clearly have insufficient HCRT1 levels but may not yet manifest cataplexy, since a third of patients with low HCRT1 do not develop cataplexy until ~15 years after the onset of EDS (Andlauer et al., 2012). Classification of narcolepsy type 1 requires EDS, a positive score on the MSLT, and either cataplexy or, if cataplexy is absent, a low concentration of cerebrospinal fluid (CSF) HCRT1 (<110 pg/ml, if using a standard reference) that is 1/3 of normal levels (Baumann et al., 2014). Narcolepsy type 2 also requires EDS and a positive score on the MSLT, and is diagnosed based on exclusion of narcolepsy type 1 and differentiation from hypersomnia by the presence of sleep paralysis and hallucinations. A SOREMP on polysomnography the night preceding the MSLT may substitute for one of the ≥2 SOREMPs on the MSLT that is typically required for the diagnosis of narcolepsy type 1 and type 2. These new diagnostic criteria and the classification of narcolepsy types underscore the move away from classic tetrad symptomatology to a more refined understanding based on the etiology of the disease. 1.2.3. Metabolic and autonomic abnormalities A renewed interest in the metabolic abnormalities noted anecdotally in narcoleptic patients resulted from the discovery of the role of HCRT loss in narcolepsy (see Section 2.2 and Section 2.4) and identification of the concomitant role of HCRT in energy homeostasis (Funato et al., 2009; Hara et al., 2001; Yamanaka et al., 2003). Narcolepsy has consistently found to be associated with obesity or increased body mass index (BMI) that does not appear to be secondary to daytime inactivity from EDS or medication (Chabas et al., 2007; Dahmen et al., 2001; Heier et al., 2011; Kok et al., 2003; Nishino et al., 2001; Poli et al., 2009; Schuld et al., 2002). Although the obesity has been clearly linked to low cerebrospinal fluid (CSF) HCRT1 levels (Heier et al., 2011; Nishino et al., 2001; Poli et al., 2009) and abdominal fat deposits (Kok et al., 2003; Poli et al., 2009), the cause of the BMI increase in narcolepsy has been challenging to parse. Binge eating, irresistible food cravings, and sleep-related eating disorder have been reported in approximately a quarter of narcoleptic patients (Fortuyn et al., 2008; Palaia et al., 2011), but this population also has reduced daily caloric intake compared to healthy controls (Chabas et al., 2007; Lammers et al., 1996). Like other overweight people, patients with narcolepsy who are obese tend to have lower resting metabolic rates (Chabas et al., 2007). The higher prevalence of eating disorders and the specificity of reduced basal metabolism to narcolepsy has been challenged in studies that employed BMI-matched controls, and have led to the hypotheses that obesity in narcolepsy could be due to changes in BMI set point or to reduced sympathetic tone (Dahmen et al., 2008, 2009; Fronczek et al., 2008a). Early reports of reduced leptin, a hormone that signals energy deficiency (Kok et al., 2002; Nishino et al., 2001; Schuld et al., 2000) or impaired glucose metabolism as in type 2 diabetes mellitus (Honda et al., 1986) have not been corroborated by studies of narcolepsy patients that have employed more rigorous methodology (Beitinger et al., 2012; Donjacour et al., 2014, 2013; Ohayon, 2013). However, a recent study of narcolepsy type 1 patients sampled close to disease onset found a positive correlation between CSF leptin levels and BMI; subjects with long disease duration and low CSF HCRT1 levels had higher leptin levels and BMI (Kornum et al., 2015). Surprisingly, patients with narcolepsy have been shown to be more sensitive to insulin in peripheral tissues (Donjacour et al., 2014). Autonomic disturbances in narcolepsy have been reported, but still remain poorly understood. Whether the autonomic changes are directly due to the loss of HCRTs or are secondary to the ensuing sleep/wake fragmentation has been difficult to ascertain, particularly because Hcrt neurons influence both sympathetic and parasympathetic outflows (Plazzi et al., 2011; van den Pol, 1999). Some studies have indicated reduced sympathetic tone in narcolepsy, as revealed by attenuated cardiovascular reflexes (Sachs and Kaijser, 1982) or increased variability in heart rate and blood pressure (Fronczek et al., 2008a). Other studies in which heart rate variability was measured in patients with narcolepsy have demonstrated normal reduction in sympathetic outflow during sleep, but reduced parasympathetic tone during wakefulness (Ferini-Strambi et al., 1997) or enhanced sympathetic activity during orthostatic stress (Grimaldi et al., 2010b). Baseline heart rate has also been reported to be elevated in narcolepsy (Grimaldi et al., 2012; Sorensen et al., 2013). The normal responses of increased heart rate in relation to arousals from sleep and decreased blood pressure during sleep have been shown to be blunted in narcolepsy with cataplexy (Dauvilliers et al., 2012; Grimaldi et al., 2012), particularly in those patients with low or absent HCRT1, even compared to narcolepsy patients with normal HCRT1 levels (Sorensen et al., 2013). Together, these results support the hypothesis that HCRT insufficiency, concomitant with altered sleep architecture, leads to increased sympathetic activation. However, this hypothesis appears to be refuted by a recent study in which a correlation was found between HCRT1 level, heart rate and resting muscle sympathetic nerve activity as measured by direct microneurographic monitoring (Donadio et al., 2014). Perhaps these opposing results could be explained by the possibility that autonomic changes near disease onset differ from autonomic responses that develop over time to compensate for the initial changes, as has been hypothesized for thermoregulatory changes in narcolepsy (Black et al., 2013). There is also evidence for altered temperature regulation in both human narcoleptics and animal models. Although a circadian rhythm of core body temperature (Tb) was clearly evident in human narcoleptics (Grimaldi et al., 2010a), nighttime Tb levels were elevated relative to controls (Mosko et al., 1983; Pollak and Wagner, 1994). Whether this is a cause or a consequence of disrupted nocturnal sleep in narcoleptics was unclear. Human narcoleptics also have a larger distal-to-proximal skin temperature gradient during wakefulness and, when asleep, maintain elevated distal skin temperature whereas proximal skin temperature increases to normal levels (Fronczek et al., 2006). Proximal skin warming in narcoleptic patients significantly suppressed nocturnal wakefulness and enhanced slow wave sleep whereas distal skin warming disrupted nocturnal sleep (Fronczek et al., 2008b). These observations are consistent with the concept that the Hcrt system is a key determinant of overall energy expenditure and, thus, body weight regulation. The normal decline in Tb at sleep onset is also blunted in Hcrt-deficient mice; these mice also have normal Tb levels during wakefulness despite exhibiting less locomotor activity (Mochizuki et al., 2006). Similar results were also observed in orexin/ataxin-3 mice in which Hcrt neurons are chronically lost after birth (Black et al., 2013). In this study, the number of remaining Hcrt neurons in the transgenic mice positively correlated with the Tb change from wakefulness to NREM and REM sleep. Together, these results imply a role for the Hcrt system in either activation of heat loss mechanisms or, alternatively, loss of the HCRTs may result in compensatory elevated activity in other thermogenic systems. In contrast to chronic HCRT deficiency, acute disruption of Hcrt signaling with the dual HCRT receptor antagonist almorexant decreased Tb in wild type mice. These observations led to the hypothesis that at the onset of narcolepsy, early disruption of Hcrt signaling may reduce thermogenesis; then as more Hcrt neurons are lost, the ensuing hypothermia may activate compensatory heat conservation mechanisms (Black et al., 2013). Hypocretin neurons directly innervate the rostral raphe pallidus, site of sympathetic premotor neurons that innervate brown adipose tissue (BAT), thereby providing a neural pathway for elevating thermogenesis (Tupone et al., 2011). Hypocretin-deficient mice also exhibit impaired BAT differentiation (Sellayah et al., 2011). Hypocretin deficiency in narcolepsy could thus both increase the risk of obesity and the metabolic syndrome (Poli et al., 2009), while augmented activity in this system could contribute to a lean phenotype (Funato et al., 2009). 1.3. Prevalence and risk factors Narcolepsy occurs in North America and Europe with a prevalence of 0.03–0.05% (Ohayon et al., 2002; Silber et al., 2002). Geographic differences clearly occur; in Japan, the prevalence may be as high as 0.18% (Honda, 1979; Tashiro et al., 1992) whereas cases are rarely reported in Israel (Lavie and Peled, 1987). As of 2002, the incidence rate was determined to be 1.37/100,000 new cases per year (Silber et al., 2002) with a bimodal distribution of age of onset with peaks at about 15 and 35 years of age (Dauvilliers et al., 2001). The proportion of narcolepsy cases in epidemiology studies that could be classified as narcolepsy type 1 or type 2 is currently unknown (Baumann et al., 2014). The average time to diagnosis from symptom onset (usually EDS appears first) has usually been reported to take longer than 10 years, but the time lag may diminish with improved awareness of the disorder (Thorpy and Krieger, 2014). Diagnostic delay is longer in women than men and is associated with higher body mass index (Luca et al., 2013). A retrospective study indicated that about half of narcoleptic patients report symptom onset prior to 15 years of age and, in China, 70% had onset before age 10 (Han et al., 2011). The delay in diagnosis is greatly reduced when disease onset is in childhood (Nevsimalova, 2009) or when cataplexy presents as the first symptom or with high frequency (Luca et al., 2013). Only 1–2% of first degree relatives share narcolepsy diagnoses, which represents a 10–40× increased risk factor compared to the general population (Mignot, 1998). Monozygotic twins that have narcolepsy with cataplexy are only 25–30% concordant for the disorder (Mignot, 1997). While this evidence supports a role for environmental factors to contribute to narcolepsy onset, genetic predisposition is clearly important. Risk factors for narcolepsy can be found in genes that encode the major histocompatibility complex proteins, also known as human leukocyte antigen (HLA) genes. Narcolepsy is highly associated with HLA class II polymorphisms in the closely linked loci DQB1*06:02 and DQA1*01:02 that together form the DQ0602 heterodimer (Juji et al., 1984; Matsuki et al., 1992; Mignot et al., 1994a). Almost all narcolepsy with cataplexy patients (82–99%) are carriers of DQB1*06:02 while only 12–38% of non-narcoleptic individuals carry this allele (Mignot et al., 1997, 2001). Full length exome sequencing (Tafti et al., 2014) has indicated the DQB1*0602 allele is not mutated in narcolepsy and, thus, is only a susceptibility factor in the disorder. The susceptibility risk for narcolepsy is two-fold in individuals that are homozygous for DQB1*06:02 compared to heterozygous carriers (Pelin et al., 1998). The largest genome-wide association study to date estimated a 251-fold increased risk for narcolepsy in DQB1*0602 carriers and identified 4 protective DQB1 alleles (Tafti et al., 2014). A much smaller, but significant, predisposing effect of HLA-DPB1*05:01 in narcolepsy has recently been found in Asians (Ollila et al., 2015). The HLA genes encode molecules that present antigen fragments to the T-cell receptor in order to direct an immune response to a specific antigen. Other predisposing factors for narcolepsy are associations with a polymorphism in the T-cell receptor alpha and beta genes, whose products recognize antigens presented by HLA molecules, and Cathepsin H, which processes antigens for presentation (Faraco et al., 2013; Hallmayer et al., 2009; Han et al., 2013a). The purinergic receptor subtype P2Y11, which is expressed in cytotoxic lymphocytes and in the brain, has also been shown to be associated with narcolepsy (Kornum et al., 2011b). The tight linkage between narcolepsy and HLA subtypes and other genes involved in autoimmunity such as TNFSF4, IL10RB-INFAR1, and P2YR11/DNMT1 (Faraco et al., 2013; Han et al., 2013a) strongly suggests an autoimmune basis for narcolepsy (Arango et al., 2015; Kornum et al., 2011a) (see Section 2.6). 1.4. Comorbidities The range of symptoms in narcolepsy, including arousal state instability and cataplexy, along with other metabolic and autonomic features of the disease that are not yet fully understood, can make it difficult to understand where the entity of narcolepsy ends and where comorbidities begin. For example, night time sleep disturbance in narcolepsy, while a consequence of arousal state instability due to HCRT deficiency (see Section 2.4), can also be caused by comorbidities thought to reflect motor instability such as restless legs syndrome, REM sleep without atonia, or REM sleep behavior disorder (Frauscher et al., 2013; Plazzi et al., 2010). Comorbidities were recognized by Legrand in one of the earliest descriptions of narcolepsy: “Encephalic congestion cardiac deficiency, gastric troubles or hepatic derangements, and such diseases as gout, diabetes, and rheumatism are some of its associations” (Legrand, 1888). Investigation into the comorbidities of narcolepsy began primarily with concerns for differential diagnosis or genetic associations, particularly with diabetes, depression, sleep apnea or parasomnias (Baker et al., 1986; Broughton et al., 1983; Honda et al., 1986; Mayer and Meier-Ewert, 1993). As the metabolic and autonomic abnormalities in narcolepsy began to be uncovered, concern was raised that patients with narcolepsy might be at a greater risk for serious, chronic conditions such as diabetes or cardiovascular disease. Recently, a well-controlled, prospective epidemiological study was undertaken to determine the odds ratios of medical and psychiatric disorders associated with narcolepsy in the United States (Ohayon, 2013). According to this study, digestive diseases, upper respiratory tract diseases, heart disease, hypercholesterolemia, and hypertension represented an increased risk in narcolepsy. Neither diabetes nor the metabolic syndrome was more prevalent in narcoleptic participants compared to age-, gender- and BMI-matched controls from the general population. Among the psychiatric conditions that present with a greater risk in narcolepsy, mood disorders were present in 19.2% of people with narcolepsy and primarily began after narcolepsy onset. Panic disorder, simple phobia, agoraphobia, generalized anxiety disorder, and posttraumatic stress disorder also were found to occur at an increased frequency in narcolepsy. These phobias and anxiety disorders began after narcolepsy onset in most cases. Participants with narcolepsy were not at greater risk than the general population for alcohol dependency. Psychotic disorders were not assessed in this study because there was no case in the matched general population sample. However, schizophrenia and other psychotic disorders have been noted to begin in children after the onset of narcolepsy (Canellas et al., 2014). The consequences of these medical and psychiatric comorbidities in narcolepsy may, at least in part, underlie the 1.5-fold increase in mortality observed in narcolepsy (Jennum et al., 2013; Ohayon et al., 2014). 2. Hypocretin/Orexin and disease mechanism 2.1. Narcoleptic dogs Narcolepsy has been described in other mammals including several breeds of dogs (dachshunds, poodles, Labrador retrievers and Doberman pinschers) (Mitler et al., 1976), horses (Dreifuss and Flynn, 1984; Lunn et al., 1993; Sweeney et al., 1983) and Brahman bulls (Strain et al., 1984). Although the mode of inheritance in small breed dogs is complex as it is in humans, large breed dogs such as Doberman pinschers and Labrador retrievers provide a simplified genetic system in which the mutation in the canine narcolepsy (canarc-1) gene is transmitted as an autosomal recessive trait with full penetrance (Foutz et al., 1979). Narcolepsy is also genetically transmitted in at least one breed of horse (Ludvikova et al., 2012). The existence of a genetic animal model of narcolepsy in dogs greatly facilitated research as it enabled the creation of a breeding colony to allow experimental pharmacological (Babcock et al., 1976; Delashaw et al., 1979; Mignot et al., 1988a,b; Mitler et al., 1976; Nishino and Mignot, 1997; Nishino et al., 1989), neurochemical (Faull et al., 1982, 1986; Mefford et al., 1983; Miller et al., 1990) and neurotransmitter receptor (Boehme et al., 1984; Bowersox et al., 1987, 1986; Kilduff et al., 1986; Mignot et al., 1988a,b) studies of narcolepsy to be conducted for the first time. The Food-elicited Cataplexy Test (FECT) was devised to enable quantitative assessment of cataplexy for pharmacological studies (Babcock et al., 1976; Mitler et al., 1976) and proved to be a useful tool for the assessment of drugs that ameliorated or exacerbated cataplexy (Nishino and Mignot, 1997). The general conclusion from these studies was that narcolepsy-cataplexy was likely a result of an imbalance between the monoaminergic and cholinergic systems resulting from monoaminergic hypoactivity and cholinergic hypersensitivity in the pons (Baker and Dement, 1985; Boehme et al., 1984; Mefford et al., 1983). Subsequent studies demonstrated that: (1) the activation of presynaptic adrenergic transmission likely underlies the anticaplectic activity of antidepressants (Mignot et al., 1993; Nishino et al., 1993), (2) the wake-promoting effects of the stimulants amphetamine and modafinil are mediated by presynaptic activation of dopamine (DA) neurotransmission (Mignot et al., 1994b) and (3) mesolimbic dopaminergic hypoactivity (Reid et al., 1996) and basal forebrain cholinergic hypersensitivity (Nishino et al., 1995) occur in narcolepsy. Although cataplexy, the pathognomonic symptom of narcolepsy, was unequivocally present in dogs, documentation of the occurrence of excessive sleepiness in these mutant animals was controversial and difficult to establish a species that can apparently sleep ad libitum in most domestic and laboratory situations (Kaitin et al., 1986a,b; Kushida et al., 1985; Lucas et al., 1979; Mitler and Dement, 1977). In contrast to the ease of evaluating cataplexy using the FECT, the labor-intensive nature of scoring of sleep/wake states based on EEG and EMG recordings limited the number of sleep pharmacology studies that were undertaken in narcoleptic dogs (Shelton et al., 1995). In contrast to human narcolepsy, early studies established that the canarc-1 mutation was not associated with the dog equivalent of the HLA system (Dean et al., 1989; Motoyama et al., 1989). Identification of the genetic basis of autosomal recessive mutation in dogs in an era before genome sequencing existed was a challenging task, necessitating the creation of a canine genomic bacterial artificial chromosome (BAC) library (Li et al., 1999). Using this BAC library, canarc-1 was ultimately identified as a deletion mutation in the gene encoding Hcrt receptor 2 (Hcrtr2) (Lin et al., 1999) as described below. 2.2. Dual discovery of hypocretins and orexins HCRT1 and HCRT2 are hypothalamic neuropeptides derived from a single precursor molecule (prepro-HCRT) by proteolytic processing (de Lecea et al., 1998; Sakurai et al., 1998). Although the HCRTs were formally described early in 1998 (de Lecea et al., 1998), subtraction cloning of the Hcrt gene from mouse and rat hypothalamus, localization of the cell bodies expressing the HCRT peptides, and description of their efferent projections was first reported at the 1997 Society for Neuroscience meeting (Peyron et al., 1997; Sutcliffe et al., 1997). Because the cell bodies expressing this gene were restricted to an area of the hypothalamus centered around the perifornical nucleus (PFH) and because of a weak homology to the gut peptide secretin, these molecules were called “hypocretins.” Just 6 weeks after the description of the HCRTs, these neuropeptides were independently reported by another group of investigators as ligands binding to cell lines expressing orphan G protein-coupled receptors (Sakurai et al., 1998). The chemical isolation procedure used allowed these investigators to determine that the longer of these two peptides was 33 amino acids in length and to define the N-terminal pyroglutamyl residue, the intrachain disulfide links, and the expected C-terminal amidation of both peptides. Since intracerebroventricular injections of these peptides increased food intake in rats, these investigators called the peptides orexin-A and orexin-B from the Greek root “orexis” meaning appetite. Importantly, this paper also reported functional information on the receptors for the two peptides: the orexin-1 receptor (OX1R) was shown to preferentially bind orexin-A over orexin-B, whereas the orexin-2 receptor (OX2R) bound both peptides with similar affinity (Sakurai et al., 1998). In the subsequent issue of Cell, the HCRTs and the orexins were confirmed to be the same peptides (Sakurai et al., 1998). The nomenclature describing this neuropeptidergic system can be confusing. As indicated above, the term “hypocretin” appeared first in the literature referring to the two neuropeptides encoded by the preprohormone (de Lecea et al., 1998). However, the cognate receptors for these peptides were first called the “orexin receptors” (Sakurai et al., 1998). The Human Genome Organization (HUGO) Gene Nomenclature Committee (Gray et al., 2015) and Genbank use the term “hypocretin (orexin)” to refer to both the peptides and the receptors and employ the abbreviations HCRT, HCRTR1 and HCRTR2 to refer to the genes encoding the preprohormone and the two receptors, respectively. In contrast, “The Concise Guide to PHARMACOLOGY 2013/2014” published by Committee on Receptor Nomenclature and Drug Classification of the International Union of Basic and Clinical Pharmacology (NCIUPHAR) refers to the receptors as “OX1 receptor” and “OX2 receptor” (Alexander et al., 2013). Thus, both terminologies are used in the literature: the prepro-Hcrt gene is identical to preproorexin; HCRT1 is equivalent to orexin-A; and HCRT2 is identical to orexin-B. Here, we will use the terms “HCRT1” and “HCRT2” to denote the two peptides, “Hcrt” to refer to the HCRT/orexin-containing cells or to the Hcrt or orexin gene, and “HCRTR1” and “HCRTR2” to refer to the receptors for these peptides. 2.3. Link to animal models of narcolepsy The Hcrt system was first linked to narcolepsy in 1999 when canarc-1 was identified as a deletion mutation in the Hcrtr2 gene (Lin et al., 1999). As described above, canarc-1 is transmitted as an autosomal recessive trait with full penetrance in Doberman pinschers and Labrador retrievers. Importantly, the mutation in these two breeds is in a different region of the Hcrtr2 gene, both of which result in a truncated, non-functional receptor protein. Shortly thereafter, the prepro-orexin ligand knockout mouse was found to exhibit periods of behavioral arrest that strongly resembled cataplexy in dogs and humans (Chemelli et al., 1999). These mice also have a disrupted sleep architecture, particularly during the dark period, as evidenced by increased levels of both REM and NREM sleep, short latency REM periods, and decreased sleep bout durations. The identification of the canarc-1 mutation as a deletion mutation in the Hcrtr2 gene through positional cloning and the creation of the orexin null mutant mouse were major advances in this field that not only drew a link between narcolepsy and the Hcrt system, but also suggested that this neuropeptidergic system, which had only been described 18 months earlier, might also be involved more generally in sleep/wake control. Subsequently, HCRT1 levels were found to be normal in the CSF of Hcrtr2-mutated narcoleptic Doberman pinschers and Labrador retrievers but were reduced or undetectable in either CSF or the brain of poodles and mixed breed dogs that exhibited sporadic (i.e., non-familial) narcolepsy (Ripley et al., 2001). A mutation in Hcrtr2 in a family of dachshunds was identified in which the protein was appropriately localized to the cell membrane, but failed to bind ligand and, consequently, Ca2+ mobilization upon receptor stimulation was greatly compromised (Hungs et al., 2001). Collectively, these observations further solidified the link between the Hcrt system and narcolepsy and demonstrated that dysfunction in Hcrt neurotransmission – whether occurring presynaptically, as in the orexin ligand null mutant mouse, or postsynaptically, as in the Hcrtr2-mutated Doberman pinschers and Labrador retrievers – could result in the similar phenotype of narcolepsy with cataplexy. 2.4. CSF peptide levels and neurodegeneration in human narcoleptics Although these studies in narcoleptic dog and mouse models strongly implicated the Hcrt system, the first direct indication of an abnormality in Hcrt neurotransmission in narcoleptic humans was suggested by undetectable levels of HCRT1 in CSF from 7 of 9 narcoleptic patients (Nishino et al., 2000). At that point, it was uncertain whether the reduced CSF HCRT1 reflected a problem in release or in synthesis of the peptide. However, 9 months later, postmortem analyses demonstrated that HCRT mRNA was undetectable in two human narcoleptic brains, although melanin-concentrating hormone (MCH) mRNA, a phenotypic marker of cells in the posterolateral hypothalamus that are distinct from Hcrt neurons, was readily detectable in both controls and narcoleptics (Peyron et al., 2000). Furthermore, an independently-conducted immunohistochemical study showed an 85–95% reduction in the number of HCRT-containing cells in human narcoleptic brains with no evident change in the number of MCH cells (Thannickal et al., 2000). Increased staining for glial fibrillary acid protein in the PFH of the human narcoleptic brains suggested that neurodegeneration likely had occurred whereas the preservation of MCH cells suggested that the Hcrt neurons selectively degenerate, possibly through an autoimmune mechanism (see Section 2.5) (Thannickal et al., 2000). Along with the animal studies described above, these studies strongly suggested that degeneration of the Hcrt cells is the likely cause of human narcolepsy. Subsequent studies documented loss of substances that are colocalized with HCRT, the neuropeptides dynorphin (Crocker et al., 2005) and neuronal-activity-regulated pentraxin (Blouin et al., 2005), thereby confirming neuronal cell loss in the narcoleptic brain, at least for narcolepsy type 1, rather than an absence of HCRT expression. The etiology of narcolepsy type 2 is less understood than the pathology underlying narcolepsy type 1. Most patients without cataplexy have HCRT1 levels above the criteria for narcolepsy type 1 diagnosis (Mignot et al., 2002), suggesting some degree of Hcrt neurotransmission may still occur in narcolepsy type 2. An immunohistochemical study of a single postmortem human brain from a narcoleptic patient without cataplexy found only a 33% loss of Hcrt cells, suggesting that narcolepsy without cataplexy can result from a partial loss of the Hcrt neuron population (Thannickal et al., 2009). An MRI study found microstructural changes in the inferior frontal gyrus, postcentral gyrus, and amygdala in narcoleptic patients with cataplexy vs. normal controls, but no differences were detected between normal controls and narcoleptic patients without cataplexy (Nakamura et al., 2013). How these morphological changes might relate to Hcrt neurodegeneration is unclear. However, two independent groups have suggested that remodelling of wake-promoting systems occurs subsequent to loss of Hcrt neurons, as they reported an increase in the number of histaminergic (HA) cells in the tuberomammillary nuclei (TM) in post mortem tissue from narcoleptic patients with cataplexy (John et al., 2013; Valko et al., 2013), and the increase in HA cells correlates with the extent of Hcrt cell loss (Valko et al., 2013). 2.5. Hypothesized autoimmunity While it is now well established that narcolepsy type 1 is caused by neurodegeneration of Hcrt cells and that a strong genetic component in immune function confers susceptibility to the disease, it is less clear how these two pieces fit together to fully understand the etiology of narcolepsy. The hypothesis that narcolepsy is an autoimmune disease that targets Hcrt neurons seems plausible but the mechanism has remained elusive for decades (Carlander et al., 1993; Mignot, 2014). Autoantibodies against HCRT peptides, HCRT receptors, or antigens co-localized on Hcrt neurons have evaded detection in numerous studies (Black et al., 2005; Overeem et al., 2006; Tanaka et al., 2006). The absence of identified autoantibodies in narcolepsy contrasts with other CNS autoimmune diseases (Graus et al., 2010), although some evidence for autoantibodies exists. Sera from a small group of patients with narcolepsy has been shown to bind tribbles homolog 2 (TRIB2) (Cvetkovic-Lopes et al., 2010) but, because this protein is expressed in many tissues both in CNS and in the periphery and not just in Hcrt neurons, TRIB2 autoantibodies are unlikely to be causative of Hcrt neurodegeneration (Kornum et al., 2011a) and instead may be secondary to the destruction of Hcrt cells (Liblau et al., 2015). When sera from narcoleptics and patients with other sleep disorders were screened on rat brain tissue, three distinct patterns of immunoreactivity were observed, one of which was identified to correspond to the C-terminal epitope of neuropeptide glutamic acid-isoleucine/alpha-melanocyte-stimulating hormone peptides (Bergman et al., 2014). A recent study, in which CSF samples from narcolepsy type 1 patients with early onset (within 1–12 months) were examined for changes in 51 cytokines and chemokines, did not find a difference in the levels of these immune markers compared to healthy controls (Kornum et al., 2015) and did not replicate a previously observed elevation of the cytokine interleukin 4 in this population (Dauvilliers et al., 2014a). Infections can induce autoimmunity through a wide variety of mechanisms, including molecular mimicry, epitope spreading, bystander activation and superantigens, and a growing body of evidence suggests pathogens can trigger narcolepsy (Arango et al., 2015; Kornum et al., 2011a). Onset of narcolepsy is more frequent in spring and early summer than in winter, consistent with triggering by winter time upper airway infections (Han et al., 2011). Streptococcal throat infection has been associated with a 5.4-fold increased risk of narcolepsy (Koepsell et al., 2010), and anti-streptococcal antibodies have been detected in 65% of narcoleptic patients with recent onset compared to age-matched controls (Aran et al., 2009). Serum from children with Syndenham's Chorea or pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS) has been shown to contain cross-reactive antineuronal antibodies that can alter dopaminergic signaling pathways (Chang et al., 2015), but whether a similar cross-reactivity can affect the Hcrt system remains to be determined. Homology between an epitope specific to the 2009 H1N1 strain and HCRT peptides was initially claimed (De la Herran-Arita et al., 2013) but subsequently retracted (De la Herran-Arita et al., 2014), so how an immune response to the virus may ultimately cause Hcrt cell death remains to be demonstrated. Perhaps the most promising connection between narcolepsy and an infectious agent occurred following the 2009 H1N1 influenza pandemic. In China, the incidence of narcolepsy increased three-fold in the 6 months after the peak of the outbreak, then decreased to the normal rate of onset by 2011 after the pandemic had been contained (Han et al., 2013b, 2011). As initially reported in Finland and Sweden (Bardage et al., 2011; Nohynek et al., 2012; Partinen et al., 2012) and later in other European countries (Dauvilliers et al., 2013a; O'Flanagan et al., 2014; Winstone et al., 2014), a 6–9-fold increase in new narcolepsy cases in children was observed a few months following vaccination against H1N1 with Pandemrix®—a formulation that contained the AS03 adjuvant. In contrast, no elevations in the rate of narcolepsy were reported in the U.S. where only non-adjuvanted vaccines were used (Duffy et al., 2014) or elsewhere in Europe where a closely related adjuvant, MF59, was used in the vaccine Focetria® (Ahmed et al., 2014; Calabro et al., 2013). Although these results suggest that the adjuvant AS03 could be problematic, no elevation in narcolepsy rates were observed in Canada where AS03 was also used as a component of the vaccine Arepanrix® (Montplaisir et al., 2014). However, although Pandemrix® and Arepanrix® were produced by the same manufacturer and administered with AS03, slightly different protocols for antigen isolation were utilized which resulted in more nucleoprotein and neuraminidase in Pandemrix® and a larger diversity of viral and chicken proteins in Arepanrix® (Jacob et al., 2015). This has led to the hypothesis that differential composition of the vaccines may be the key to understanding the increased incidence in narcolepsy in the affected populations (Ahmed et al., 2014; Jacob et al., 2015). A recent study (Ahmed et al., 2015) has shown that Pandemrix® contained 72.7% more influenza nucleoprotein than Focetria®, and sera from HLA-DQB1*0602-positive patients with Pandemrix®-associated narcolepsy contained antibodies that cross-reacted with the influenza nucleoprotein and HCRTR2. These antibodies were found in a significantly greater proportion of sera from patients with Pandemrix®-associated narcolepsy compared to non-narcoleptic individuals with either a history of H1NI infection or Focetria® vaccination. Whether a similar mechanism of molecular mimicry against HCRT-related targets occurs in sporadic narcolepsy remains to be determined. 2.6. Rodent models of narcolepsy Since the recognition that the prepro-orexin ligand knockout mouse has a behavioral phenotype that strongly resembles cataplexy (Chemelli et al., 1999), a number of rodent models have been produced to further understanding of the link between the Hcrt system and narcolepsy and as tools to probe the Hcrt system (Ch'ng and Lawrence, 2015; Darwinkel et al., 2014; De La Herran-Arita et al., 2011; Makela et al., 2010; Muraki et al., 2004; Tanaka et al., 2012; Tsunematsu et al., 2011, 2013) (Table 1). In mice, knockout of the HCRT precursor protein (Chemelli et al., 1999) or HCRT receptors (Kalogiannis et al., 2011; Kisanuki et al., 2001; Mieda et al., 2011; Willie et al., 2003), or genetic ablation of Hcrt neurons (Beuckmann et al., 2004; Black et al., 2013; Hara et al., 2001; Tabuchi et al., 2014) result in a narcoleptic phenotype including cataplexy, sleep/wake fragmentation and increased REM sleep propensity. The Hcrtr2 knockout mouse exhibits behavioral arrests resembling cataplexy but the phenotype is less severe than the Hcrt ligand knockout or the Hcrtr2-mutated narcoleptic dog (Willie et al., 2003). Although description of the Hcrtr1 knockout mouse has only appeared in abstract form (Kisanuki et al., 2001), cataplexy is thought to be mild (if it occurs at all) in these animals. Like their human counterparts (Besset et al., 1994; Dantz et al., 1994), narcoleptic mice exhibit normal homeostatic responses to sleep loss (Mochizuki et al., 2004). A rat model has been produced in which the Hcrt neurons are targeted with a HCRT2-saporin conjugate (Gerashchenko et al., 2003, 2001). Although these animals exhibit sleepiness and sleep attacks, neurons other than the Hcrt cells are likely also destroyed. In the orexin/ataxin-3 transgenic mouse model, the HCRT promoter drives expression of the polyglutamine neurodegenerative ataxin-3 protein, resulting in a postnatal loss of Hcrt neurons (Hara et al., 2001). Degeneration of the Hcrt neurons occurs postnatally with cataplexy occurring by 6 weeks of age. Orexin/ataxin-3 mice exhibit a phenotype strikingly similar to the Hcrtr2 mutant canine model (Lin et al., 1999) and the prepro-orexin knockout mouse (Chemelli et al., 1999) with deficits that include cataplexy (Black et al., 2013), premature entries into rapid eye movement (REM) sleep, and poorly consolidated sleep patterns. These animals also become obese relative to wild type mice, emulating the increased BMI characteristic of human narcoleptics (Dahmen et al., 2001; Kok et al., 2003; Schuld et al., 2002). A similar model has been produced in the rat, which exhibits episodes resembling cataplexy as well as fragmented vigilance states, decreased latency to REM sleep, and increased REM sleep time and decreased wakefulness during the dark (active) phase (Beuckmann et al., 2004). Despite profound cell loss, HCRT1 can be detected in CSF from these rats and prolonged wakefulness can further increase CSF HCRT1, indicating that the remaining Hcrt neurons can be activated (Zhang et al., 2007). A conditional model of Hcrt neuron ablation has been introduced that in many ways is superior to the orexin/ataxin-3 transgenic mouse model (Tabuchi et al., 2014). In the orexin/tTA; TetO diphtheria toxin model (the DTA mouse), degeneration of Hcrt neurons is controlled through the tetracycline transactivator (tTA) Tet-off (TetO) system. Dietary doxycycline (DOX) binds to the tetracycline transactivator (tTA) and prevents the synthesis of diphtheria toxin A fragment (DTA) through the TetO regulatory sequence. Because the tTA is attached to the HCRT promoter, removal of DOX from the diet (DOX(−) condition) initiates the synthesis of DTA for neurotoxic degeneration exclusively in Hcrt neurons. Thus, unlike orexin/ataxin-3 mice, DTA mice can be used in pre-post study designs to serve as their own Hcrt-intact controls. Hcrt neurodegeneration can be induced postpuberty to create a model with greater fidelity to most cases of human narcolepsy. The narcoleptic phenotype of these mice appears to parallel the symptomatic progression of human narcolepsy. As described in our original report (Tabuchi et al., 2014), DTA mice lost ~86% of their Hcrt neurons and exhibited fragmented sleep/wake states by 1 week of DOX(−) that continued to worsen over subsequent weeks (Fig. 1). By 2 weeks of DOX(−), DTA mice lost ~95% of their Hcrt neurons; cataplexy first appeared and progressively increased in frequency until at least 11 weeks DOX(−). Compared to other mouse models of narcolepsy, cataplexy levels are high in these mice after 4 weeks of DOX(−) (>97% Hcrt cell loss), as they exhibit approximately 20 to 60 cataplexy episodes on average during the dark period. The behavioral and electrophysiological aspects of cataplexy in DTA mice (Fig. 2) are similar to those observed in orexin/ataxin-3 and prepro-orexin knockout mice, including hyper-synchronous theta activity in the EEG and abrupt behavioral immobilization (Bastianini et al., 2012; Vassalli et al., 2013). Partial Hcrt neuron ablation can also be induced by reinstatement of DOX after variable durations of DOX(−) to permit examination of the physiological consequences of reduced Hcrt cell numbers, and to perhaps model narcolepsy type 2. The DTA mouse will certainly expand the tool box for addressing a number of questions in narcolepsy, including developmental aspects and network reorganization in terminal fields as Hcrt input degenerates. 2.7. Hypocretin and the neural circuitry of arousal states Because narcolepsy is characterized by EDS as well as cataplexy and since the Hcrt cells are known to innervate the monoaminergic and cholinergic cell groups involved in the promotion of wakefulness (Date et al., 1999; Peyron et al., 1998), the Hcrt system was proposed soon after its discovery to be a wake-promoting system that provides excitatory input to the monoaminergic and cholinergic systems. In this model, the absence of excitatory input from the HCRT peptides in narcolepsy results in an imbalance between the monoaminergic and cholinergic systems and the consequent behavioral state instability (Kilduff and Peyron, 2000). This concept has been elaborated further in the “flip flop” switch model of arousal state control in which the Hcrt system is proposed to pivot the arousal systems from sleep to wake (Saper et al., 2001, 2010). A “flip flop” switch for REM sleep control has also been posited in which GABAergic populations that are active during REM sleep (“REM on”) or that are silent during REM sleep and active during wakefulness (“REM off”) mutually inhibit each other (Lu et al., 2006). In both of these models, the absence of stabilizing input from the HCRTs in narcolepsy causes rapid switching between wakefulness and sleep, especially REM sleep. Models of arousal state circuitry now incorporate excitatory input from Hcrt neurons to populations that are active during wakefulness (Fig. 3), including the noradrenergic locus coeruleus (LC; which receives the densest innervation), serotonergic dorsal raphe (DR), histaminergic TM nucleus, cholinergic basal forebrain (BF) and pontine reticular formation, and GABAergic ventrolateral periaqueductal gray (vlPAG) (Brown et al., 2012; Burgess and Scammell, 2012; Sakurai, 2014). HCRT facilitates motor tone during wakefulness through direct excitation of motor neurons (Peever et al., 2003; Yamuy et al., 2004) and by suppression of the circuitry necessary for REM sleep atonia (Brown et al., 2012; Burgess and Scammell, 2012; Dauvilliers et al., 2014b; Luppi et al., 2011) (Fig. 3). During REM sleep, motor neurons are inhibited by glycine (Chase et al., 1989) and GABA (Brooks and Peever, 2008, 2012) from neurons in the medial medulla (MM) and by spinal interneurons (Krenzer et al., 2011). These medial medullary and spinal interneurons are stimulated by glutamatergic input from the sublaterodorsal nucleus (SLD, functionally equivalent to the subcoeruleus in cats) to induce atonia during REM sleep (Krenzer et al., 2011). Neurons in the SLD receive strong GABAergic inhibition during wakefulness that must be disinhibited for atonia to occur (Boissard et al., 2003, 2002; Pollock and Mistlberger, 2003). REM sleep atonia can also be induced via cholinergic excitation of the SLD (Brown et al., 2006; Heister et al., 2009). These GABAergic and cholinergic inputs to the SLD are controlled by monoaminergic (Aston-Jones and Bloom, 1981; Crochet et al., 2006; Hobson et al., 1975; McGinty and Harper, 1976) and hypocretinergic (Burlet et al., 2002; Mileykovskiy et al., 2005; Peyron et al., 1998) neurons that are active during wakefulness. In the absence of HCRT in narcolepsy, cataplexy results from disinhibition of the REM sleep atonia circuitry (Luppi et al., 2011) and disfacilitation of noradrenergic activation of motor neurons (Burgess and Scammell, 2012; Dauvilliers et al., 2014b) (Fig. 4). Genetic knockout and transgenic mouse models of narcolepsy have enabled some of the neural circuitry underlying cataplexy mechanisms to be discerned. Building upon electrophysiological studies in narcoleptic dogs which showed that a population of cells in the central nucleus of the amygdala (CeA) was active during cataplexy (Gulyani et al., 2002), lesions of the CeA in prepro-orexin knockout mice markedly reduced cataplexy, especially in the context of strong emotional cues (Burgess et al., 2013). Neuronal activity, as indicated by Fos immunostaining in the dorsal portion of the medial prefrontal cortex (mPFC), positively correlated with the number of chocolate-elicited cataplexy bouts in prepro-orexin knockout mice (Oishi et al., 2013). Reversible suppression of mPFC activity with a conditional glutamate-gated chloride channel decreased this chocolate-elicited cataplexy (Oishi et al., 2013). Both the CeA and the mPFC (via the basolateral amygdala-CeA circuit) send inhibitory GABAergic projections to brainstem regions that gate cataplexy, such as the vlPAG and the lateral pontine tegmentum (Burgess et al., 2013; Oishi et al., 2013). Focal restoration of HCRT by adeno-associated virus (AAV) gene transfer into the dorsolateral pons reduced cataplexy (Blanco-Centurion et al., 2013). Together, these studies provide evidence that activation of forebrain limbic circuitry by emotional stimulation underlies cataplexy triggering and support the hypothesis that Hcrt excitation may balance this limbic inhibition to maintain muscle tone. Brainstem mechanisms underlie the direct inhibition and reduced excitation of motor neurons that defines cataplexy (Luppi et al., 2011; Peever, 2011). In narcoleptic dogs during cataplexy, neurons in the atonia circuitry of the MM become active (Siegel et al., 1991) and noradrenergic neurons of the LC become quiescent (Wu et al., 1999), but a causal role for noradrenergic neurons in cataplexy induction was uncertain. A recent study in prepro-orexin knockout mice has provided direct evidence that manipulation of adrenergic tone controls atonia during cataplexy (Burgess and Peever, 2013). In this study, antagonism of α1 receptors on trigeminal motor neurons via microdialysis induced atonia of their masseter muscle targets during wakefulness. During cataplexy, the effects of this antagonism were absent, indicating lost adrenergic tone and cataplectic attacks were halted by application of an α1 agonist (Burgess and Peever, 2013). However, focal restoration of Hcrtr1 in the LC of narcoleptic mice that lack both HCRT receptors failed to inhibit cataplexy, whereas cataplexy was blocked by targeted expression of Hcrtr2 in the dorsal raphe (Hasegawa et al., 2014). These studies highlight the complexity of the circuitry that governs motor control during arousal states and exemplify new experimental approaches to discern these mechanisms. 3. Therapeutics 3.1. Historical approaches Many health professionals in the late 1800s viewed the narcolepsy syndrome as a form of epilepsy, despite Gelineau's and Westphal's arguments to the contrary (see “history” above). Others misinterpreted narcolepsy in Freudian terms as a hysterical defense mechanism, perhaps because Westphal's index case was a rapist who suffered from pathological sleepiness (Mignot, 2001). It is not surprising then that attempted treatments for the narcolepsy syndrome, as historically misunderstood at the turn of the 20th century, were ineffective and included the anti-epileptic potassium bromide, vasodilators picrotoxin and amyl nitrate vapors, agents aimed to improve motor abnormalities such apomorphine and strychnine, and, lastly, hydrotherapy, electricity and cauterization of the nape of the neck (Dement, 2007; Schenck et al., 2007). Even caffeine granules administered by Gelineau were found to be insufficient to counter narcoleptic sleepiness (Schenck et al., 2007). In the 1930s, mild success was found with the CNS stimulant ephedrine (Doyle and Daniels, 1932). Amphetamine was introduced to treat narcolepsy in 1935 and found to be more effective at controlling sleepiness (Prinzmetal and Bloomberg, 1935); it is still prescribed for this purpose (Hishikawa and Shimizu, 2007). Pharmacological alleviation of cataplexy was first documented in 1960 with the unexpected discovery that the tricyclic antidepressant imipramine significantly decreased the occurrence of cataplexy (Akimoto et al., 1960). Neuropsychiatrists at the time hypothesized that the effectiveness of antidepressants to improve the mood and increase the activity levels of depressed and psychotic patients might also confer therapeutic potential of antidepressants to ameliorate the EDS of narcolepsy (Hishikawa and Shimizu, 2007). Although imipramine had no effect on EDS and irresistible sleep attacks, it did improve cataplexy (Akimoto et al., 1960) through its active metabolite desipramine (Hishikawa et al., 1966). These and other tricyclic antidepressants, such as protripty-line and clomipramine, were found to inhibit REM sleep and thereby control sleep paralysis and hypnogogic hallucinations in addition to cataplexy (Hishikawa and Shimizu, 2007). While tricyclic antidepressants non-selectively block serotonergic and noradrenergic reuptake as the therapeutic mechanism of action, they also antagonize histamine and muscarinic acetylcholine receptors, leading to frequently intolerable side effects such as sedation, dry mouth, sweating, constipation, blurred vision, sexual dysfunction, tachycardia, orthostatic hypotension and are contra-indicated in cardiovascular disease (Guilleminault and Cao, 2011; Mignot, 2012). Withdrawal from tricyclic antidepressants can also induce rebound cataplexy for several weeks after discontinuation of the medication (Ristanovic et al., 2009). The monoamine oxidase inhibitor selegiline, which increases synaptic DA in addition to serotonin and norepinephrine, has been noted to also strongly suppress REM sleep and improve EDS but has sympathomimetic side effects that limit its utility (Hublin et al., 1994). 3.2. Current treatments Current treatments for narcolepsy are based on symptomatic management of sleepiness and cataplexy. At least half of all narcolepsy patients take medication to manage their condition (Ohayon, 2013) and may need to do so for the rest of their lives as there is no cure yet. Medication in combination with lifestyle adaptations helps return an estimated 80% of patients with narcolepsy back to near normal functioning (Mignot, 2012). Behavioral modifications such as scheduled bedtimes, wake-up times, and napping, either as long naps in the afternoon or several brief naps distributed throughout the day, can benefit daytime performance and sometimes reduce the doses of stimulants needed (Thorpy and Dauvilliers, 2015). Behavioral management of cataplexy has not been well developed, but patients themselves may choose to avoid or control known emotional triggers (CDER, 2013; Dauvilliers et al., 2014b). The US Food and Drug Administration (FDA) has approved five pharmacotherapeutics for the treatment of narcolepsy: sodium oxybate (Xyrem®), modafinil (Provigil®), armodafinil (Nuvigil®), methylphenidate, and amphetamine (CDER, 2013). 3.2.1. Sodium oxybate Sodium oxybate, the sodium salt of gamma-hydroxybutyrate (GHB), is currently the only approved drug to treat both cataplexy and EDS in narcolepsy (Bosch et al., 2012; Boscolo-Berto et al., 2012) and, as such, is favored as the first-line therapeutic. A randomized, double blind, placebo-controlled multi-center trial compared the effects of three doses of orally administered sodium oxybate with placebo for the treatment of narcolepsy in 136 patients (Study Group, 2002). Subjects received 3, 6, or 9 g doses of sodium oxybate or placebo taken in 2 doses, the first upon retiring and the second 2.5–4 h later, for 4 weeks. Compared to placebo, weekly cataplexy attacks were decreased by sodium oxybate at the 6 g dose (p = 0.0529) and significantly at the 9 g dose (p = 0.0008). The frequency of inadvertent naps or sleep attacks and nighttime awakenings showed similar dose-response trends, becoming significant at the 9 g dose (Study Group, 2002). In a follow-up study, 55 narcoleptic patients with cataplexy who had received continuous treatment with sodium oxybate for 7–44 months (mean 21 months) were enrolled in a double-blind treatment withdrawal paradigm (Study Group, 2003a). During the 2-week double-blind phase, the abrupt cessation of sodium oxybate therapy in the placebo patients resulted in a significant increase in the number of cataplexy attacks (median = 21; p < 0.001) compared to patients who remained on sodium oxybate (median = 0). Cataplexy attacks returned gradually with placebo patients reporting a median of 4.2 and 11.7 cataplexy attacks during the first and second weeks, respectively (Study Group, 2004). Interestingly, there was no evidence of withdrawal following abrupt cessation of chronic sodium oxybate dosing in the therapeutic range (Study Group, 2003b), which is consistent with the observation that the minimum daily dose of sodium oxybate associated with withdrawal is 18 g (Miro et al., 2002). In a 12 month open label study, 118 patients exhibited a 90% reduction in cataplexy without evidence of tolerance (Study Group, 2003a). Lastly, a double-blind, placebo-controlled study in 228 patients across 42 clinics found that nightly doses of 4.5, 6 and 9 g sodium oxybate for 8 weeks resulted in statistically significant median decreases in weekly cataplexy attacks of 57.0, 65.0 and 84.7%, respectively (Study Group, 2005). Post FDA approval of sodium oxybate for cataplexy and EDS in a double-blind, placebo-controlled study showed that sodium oxybate improved nighttime sleep with a dose-related increase in slow wave sleep and total sleep time and a reduction in nocturnal awakening (Black et al., 2010). The mechanism of action of sodium oxybate as a narcolepsy therapeutic has not yet been elucidated. Some researchers have proposed that the therapeutic efficacy may be mediated by the increase in slow wave activity during NREM sleep that has been seen after acute administration of sodium oxybate (Black et al., 2010; Boscolo-Berto et al., 2012; Van Cauter et al., 1997; Walsh et al., 2010), although others have questioned the physiological relevance of GHB-induced slow waves in the EEG (Meerlo et al., 2004; Vienne et al., 2010, 2012). Despite its acute sedating effects, the therapeutic efficacy of sodium oxybate gradually emerges over time and persists after the drug has been metabolized (Boscolo-Berto et al., 2012; Mamelak, 2009). This delayed effect suggests that the therapeutic mechanism may be indirect and possibly involves secondary interactions of GABAB receptors and DA transmission (Guilleminault and Cao, 2011; Huang and Guilleminault, 2009), perhaps through disinhibition of DA release via G-protein coupled inwardly rectifying potassium channels (Cruz et al., 2004). GHB is a low affinity GABAB agonist that has been shown to preferentially inhibit GABAergic interneurons in the ventral tegmental area and consequently disinhibit dopaminergic cells—unlike high affinity GABAB agonists such as baclofen that directly inhibit both GABAergic interneurons and catecholamine neurons (Cruz et al., 2004). The effects of GHB are known to at least partially depend on the GABAB receptor (Brown and Guilleminault, 2011), as GABAB antagonists block GHB-induced inhibition of neurons (Jensen and Mody, 2001). The GHB prodrug gamma-butyrolactone does not produce behavioral or EEG effects in mice lacking GABAB1 and GABAB2 subunits (Kaupmann et al., 2003; Vienne et al., 2010). Sodium oxybate may also exert therapeutic effects via the α4βδ subunit of the GABAA receptor, which has been identified as a high-affinity target for GHB and is likely the elusive, endogenous GHB receptor (Absalom et al., 2012). While sodium oxybate is often well-tolerated and has been reported to have a positive impact on the lives of many patients with narcolepsy, it is not a medication without challenges. GHB is a controlled substance with abuse potential and possible neurotoxic side effects (Langford and Gross, 2011; van Amsterdam et al., 2012). Other side effects, even at recommended doses, include nausea, confusion, CNS and respiratory depression, neuropsychiatric depression and confusion, incontinence, sleepwalking, automatic behaviors, and involuntary movements (Xyrem®, 2005). Adverse events associated with GHB abuse consist of seizures, loss of consciousness and death (Xyrem®, 2005). Titration up to the final dose can help manage most side effects (Thorpy and Dauvilliers, 2015). Sodium oxybate has a short half-life and is usually administered in a split dose, once at bedtime and again 2.5–4 h later (Black et al., 2010), which can be difficult to manage. Sodium oxybate is contraindicated in patients with succinic semialdehyde dehydrogenase deficiency or those patients who are also taking sedative hypnotics or other CNS depressants. 3.2.2. Modafinil and armodafinil For patients who cannot take sodium oxybate, racemic modafinil or the extended release formulation of enantiomerspecific armodafinil may be suitable alternatives. Modafinil was approved for the promotion of wakefulness in narcolepsy by the US FDA in 1998. A meta-analysis comparison of modafinil and sodium oxybate has indicated similar therapeutic efficacy of the two medications in controlling EDS; however, modafinil is ineffective at reducing cataplexy when compared to placebo (Golicki et al., 2010). In contrast to amphetamines, modafinil has limited dependency and abuse potential, induces wakefulness with less psychomotor agitation (Bastuji and Jouvet, 1988) and does not cause rebound hypersomnolence (Edgar and Seidel, 1997). However, some studies have shown that modafinil has rewarding properties as indicated by reinforcing, discriminative stimulus effects (Paterson et al., 2010) or conditioned place preference in mice (Shuman et al., 2012). Similarly, modafinil has been shown to increase motivation in mice as measured by progressive ratio breakpoint (Young and Geyer, 2010). Nevertheless, modafinil and armodafinil lack the symphathomimetic side effects of amphetamines and, as such, are the first-line choice of treatment to promote wakefulness (Guilleminault and Cao, 2011). The mechanism of action of modafinil is still not fully understood but likely involves the dopamine transporter (DAT) to inhibit DA reuptake (Wisor, 2013). Modafinil has been shown to bind the DAT with very low affinity but high selectivity, with no binding at norepinephrine or serotonin transporters (Mignot et al., 1994b). Mice with genetic knockout of the DAT failed to increase wakefulness in response to modafinil (Wisor et al., 2001). Extracellular DA levels were increased in the caudate nuclei of narcoleptic dogs after modafinil administration, as would be expected by diminished activity of the DAT (Wisor et al., 2001). Whether modafinil, and particularly armodafinil, might also act postsynaptically as D1 or D2 agonists is still equivocal (Mignot et al., 1994b; Seeman et al., 2009; Young and Geyer, 2010; Zolkowska et al., 2009). Interestingly, modafinil has been shown to induce wakefulness more effectively in narcoleptic mice that lack either the HCRT ligand or Hcrt neurons vs. wild type littermates, which suggests that modafinil may act on components of wake-promoting systems that are facilitated as a compensatory response to disrupted Hcrt signaling (Willie et al., 2005), perhaps such as the mesolimbic D2 receptors that have been observed to increase in narcoleptic dogs (Bowersox et al., 1987). A proposed mechanism by which armodafinil may increase wakefulness through inhibition of sleep-active ventro-lateral preoptic neurons has been ruled out because large, cell-specific lesions to this area do not inhibit wake promotion by armodafinil (Vetrivelan et al., 2014). 3.2.3. Methylphenidate and amphetamine Methylphenidate and amphetamines are now second and third-line therapies, respectively, used to counter EDS in narcolepsy, usually to supplement modafinil or sodium oxybate (Thorpy and Dauvilliers, 2015). Like modafinil, methylphenidate and amphetamine promote wakefulness through presynaptic enhancement of DA release (Burgess and Scammell, 2012; Burgess et al., 2010; Nishino and Mignot, 2011). Amphetamine and methylphenidate both increase the release of DA and to a lesser extent, norepinephrine and serotonin (Mignot, 2012). Amphetamine also increases synaptic DA by reverse efflux through the DAT and, at higher doses, inhibits the vesicular monoamine transporter (VMAT), to increase intracellular DA concentration presynaptically (Leviel, 2011; Mignot, 2012). Methylphenidate primarily inhibits the DAT (Leonard et al., 2004). The d-isomer of these stimulants is more potent than the l-isomer in increasing wakefulness because it is more specific to DA neurotransmission (Nishino and Mignot, 2011). Although both enantiomers of amphetamine are equipotent at reducing REM sleep (Nishino and Mignot, 2011), l-amphetamine is more effective than d-amphetamine in decreasing cataplexy (Mignot, 2012). The anticataplectic effects of amphetamine may relate to its action at transporters for norepinephrine and/or serotonin, as selective DAT inhibition fails to reduce cataplexy (Wisor, 2013). Unlike modafinil, amphetamine has several other actions that lead to an unfavorable side effect profile, such as vasoconstriction and cardiac stimulation resulting from increased noradrenaline in the periphery. At high doses, amphetamines inhibit monamine oxidase to further increase synaptic DA, which can lead to cytotoxicity or psychosis (Leviel, 2011; Mignot, 2012). Rapid increases in DA concentration and high levels of DA release and reverse efflux probably contribute to amphetamine addiction (Volkow et al., 2009). 3.2.4. Antidepressants Antidepressants have not been approved by the FDA for the treatment of cataplexy, as there are currently no clinical trials on the efficacy and safety of these compounds for the indication of narcolepsy. In practice, however, both older tricyclic antidepressants and newer, selective monoaminergic reuptake inhibitors have been used off label and are reported to be very effective at reducing cataplexy (Dauvilliers et al., 2014b; Mignot, 2012; Thorpy and Dauvilliers, 2015). Because antidepressants generally do not promote wakefulness, they are usually prescribed in combination with modafinil or other stimulants (Mignot, 2012). A series of pharmacological studies in narcoleptic dogs led to the conclusion that monoaminergic reuptake inhibitors alleviate cataplexy to the extent that they, or their metabolites, activate presynaptic noradrenergic neurotransmission (Nishino and Mignot, 1997) (see Fig. 3). For this reason, the highly selective serotonergic and noradrenergic reuptake inhibitor (SNRI) venlafaxine is the first-line choice of treatment for cataplexy among the antidepressants, and low doses (below the level required to treat depression) are sufficient for effective cataplexy treatment (Dauvilliers et al., 2014b; Mignot, 2012). The selective serotonergic reuptake inhibitor (SSRI) fluoxetine is also commonly used to alleviate cataplexy, but higher doses than the level needed to control depression are required, probably because its anticataplectic effect is likely mediated by its noradrenergic-promoting metabolite norfluoxetine (Langdon et al., 1986). SNRIs are advantageous over tricyclic antidepressants because, unlike the latter, they do not block muscarinic cholinergic, H1 histaminergic, nor α1-adrenergic receptors, thereby resulting in fewer side effects (Dauvilliers et al., 2014b; Guilleminault and Cao, 2011). However, similar to tricyclic antidepressants, SNRIs are associated with an increase in the number and severity of cataplexy attacks after withdrawal, even with gradual tapering of the drugs (Ristanovic et al., 2009; Wang and Greenberg, 2013). Antidepressants have also been linked to the development of REM behavior disorder (Ju et al., 2011), although it is not clear if narcolepsy predisposes for this risk. 3.3. Use of narcoleptic mouse models for therapeutic development The pharmacology of many of the narcolepsy therapeutics used today has been examined in narcoleptic dogs in studies aimed at elucidating the receptor subtypes, active metabolites and brain regions involved the therapeutic control of cataplexy (see Section 2.1). Prior to the discovery of HCRT, these neurochemical studies in narcoleptic dogs established that an imbalance between monoaminergic and cholinergic transmission contributes to cataplexy induction (Baker and Dement, 1985; Nishino, 2007; Nishino and Mignot, 1997). This imbalance was inferred from studies that showed a reduction of monoaminergic tone and/or cholinergic stimulation exacerbated cataplexy in narcoleptic dogs (Faull et al., 1986; Mefford et al., 1983; Nishino et al., 1991, 1990; Reid et al., 1994). Although it is unclear how the loss of Hcrt signaling leads to a cholinergic/aminergic imbalance, the absence of HCRT-induced excitation of monoaminergic cells in narcolepsy may underlie the observed compensatory increase in the sensitivity of these systems to wake-promoting compounds (see Section 3.2.2. and Fig. 4). Pharmacological studies in mouse models of narcolepsy recapitulate the monoaminergic/cholinergic imbalance first described in narcoleptic dogs. Compared to wild type controls, HCRT-deficient mice exhibit reduced DA turnover (Mori et al., 2010) and increased responsiveness to modafinil (Willie et al., 2005). Cortical release of DA, NE and 5-HT can be more potently evoked in Hcrtr2 knockout mice vs. wild types (Ortega et al., 2012). In prepro-orexin ligand knockout mice, activation of dopamine D1 receptors with SKF38393 or amphetamine decreased sleep attacks while blockade of this receptor with SCH23390 had the opposite effect (Burgess et al., 2010). This study also showed that cataplexy could be suppressed by amphetamine or blockade of dopamine D2-like receptors with eticlopride and exacerbated by the D2/D3 agonist quinpirole (Burgess et al., 2010), but these effects were not replicated by others (Black et al., 2013; Fujiki et al., 2009). Cholinergic hypersensitivity has been demonstrated in mice that lack both HCRT receptors, as the enzymes necessary for acetylcholine synthesis, vesicular transport and metabolite reup-take were upregulated in laterodorsal tegmental neurons (Kalogiannis et al., 2010). Enhancement of cholinergic tone with physostigmine and muscarinic antagonism with atropine increased and decreased cataplexy, respectively, in mice that lack both HCRT receptors (Kalogiannis et al., 2011). In orexin/ataxin-3 mice, exogenous HCRT increased wakefulness more effectively than in wild type controls (Mieda et al., 2004) whereas the HCRTR1 and HCRTR2 antagonist almorexant induced sleep less effectively (Black et al., 2013). Collectively, these findings suggest that, in murine narcolepsy, compensatory changes in monoaminergic and cholinergic pathways downstream from HCRT receptors are primed to facilitate wake promotion and play a role in cataplexy induction. A challenge in the use of mouse models of narcolepsy to study cataplexy is the high inter-individual variability in the expression of this phenotype. Although not well documented, the genetic background of the mouse model and environmental factors, such as physical restrictions from EEG recording tethers and handling stress, have been thought to reduce the frequency of cataplexy episodes (Hara et al., 2005; Scammell et al., 2009). To help foster uniformity between laboratories that had been attempting to capture the same presumed cataplexy behavior using different methodologies (variously termed “narcoleptic episodes”, “abrupt arrests”, “behavioral arrests”, “direct transitions into REM”, or “cataplexy-like events”), consensus criteria were established to define murine cataplexy (Scammell et al., 2009). An episode of murine cataplexy requires (1) abrupt nuchal atonia for ≥10 s, (2) video-confirmed behavioral immobility, (3) theta-predominated EEG activity, and (4) ≥40 s of immediately prior wakefulness (Scammell et al., 2009). Individual prepro-orexin ligand knockout mice have been reported to exhibit a wide range in the frequency of presumed cataplexy during the first 4 h of the dark period, with episode counts ranging from 8 to 27 (Chemelli et al., 1999), 2 to 87 (Willie et al., 2003), and 1 to 31 (Morawska et al., 2011). Other studies have reported presumed cataplexy in the prepro-orexin ligand knockout mice during the first 4 h of the dark period to occur with a mean (±S.E.) frequency ranging from 25.1 ± 8.7 (Kisanuki et al., 2001), 18 ± 9 (Espana et al., 2007), 8.5 ± 2.5 (Liu et al., 2008), and 2 ± 1 (Burgess et al., 2010). In orexin/ataxin-3 transgenic mice, presumed cataplexy during the entire 12 h dark period has occurred with a mean (±S.E.) frequency ranging from 4.8 ± 3.2 (Fujiki et al., 2009), 16.4 ± 2.3 (Liu et al., 2011), 9.3 ± 2.2 (Black et al., 2013), 12.4 ± 3.9 (Kantor et al., 2013), 7 ± 2 (Tabuchi et al., 2014) and 6.5 ± 1 (Hasegawa et al., 2014). In a survey of 39 hemizygous orexin/ataxin-3 mice instrumented for telemetric EEG and EMG recording, 34% of the mice exhibited < 3 episodes of consensus-defined cataplexy during the dark period (Black, unpublished observation). This wide variability in basal cataplexy expression can challenge interpretation of studies that use between-groups study designs (Blanco-Centurion et al., 2013; Hasegawa et al., 2014; Kantor et al., 2013; Liu et al., 2011). Interventions that significantly increase cataplexy frequency have included stimuli intended to evoke strong emotions, such as novel environments (Mieda et al., 2004), controlled access to running wheels (Espana et al., 2007), highly palatable food (Clark et al., 2009) including chocolate (Oishi et al., 2013; Tabuchi et al., 2014), the combination of chocolate and running wheels (Burgess et al., 2013), and attractive or aversive odors (Morawska et al., 2011). While chocolate has been shown to be highly effective at increasing cataplexy from 1.6-fold (Tabuchi et al., 2014) to 8.1-fold (Oishi et al., 2013), it could potentially confound studies by introducing dietary changes or secondary effects of prolonged wakefulness. For example, the three-fold increase in cataplexy observed after chocolate with a running wheel increased wakefulness to 90% of the 12 h dark period (Burgess et al., 2013). A manipulation that stays within the bounds of the Hcrt signaling system could represent a more pure means of amplifying cataplexy. For example, cataplexy has been shown to be provoked in orexin/ataxin-3 mice in a dose × time related manner with the dual HCRT receptor antagonist almorexant (Black et al., 2013), which has led to its use in a murine narcolepsy/cataplexy assay for pharmacological studies. This assay combines a pretreatment of almorexant at the start of the dark period with a dose 30 min later of either a test compound or desipramine (5 mg/kg) as a positive control for cataplexy reduction (see Section 3.1) in the presence of a running wheel. The combination of almorexant (30 mg/kg) and desipramine (5 mg/kg) does not change the amount of time spent awake, but enables both an increase and a decrease in cataplexy to be observed in orexin/ataxin-3 mice (Fig. 5). A separate control condition in the dual dosing paradigm with a wake-promoting therapeutic such as modafinil (Fujiki et al., 2009) (see Section 3.2.2) can be added to the assay to enable full assessment of both anticataplectic and wake-promoting efficacy of test compounds. 3.4. Future therapies 3.4.1. Pharmacotherapeutics on the horizon Histaminergic neurons in the tuberomammillary nucleus of the posterior hypothalamus are one of the targets of Hcrt innervation (Peyron et al., 1998) that mediate HCRT-induced wakefulness (Huang et al., 2001; Mochizuki et al., 2011). In narcolepsy, CSF histamine levels are reduced compared to healthy controls, especially in patients with HCRT1 deficiency (Nishino et al., 2009). The activity of histaminergic neurons is preserved during cataplexy which is thought to underlie the maintenance of consciousness (John et al., 2004). For these reasons, it has been hypothesized that increasing histaminergic tone by antagonism of the histamine H3 autoreceptor could exert therapeutic effects on EDS in narcolepsy. The histamine H3 inverse agonist pitolisant has recently been tested for efficacy as a wakefulness promoter in 95 narcoleptic patients in a double-blind, randomized, parallel-group controlled clinical trial (Dauvilliers et al., 2013b). Pitolisant was found to increase wakefulness more than placebo and to a level indistinguishable from modafinil as measured by the Epworth Sleepiness Scale (Dauvilliers et al., 2013b). While pitolisant may be ultimately become a useful wake-promoting therapeutic in practice, it is not expected to show anticataplectic effects as histaminergic cells remain active during cataplexy (John et al., 2004). ADX-N05 is a phenylalanine derivative with dopaminergic and noradrenergic activity that has been assessed for treatment of daytime EDS in narcolepsy (Bogan et al., 2013). In a 10-center double-blind, placebo-controlled, randomized, cross-over study, ADX-N05 (150–300 mg/day) increased mean sleep latency by 11.8 min on the Maintenance of Wakefulness Test (MWT) compared to placebo, as well as on the Epworth Sleepiness Scale (ESS) and the Clinical Global Impressions-Change (CGI-C) after one week of treatment. A subsequent Phase 2 study confirmed the results on the primary (sleep onset latency on the MWT) and secondary (ESS and CGI-C) efficacy endpoints in patients with narcolepsy (Black et al., 2014a). To this point, there is no evidence that ADX-N05 (now known as JZP-110) is effective in cataplexy. GABAB agonism with R-baclofen, an enantiomer-specific form of racemic baclofen that has been used for decades in the treatment of muscle spasticity (Hudgson and Weightman, 1971), has recently been evaluated as a narcolepsy therapeutic in a preclinical study using two mouse models of Hcrt neuron ablation (Black et al., 2014b). The original goal of the research was to use the GABAB agonist baclofen in its most bioactive isoform as a tool compound to probe the therapeutic mechanism of action of sodium oxybate, which is known to be a partial agonist at GABAB receptors (Xie and Smart, 1992). It was hypothesized that if sodium oxybate acts on the GABAB receptor to normalize arousal states, then R-baclofen would mimic the effects of sodium oxybate on the symptoms of narcolepsy. At the doses tested (in a paradigm to model the chronic, twice-nightly administration of sodium oxybate in humans), R-baclofen was more effective than sodium oxybate at increasing the duration, intensity and consolidation of NREM sleep. As a consequence, the narcoleptic mice spent more time in consolidated bouts of wakefulness during the subsequent active phase. R-baclofen was also more effective than sodium oxybate at reducing cataplexy. Whether the high-affinity GABAB receptor agonist R-baclofen exerts greater anticataplectic efficacy via GABAB receptors on glutamatergic SLD neurons (Fig. 4) than the low affinity agonist sodium oxybate needs to be determined. The evaluation of R-baclofen and racemic baclofen in a mouse model of Fragile × Syndrome has led to clinical trials of arbaclofen (STX209) (Berry-Kravis et al., 2012; Pacey et al., 2011) and, more broadly, to preclinical efficacy tests of baclofen isoforms in mouse models of autism (Silverman et al., 2015). These paths of research encourage clinical trials of arbaclofen for narcolepsy. 3.4.2. Hypocretin replacement therapy Replacement of the HCRT that is lost from Hcrt neurodegeneration, like replacement of lost DA in Parkinson's disease, is an obvious therapeutic approach, but a number of obstacles must be overcome before being realized. However, since cataplexy and sleep fragmentation are exacerbated in narcoleptic orexin/ataxin-3 mice treated with the dual HCRTR antagonist almorexant, only small amounts of HCRT may be necessary for therapeutic benefit (Black et al., 2013). The proof of principle that HCRT replacement therapy could be effective stems from a study in which the narcoleptic mouse phenotype was independently rescued by pharmacological and genetic means. In the first approach, intracerebroventricular administration of HCRT1 into orexin/ataxin-3 mice (in which the vast majority of Hcrt neurons are ablated) rescued the narcoleptic phenotype (Mieda et al., 2004). Neither the HCRT2 peptide nor any HCRT fragments were tested for efficacy so whether other molecules related to HCRT1 can also rescue the phenotype is currently unknown. In the second approach, transgenic mice that express prepro-orexin under control of CAG promoter were bred with Hcrt neuron-ablated orexin/ataxin-3 mice to generate CAG/orexin; orexin/ataxin-3 mice. In these double transgenic mice, prepro-orexin is ectopically expressed throughout the brain but the Hcrt neurons are ablated. Narcolepsy symptoms such as sleep onset REM sleep, fragmentation of wakefulness in the dark period and cataplexy were completely inhibited in CAG/orexin; orexin/ataxin-3 mice, indicating that ectopic HCRT production could prevent narcolepsy. Given the efficacy of HCRT treatment for cataplexy suppression in the proof-of-principle study described above, a small molecule, brain penetrable HCRT receptor agonist would be ideal for symptomatic treatment of narcolepsy. Although the dual orexin receptor antagonist suvorexant has recently been approved by the FDA for the treatment of insomnia, development of agonists for G protein-coupled receptors is much more difficult than developing antagonists. Among other factors, the HCRT peptides are 33 and 28 amino acids in length and each peptide assumes a particular tertiary conformation in vivo. Screening for a small molecule that fits the binding pocket of HCRTR1 and/or HCRTR2 and contacts the necessary residues for receptor activation is a challenging task. However, the recently-described crystal structure of the human HCRTR2 (Yin et al., 2015) may accelerate development of orexin receptor agonists. Indeed, a non-peptide agonist that is selective for HCRTR2 with submicromolar potency has recently been introduced (Nagahara et al., 2015); however, the potential clinical utility of the compound remains to be determined. A major hindrance in the development of HCRT replacement therapy, whether the replacement molecule is the exogenous peptide, a prodrug precursor, or a small molecule agonist, is penetration of the blood–brain barrier. Delivery of HCRT1 through an alternative entry to the CNS along olfactory nerves has been attempted with intranasal administration in animals (Deadwyler et al., 2007; Dhuria et al., 2009) and narcoleptic patients (Baier et al., 2011; Weinhold et al., 2014) with limited success. A new technology that may aid delivery of HCRT peptides through the blood–brain barrier is receptor-mediated transcytosis using “Brain Shuttle” constructs (Niewoehner et al., 2014). The current Brain Shuttle technology consists of an anti-transferrin antibody that has been modified to monovalently bind the transferrin receptor (thereby resisting lysosomal degradation) which can also be fused to either another antibody (Niewoehner et al., 2014) or, hopefully in the future, to macromolecules such as HCRT1 as the therapeutic cargo. 3.4.3. Genetic and stem cell therapies Another potential approach is viral vector-based delivery of the prepro-HCRT gene. Expression of HCRT using recombinant adeno-associated virus (rAAV) in neurons of the zona incerta or lateral hypothalamus effectively prevented cataplexy in narcoleptic orexin/ataxin-3 mice (Liu et al., 2011); transfer into the striatum was ineffective. Orexin gene transfer into the dorsolateral pons of orexin KO mice also significantly decreased cataplexy and modestly improved wake maintenance compared to orexin KO mice that did not receive rAAV (Blanco-Centurion et al., 2013). Orexin gene transfer into the mediobasal hypothalamus only improved the timing and consolidation of sleep and wakefulness and did not change cataplexy in orexin/ataxin-3 mice (Kantor et al., 2013). Together, these results strongly suggest that viral-based HCRT replacement may be another avenue for treatment of narcolepsy. Novel approaches using pluripotent stem cells for treatment of narcolepsy are also under development. The use of induced pluripotent stem cells as in vitro models specific to individuals with narcolepsy has recently been proposed as a means to identify potential autoantigens as well as cell degenerative and survival responses (Liblau et al., 2015). Several groups have recently succeeded to generate hypothalamic neuropeptidergic neurons, including Hcrt neurons from human pluripotent stem cells (Merkle et al., 2015; Wang et al., 2015). Although much work is yet to be done, transplantation of HCRT-producing neurons generated from pluripotent stem cell from narcolepsy patients might become a useful treatment for narcolepsy in the future. 3.4.4. Immuno or neuroprotective therapies Assuming that narcolepsy is indeed due to an autoimmune attack on Hcrt neurons, the most fundamental treatment would be prevention of its development using immunotherapies or neuro-protective strategies. This approach is challenging due to the difficulty in finding suitable numbers of cases of narcolepsy close to disease onset for well-controlled studies. In a case of childhood narcolepsy, the immunosuppressant prednisone failed to improve EDS and to prevent the development of cataplexy when administered 3 months after abrupt onset (Mignot and Nishino, 2007). Attempts to remove hypothesized autoantibodies with intravenous immunoglobin (IVIg) have produced mixed results. Early attempts at IVIg therapy resulted in subjective improvement in EDS and cataplexy despite persistently low HCRT1 levels, but effects were not seen in cases in which narcolepsy onset was a year or more prior to IVIg (Dauvilliers et al., 2004). However, IVIg administered very close to disease onset (15 d) normalized HCRT1 levels and resulted in clinical improvement in cataplexy (Dauvilliers et al., 2009). More recent attempts using IVIg found that HCRT1 levels remained abnormal and any symptom improvements tended to be temporary (Knudsen et al., 2012, 2010; Valko et al., 2008). 4. Perspective 4.1. Voice of the patient In 2013, the US. FDA issued a patient-focused drug development initiative in which public meetings were held to gather information from the patients’ perspective in 20 specific disease areas. One of these meetings focused on narcolepsy in order to directly hear from patients, their caregivers and advocates about the impact of the disease on their lives and their experiences with currently available therapies. Several key themes emerged from the dialog with over 120 participants and highlighted the chronic, debilitating toll that narcolepsy takes on patients’ lives and the challenges they face in managing their condition (CDER, 2013). First, patients emphasized that EDS is the most problematic symptom that affects their lives, mostly because of the difficulty they face in fighting against it, or the consequences they face from chronic sleep deprivation such as cognitive impairments, the feeling of “brain fog”, and automatic behaviors. Second, the unpredictable loss of control that accompanies cataplexy, hallucinations and sleep paralysis can be terrifying. Other aspects of the disease, such as weight gain, insomnia, mood changes and depression were specifically noted as detrimental to patients’ lives. Third, patients reported that their symptoms can change over time. For example, a seasonal or monthly change occurs in their ability to sleep, or new symptoms emerge, such as cataplexy. Some symptoms, even when treated, worsen with time. Fourth, many patients reported that challenges often arise with medication use, including variable effectiveness, side effects, the development of tolerance, and access to currently-available treatments. These challenges necessitate a continued need to switch therapeutics, which is itself problematic. Last but not least, patients emphasized the significant impact that narcolepsy exerts on their social, emotional, and financial well-being and the frustration they experience as their disorder is misunderstood by colleagues, health care providers, and other people in their lives. This report from the patients’ perspective on narcolepsy underscores the critical need for innovative new therapies to improve patients’ lives. It also highlights some areas of unmet needs faced by patients, such as cognitive impairment, automatic behavior and anxiety, that have not yet been specifically addressed in drug development for narcolepsy. 4.2. Challenges in the development of narcolepsy therapeutics Before discussing the challenges in the development of narcolepsy therapeutics, it is appropriate to assess the advantages in targeting this disorder. First, despite the earlier confusion with epilepsy, narcolepsy type 1 is readily identified as a distinct diagnostic entity. Although other sleep disorders such as sleep apnea can result in EDS and even disrupted nighttime sleep, cataplexy is the pathognomonic symptom that is diagnostic of this disorder. Second, narcolepsy has clearly defined criteria used in both research and clinical (Eichler, 2014) settings. As described above (see Section 1.2.2), one of the first applications of the MSLT was to aid in the diagnosis of narcolepsy (Richardson et al., 1978) and the MSLT has proven to be useful as an objective measure of EDS for the subsequent ~40 years. Third, the loss of Hcrt neurons in the hypothalamus is now widely accepted as the cause of narcolepsy type 1. Although the reason for this cell loss that would complete our understanding of the etiology of this disorder remains obscure, the consequent interruption of Hcrt neurotrans-mission resulted in another asset for study of this disorder: a decline of CSF HCRT1 levels that has thereby enabled a diagnostic test. Fifth, recognition of Hcrt cell loss and interruption of Hcrt neurotransmission as an essential component of human narcolepsy type 1 has allowed creation of numerous animal models that recapitulate various symptoms of narcolepsy. These animal models will be extremely valuable for testing of potential therapeutics, as exemplified by a recent study (Black et al., 2014b). Particularly critical for future therapeutic development have been the proof of principle studies that have utilized some of these animal models to demonstrate mitigation of cataplexy and normalization of REM sleep levels after HCRT replacement by either pharmacological or genetic means (Blanco-Centurion et al., 2013; Liu et al., 2011; Mieda et al., 2004). Sixth, recognition of Hcrt cell loss as the source of interruption of Hcrt neurotransmission suggests that, in contrast to Hcrtr2-mutated narcoleptic Doberman pinschers and Labrador retrievers, HCRT receptors in human narcoleptics are intact and functional and thus represent viable pharmacological targets for HCRT replacement therapy whether by small molecule agonists or other innovative therapeutic means. Lastly, the search for new therapeutics will be greatly aided by the rich database of information that has resulted from extensive pharmacological testing in narcoleptic Doberman pinschers (Crocker et al., 2005; Nishino and Mignot, 1997), as these studies have suggested the downstream targets of Hcrt innervation that become unbalanced with Hcrt cell loss and should be normalized by effective therapeutics. Despite the advantages stated above, a number of obstacles remain problematic for future development of narcolepsy therapeutics. Perhaps the paramount obstacle is the relatively low prevalence of this disorder which may limit the number of companies that are willing to invest in new therapeutic development for a disorder with a small market. With a prevalence of 0.05%, there are approximately 150,000 individuals with narcolepsy in the US. This is obviously a small fraction of the population of patients that are suffering from depression, autism or Alzheimer's disease, for example. On the other hand, the US. Orphan Drug Act of 1983 was specifically designed to facilitate the development and commercialization of drugs to treat rare diseases such as narcolepsy. The FDA's Office of Orphan Products Development administers the Orphan Drug Designation program that provides orphan status to drugs and biologics for the treatment diseases and disorders that affect fewer than 200,000 people in the US. Government intervention on behalf of orphan drug development can be in the form of tax incentives, enhanced patent protection and marketing rights, subsidization of clinical research and even the creation of a government-run enterprise to engage in research and development. Since 1984, seven applications for orphan drug status for the treatment of narcolepsy have been successful (Table 2). As the era of the multi-billion dollar blockbuster drug recedes and precision medicine segments diseases and disorders into smaller markets, rare disorders and diseases, particularly life-long diseases with early age of onset such as narcolepsy, may become more attractive to large pharmaceutical companies specifically because of the possibility of orphan drug designation and the resultant market exclusivity. Another major challenge in the development of narcolepsy therapeutics is the incomplete understanding of the etiology of this disorder. Although loss of Hcrt cells is widely accepted the final endpoint of narcolepsy type 1, it is unclear why Hcrt neurons are lost in the first place. While some evidence suggests that narcolepsy type 2 may result from less severe loss of Hcrt neurons than in narcolepsy type 1 (Thannickal et al., 2009), whether these two types of narcolepsy differ in patterns of neurodegeneration, mechanisms of cell loss, compensatory preservation of remaining cells or Hcrt function (e.g., axonal sprouting, increased HCRT release, decreased degredation, etc.) is unknown. Clearly, a genetic predisposition involving the HLA system has been established as an important component across multiple ethnic populations and in family studies, but only about 25% of monozygotic twins are concordant for narcolepsy (Mignot, 1997), implicating unspecified environmental factors that may vary among patients. Although an autoimmune mechanism was suggested by the original association with HLA-DR2 over 30 years ago (Juji et al., 1984), as indicated in Section 2.5, the search for autoantibodies in sporadic narcolepsy has been fruitless to date, which stands in stark contrast to well-established autoimmune neurological disorders such as myasthenia gravis (Graus et al., 2010). Nonetheless, it is appropriate here to keep in mind the old adage that “absence of evidence is not evidence of absence.” The existence of streptococcal infections in a large proportion of recent narcolepsy onset cases may be an important clue both in the search for an environmental trigger and in the mechanism of the autoimmune response (Aran et al., 2009). Surmounting the blood–brain barrier is a formidable challenge to any drug development effort directed toward a CNS target. In the case of narcolepsy, as indicated above, the most straightforward therapeutic strategy for symptomatic treatment would be HCRT replacement therapy. However, because the HCRTs are neuropep-tides, peripheral administration renders them vulnerable to enzymatic hydrolysis, although the disulfide bridges and two alpha-helices in HCRT1 (Kim et al., 2004) confers greater stability than that of HCRT2 (Lee et al., 1999). For small molecules, the size and charge limitations encapsulated in Lipinski's rule of 5 applies (Lipinski et al., 2001). Hypocretin replacement therapy enabled by gene transfer may be one means of bypassing the blood–brain barrier, however, controlling the diurnal timing of ectopic HCRT expression and its restriction to wakefulness is an important problem that has not yet been solved (Willie et al., 2011). Until a solution is found, the nighttime use of HCRTR1 and HCRTR2 antagonists, such as suvorexant (Michelson et al., 2014), could conceivably be employed, but the benefit of increased sleep propensity must be weighed against the possibility of cataplexy exacerbation in patients with narcolepsy (Black et al., 2013). Construction of a “Brain Shuttle” for receptor-mediated transcytosis across the blood–brain barrier (see Section 3.4.2) is a promising approach, but the technology is still in its infancy for peptide delivery. Determination of the minimal peptide length of the HCRTs that can both bind to the shuttle antibody and engage HCRTR2 to effect a cellular response is a difficult challenge. With respect to small molecule development, HCRT replacement therapy implies synthesis of an agonist at the HCRT receptors. In contrast to antagonists, identification of receptor agonists is much more difficult, likely due to the constraints on identifying molecules that can both fit the binding pocket of a particular receptor and contact the appropriate amino acid residues for receptor activation. The recently-described crystal structure of the human HCRTR2 (Yin et al., 2015) may facilitate development of receptor agonists by enabling virtual screening of compounds. Another approach would be the development of a positive allosteric modulator at the HCRT receptors, but the efficacy of this approach will be limited by the availability of enough endogenous ligand to activate the receptor. On the other hand, postmortem human studies suggest that 5–15% of Hcrt neurons may remain even in patients who have experienced narcolepsy symptoms for over 50 years (Thannickal et al., 2000). Furthermore, only a small degree of HCRT receptor activation may be necessary for therapeutic benefit since cataplexy and sleep fragmentation are exacerbated in narcoleptic orexin/ataxin-3 mice treated with almorexant (Black et al., 2013). In conclusion, narcolepsy represents a model neurodegenerative disease for the development of therapeutics. The loss of a small, discrete and homogeneous population of neurons translates to a relatively simple system in which to develop gene- and pharmacology-based therapies. Although the symptomatic consequences of Hcrt cell loss are wide and varied, rodent animal models recapitulate the cardinal features of arousal state boundary dysregulation and cataplexy with clearly defined end points. Restoration of Hcrt neurotransmission, whether by small molecule agonists or by exogenous peptides, may be the most parsimonious means of treating the many disparate features of narcolepsy. Whether the compensatory neurochemical and molecular changes that accompany Hcrt cell loss can be fully surmounted, or even reversed, remains to be determined. Until the cause of Hcrt neurodegeneration can be understood and prevented, development of therapeutics to provide symptomatic relief by improving Hcrt neurotransmission remains a worthy goal. Acknowledgments This work was supported by NIH R01 HL059658, R01 NS077408, R21 NS087550, R01 NS082876, R21 NS083639 and R21 NS085757. We thank Drs. Michael Schwartz, Theresa Steininger and Jed Black for helpful comments on the manuscript. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Abbreviations 5-HT 5-hydroxytrytamine (serotonin) α1 noradrenergic alpha 1 receptor α2 noradrenergic alpha 2 receptor ACh acetylcholine Atax orexin/ataxin-3 transgenic BAC bacterial artificial chromosome BAT brown adipose tissue BF basal forebrain BLA basolateral amygdala BMI body mass index CeA central nucleus of the amygdala CGI-C Clinical Global Impressions-Change CSF cerebrospinal fluid Ctx cerebral cortex D1 dopamine 1 receptor D2 dopamine 2 receptor D3 dopamine 3 receptor DA dopamine DAT dopamine transporter DOX doxycycline DR dorsal raphe nucleus DTA orexin/tTA TetO DTA transgenic EDS excessive daytime sleepiness EEG electroencephalograph EMG electromyograph ESS Epworth Sleepiness Scale FDA Food and Drug Administration FECT Food-elicited Cataplexy Test GABA gamma-aminobutyric acid GABAA gamma-aminobutyric acid A receptor GABAB gamma-aminobutyric acid B receptor Glu glutamate Gly glycine H3 histamine receptor 3 HA histamine Hcrt hypocretin (orexin) HCRT1 hypocretin 1 peptide (orexin-A peptide) HCRT2 hypocretin 2 peptide (orexin-B peptide) HCRTR1 hypocretin receptor 1 (orexin-1 receptor, OX1R) HCRTR2 hypocretin receptor 2 (orexin-2 receptor, OX2R) GHB gammahydroxybutyrate (sodium oxybate) HLA human leukocyte antigen IVIg intravenous immunoglobin LC locus coeruleus LDT laterodorsal tegmental nucleus MCH melanin-concentrating hormone MM medial medulla mPFC medial prefrontal cortex MSLT Multiple Sleep Latency Test MWT Maintenance of Wakefulness Test NE norepinephrine NREM non-rapid eye movement sleep PANDAS pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections PFH perifornical nucleus PPT pedunculopontine tegmental nucleus rAAV recombinant adeno-associated virus REM Rapid eye movement sleep SLD sublaterodorsal nucleus SNRI serotonergic and noradrenergic reuptake inhibitor SOREMP sleep-onset REM period SSRI selective serotonergic reuptake inhibitor T b core body temperature TetO tetracycline operator TM tuberomammillary nucleus TRIB2 tribbles homologue 2 tTA tetracycline transactivator VEH Vehicle vlPAG ventrolateral periaqueductal gray WT Wild type ZT Zeitgeber time Fig. 1 Hypnograms depicting arousal state (C, cataplexy; R, REM sleep; W, wakefulness; S, NREM sleep) changes over time in a representative orexin/tTA; TetO diphtheria toxin (DTA) mouse before hypocretin neuron ablation, while maintained on doxycycline chow (DOX(+)), and 1, 2, 4 and 13 weeks (wk) after return to standard mouse chow (DOX(−)) in comparison to an orexin-ataxin3 mouse. Hypnograms are plotted for 24 h beginning at lights off at 20:00, with the 12 h dark period in black and 12 h light period in orange. Sleep/wake fragmentation occurred within 1 wk after DOX(−) and progressively worsened throughout the recording period. Cataplexy appeared after 2 wks DOX(−), at which point arousal states resembled those of the orexin-ataxin3 mouse. Cataplexy was highly enriched at 13 wks DOX(−) when compared to earlier ages and to the orexin-ataxin3 mouse. Fig. 2 Electrophysiological and behavioral indices of cataplexy in a representative DTA mouse at 9 weeks DOX(−). (A) 80 s recording (demarcated in 10 s epochs by vertical lines) shows ~40 s cataplexy episode (bracket) and prior wakefulness with wheel-running activity revealed by EEG (blue), EMG (black), EEG periodogram (blue area under curve, 0–25 Hz), and gross motor activity (red, as inferred by telemetry unit signal strength). (B) Frames from video in the seconds before cataplexy onset, at onset and during the bout of cataplexy, and 2 s after cataplexy termination. Bars = 200 μV (EEG and EMG), 200 μV2 (EEG periodogram), and 5 arbitrary units (signal strength measure of gross motor activity). Fig. 3 Schematic illustrating the brain regions currently known to be involved in the control of wakefulness and muscle tone. Neuronal populations that are active during wakefulness (green) consist of hypocretin neurons (Hcrt) that project most densely and provide excitatory input (solid arrowheads) to the locus coeruleus (LC) and other wake-promoting areas: the basal forebrain (BF), tuberomamillary nucleus (TM), dorsal raphe (DR), laterodorsal and pedunculopontine tegmental nuclei (LDT/PPT), ventrolateral periaqueductal gray (vlPAG), and directly to cortex and spinal motor neurons. Hcrt neurons also increase motor tone through suppression of REM sleep atonia circuitry, which consists of sublaterodorsal nucleus (SLD) excitation of medial medulla (MM) and spinal interneurons that inhibit motor neurons. Active inhibition of atonia circuitry is mediated by HCRT excitation of monoaminergic and GABAergic pathways that inhibit (blunt terminals) the SLD. Disinhibition of REM sleep atonia circuitry during wakefulness may underlie cataplexy in narcolepsy (see Fig. 4). Key: solid lines/arrows, active excitation; solid lines/blunt terminals, active inhibition; dashed lines/arrows, disfacilitation; dashed lines/blunt terminals, disinhibition; thick black lines, Hcrt projections; red, neurons and pathways that are active in REM sleep; REMoff, neurons that are silent in REM sleep; Ctx, Cerebral cortex; mPFC, medial prefrontal cortex; BLA, basolateral amygdala; CeA, central nucleus of the amygdala; MCH, melanin-concentrating hormone cells; ACh, acetylcholine; HA, histamine; Glu, glutamate; GABA, gamma-aminobutyric acid; Gly, glycine; 5-HT, serotonin; NE, norepinephrine; α2, noradrenergic autoreceptor; α1, noradrenergic α1 receptor. Fig. 4 Schematic illustrating arousal-state circuitry in narcolepsy and cataplexy. Hypocretin (Hcrt) neurons and their projections to wake-promoting regions (see Fig. 3) are absent in narcolepsy. Presumably as a compensatory consequence, pontine cholinergic supersensitivity (thick lined box), monoaminergic hypoactivity in the amygdala (thin lined boxes) and increased numbers of HA neurons develop. Neuronal populations that are active during cataplexy (yellow) include regions that are also active during wakefulness and REM sleep. Cataplexy can be triggered by positive emotional stimuli that activate neurons in the mPFC and amygdala, which then may disinhibit REM sleep atonia circuitry. In individuals without narcolepsy, HCRT excitation onto LC neurons could balance the GABAergic inhibition from CeA to maintain normal muscle tone. Activation of the atonia circuitry is mediated by withdrawal of GABAergic inhibition of SLD neurons from the vlPAG and monoaminergic inhibition from the DR and LC. Direct activation of SLD neurons could result from upregulated cholinergic mechanisms in the LDT/PPT. Cataplexy can be alleviated by antidepressants that increase noradrenergic tone via blockade of α2 autoreceptors; mechanisms of GABAB therapeutics for cataplexy are unknown, but may involve inhibition of SLD neurons to suppress atonia. Key: see Fig. 3. Fig. 5 The narcolepsy/cataplexy assay for orexin/ataxin-3 transgenic mice. (A) Cataplexy density (the number of cataplexy bouts per time awake) and (B) percent time awake during the 6 h following pretreatment with almorexant (ALM, black) vs. vehicle (VEH, white) at ZT12 and treatment with desipramine (DES 0–5 mg/kg) 30 min later. Two-way repeated measures ANOVA: * p < 0.05 vs. VEH pretreatment, *p < 0.05 vs. DES (0 mg/kg); n = 4. Table 1 Rodent models of narcolepsy, hypocretin dysfunction or transgenic tools. Strain & alternative names Structural mutation Phenotype Reference Mouse Prepro-orexin knockout; orexin−/− Constitutively absent Hcrt precursor gene Sleep/wake fragmentation, reduced REM onset, increased REM time, decreased wake during dark period, frequent cataplexy; obesity Chemelli et al. (1999); Mochizuki et al. (2004) HcrtR1 knockout; OX1R−/− Constitutively absent HcrtR1 Mild sleep/wake fragmentation, no cataplexy Kisanuki et al. (2001) HcrtR2 knockout; OX2R−/− Constitutively absent HcrtR2 Mild sleep/wake fragmentation, sleep attacks, mild cataplexy Willie et al. (2003) HcrtR1 & HcrtR2 double knockout; OX1R−/−/OX2R−/− Constitutively absent HcrtR1 and HcrtR2 Sleep/wake fragmentation, reduced REM onset, increase REM time, decreased wake during dark period, cataplexy Kisanuki et al. (2001); Kalogiannis et al. (2011) Orexin/ataxin-3 transgenic, Ataxin Postnatal Hcrt neuron ablation Sleep/wake fragmentation, reduced REM onset, increased REM time, decreased wake during dark period, variable cataplexy; obesity Hara et al. (2001) Orexin/eGFP transgenic eGFP reporter driven by Hcrt promoter None; permits visualization of Hcrt cells Yamanaka et al. (2003); Muraki et al. (2004) Prepro-Hcrt overexpressor Constitutive overexpression of prepro-orexin, decreased HcrtR2 in hypothalamus Small decrease in REM during recovery from sleep deprivation Makela et al. (2010) O/E3 null mutant Constitutively absent helix-loop-helix transcription factor, decreased Hcrt cell number Sleep/wake fragmentation, reduced REM onset, increased REM time, decreased wake during dark period, cataplexy De la Herran-Arita et al. (2011) Orexin/Halo transgenic Halorhodopsin expressed in Hcrt cells None; permits inhibition of Hcrt cells Tsunematsu et al. (2011) Orexin/tTA;TetO Chr2 (C128S) double transgenic Conditional step-function opsin expression in Hcrt cells None; permits widespread expression of ChR2 in Hcrt cells under Dox(-) control & excitation of Hcrt cells Tanaka et al. (2012) Orexin/tTA;TetO ArchT double transgenic Conditional archaerodopsin expression in Hcrt cells Unknown; permits wide-spread expression under Dox(-) control & inhibition of Hcrt cells Tsunematsu et al. (2013) OX/tTA transgenic Tetracycline transactivator driven by Hcrt promoter None; useful with Tet-on or Tet-off expression systems, or opsins through breeding or viral delivery Tsunematsu et al. (2013) Orexin/tTA;TetO DTA double transgenic, DTA Conditional Hcrt neuron ablation Extreme sleep/wake fragmentation, reduced REM onset, increased REM time, decreased wake during dark period, extreme cataplexy; obesity Tabuchi et al. (2014) OXlR/ eGFP transgenic eGFP reporter driven by first coding exon of Hcrtr1 gene None; permits visualization of cells that express Hcrtr1 Darwinkel et al. (2014); Ch'ng and Lawrence (2015) Rat Hcrt2-saporin conjugate Ablation of Hcrt, MCH and ADA cells Sleep/wake fragmentation, reduced REM onset, increased REM time, decreased wake during dark period, cataplexy, attenuated sleep/wake circadian rhythm Gerashchenko et al. (2001) Orexin/ataxin-3 transgenic Postnatal Hcrt neuron ablation Sleep/wake fragmentation, reduced REM onset, increased REM time, decreased wake during dark period, cataplexy Beuckmann et al. (2004) Table 2 Drugs reviewed for orphan designation for narcolepsy by the US Food and Drug Administration. Generic name Trade name Sponsor Designation date Orphan designation FDA marketing approval date Exclusivity end date Viloxazine HCL Stuart Pharmaceuticals 6/11/1984 Treatment of cataplexy and narcolepsy Not FDA Approved for Orphan Indication NA Gamma-hydroxybutyric acid Sigma Chemical Company 1/22/1985 Treatment of narcolepsy and the auxiliary symptoms of cataplexy, sleep paralysis, hypnagogic hallucinations, and automatic behavior Not FDA Approved for Orphan Indication NA Gamma hydroxybutyrate Biocraft Laboratories, Inc. 12/3/1987 Treatment of narcolepsy and the auxiliary symptoms of cataplexy, sleep paralysis, hypnagogic hallucinations, and automatic behavior Not FDA Approved for Orphan Indication NA Modafinil Provigil1 Cephalon, Inc. 3/15/1993 Treatment of excessive daytime sleepiness in narcolepsy 12/24/1998 12/24/2005 Oxybate Xyrem1 Jazz Pharmaceuticals 11/7/1994 Treatment of narcolepsy 11/18/2005 11/18/2012 BF2.649 (Pitolisant) Bioprojet Pharma 5/17/2010 Treatment of narcolepsy Not FDA Approved for Orphan Indication NA Optically pure phenylalanine derivative Jazz Pharmaceuticals International III Limited 8/20/2012 Treatment of narcolepsy Not FDA Approved for Orphan Indication NA Disclosure statement Within the last 12 months, Dr. Kilduff and Dr. Black have received research support from Hoffmann La-Roche and Dr. Kilduff has received honoraria from Merck Pharmaceuticals and Pfizer. References Absalom N Eghorn LF Villumsen IS Karim N Bay T Olsen JV Knudsen GM Brauner-Osborne H Frolund B Clausen RP Chebib M Wellendorph P alpha4betadelta GABAA receptors are high-affinity targets for gamma-hydroxybutyric acid (GHB). Proc. Natl. Acad. Sci. U.S.A 2012 109 13404 13409 22753476 Ahmed SS Schur PH MacDonald NE Steinman L Narcolepsy, 2009 A(H1N1) pandemic influenza, and pandemic influenza vaccinations: what is known and unknown about the neurological disorder, the role for autoimmunity, and vaccine adjuvants. J. Autoimmun 2014 50 1 11 24559657 Ahmed SS Volkmuth W Duca J Corti L Pallaoro M Pezzicoli A Karle A Rigat F Rappuoli R Narasimhan V Julkunen I Vuorela A Vaarala O Nohynek H Pasini FL Montomoli E Trombetta C Adams CM Rothbard J Steinman L Antibodies to influenza nucleoprotein cross-react with human hypocretin receptor 2. Sci. Transl. Med 2015 7 294ra105 Akimoto H Honda Y Takahashi Y Pharmacotherapy in narcolepsy. Dis. Nerv. Syst 1960 21 704 706 13681922 Alexander SP Benson HE Faccenda E Pawson AJ Sharman JL Spedding M Peters JA Harmar AJ Collaborators C The Concise Guide to PHARMACOLOGY 2013/14: G protein-coupled receptors. Br. J. Pharmacol 2013 170 1459 1581 24517644 Andlauer O Moore H Hong t. Dauvilliers SC Kanbayashi Y Nishino T Han S Silber F Rico MH Einen T Kornum M Jennum BR Knudsen P Nevsimalova S Poli S Plazzi F Mignot GE Predictors of hypocretin (orexin) deficiency in narcolepsy without cataplexy. Sleep 2012 35 1247 1255 22942503 Anic-Labat S Guilleminault C Kraemer HC Meehan J Arrigoni J Mignot E Validation of a cataplexy questionnaire in 983 sleep-disorders patients. Sleep 1999 22 77 87 9989368 Aran A Lin L Nevsimalova S Plazzi G Hong SC Weiner K Zeitzer J Mignot E Elevated anti-streptococcal antibodies in patients with recent narcolepsy onset. Sleep 2009 32 979 983 19725248 Arango MT Kivity S Shoenfeld Y Is narcolepsy a classical autoimmune disease? Pharmacol. Res 2015 92 6 12 25447795 Aserinsky E Kleitman N Regularly occurring periods of eye motility, and concomitant phenomena, during sleep. Science 1953 118 273 274 13089671 Aston-Jones G Bloom FE Activity of norepinephrine-containing locus coeruleus neurons in behaving rats anticipates fluctuations in the sleep-waking cycle. J. Neurosci 1981 1 876 886 7346592 Babcock DA Narver EL Dement WC Mitler MM Effects of imipramine, chlorimipramine, and fluoxetine on cataplexy in dogs. Pharmacol. Biochem. Behav 1976 5 599 602 1035802 Baier PC Hallschmid M Seeck-Hirschner M Weinhold SL Burkert S Diessner N Goder R Aldenhoff JB Hinze-Selch D Effects of intranasal hypocretin-1 (orexin A) on sleep in narcolepsy with cataplexy. Sleep Med 2011 12 941 946 22036605 Baker TL Dement WC McGinty DJ Drucker-Colín RR Morrison A Parmeggiani PL Canine narcolepsy-cataplexy syndrome: evidence for an inherited monoaminergic-cholinergic imbalance. Brain Mechanisms of Sleep 1985 199 234 Raven Press New York Baker TL Guilleminault C Nino-Murcia G Dement WC Comparative polysomnographic study of narcolepsy and idiopathic central nervous system hypersomnia. Sleep 1986 9 232 242 3704448 Bardage C Persson I Ortqvist A Bergman U Ludvigsson JF Granath F Neurological and autoimmune disorders after vaccination against pandemic influenza A (H1N1) with a monovalent adjuvanted vaccine: population based cohort study in Stockholm, Sweden. Br. Med. J 2011 343 d5956 21994316 Bassetti C Aldrich MS Narcolepsy Neurol. Clin 1996 14 545 571 8871976 Bastianini S Silvani A Berteotti C Lo Martire V Zoccoli G High-amplitude theta wave bursts during REM sleep and cataplexy in hypocretin-deficient narcoleptic mice. J. Sleep Res 2012 21 185 188 21883592 Bastuji H Jouvet M Successful treatment of idiopathic hypersomnia and narcolepsy with modafinil. Prog. Neuropsychopharmacol. Biol. Psychiatry 1988 12 695 700 2906157 Baumann CR Mignot E Lammers GJ Overeem S Arnulf I Rye D Dauvilliers Y Honda M Owens JA Plazzi G Scammell TE Challenges in diagnosing narcolepsy without cataplexy: a consensus statement. Sleep 2014 37 1035 1042 24882898 Beitinger PA Fulda S Dalal MA Wehrle R Keckeis M Wetter TC Han F Pollmacher T Schuld A Glucose tolerance in patients with narcolepsy. Sleep 2012 35 231 236 22294813 Bergman P Adori C Vas S Kai-Larsen Y Sarkanen T Cederlund A Agerberth B Julkunen I Horvath B Kostyalik D Kalmar L Bagdy G Huutoniemi A Partinen M Hokfelt T Narcolepsy patients have antibodies that stain distinct cell populations in rat brain and influence sleep patterns. Proc. Natl. Acad. Sci. U.S.A 2014 111 E3735 E3744 25136085 Berry-Kravis EM Hessl D Rathmell B Zarevics P Cherubini M Walton-Bowen K Mu Y Nguyen DV Gonzalez-Heydrich J Wang PP Carpenter RL Bear MF Hagerman RJ Effects of STX209 (arbaclofen) on neurobehavioral function in children and adults with fragile X syndrome: a randomized, controlled, phase 2 trial. Sci. Transl. Med 2012 4 152 ra127 Besset A Tafti M Nobile L Billiard M Homeostasis and narcolepsy. Sleep 1994 17 S29 S34 7701197 Beuckmann CT Sinton CM Williams SC Richardson JA Hammer RE Sakurai T Yanagisawa M Expression of a poly-glutamine-ataxin-3 transgene in orexin neurons induces narcolepsy-cataplexy in the rat. J. Neurosci 2004 24 4469 4477 15128861 Black J Pardi D Hornfeldt CS Inhaber N The nightly use of sodium oxybate is associated with a reduction in nocturnal sleep disruption: a double-blind, placebo-controlled study in patients with narcolepsy. J. Clin. Sleep Med 2010 6 596 602 21206549 Black JL 3rd Silber MH Krahn LE Avula RK Walker DL Pankratz VS Fredrickson PA Slocumb NL Studies of humoral immunity to preprohypocretin in human leukocyte antigen DQB1*0602-positive narcoleptic subjects with cataplexy. Biol Psychiatry 2005 58 504 509 16043129 Black J Swick T Feldman N Doekel J Khayrallah R Bream M Ruoff GC Oral JZP-110 (ADX-N05) for the treatment of excessive daytime sleepiness in adults with narcolepsy: results of a randomised, double-blind, placebo-controlled trial. J. Sleep Res 2014a 23 Supp. 1 S32 S33 Black SW Morairty SR Chen TM Leung AK Wisor JP Yamanaka A Kilduff TS GABAB agonism promotes sleep and reduces cataplexy in murine narcolepsy. J. Neurosci 2014b 34 6485 6494 24806675 Black SW Morairty SR Fisher SP Chen TM Warrier DR Kilduff TS Almorexant promotes sleep and exacerbates cataplexy in a murine model of narcolepsy. Sleep 2013 36 325 336 23449602 Blanco-Centurion C Liu M Konadhode R Pelluru D Shiromani PJ Effects of orexin gene transfer in the dorsolateral pons in orexin knockout mice. Sleep 2013 36 31 40 23288969 Blouin AM Thannickal TC Worley PF Baraban JM Reti IM Siegel JM Narp immunostaining of human hypocretin (orexin) neurons: loss in narcolepsy. Neurology 2005 65 1189 1192 16135770 Boehme RE Baker TL Mefford IN Barchas JD Dement WC Ciaranello RD Narcolepsy: cholinergic receptor changes in an animal model. Life Sci 1984 34 1825 1828 6539848 Bogan RK Feldman NT Lankford A Khayrallah MA A double-blind, placebo-controlled, randomized, cross-over study of the efficacy and safety of ADX-N05 for the treatment of excessive daytime sleepiness in adult subjects with narcolepsy. Sleep 2013 36 A257 Boissard R Fort P Gervasoni D Barbagli B Luppi PH Localization of the GABAergic and non-GABAergic neurons projecting to the sublaterodorsal nucleus and potentially gating paradoxical sleep onset. Eur. J. Neurosci 2003 18 1627 1639 14511341 Boissard R Gervasoni D Schmidt MH Barbagli B Fort P Luppi PH The rat ponto-medullary network responsible for paradoxical sleep onset and maintenance: a combined microinjection and functional neuroanatomical study. Eur. J. Neurosci 2002 16 1959 1973 12453060 Borbely AA Tobler I Hanagasioglu M Effect of sleep deprivation on sleep and EEG power spectra in the rat. Behav. Brain Res 1984 14 171 182 6525241 Bosch OG Quednow BB Seifritz E Wetter TC Reconsidering GHB: orphan drug or new model antidepressant? J. Psychopharmacol 2012 26 618 628 21926421 Boscolo-Berto R Viel G Montagnese S Raduazzo DI Ferrara SD Dauvilliers Y Narcolepsy and effectiveness of gamma-hydroxybutyrate (GHB): a systematic review and meta-analysis of randomized controlled trials. Sleep Med. Rev 2012 16 431 443 22055895 Bowersox SS Kilduff TS Faull KF Zeller-DeAmicis L Dement WC Ciaranello RD Brain dopamine receptor levels elevated in canine narcolepsy. Brain Res 1987 402 44 48 3828787 Bowersox SS Kilduff TS Kaitin KI Dement WC Ciaranello RD Brain benzodiazepine receptor characteristics in canine narcolepsy. Sleep 1986 9 111 115 3010425 Brooks PL Peever JH Glycinergic and GABA(A)-mediated inhibition of somatic motoneurons does not mediate rapid eye movement sleep motor atonia. J. Neurosci 2008 28 3535 3545 18385312 Brooks PL Peever JH Identification of the transmitter and receptor mechanisms responsible for REM sleep paralysis. J. Neurosci 2012 32 9785 9795 22815493 Broughton R Dunham W Newman J Lutley K Duschesne P Rivers M Ambulatory 24 hour sleep-wake monitoring in narcolepsy-cataplexy compared to matched controls. Electroencephalogr. Clin. Neurophysiol 1988 70 473 481 2461281 Broughton R Ghanem Q Hishikawa Y Sugita Y Nevsimalova S Roth B Life effects of narcolepsy: relationships to geographic origin (North American Asian or European) and to other patient and illness variables. Can. J. Neurol. Sci 1983 10 100 104 6861006 Broughton R Valley V Aguirre M Roberts J Suwalski W Dunham W Excessive daytime sleepiness and the pathophysiology of narcolepsy-cataplexy: a laboratory perspective. Sleep 1986 9 205 215 3704444 Brown MA Guilleminault C A review of sodium oxybate and baclofen in the treatment of sleep disorders. Curr. Pharm. Des 2011 17 1430 1435 21476957 Brown RE Basheer R McKenna JT Strecker RE McCarley RW Control of sleep and wakefulness. Physiol. Rev 2012 92 1087 1187 22811426 Brown RE Winston S Basheer R Thakkar MM McCarley RW Electrophysiological characterization of neurons in the dorsolateral pontine rapid-eye-movement sleep induction zone of the rat: Intrinsic membrane properties and responses to carbachol and orexins. Neuroscience 2006 143 739 755 17008019 Burgess CR Oishi Y Mochizuki T Peever JH Scammell TE Amygdala lesions reduce cataplexy in orexin knock-out mice. J. Neurosci 2013 33 9734 9742 23739970 Burgess CR Peever JH A noradrenergic mechanism functions to couple motor behavior with arousal state. Curr. Biol 2013 23 1719 1725 23993842 Burgess CR Scammell TE Narcolepsy: neural mechanisms of sleepiness and cataplexy. J. Neurosci 2012 32 12305 12311 22956821 Burgess CR Tse G Gillis L Peever JH Dopaminergic regulation of sleep and cataplexy in a murine model of narcolepsy. Sleep 2010 33 1295 1304 21061851 Burlet S Tyler CJ Leonard CS Direct and indirect excitation of laterodorsal tegmental neurons by Hypocretin/Orexin peptides: implications for wakefulness and narcolepsy. J. Neurosci 2002 22 2862 2872 11923451 Calabro S Tritto E Pezzotti A Taccone M Muzzi A Bertholet S De Gregorio E O'Hagan DT Baudner B Seubert A The adjuvant effect of MF59 is due to the oil-in-water emulsion formulation, none of the individual components induce a comparable adjuvant effect. Vaccine 2013 31 3363 3369 23684834 Canellas F Lin L Julia MR Clemente A Vives-Bauza C Ollila HM Hong SC Arboleya SM Einen MA Faraco J Fernandez-Vina M Mignot E Dual cases of type 1 narcolepsy with schizophrenia and other psychotic disorders. J. Clin. Sleep Med 2014 10 1011 1018 25142772 Carlander B Eliaou JF Billiard M Autoimmune hypothesis in narcolepsy. Neurophysiol. Clin 1993 23 15 22 8446070 CDER The Voice of the Patient: Narcolepsy, Ed 2013 1 25 Silver Spring MD, U.S. FDA Ch'ng SS Lawrence AJ Distribution of the orexin-1 receptor (OX1R) in the mouse forebrain and rostral brainstem: a characterisation of OX1R-eGFP mice. J. Chem. Neuroanat 2015 66–67V– 1 9 Chabas D Foulon C Gonzalez J Nasr M Lyon-Caen O Willer JC Derenne JP Arnulf I Eating disorder and metabolism in narcoleptic patients. Sleep 2007 30 1267 1273 17969460 Chang K Frankovich J Cooperstock M Cunningham MW Latimer ME Murphy TK Pasternack M Thienemann M Williams K Walter J Swedo SE Clinical evaluation of youth with pediatric acute-onset neuropsychiatric syndrome (PANS): recommendations from the 2013 PANS Consensus Conference. J. Child Adolesc. Psychopharmacol 2015 25 3 13 25325534 Chase MH Soja PJ Morales FR Evidence that glycine mediates the postsynaptic potentials that inhibit lumbar motoneurons during the atonia of active sleep. J. Neurosci 1989 9 743 751 2926479 Chemelli RM Willie JT Sinton CM Elmquist JK Scammell T Lee C Richardson JA Williams SC Xiong Y Kisanuki Y Fitch TE Nakazato M Hammer RE Saper CB Yanagisawa M Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 1999 98 437 451 10481909 Clark EL Baumann CR Cano G Scammell TE Mochizuki T Feeding-elicited cataplexy in orexin knockout mice. Neuroscience 2009 161 970 977 19362119 Crochet S Onoe H Sakai K A potent non-monoaminergic paradoxical sleep inhibitory system: a reverse microdialysis and single-unit recording study. Eur. J. Neurosci 2006 24 1404 1412 16987225 Crocker A Espana RA Papadopoulou M Saper CB Faraco J Sakurai T Honda M Mignot E Scammell TE Concomitant loss of dynorphin NARP, and orexin in narcolepsy. Neurology 2005 65 1184 1188 16247044 Cruz HG Ivanova T Lunn ML Stoffel M Slesinger PA Luscher C Bidirectional effects of GABA(B) receptor agonists on the mesolimbic dopamine system. Nat. Neurosci 2004 7 153 159 14745451 Cvetkovic-Lopes V Bayer L Dorsaz S Maret S Pradervand S Dauvilliers Y Lecendreux M Lammers GJ Donjacour CE Du Pasquier RA Pfister C Petit B Hor H Muhlethaler M Tafti M Elevated Tribbles homolog 2-specific antibody levels in narcolepsy patients. J. Clin. Invest 2010 120 713 719 20160349 Daan S Beersma DG Borbely AA Timing of human sleep: recovery process gated by a circadian pacemaker. Am. J. Physiol 1984 246 R161 R183 6696142 Dahmen N Becht J Engel A Thommes M Tonn P Prevalence of eating disorders and eating attacks in narcolepsy. Neuropsychiatr. Dis. Treat 2008 4 257 261 18728824 Dahmen N Bierbrauer J Kasten M Increased prevalence of obesity in narcoleptic patients and relatives. Eur. Arch. Psychiatry Clin. Neurosci 2001 251 85 89 11407443 Dahmen N Tonn P Messroghli L Ghezel-Ahmadi D Engel A Basal metabolic rate in narcoleptic patients. Sleep 2009 32 962 964 19639760 Dantz B Edgar DM Dement WC Circadian rhythms in narcolepsy: studies on a 90 min day. Electroencephalogr. Clin. Neurophysiol 1994 90 24 35 7509271 Darwinkel A Stanic D Booth LC May CN Lawrence AJ Yao ST Distribution of orexin-1 receptor-green fluorescent protein- (OX1-GFP) expressing neurons in the mouse brain stem and pons: co-localization with tyrosine hydroxylase and neuronal nitric oxide synthase. Neuroscience 2014 278 253 264 25168728 Date Y Ueta Y Yamashita H Yamaguchi H Matsukura S Kangawa K Sakurai T Yanagisawa M Nakazato M Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proc. Natl. Acad. Sci. U.S.A 1999 96 748 753 9892705 Dauvilliers Y Abril B Mas E Michel F Tafti M Normalization of hypocretin-1 in narcolepsy after intravenous immunoglobulin treatment. Neurology 2009 73 1333 1334 19841387 Dauvilliers Y Arnulf I Lecendreux M Monaca Charley C Franco P Drouot X d'Ortho MP Launois S Lignot S Bourgin P Nogues B Rey M Bayard S Scholz S Lavault S Tubert-Bitter P Saussier C Pariente A a Increased risk of narcolepsy in children and adults after pandemic H1N1 vaccination in France. Brain 2013a 136 2486 2496 23884811 Dauvilliers Y Bassetti C Lammers GJ Arnulf I Mayer G Rodenbeck A Lehert P Ding CL Lecomte JM Schwartz JC Pitolisant versus placebo or modafinil in patients with narcolepsy: a double-blind, randomised trial. Lancet Neurol 2013b 12 1068 1075 24107292 Dauvilliers Y Carlander B Rivier F Touchon J Tafti M Successful management of cataplexy with intravenous immunoglobulins at narcolepsy onset. Ann. Neurol 2004 56 905 908 15562415 Dauvilliers Y Jaussent I Krams B Scholz S Lado S Levy P Pepin JL Non-dipping blood pressure profile in narcolepsy with cataplexy. PLoS One 2012 7 e38977 22768053 Dauvilliers Y Montplaisir J Molinari N Carlander B Ondze B Besset A Billiard M Age at onset of narcolepsy in two large populations of patients in France and Quebec. Neurology 2001 57 2029 2033 11739821 Dauvilliers Y Jaussent I Lecendreux M Scholz S Bayard S Cristol JP Blain H Dupuy A.M.a. Cerebrospinal fluid and serum cytokine profiles in narcolepsy with cataplexy: a case-control study. Brain Behav. Immun 2014a 37 260 266 24394344 Dauvilliers Y Siegel JM Lopez R Torontali ZA Peever J.H.b. Cataplexy-clinical aspects, pathophysiology and management strategy. Nat. Rev. Neurol 2014b 10 7 386 395 24890646 De la Herran-Arita AK Kornum BR Mahlios J Jiang W Lin L Hou T Macaubas C Einen M Plazzi G Crowe C Newell EW Davis MM Mellins ED Mignot E CD4+ T cell autoimmunity to hypocretin/orexin and cross-reactivity to a 2009 H1N1 influenza A epitope in narcolepsy. Sci. Transl. Med 2013 5 216ra176 De la Herran-Arita AK Kornum BR Mahlios J Jiang W Lin L Hou T Macaubas C Einen M Plazzi G Crowe C Newell EW Davis MM Mellins ED Mignot E Retraction of the research article: “CD4(+) T cell autoimmunity to hypocretin/orexin and cross-reactivity to a 2009 H1N1 influenza A epitope in narcolepsy”. Sci. Transl. Med 2014 6 247rt241 De La Herran-Arita AK Zomosa-Signoret VC Millan-Aldaco DA Palomero-Rivero M Guerra-Crespo M Drucker-Colin R Vidaltamayo R Aspects of the narcolepsy-cataplexy syndrome in O/E3-null mutant mice. Neuroscience 2011 183 134 143 21435382 de Lecea L Kilduff TS Peyron C Gao X Foye PE Danielson PE Fukuhara C Battenberg EL Gautvik VT Bartlett FS 2nd Frankel WN van den Pol AN Bloom FE Gautvik KM Sutcliffe JG The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc. Natl. Acad. Sci. U.S.A 1998 95 322 327 9419374 Deadwyler SA Porrino L Siegel JM Hampson RE Systemic and nasal delivery of orexin-A (Hypocretin-1) reduces the effects of sleep deprivation on cognitive performance in nonhuman primates. J. Neurosci 2007 27 14239 14247 18160631 Dean RR Kilduff TS Dement WC Grumet FC Narcolepsy without unique MHC class II antigen association: studies in the canine model. Hum. Immunol 1989 25 27 35 2523880 Delashaw JB Jr. Foutz AS Guilleminault C Dement WC Cholinergic mechanisms and cataplexy in dogs. Exp. Neurol 1979 66 745 757 573697 Dement W Kleitman N The relation of eye movements during sleep to dream activity: an objective method for the study of dreaming. J. Exp. Psychol 1957 53 339 346 13428941 Dement W Rechtschaffen A Gulevich G The nature of the narcoleptic sleep attack. Neurology 1966 16 18 33 5948003 Dement WC Bassetti CL Billiard M Mignot E Historical aspects of narcolepsy. Narcolepsy and Hypersomnia 2007 1 5 Informa Healthcare USA New York, NY Dhuria SV Hanson LR Frey WH 2nd Intranasal drug targeting of hypocretin-1 (orexin-A) to the central nervous system. J. Pharm. Sci 2009 98 2501 2515 19025760 Donadio V Liguori R Vandi S Pizza F Dauvilliers Y Leta V Giannoccaro MP Baruzzi A Plazzi G Lower wake resting sympathetic and cardiovascular activities in narcolepsy with cataplexy. Neurology 2014 83 1080 1086 25098533 Donjacour CE Aziz NA Overeem S Kalsbeek A Pijl H Lammers GJ Glucose and fat metabolism in narcolepsy and the effect of sodium oxybate: a hyperinsulinemic-euglycemic clamp study. Sleep 2014 37 795 801 24899766 Donjacour CE Pardi D Aziz NA Frolich M Roelfsema F Overeem S Pijl H Lammers GJ Plasma total ghrelin and leptin levels in human narcolepsy and matched healthy controls: basal concentrations and response to sodium oxybate. J. Clin. Sleep Med 2013 9 797 803 23946710 Doyle JB Daniels LE Narcolepsy: results of treatment with ephedrine sulphate. J. Am. Med. Assoc 1932 98 542 545 Dreifuss FE Flynn DV Narcolepsy in a horse. J. Am. Vet. Med. Assoc 1984 184 131 132 6538191 Duffy J Weintraub E Vellozzi C DeStefano F Vaccine Safety D Narcolepsy and influenza A(H1N1) pandemic 2009 vaccination in the United States. Neurology 2014 83 1823 1830 25320099 Edgar DM Dement WC Fuller CA Effect of SCN lesions on sleep in squirrel monkeys: evidence for opponent processes in sleep-wake regulation. J. Neurosci 1993 13 1065 1079 8441003 Edgar DM Seidel WF Modafinil induces wakefulness without intensifying motor activity or subsequent rebound hypersomnolence in the rat. J. Pharmacol. Exp. Ther 1997 283 757 769 9353396 Eichler AF International Classification of Sleep Disorders, Ed 2014 American Academy of Sleep Medicine Darien, IL Espana RA McCormack SL Mochizuki T Scammell TE Running promotes wakefulness and increases cataplexy in orexin knockout mice. Sleep 2007 30 1417 1425 18041476 Faraco J Lin L Kornum BR Kenny EE Trynka G Einen M Rico TJ Lichtner P Dauvilliers Y Arnulf I Lecendreux M Javidi S Geisler P Mayer G Pizza F Poli F Plazzi G Overeem S Lammers GJ Kemlink D Sonka K Nevsimalova S Rouleau G Desautels A Montplaisir J Frauscher B Ehrmann L Hogl B Jennum P Bourgin P Peraita-Adrados R Iranzo A Bassetti C Chen WM Concannon P Thompson SD Damotte V Fontaine B Breban M Gieger C Klopp N Deloukas P Wijmenga C Hallmayer J Onengut-Gumuscu S Rich SS Winkelmann J Mignot E ImmunoChip study implicates antigen presentation to T cells in narcolepsy. PLoS Genet 2013 9 e1003270 23459209 Faull KF Barchas JD Foutz AS Dement WC Holman RB Monoamine metabolite concentrations in the cerebrospinal fluid of normal and narcoleptic dogs. Brain Res 1982 242 137 143 6179569 Faull KF Zeller-DeAmicis LC Radde L Bowersox SS Baker TL Kilduff TS Dement WC Biogenic amine concentrations in the brains of normal and narcoleptic canines: current status. Sleep 1986 9 107 110 3704432 Ferini-Strambi L Spera A Oldani A Zucconi M Bianchi A Cerutti S Smirne S Autonomic function in narcolepsy: power spectrum analysis of heart rate variability. J. Neurol 1997 244 252 255 9112594 Fortuyn HA Swinkels S Buitelaar J Renier WO Furer JW Rijnders CA Hodiamont PP Overeem S High prevalence of eating disorders in narcolepsy with cataplexy: a case-control study. Sleep 2008 31 335 341 18363309 Foutz AS Mitler MM Cavalli-Sforza LL Dement WC Genetic factors in canine narcolepsy. Sleep 1979 1 413 421 574310 Frauscher B Ehrmann L Mitterling T Gabelia D Gschliesser V Brandauer E Poewe W Hogl B Delayed diagnosis, range of severity, and multiple sleep comorbidities: a clinical and polysomnographic analysis of 100 patients of the innsbruck narcolepsy cohort. J. Clin. Sleep Med 2013 9 805 812 23946711 Fronczek R Overeem S Lammers GJ van Dijk JG Van Someren EJ Altered skin-temperature regulation in narcolepsy relates to sleep propensity. Sleep 2006 29 1444 1449 17162991 Fronczek R Overeem S Reijntjes R Lammers GJ van Dijk JG Pijl H Increased heart rate variability but normal resting metabolic rate in hypocretin/orexin-deficient human narcolepsy. J. Clin. Sleep Med 2008a 4 248 254 18595438 Fronczek R Raymann RJ Overeem S Romeijn N van Dijk JG Lammers GJ Van Someren EJ Manipulation of skin temperature improves nocturnal sleep in narcolepsy. J. Neurol. Neurosurg. Psychiatry 2008b 79 1354 1357 18653548 Fujiki N Cheng T Yoshino F Nishino S Specificity of direct transition from wake to REM sleep in orexin/ataxin-3 transgenic narcoleptic mice. Exp. Neurol 2009 217 46 54 19416673 Funato H Tsai AL Willie JT Kisanuki Y Williams SC Sakurai T Yanagisawa M Enhanced orexin receptor-2 signaling prevents diet-induced obesity and improves leptin sensitivity. Cell Metab 2009 9 64 76 19117547 Gélineau J De la Narcolepsie. Gaz Hop (Paris) 1880 53 535 637 Gerashchenko D Murillo-Rodriguez E Lin L Xu M Hallett L Nishino S Mignot E Shiromani PJ Relationship between CSF hypocretin levels and hypocretin neuronal loss. Exp. Neurol 2003 184 1010 1016 14769395 Gerashchenko D Salin-Pascual R Shiromani PJ Effects of hypocretinsaporin injections into the medial septum on sleep and hippocampal theta. Brain Res 2001 913 106 115 11532254 Golicki D Bala MM Niewada M Wierzbicka A Modafinil for narcolepsy: systematic review and meta-analysis. Med. Sci. Monit 2010 16 Ra177 Ra186 20671626 Graus F Saiz A Dalmau J Antibodies and neuronal autoimmune disorders of the CNS. J. Neurol 2010 257 509 517 20035430 Gray KA Yates B Seal RL Wright MW Bruford EA 2015 Genenames.org: the HGNC resources in. Nucleic Acids Res 2015 43 D1079 D1085 25361968 Grimaldi D Agati P Pierangeli G Franceschini C Guaraldi P Barletta G Vandi S Cevoli S Plazzi G Montagna P Cortelli P a Hypocretin deficiency in narcolepsy with cataplexy is associated with a normal body core temperature modulation. Chronobiol. Int 2010a 27 1596 1608 20854137 Grimaldi D Pierangeli G Barletta G Terlizzi R Plazzi G Cevoli S Franceschini C Montagna P Cortelli P b Spectral analysis of heart rate variability reveals an enhanced sympathetic activity in narcolepsy with cataplexy. Clin. Neurophysiol 2010b 121 1142 1147 20181520 Grimaldi D Calandra-Buonaura G Provini F Agati P Pierangeli G Franceschini C Barletta G Plazzi G Montagna P Cortelli P Abnormal sleep-cardiovascular system interaction in narcolepsy with cataplexy: effects of hypocretin deficiency in humans. Sleep 2012 35 519 528 22467990 Guilleminault C Cao MT Kryger MH Roth T Dement WC Narcolepsy: Diagnosis and management. Principles and Practice of Sleep Medicine 2011 957 968 Elsevier Saunders St. Louis, Missouri Guilleminault C Lee JH Arias V Bassetti CL Billiard M Mignot E Cataplexy. Narcolepsy and Hypersomnia 2007 49 62 Informa Healthcare USA New York, NY Gulyani S Wu MF Nienhuis R John J Siegel JM Cataplexy-related neurons in the amygdala of the narcoleptic dog. Neuroscience 2002 112 355 365 12044453 Hallmayer J Faraco J Lin L Hesselson S Winkelmann J Kawashima M Mayer G Plazzi G Nevsimalova S Bourgin P Hong SC Honda Y Honda M Hogl B Longstreth WT Jr. Montplaisir J Kemlink D Einen M Chen J Musone SL Akana M Miyagawa T Duan J Desautels A Erhardt C Hesla PE Poli F Frauscher B Jeong JH Lee SP Ton TG Kvale M Kolesar L Dobrovolna M Nepom GT Salomon D Wichmann HE Rouleau GA Gieger C Levinson DF Gejman PV Meitinger T Young T Peppard P Tokunaga K Kwok PY Risch N Mignot E Narcolepsy is strongly associated with the T-cell receptor alpha locus. Nat. Genet 2009 41 708 711 19412176 Han F Faraco J Dong XS Ollila HM Lin L Li J An P Wang S Jiang KW Gao ZC Zhao L Yan H Liu YN Li QH Zhang XZ Hu Y Wang JY Lu YH Lu CJ Zhou W Hallmayer J Huang YS Strohl KP Pollmacher T Mignot E Genome wide analysis of narcolepsy in China implicates novel immune loci and reveals changes in association prior to versus after the 2009 H1N1 influenza pandemic. PLoS Genet 2013a 9 e1003880 24204295 Han F Lin L Li J Dong XS Mignot E Decreased incidence of childhood narcolepsy 2 years after the 2009 H1N1 winter flu pandemic. Ann. Neurol 2013b 73 560 23225098 Han F Lin L Warby SC Faraco J Li J Dong SX An P Zhao L Wang LH Li QY Yan H Gao ZC Yuan Y Strohl KP Mignot E Narcolepsy onset is seasonal and increased following the 2009 H1N1 pandemic in China. Ann. Neurol 2011 70 410 417 21866560 Hara J Beuckmann CT Nambu T Willie JT Chemelli RM Sinton CM Sugiyama F Yagami K Goto K Yanagisawa M Sakurai T Genetic ablation of orexin neurons in mice results in narcolepsy, hypophagia, and obesity. Neuron 2001 30 345 354 11394998 Hara J Yanagisawa M Sakurai T Difference in obesity phenotype between orexin-knockout mice and orexin neuron-deficient mice with same genetic background and environmental conditions. Neurosci. Lett 2005 380 239 242 15862893 Harsh J Peszka J Hartwig G Mitler M Night-time sleep and daytime sleepiness in narcolepsy. J. Sleep Res 2000 9 309 316 11012872 Hasegawa E Yanagisawa M Sakurai T Mieda M Orexin neurons suppress narcolepsy via 2 distinct efferent pathways. J. Clin. Invest 2014 124 604 616 24382351 Heier MS Jansson TS Gautvik KM Cerebrospinal fluid hypocretin 1 deficiency, overweight, and metabolic dysregulation in patients with narcolepsy. J. Clin. Sleep Med 2011 7 653 658 22171205 Heister DS Hayar A Garcia-Rill E Cholinergic modulation of GABAergic and glutamatergic transmission in the dorsal subcoeruleus: mechanisms for REM sleep control. Sleep 2009 32 1135 1147 19750918 Hishikawa Y Ida H Nakai K Kaneko Z Treatment of narcolepsy with imipramine (tofranil) and desmethylimipramine (pertofran). J. Neurol. Sci 1966 3 453 461 4380880 Hishikawa Y Shimizu T Bassetti CL Billiard M Mignot E Historical aspects of the treatment for narcolepsy. Narcolepsy and Hypersomnia 2007 25 30 Informa Healthcare USA New York, NY Hobson JA McCarley RW Wyzinski PW Sleep cycle oscillation: reciprocal discharge by two brainstem neuronal groups. Science 1975 189 55 58 1094539 Honda Y Census of narcolepsy, cataplexy and sleep life among teenagers in Fujisawa City. Sleep Res 1979 8 191 Honda Y Doi Y Ninomiya R Ninomiya C Increased frequency of non-insulin-dependent diabetes mellitus among narcoleptic patients. Sleep 1986 9 254 259 3518018 Huang YS Guilleminault C Narcolepsy: action of two gamma-aminobutyric acid type B agonists, baclofen and sodium oxybate. Pediatr. Neurol 2009 41 9 16 19520267 Huang ZL Qu WM Li WD Mochizuki T Eguchi N Watanabe T Urade Y Hayaishi O Arousal effect of orexin A depends on activation of the histaminergic system. Proc. Natl. Acad. Sci. U.S.A 2001 98 9965 9970 11493714 Hublin C Partinen M Heinonen EH Puukka P Salmi T Selegiline in the treatment of narcolepsy. Neurology 1994 44 2095 2101 7969965 Hudgson P Weightman D Baclofen in the treatment of spasticity. Br. Med. J 1971 4 15 17 4938243 Hungs M Fan J Lin L Lin X Maki RA Mignot E Identification and functional analysis of mutations in the hypocretin (orexin) genes of narcoleptic canines. Genome Res 2001 11 531 539 11282968 Jacob L Leib R Ollila HM Bonvalet M Adams CM Mignot E Comparison of Pandemrix and Arepanrix, two pH1N1 AS03-adjuvanted vaccines differentially associated with narcolepsy development. Brain Behav. Immun 2015 47 44 57 25452148 Jennum P Ibsen R Knudsen S Kjellberg J Comorbidity and mortality of narcolepsy: a controlled retro- and prospective national study. Sleep 2013 36 835 840 23729926 Jensen K Mody I GHB depresses fast excitatory and inhibitory synaptic transmission via GABA(B) receptors in mouse neocortical neurons. Cereb. Cortex 2001 11 424 429 11313294 John J Thannickal TC McGregor R Ramanathan L Ohtsu H Nishino S Sakai N Yamanaka A Stone C Cornford M Siegel JM Greatly increased numbers of histamine cells in human narcolepsy with cataplexy. Ann. Neurol 2013 74 786 793 23821583 John J Wu MF Boehmer LN Siegel JM Cataplexy-active neurons in the hypothalamus: implications for the role of histamine in sleep and waking behavior. Neuron 2004 42 619 634 15157423 Jouvet M Michel F Courjon J [On a stage of rapid cerebral electrical activity in the course of physiological sleep]. C R Seances Soc. Biol. Fil 1959 153 1024 1028 14408003 Ju YE Larson-Prior L Duntley S Changing demographics in REM sleep behavior disorder: possible effect of autoimmunity and antidepressants. Sleep Med 2011 12 278 283 21317035 Juji T Satake M Honda Y Doi Y HLA antigens in Japanese patients with narcolepsy: all the patients were DR2 positive. Tissue Antigens 1984 24 316 319 6597978 Kaitin KI Kilduff TS Dement WC Evidence for excessive sleepiness in canine narcoleptics. Electroencephalogr. Clin. Neurophysiol 1986a 64 447 454 2428595 Kaitin KI Kilduff TS Dement WC Sleep fragmentation in canine narcolepsy. Sleep 1986b 9 116 119 3704433 Kalogiannis M Grupke SL Potter PE Edwards JG Chemelli RM Kisanuki YY Yanagisawa M Leonard CS Narcoleptic orexin receptor knockout mice express enhanced cholinergic properties in laterodorsal tegmental neurons. Eur. J. Neurosci 2010 32 130 142 20576035 Kalogiannis M Hsu E Willie JT Chemelli RM Kisanuki YY Yanagisawa M Leonard CS Cholinergic modulation of narcoleptic attacks in double orexin receptor knockout mice. PLoS One 2011 6 e18697 21533254 Kantor S Mochizuki T Lops SN Ko B Clain E Clark E Yamamoto M Scammell TE Orexin gene therapy restores the timing and maintenance of wakefulness in narcoleptic mice. Sleep 2013 36 1129 1138 23904672 Kaupmann K Cryan JF Wellendorph P Mombereau C Sansig G Klebs K Schmutz M Froestl W van der Putten H Mosbacher J Brauner-Osborne H Waldmeier P Bettler B Specific gamma-hydroxybutyrate-binding sites but loss of pharmacological effects of gamma-hydroxybutyrate in GABA(B)(1)-deficient mice. Eur. J. Neurosci 2003 18 2722 2730 14656321 Kilduff TS Bowersox SS Kaitin KI Baker TL Ciaranello RD Dement WC Muscarinic cholinergic receptors and the canine model of narcolepsy. Sleep 1986 9 102 106 3704431 Kilduff TS Peyron C The hypocretin/orexin ligand-receptor system: Implications for sleep and sleep disorders. Trends Neurosci 2000 23 359 365 10906799 Kim HY Hong E Kim JI Lee W Solution structure of human orexin-A: regulator of appetite and wakefulness. J. Biochem. Mol. Biol 2004 37 565 573 15479620 Kisanuki YY Chemelli RM Willie JT Sinton CM Yanagisawa M Behavioral and polysomnographic characterization of orexin-1 receptor and orexin-2 receptor double knockout mice. Sleep 2001 24 A22 Knudsen S Biering-Sorensen B Kornum BR Petersen ER Ibsen JD Gammeltoft S Mignot E Jennum PJ Early IVIg treatment has no effect on post-H1N1 narcolepsy phenotype or hypocretin deficiency. Neurology 2012 79 102 103 22722630 Knudsen S Mikkelsen JD Bang B Gammeltoft S Jennum PJ Intravenous immunoglobulin treatment and screening for hypocretin neuron-specific autoantibodies in recent onset childhood narcolepsy with cataplexy. Neuropediatrics 2010 41 217 222 21210337 Koepsell TD Longstreth WT Ton TG Medical exposures in youth and the frequency of narcolepsy with cataplexy: a population-based case-control study in genetically predisposed people. J. Sleep Res 2010 19 80 86 19732319 Kok SW Meinders AE Overeem S Lammers GJ Roelfsema F Frolich M Pijl H Reduction of plasma leptin levels and loss of its circadian rhythmicity in hypocretin (orexin)-deficient narcoleptic humans. J. Clin. Endocrinol. Metab 2002 87 805 809 11836325 Kok SW Overeem S Visscher TL Lammers GJ Seidell JC Pijl H Meinders AE Hypocretin deficiency in narcoleptic humans is associated with abdominal obesity. Obes. Res 2003 11 1147 1154 12972686 Kornum BR Faraco J Mignot E Narcolepsy with hypocretin/orexin deficiency, infections and autoimmunity of the brain. Curr. Opin. Neurobiol 2011a 21 897 903 21963829 Kornum BR Kawashima M Faraco J Lin L Rico TJ Hesselson S Axtell RC Kuipers H Weiner K Hamacher A Kassack MU Han F Knudsen S Li J Dong X Winkelmann J Plazzi G Nevsimalova S Hong SC Honda Y Honda M Hogl B Ton TG Montplaisir J Bourgin P Kemlink D Huang YS Warby S Einen M Eshragh JL Miyagawa T Desautels A Ruppert E Hesla PE Poli F Pizza F Frauscher B Jeong JH Lee SP Strohl KP Longstreth WT Jr. Kvale M Dobrovolna M Ohayon MM Nepom GT Wichmann HE Rouleau GA Gieger C Levinson DF Gejman PV Meitinger T Peppard P Young T Jennum P Steinman L Tokunaga K Kwok PY Risch N Hallmayer J Mignot E Common variants in P2RY11 are associated with narcolepsy. Nat. Genet 2011b 43 66 71 21170044 Kornum BR Pizza F Knudsen S Plazzi G Jennum P Mignot E Cerebrospinal fluid cytokine levels in type 1 narcolepsy patients very close to onset. Brain Behav. Immun 2015 49 54 58 25771509 Krenzer M Anaclet C Vetrivelan R Wang N Vong L Lowell BB Fuller PM Lu J Brainstem and spinal cord circuitry regulating REM sleep and muscle atonia. PLoS One 2011 6 e24998 22043278 Kushida CA Baker TL Dement WC Electroencephalographic correlates of cataplectic attacks in narcoleptic canines. Electroencephalogr. Clin. Neurophysiol 1985 61 61 70 2408864 Lammers GJ Pijl H Iestra J Langius JA Buunk G Meinders AE Spontaneous food choice in narcolepsy. Sleep 1996 19 75 76 8650468 Langdon N Shindler J Parkes JD Bandak S Fluoxetine in the treatment of cataplexy. Sleep 1986 9 371 373 3509809 Langford J Gross WL Psychosis in the context of sodium oxybate therapy. J. Clin. Sleep Med 2011 7 665 666 22171207 Lavie P Peled R Narcolepsy is a rare disease in Israel. Sleep 1987 10 608 609 3432862 Lee JH Bang E Chae KJ Kim JY Lee DW Lee W Solution structure of a new hypothalamic neuropeptide, human hypocretin-2/orexin-B. Eur. J. Biochem 1999 266 831 839 10583376 Legrand Narcolepsy. Lancet 1888 131 Leonard BE McCartan D White J King DJ Methylphenidate: a review of its neuropharmacological, neuropsychological and adverse clinical effects. Hum. Psychopharmacol 2004 19 151 180 15079851 Leviel V Dopamine release mediated by the dopamine transporter, facts and consequences. J. Neurochem 2011 118 475 489 21644994 Li R Mignot E Faraco J Kadotani H Cantanese J Zhao B Lin X Hinton L Ostrander EA Patterson DF de Jong PJ Construction and characterization of an eightfold redundant dog genomic bacterial artificial chromosome library. Genomics 1999 58 9 17 10331940 Liblau RS Vassalli A Seifinejad A Tafti M Hypocretin (orexin) biology and the pathophysiology of narcolepsy with cataplexy. Lancet Neurol 2015 14 318 328 25728441 Lin L Faraco J Li R Kadotani H Rogers W Lin X Qiu X de Jong PJ Nishino S Mignot E The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 1999 98 365 376 10458611 Lipinski CA Lombardo F Dominy BW Feeney PJ Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev 2001 46 3 26 11259830 Liu M Blanco-Centurion C Konadhode R Begum S Pelluru D Gerashchenko D Sakurai T Yanagisawa M van den Pol AN Shiromani PJ Orexin gene transfer into zona incerta neurons suppresses muscle paralysis in narcoleptic mice. J. Neurosci 2011 31 6028 6040 21508228 Liu M Thankachan S Kaur S Begum S Blanco-Centurion C Sakurai T Yanagisawa M Neve R Shiromani PJ Orexin (hypocretin) gene transfer diminishes narcoleptic sleep behavior in mice. Eur. J. Neurosci 2008 28 1382 1393 18973565 Löwenfeld L Über narkolepsie. Münch Med Wochenschr 1902 49 1041 1045 Lu J Sherman D Devor M Saper CB A putative flip-flop switch for control of REM sleep. Nature 2006 441 589 594 16688184 Luca G Haba-Rubio J Dauvilliers Y Lammers GJ Overeem S Donjacour CE Mayer G Javidi S Iranzo A Santamaria J Peraita-Adrados R Hor H Kutalik Z Plazzi G Poli F Pizza F Arnulf I Lecendreux M Bassetti C Mathis J Heinzer R Jennum P Knudsen S Geisler P Wierzbicka A Feketeova E Pfister C Khatami R Baumann C Tafti M Clinical, polysomnographic and genome-wide association analyses of narcolepsy with cataplexy: a European Narcolepsy Network study. J. Sleep Res 2013 22 482 495 23496005 Lucas EA Foutz AS Dement WC Mitler MM Sleep cycle organization in narcoleptic and normal dogs. Physiol. Behav 1979 23 737 743 228334 Ludvikova E Nishino S Sakai N Jahn P Familial narcolepsy in the Lipizzaner horse: a report of three fillies born to the same sire. Vet. Q 2012 32 99 102 22889297 Lunn DP Cuddon PA Shaftoe S Archer RM Familial occurrence of narcolepsy in miniature horses. Equine Vet. J 1993 25 483 487 7903939 Luppi PH Clement O Sapin E Gervasoni D Peyron C Leger L Salvert D Fort P The neuronal network responsible for paradoxical sleep and its dysfunctions causing narcolepsy and rapid eye movement (REM) behavior disorder. Sleep Med. Rev 2011 15 153 163 21115377 Makela KA Wigren HK Zant JC Sakurai T Alhonen L Kostin A Porkka-Heiskanen T Herzig KH Characterization of sleep-wake patterns in a novel transgenic mouse line overexpressing human prepro-orexin/hypocretin. Acta Physiol. (Oxf) 2010 198 237 249 20003098 Mamelak M Narcolepsy and depression and the neurobiology of gamma-hydroxybutyrate. Prog. Neurobiol 2009 89 193 219 19654034 Mathis J Hess CW Bassetti CL Billiard M Mignot E Vigilance tests in narcolepsy. Narcolepsy and Hypersomnia 2007 243 256 Informa Healthcare USA New York, NY Matsuki K Grumet FC Lin X Gelb M Guilleminault C Dement WC Mignot E DQ (rather than DR) gene marks susceptibility to narcolepsy. Lancet 1992 339 1052 Mayer G Meier-Ewert K Motor dyscontrol in sleep of narcoleptic patients (a lifelong development?). J. Sleep Res 1993 2 143 148 10607086 McGinty DJ Harper RM Dorsal raphe neurons: depression of firing during sleep in cats. Brain Res 1976 101 569 575 1244990 Meerlo P Westerveld P Turek FW Koehl M Effects of gamma-hydroxybutyrate (GHB) on vigilance states and EEG in mice. Sleep 2004 27 899 904 15453548 Mefford IN Baker TL Boehme R Foutz AS Ciaranello RD Barchas JD Dement WC Narcolepsy: biogenic amine deficits in an animal model. Science 1983 220 629 632 6188216 Merkle FT Maroof A Wataya T Sasai Y Studer L Eggan K Schier AF Generation of neuropeptidergic hypothalamic neurons from human pluripotent stem cells. Development 2015 142 633 643 25670790 Michelson D Snyder E Paradis E Chengan-Liu M Snavely DB Hutzelmann J Walsh JK Krystal AD Benca RM Cohn M Lines C Roth T Herring WJ Safety and efficacy of suvorexant during 1-year treatment of insomnia with subsequent abrupt treatment discontinuation: a phase 3 randomised, double-blind, placebo-controlled trial. Lancet Neurol 2014 13 461 471 24680372 Mieda M Hasegawa E Kisanuki YY Sinton CM Yanagisawa M Sakurai T Differential roles of orexin receptor-1 and -2 in the regulation of non-REM and REM sleep. J. Neurosci 2011 31 6518 6526 21525292 Mieda M Willie JT Hara J Sinton CM Sakurai T Yanagisawa M Orexin peptides prevent cataplexy and improve wakefulness in an orexin neuronablated model of narcolepsy in mice. Proc. Natl. Acad. Sci. U.S.A 2004 101 4649 4654 15070772 Mignot E Genetics of narcolepsy and other sleep disorders. Am. J. Hum. Genet 1997 60 1289 1302 9199548 Mignot E Genetic and familial aspects of narcolepsy. Neurology 1998 50 S16 S22 9484418 Mignot E A hundred years of narcolepsy research. Arch. Ital. Biol 2001 139 207 220 11330202 Mignot E Guilleminault C Bowersox S Rappaport A Dement WC Effect of alpha 1-adrenoceptors blockade with prazosin in canine narcolepsy. Brain Res 1988a 444 184 188 2834022 Mignot E Guilleminault C Bowersox S Rappaport A Dement WC Role of central alpha-1 adrenoceptors in canine narcolepsy. J. Clin. Investig 1988b 82 885 894 2843574 Mignot E Hayduk R Black J Grumet FC Guilleminault C HLA DQB1*0602 is associated with cataplexy in 509 narcoleptic patients. Sleep 1997 20 1012 1020 9456467 Mignot E Lammers GJ Ripley B Okun M Nevsimalova S Overeem S Vankova J Black J Harsh J Bassetti C Schrader H Nishino S The role of cerebrospinal fluid hypocretin measurement in the diagnosis of narcolepsy and other hypersomnias. Arch. Neurol 2002 59 1553 1562 12374492 Mignot E Lin L Rogers W Honda Y Qiu X Lin X Okun M Hohjoh H Miki T Hsu S Leffell M Grumet F Fernandez-Vina M Honda M Risch N Complex HLA-DR and -DQ interactions confer risk of narcolepsy-cataplexy in three ethnic groups. Am. J. Hum. Genet 2001 68 686 699 11179016 Mignot E Nishino S Bassetti CL Billiard M Mignot E Perspectives for new treatments: an update. Narcolepsy and Hypersomnia 2007 641 654 Informa Healthcare USA New York, NY Mignot E Lin X Arrigoni J Macaubas C Olive F Hallmayer J Underhill P Guilleminault C Dement WC Grumet FC a DQB1*0602 and DQA1*0102 (DQ1) are better markers than DR2 for narcolepsy in Caucasian and black Americans. Sleep 1994a 17 S60 S67 7701202 Mignot E Nishino S Guilleminault C Dement WC Modafinil binds to the dopamine uptake carrier site with low affinity. Sleep 1994b 17 436 437 7991954 Mignot E Renaud A Nishino S Arrigoni J Guilleminault C Dement WC Canine cataplexy is preferentially controlled by adrenergic mechanisms: evidence using monoamine selective uptake inhibitors and release enhancers. Psychopharmacology (Berl) 1993 113 76 82 7862832 Mignot EJ A practical guide to the therapy of narcolepsy and hypersomnia syndromes. Neurotherapeutics 2012 9 739 752 23065655 Mignot EJ History of narcolepsy at Stanford University. Immunol. Res 2014 58 315 339 24825774 Mileykovskiy BY Kiyashchenko LI Siegel JM Behavioral correlates of activity in identified hypocretin/orexin neurons. Neuron 2005 46 787 798 15924864 Miller JD Faull KF Bowersox SS Dement WC CNS monoamines and their metabolites in canine narcolepsy: a replication study. Brain Res 1990 509 169 171 1689603 Miro O Nogue S Espinosa G To-Figueras J Sanchez M Trends in illicit drug emergencies: the emerging role of gamma-hydroxybutyrate. J. Toxicol. Clin. Toxicol 2002 40 129 135 12126184 Mitler MM Dement WC Sleep studies on canine narcolepsy: pattern and cycle comparisons between affected and normal dogs. Electroencephalogr. Clin. Neurophysiol 1977 43 691 699 72649 Mitler MM Soave O Dement WC Narcolepsy in seven dogs. J. Am. Vet. Med. Assoc 1976 168 1036 1038 945254 Mitler MM Van den Hoed J Carskadon MA Richardson G Park R Guilleminault C Dement WC REM sleep episodes during the Multiple Sleep Latency Test in narcoleptic patients. Electroencephalogr. Clin. Neurophysiol 1979 46 479 481 85544 Mochizuki T Arrigoni E Marcus JN Clark EL Yamamoto M Honer M Borroni E Lowell BB Elmquist JK Scammell TE Orexin receptor 2 expression in the posterior hypothalamus rescues sleepiness in narcoleptic mice. Proc. Natl. Acad. Sci. U.S.A 2011 108 4471 4476 21368172 Mochizuki T Crocker A McCormack S Yanagisawa M Sakurai T Scammell TE Behavioral state instability in orexin knock-out mice. J. Neurosci 2004 24 6291 6300 15254084 Mochizuki T Klerman EB Sakurai T Scammell TE Elevated body temperature during sleep in orexin knockout mice. Am. J. Physiol. Regul. Integr. Comp. Physiol 2006 291 R533 R540 16556901 Montplaisir J Petit D Quinn MJ Ouakki M Deceuninck G Desautels A Mignot E De Wals P Risk of narcolepsy associated with inactivated adjuvanted (AS03) A/H1N1 (2009) pandemic influenza vaccine in Quebec. PLoS One 2014 9 e108489 25264897 Morawska M Buchi M Fendt M Narcoleptic episodes in orexin-deficient mice are increased by both attractive and aversive odors. Behav. Brain Res 2011 222 397 400 21510981 Mori T Ito S Kuwaki T Yanagisawa M Sakurai T Sawaguchi T Monoaminergic neuronal changes in orexin deficient mice. Neuropharmacology 2010 58 826 832 19703479 Mosko SS Holowach JB Sassin JF The 24-hour rhythm of core temperature in narcolepsy. Sleep 1983 6 137 146 6878983 Motoyama M Kilduff TS Lee BS Dement WC McDevitt HO Restriction fragment length polymorphism in canine narcolepsy. Immunogenetics 1989 29 124 126 2563354 Muraki Y Yamanaka A Tsujino N Kilduff TS Goto K Sakurai T Serotonergic regulation of the orexin/hypocretin neurons through the 5-HT1A receptor. J. Neurosci 2004 24 7159 7166 15306649 Nagahara T Saitoh T Kutsumura N Irukayama-Tomobe Y Ogawa Y Kuroda D Gouda H Kumagai H Fujii H Yanagisawa M Nagase H Design and synthesis of non-peptide selective orexin receptor 2 agonists. J. Med. Chem 2015 58 7931 7937 26267383 Nakamura M Nishida S Hayashida K Ueki Y Dauvilliers Y Inoue Y Differences in brain morphological findings between narcolepsy with and without cataplexy. PLoS One 2013 8 e81059 24312261 Nevsimalova S Narcolepsy in childhood. Sleep Med. Rev 2009 13 169 180 19153053 Niewoehner J Bohrmann B Collin L Urich E Sade H Maier P Rueger P Stracke JO Lau W Tissot AC Loetscher H Ghosh A Freskgard PO Increased brain penetration and potency of a therapeutic antibody using a monovalent molecular shuttle. Neuron 2014 81 49 60 24411731 Nishino S Clinical and neurobiological aspects of narcolepsy. Sleep Med 2007 8 373 399 17470414 Nishino S Arrigoni J Valtier D Miller JD Guilleminault C Dement WC Mignot E Dopamine D2 mechanisms in canine narcolepsy. J. Neurosci 1991 11 2666 2671 1831837 Nishino S Fruhstorfer B Arrigoni J Guilleminault C Dement WC Mignot E Further characterization of the alpha-1 receptor subtype involved in the control of cataplexy in canine narcolepsy. J. Pharmacol. Exp. Ther 1993 264 1079 1084 8095546 Nishino S Haak L Shepherd H Guilleminault C Sakai T Dement WC Mignot E Effects of central alpha-2 adrenergic compounds on canine narcolepsy, a disorder of rapid eye movement sleep. J. Pharmacol. Exp. Ther 1990 253 1145 1152 1972749 Nishino S Mignot E Pharmacological aspects of human and canine narcolepsy. Prog. Neurobiol 1997 52 27 78 9185233 Nishino S Mignot E Kryger MH Roth T Dement WC Wake-promoting medications: basic mechanisms and pharmacology. Principles and Practice of Sleep Medicine 2011 510 526 Elsevier Saunders St. Louis, Missouri Nishino S Mignot E Fruhstorfer B Dement WC Hayaishi O Prostaglandin E2 and its methyl ester reduce cataplexy in canine narcolepsy. Proc. Natl. Acad. Sci. U.S.A 1989 86 2483 2487 2928344 Nishino S Ripley B Overeem S Lammers GJ Mignot E Hypocretin (orexin) deficiency in human narcolepsy. Lancet 2000 355 39 40 10615891 Nishino S Ripley B Overeem S Nevsimalova S Lammers GJ Vankova J Okun M Rogers W Brooks S Mignot E Low cerebrospinal fluid hypocretin (Orexin) and altered energy homeostasis in human narcolepsy. Ann. Neurol 2001 50 381 388 11558795 Nishino S Sakurai E Nevsimalova S Yoshida Y Watanabe T Yanai K Mignot E Decreased CSF histamine in narcolepsy with and without low CSF hypocretin-1 in comparison to healthy controls. Sleep 2009 32 175 180 19238804 Nishino S Tafti M Reid MS Shelton J Siegel JM Dement WC Mignot E Muscle atonia is triggered by cholinergic stimulation of the basal fore-brain: implication for the pathophysiology of canine narcolepsy. J. Neurosci 1995 15 4806 4814 7623112 Nohynek H Jokinen J Partinen M Vaarala O Kirjavainen T Sundman J Himanen SL Hublin C Julkunen I Olsen P Saarenpaa-Heikkila O Kilpi T AS03 adjuvanted AH1N1 vaccine associated with an abrupt increase in the incidence of childhood narcolepsy in Finland. PLoS One 2012 7 e33536 22470453 O'Flanagan D Barret AS Foley M Cotter S Bonner C Crowe C Lynch B Sweeney B Johnson H McCoy B Purcell E Investigation of an association between onset of narcolepsy and vaccination with pandemic influenza vaccine Ireland April 2009–December 2010. EuroSurveillance 2014 19 15 25 24821121 Ohayon MM Narcolepsy is complicated by high medical and psychiatric comorbidities: a comparison with the general population. Sleep Med 2013 14 488 492 23643648 Ohayon MM Black J Lai C Eller M Guinta D Bhattacharyya A Increased mortality in narcolepsy. Sleep 2014 37 439 444 24587565 Ohayon MM Priest RG Zulley J Smirne S Paiva T Prevalence of narcolepsy symptomatology and diagnosis in the European general population. Neurology 2002 58 1826 1833 12084885 Oishi Y Williams RH Agostinelli L Arrigoni E Fuller PM Mochizuki T Saper CB Scammell TE Role of the medial prefrontal cortex in cataplexy. J. Neurosci 2013 33 9743 9751 23739971 Ollila HM Ravel JM Han F Faraco J Lin L Zheng X Plazzi G Dauvilliers Y Pizza F Hong SC Jennum P Knudsen S Kornum BR Dong XS Yan H Hong H Coquillard C Mahlios J Jolanki O Einen M Arnulf I Hogl B Frauscher B Crowe C Partinen M Huang YS Bourgin P Vaarala O Desautels A Montplaisir J Mack SJ Mindrinos M Fernandez-Vina M Mignot E HLA-DPB1 and HLA class I confer risk of and protection from narcolepsy. Am. J. Hum. Genet 2015 96 136 146 25574827 Ortega JE Katner J Davis R Wade M Nisenbaum L Nomikos GG Svensson KA Perry KW Modulation of neurotransmitter release in orexin/hypocretin-2 receptor knockout mice: a microdialysis study. J. Neurosci. Res 2012 90 588 596 22038504 Overeem S Verschuuren JJ Fronczek R Schreurs L den Hertog H Hegeman-Kleinn IM van Duinen SG Unmehopa UA Swaab DF Lammers GJ Immunohistochemical screening for autoantibodies against lateral hypothalamic neurons in human narcolepsy. J. Neuroimmunol 2006 174 187 191 16563524 Pacey LK Tharmalingam S Hampson DR Subchronic administration and combination metabotropic glutamate and GABAB receptor drug therapy in fragile X syndrome. J. Pharmacol. Exp. Ther 2011 338 897 905 21636656 Palaia V Poli F Pizza F Antelmi E Franceschini C Moghadam KK Provini F Pagotto U Montagna P Schenck CH Mignot E Plazzi G Narcolepsy with cataplexy associated with nocturnal compulsive behaviors: a case-control study. Sleep 2011 34 1365 1371 21966068 Parkes JD Chen SY Clift SJ Dahlitz MJ Dunn G The clinical diagnosis of the narcoleptic syndrome. J. Sleep Res 1998 7 41 52 9613427 Partinen M Saarenpaa-Heikkila O Ilveskoski I Hublin C Linna M Olsen P Nokelainen P Alen R Wallden T Espo M Rusanen H Olme J Satila H Arikka H Kaipainen P Julkunen I Kirjavainen T Increased incidence and clinical picture of childhood narcolepsy following the 2009 H1N1 pandemic vaccination campaign in Finland. PLoS One 2012 7 e33723 22470463 Paterson NE Fedolak A Olivier B Hanania T Ghavami A Caldarone B Psychostimulant-like discriminative stimulus and locomotor sensitization properties of the wake-promoting agent modafinil in rodents. Pharmacol. Biochem. Behav 2010 95 449 456 20346966 Peever J Control of motoneuron function and muscle tone during REM sleep REM sleep behavior disorder and cataplexy/narcolepsy. Arch. Ital. Biol 2011 149 454 466 22205591 Peever JH Lai YY Siegel JM Excitatory effects of hypocretin-1 (orexin-A) in the trigeminal motor nucleus are reversed by NMDA antagonism. J. Neurophysiol 2003 89 2591 2600 12611960 Pelin Z Guilleminault C Risch N Grumet FC Mignot E HLA-DQB1*0602 homozygosity increases relative risk for narcolepsy but not disease severity in two ethnic groups. US Modafinil in Narcolepsy Multicenter Study Group. Tissue Antigens 1998 51 96 100 9459509 Peyron C Faraco J Rogers W Ripley B Overeem S Charnay Y Nevsimalova S Aldrich M Reynolds D Albin R Li R Hungs M Pedrazzoli M Padigaru M Kucherlapati M Fan J Maki R Lammers GJ Bouras C Kucherlapati R Nishino S Mignot E A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat. Med 2000 6 991 997 10973318 Peyron C Tighe DK Lee BS de Lecea L Heller HC Sutcliffe JG Kilduff TS Distribution of immunoreactive neurons and fibers for a hypothalamic neuropeptide precursor related to secretin. Soc. Neurosci. Abs 1997 23 2032 Peyron C Tighe DK van den Pol AN de Lecea L Heller HC Sutcliffe JG Kilduff TS Neurons containing hypocretin (orexin) project to multiple neuronal systems. J. Neurosci 1998 18 9996 10015 9822755 Plazzi G Ferri R Antelmi E Bayard S Franceschini C Cosentino FI Abril B Spruyt K Provini F Montagna P Dauvilliers Y Restless legs syndrome is frequent in narcolepsy with cataplexy patients. Sleep 2010 33 689 694 20469811 Plazzi G Moghadam KK Maggi LS Donadio V Vetrugno R Liguori R Zoccoli G Poli F Pizza F Pagotto U Ferri R Autonomic disturbances in narcolepsy. Sleep Med. Rev 2011 15 187 196 20634114 Poli F Plazzi G Di Dalmazi G Ribichini D Vicennati V Pizza F Mignot E Montagna P Pasquali R Pagotto U Body mass index-independent metabolic alterations in narcolepsy with cataplexy. Sleep 2009 32 1491 1497 19928388 Pollak CP Wagner DR Core body temperature in narcoleptic and normal subjects living in temporal isolation. Pharmacol. Biochem. Behav 1994 47 65 71 8115430 Pollock MS Mistlberger RE Rapid eye movement sleep induction by microinjection of the GABA-A antagonist bicuculline into the dorsal subcoeruleus area of the rat. Brain Res 2003 962 68 77 12543457 Prinzmetal M Bloomberg W The use of benzedrine for the treatment of narcolepsy. J. Am. Med. Assoc 1935 105 2051 2054 Rechtschaffen A Wolpert EA Dement WC Mitchell SA Fisher C Nocturnal sleep of narcoleptics. Electroencephalogr. Clin. Neurophysiol 1963 15 599 609 14161512 Reid MS Tafti M Nishino S Sampathkumaran R Siegel JM Mignot E Local administration of dopaminergic drugs into the ventral tegmental area modulates cataplexy in the narcoleptic canine. Brain Res 1996 733 83 100 8891251 Reid MS Tafti M Nishino S Siegel JM Dement WC Mignot E Cholinergic regulation of cataplexy in canine narcolepsy in the pontine reticular formation is mediated by M2 muscarinic receptors. Sleep 1994 17 424 435 7991953 Richardson GS Carskadon MA Flagg W Van den Hoed J Dement WC Mitler MM Excessive daytime sleepiness in man: multiple sleep latency measurement in narcoleptic and control subjects. Electroencephalogr. Clin. Neurophysiol 1978 45 621 627 81764 Ripley B Fujiki N Okura M Mignot E Nishino S Hypocretin levels in sporadic and familial canine narcolepsy. Neurobiol. Dis 2001 8 525 534 11442359 Ristanovic RK Liang H Hornfeldt CS Lai C Exacerbation of cataplexy following gradual withdrawal of antidepressants: manifestation of probable protracted rebound cataplexy. Sleep Med 2009 10 416 421 18753005 Roth T Dauvilliers Y Mignot E Montplaisir J Paul J Swick T Zee P Disrupted nighttime sleep in narcolepsy. J. Clin. Sleep Med 2013 9 955 965 23997709 Sachs C Kaijser L Autonomic regulation of cardiopulmonary functions in sleep apnea syndrome and narcolepsy. Sleep 1982 5 227 238 6813932 Sakurai T The role of orexin in motivated behaviours. Nat. Rev. Neurosci 2014 15 719 731 25301357 Sakurai T Amemiya A Ishii M Matsuzaki I Chemelli RM Tanaka H Williams SC Richardson JA Kozlowski GP Wilson S Arch JR Buckingham RE Haynes AC Carr SA Annan RS McNulty DE Liu WS Terrett JA Elshourbagy NA Bergsma DJ Yanagisawa M Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 1998 92 573 585 9491897 Saper CB Chou TC Scammell TE The sleep switch: hypothalamic control of sleep and wakefulness. Trends Neurosci 2001 24 726 731 11718878 Saper CB Fuller PM Pedersen NP Lu J Scammell TE Sleep state switching. Neuron 2010 68 1023 1042 21172606 Scammell TE Willie JT Guilleminault C Siegel JM A consensus definition of cataplexy in mouse models of narcolepsy. Sleep 2009 32 111 116 19189786 Schenck CH Bassetti CL Arnulf I Mignot E English translations of the first clinical reports on narcolepsy and cataplexy by Westphal and Gelineau in the late 19th century, with commentary. J. Clin. Sleep Med 2007 3 301 311 17561602 Schuld A Beitinger PA Dalal M Geller F Wetter TC Albert ED Hebebrand J Pollmacher T Increased body mass index (BMI) in male narcoleptic patients, but not in HLA-DR2-positive healthy male volunteers. Sleep Med 2002 3 335 339 14592196 Schuld A Blum WF Uhr M Haack M Kraus T Holsboer F Pollmacher T Reduced leptin levels in human narcolepsy. Neuroendocrinology 2000 72 195 198 11070422 Seeman P Guan HC Hirbec H Dopamine D2High receptors stimulated by phencyclidines, lysergic acid diethylamide, salvinorin A, and modafinil. Synapse 2009 63 698 704 19391150 Sellayah D Bharaj P Sikder D Orexin is required for brown adipose tissue development, differentiation, and function. Cell Metab 2011 14 478 490 21982708 Shelton J Nishino S Vaught J Dement WC Mignot E Comparative effects of modafinil and amphetamine on daytime sleepiness and cataplexy of narcoleptic dogs. Sleep 1995 18 817 826 8746387 Shuman T Cai DJ Sage JR Anagnostaras SG Interactions between modafinil and cocaine during the induction of conditioned place preference and locomotor sensitization in mice: implications for addiction. Behav. Brain Res 2012 235 105 112 22963989 Siegel JM Nienhuis R Fahringer HM Paul R Shiromani P Dement WC Mignot E Chiu C Neuronal activity in narcolepsy: identification of cataplexy-related cells in the medial medulla. Science 1991 252 1315 1318 1925546 Silber MH Krahn LE Olson EJ Pankratz VS The epidemiology of narcolepsy in Olmsted County Minnesota: a population-based study. Sleep 2002 25 197 202 11902429 Silverman JL Pride MC Hayes JE Puhger KR Butler-Struben HM Baker S Crawley JN GABA receptor agonist R-baclofen reverses social deficits and reduces repetitive behavior in two mouse models of autism. Neuropsychopharmacology 2015 40 9 2228 2239 25754761 Sorensen GL Knudsen S Petersen ER Kempfner J Gammeltoft S Sorensen HB Jennum P Attenuated heart rate response is associated with hypocretin deficiency in patients with narcolepsy. Sleep 2013 36 91 98 23288975 Strain GM Olcott BM Archer RM McClintock BK Narcolepsy in a Brahman bull. J. Am. Vet. Med. Assoc 1984 185 538 541 6541218 Study Group U.S.X.M. A randomized, double blind, placebo-controlled multicenter trial comparing the effects of three doses of orally administered sodium oxybate with placebo for the treatment of narcolepsy. Sleep 2002 25 42 49 11833860 Study Group, U.S.M.X A 12-month, open-label, multicenter extension trial of orally administered sodium oxybate for the treatment of narcolepsy. Sleep 2003a 26 31 35 12627729 Study Group U.S.X.M. The abrupt cessation of therapeutically administered sodium oxybate (GHB) does not cause withdrawal symptoms. J. Toxicol. Clin. Toxicol 2003b 41 131 135 12733850 Study Group U.S.X.M. Sodium oxybate demonstrates long-term efficacy for the treatment of cataplexy in patients with narcolepsy. Sleep Med 2004 5 119 123 15033130 Study Group X.I. Further evidence supporting the use of sodium oxybate for the treatment of cataplexy: a double-blind, placebo-controlled study in 228 patients. Sleep Med 2005 6 415 421 16099718 Sutcliffe JG Gautvik KM Kilduff TS Horn T Foye PE Danielson PE Frankel WN Bloom FE de Lecea L Two novel hypothalamic peptides related to secretin derived from a single neuropeptide precursor. Soc. Neurosci. Abs 1997 23 2032 Sweeney CR Hendricks JC Beech J Morrison AR Narcolepsy in a horse. J. Am. Vet. Med. Assoc 1983 183 126 128 6683719 Tabuchi S Tsunematsu T Black SW Tominaga M Maruyama M Takagi K Minokoshi Y Sakurai T Kilduff TS Yamanaka A Conditional ablation of orexin/hypocretin neurons: a new mouse model for the study of narcolepsy and orexin system function. J. Neurosci 2014 34 6495 6509 24806676 Tafti M Hor H Dauvilliers Y Lammers GJ Overeem S Mayer G Javidi S Iranzo A Santamaria J Peraita-Adrados R Vicario JL Arnulf I Plazzi G Bayard S Poli F Pizza F Geisler P Wierzbicka A Bassetti CL Mathis J Lecendreux M Donjacour CE van der Heide A Heinzer R Haba-Rubio J Feketeova E Hogl B Frauscher B Beneto A Khatami R Canellas F Pfister C Scholz S Billiard M Baumann CR Ercilla G Verduijn W Claas FH Dubois V Nowak J Eberhard HP Pradervand S Hor CN Testi M Tiercy JM Kutalik Z DQB1 locus alone explains most of the risk and protection in narcolepsy with cataplexy in Europe. Sleep 2014 37 19 25 24381371 Tafti M Rondouin G Besset A Billiard M Sleep deprivation in narcoleptic subjects: effect on sleep stages and EEG power density. Electroencephalogr. Clin. Neurophysiol 1992 83 339 349 1281079 Tanaka KF Matsui K Sasaki T Sano H Sugio S Fan K Hen R Nakai J Yanagawa Y Hasuwa H Okabe M Deisseroth K Ikenaka K Yamanaka A Expanding the repertoire of optogenetically targeted cells with an enhanced gene expression system. Cell Rep 2012 2 397 406 22854021 Tanaka S Honda Y Inoue Y Honda M Detection of autoantibodies against hypocretin, hcrtrl, and hcrtr2 in narcolepsy: anti-Hcrt system antibody in narcolepsy. Sleep 2006 29 633 638 16774153 Tashiro T Kanbayashi T Iijima S Hishikawa Y An epidemiological study on prevalence of narcolpsy in Japanese. J. Sleep Res 1992 1 228 Thannickal T Moore R Nienhuis Y Ramanathan R Gulyani L Aldrich S Cornford M Siegel MJM Reduced number of hypocretin neurons in human narcolepsy. Neuron 2000 27 469 474 11055430 Thannickal TC Nienhuis R Siegel JM Localized loss of hypocretin (orexin) cells in narcolepsy without cataplexy. Sleep 2009 32 993 998 19725250 Thorpy MJ Dauvilliers Y Clinical and practical considerations in the pharmacologic management of narcolepsy. Sleep Med 2015 16 9 18 25458251 Thorpy MJ Krieger AC Delayed diagnosis of narcolepsy: characterization and impact. Sleep Med 2014 15 502 507 24780133 Tsunematsu T Kilduff TS Boyden ES Takahashi S Tominaga M Yamanaka A Acute optogenetic silencing of orexin/hypocretin neurons induces slow-wave sleep in mice. J. Neurosci 2011 31 10529 10539 21775598 Tsunematsu T Tabuchi S Tanaka KF Boyden ES Tominaga M Yamanaka A Long-lasting silencing of orexin/hypocretin neurons using archaerhodopsin induces slow-wave sleep in mice. Behav. Brain Res 2013 255 64 74 23707248 Tupone D Madden CJ Cano G Morrison SF An orexinergic projection from perifornical hypothalamus to raphe pallidus increases rat brown adipose tissue thermogenesis. J. Neurosci 2011 31 15944 15955 22049437 Valko PO Gavrilov YV Yamamoto M Reddy H Haybaeck J Mignot E Baumann CR Scammell TE Increase of histaminergic tuberomammillary neurons in narcolepsy. Ann. Neurol 2013 74 794 804 24006291 Valko PO Khatami R Baumann CR Bassetti CL No persistent effect of intravenous immunoglobulins in patients with narcolepsy with cataplexy. J. Neurol 2008 255 1900 1903 18825431 van Amsterdam JG Brunt TM McMaster MT Niesink RJ Possible long-term effects of gamma-hydroxybutyric acid (GHB) due to neurotoxicity and overdose. Neurosci. Biobehav. Rev 2012 36 1217 1227 22342779 Van Cauter E Plat L Scharf MB Leproult R Cespedes S L'Hermite-Baleriaux M Copinschi G Simultaneous stimulation of slow-wave sleep and growth hormone secretion by gamma-hydroxybutyrate in normal young Men. J. Clin. Investig 1997 100 745 753 9239423 van den Pol AN Hypothalamic hypocretin (orexin): robust innervation of the spinal cord. J. Neurosci 1999 19 3171 3182 10191330 Vassalli A Dellepiane JM Emmenegger Y Jimenez S Vandi S Plazzi G Franken P Tafti M Electroencephalogram paroxysmal theta characterizes cataplexy in mice and children. Brain 2013 136 1592 1608 23616586 Vetrivelan R Saper CB Fuller PM Armodafinil-induced wakefulness in animals with ventrolateral preoptic lesions. Nat. Sci. Sleep 2014 6 57 63 24833927 Vienne J Bettler B Franken P Tafti M Differential effects of GABAB receptor subtypes {gamma}-hydroxybutyric acid, and baclofen on EEG activity and sleep regulation. J. Neurosci 2010 30 14194 14204 20962240 Vienne J Lecciso G Constantinescu I Schwartz S Franken P Heinzer R Tafti M Differential effects of sodium oxybate and baclofen on EEG, sleep, neurobehavioral performance, and memory. Sleep 2012 35 1071 1083 22851803 Volkow ND Fowler JS Wang GJ Baler R Telang F Imaging dopamine's role in drug abuse and addiction. Neuropharmacology 2009 56 Suppl 1 S3 S8 Walsh JK Hall-Porter JM Griffin KS Dodson ER Forst EH Curry DT Eisenstein RD Schweitzer PK Enhancing slow wave sleep with sodium oxybate reduces the behavioral and physiological impact of sleep loss. Sleep 2010 33 1217 1225 20857869 Wang J Greenberg H Status cataplecticus precipitated by abrupt withdrawal of venlafaxine. J. Clin. Sleep Med 2013 9 715 716 23853567 Wang L Meece K Williams DJ Lo KA Zimmer M Heinrich G Martin Carli J Leduc CA Sun L Zeltser LM Freeby M Goland R Tsang SH Wardlaw SL Egli D Leibel RL Differentiation of hypothalamic-like neurons from human pluripotent stem cells. J. Clin. Investig 2015 125 796 808 25555215 Weinhold SL Seeck-Hirschner M Nowak A Hallschmid M Goder R Baier PC The effect of intranasal orexin-A (hypocretin-1) on sleep, wakefulness and attention in narcolepsy with cataplexy. Behav. Brain Res 2014 262 8 13 24406723 Westphal C Eigenhümliche mit Einschläfen verbundene Anfälle. Arch. Psychiatry 1877 7 631 635 Willie JT Chemelli RM Sinton CM Tokita S Williams SC Kisanuki YY Marcus JN Lee C Elmquist JK Kohlmeier KA Leonard CS Richardson JA Hammer RE Yanagisawa M Distinct narcolepsy syndromes in Orexin receptor-2 and Orexin null mice: molecular genetic dissection of Non-REM and REM sleep regulatory processes. Neuron 2003 38 715 730 12797957 Willie JT Renthal W Chemelli RM Miller MS Scammell TE Yanagisawa M Sinton CM Modafinil more effectively induces wakefulness in orexin-null mice than in wild-type littermates. Neuroscience 2005 130 983 995 15652995 Willie JT Takahira H Shibahara M Hara J Nomiyama M Yanagisawa M Sakurai T Ectopic overexpression of orexin alters sleep/wakefulness states and muscle tone regulation during REM sleep in mice. J. Mol. Neurosci 2011 43 155 161 20711757 Winstone AM Stellitano L Verity C Andrews N Miller E Stowe J Shneerson J Clinical features of narcolepsy in children vaccinated with AS03 adjuvanted pandemic A/H1N1 2009 influenza vaccine in England. Dev. Med. Child Neurol 2014 56 1117 1123 25041214 Wisor J Modafinil as a catecholaminergic agent: empirical evidence and unanswered questions. Front. Neurol 2013 4 139 24109471 Wisor JP Nishino S Sora I Uhl GH Mignot E Edgar DM Dopaminergic role in stimulant-induced wakefulness. J. Neurosci 2001 21 1787 1794 11222668 Wu MF Gulyani SA Yau E Mignot E Phan B Siegel JM Locus coeruleus neurons: cessation of activity during cataplexy. Neuroscience 1999 91 1389 1399 10391445 Xie X Smart TG Gamma-hydroxybutyrate hyperpolarizes hippocampal neurones by activating GABAB receptors. Eur. J. Pharmacol 1992 212 291 294 1318216 Xyrem® Xyrem(R) (sodium oxybate oral solution) Prescribing Information [Package Insert] 2005 Jazz Pharmaceuticals, Inc. Palo Alto, CA Yamanaka A Muraki Y Tsujino N Goto K Sakurai T Regulation of orexin neurons by the monoaminergic and cholinergic systems. Biochem. Biophys. Res. Commun 2003 303 120 129 12646175 Yamuy J Fung SJ Xi M Chase MH Hypocretinergic control of spinal cord motoneurons. J. Neurosci 2004 24 5336 5345 15190106 Yin J Mobarec JC Kolb P Rosenbaum DM Crystal structure of the human OX2 orexin receptor bound to the insomnia drug suvorexant. Nature 2015 519 247 250 25533960 Yoss RE Daly DD Criteria for the diagnosis of the narcoleptic syndrome. Proc. Staff Meet. Mayo Clin 1957 32 320 328 13441766 Young JW Geyer MA Action of modafinil–increased motivation via the dopamine transporter inhibition and D1 receptors? Biol. Psychiatry 2010 67 784 787 20132929 Zhang S Lin L Kaur S Thankachan S Blanco-Centurion C Yanagisawa M Mignot E Shiromani PJ The development of hypocretin (orexin) deficiency in hypocretin/ataxin-3 transgenic rats. Neuroscience 2007 148 34 43 17618058 Zolkowska D Jain R Rothman RB Partilla JS Roth BL Setola V Prisinzano TE Baumann MH Evidence for the involvement of dopamine transporters in behavioral stimulant effects of modafinil. J. Pharmacol. Exp. Ther 2009 329 738 746 19197004

PMC005xxxxxx/PMC5114177.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9607835 20545 Mol Psychiatry Mol. Psychiatry Molecular psychiatry 1359-4184 1476-5578 27184122 5114177 10.1038/mp.2016.75 NIHMS770897 Article Haploinsufficiency of the 22q11.2-microdeletion gene Mrpl40 disrupts short-term synaptic plasticity and working memory through dysregulation of mitochondrial calcium Devaraju Prakash 1 Yu Jing 1 Eddins Donnie 1 Mellado-Lagarde Marcia M. 1 Earls Laurie R. 13 Westmoreland Joby J. 13 Quarato Giovanni 2 Green Douglas R. 2 Zakharenko Stanislav S. 1* 1 Department of Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 2 Department of Immunology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA * Contact information: Stanislav S. Zakharenko M.D., Ph.D., Department of Developmental Neurobiology, Mail Stop 323, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA, stanislav.zakharenko@stjude.org 3 Present address: Department of Cell and Molecular Biology, Tulane University, New Orleans, LA, 70118, USA 22 3 2016 17 5 2016 9 2017 18 11 2016 22 9 13131326 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Hemizygous deletion of a 1.5- to 3-megabase region on chromosome 22 causes 22q11.2 deletion syndrome (22q11DS), which constitutes one of the strongest genetic risks for schizophrenia. Mouse models of 22q11DS have abnormal short-term synaptic plasticity (STP) that contributes to working memory deficiencies similar to those in schizophrenia. We screened mutant mice carrying hemizygous deletions of 22q11DS genes and identified haploinsufficiency of Mrpl40 (mitochondrial large ribosomal subunit protein 40) as a contributor to abnormal STP. Two-photon imaging of the genetically encoded fluorescent calcium indicator GCaMP6, expressed in presynaptic cytosol or mitochondria, showed that Mrpl40 haploinsufficiency deregulates STP via impaired calcium extrusion from the mitochondrial matrix through the mitochondrial permeability transition pore. This led to abnormally high cytosolic calcium transients in presynaptic terminals and deficient working memory but did not affect long-term spatial memory. Thus, we propose that mitochondrial calcium deregulation is a novel pathogenic mechanism of cognitive deficiencies in schizophrenia. Introduction Schizophrenia (SCZ) is a catastrophic disease that affects approximately 1% of the world’s population and is characterized by multiple symptoms that include cognitive abnormalities such as deficits in working memory, executive function, and learning1. Mechanisms of cognitive symptoms of SCZ are poorly understood, partly because only weak associations have been identified between any single gene and the disease, and valid animal models have been lacking2. Mouse models of 22q11 deletion syndrome (22q11DS) are among the few animal models that replicate abnormalities associated with SCZ. The 22q11DS is the most common multi-gene syndrome in humans and is considered a genetic risk factor for SCZ. The 22q11DS is caused by the hemizygous deletion of a 1.5- to 3-megabase region on the q arm of chromosome 22, resulting in the haploinsufficiency of multiple genes3. Approximately 30% of children with 22q11DS experience SCZ during late adolescence or early adulthood4, 5. Symptoms of 22q11DS-related SCZ are indistinguishable from those of the idiopathic disease5, suggesting that the biological mechanisms involved in SCZ arising from the 22q11 deletion are similar to those involved in non–deletion-related SCZ. The diagnosis of SCZ usually includes positive symptoms (i.e., disorderly thinking, hallucinations, and delusional ideas), negative symptoms (i.e., low levels of emotional arousal or social activity), and cognitive symptoms (i.e., deficits in attention, working memory, executive function, and learning and memory). Recognition of cognitive deficits as a core feature of SCZ and 22q11DS is increasing, as these deficits better predict disease progression than do the other symptoms6, 7. Many cognitive symptoms of SCZ are thought to originate in the hippocampus8, 9, a key brain region involved in learning and memory. Spatial working-memory deficits occur in patients with 22q11DS10, 11 and are also seen in 22q11DS mouse models. Mouse models of 22q11DS exhibit abnormal hippocampal short- and long-term synaptic plasticity12, 13, which is consistent with the notion that synaptic plasticity is a cellular mechanism of learning and memory16. Short-term synaptic plasticity (STP) acting on the millisecond-to-minute time scale is believed to underlie reliable information transfer between hippocampal excitatory synapses in an activity-dependent manner14–17, working memory18, and decision making19. STP predominantly occurs in presynaptic neurons20. Several studies have shown that presynaptic abnormalities can be attributed to dysregulation of presynaptic calcium (Ca2+). For example, altered STP resulting from deregulated presynaptic Ca2+ are seen in models of neuropsychiatric diseases such as FMRP-related autism, Alzheimer disease, and 22q11DS12, 21, 22. Because 22q11DS is a multi-gene deletion syndrome, more than one gene may affect STP. Initially, STP dysregulation in Df(16)1+/− models of 22q11DS was linked to haploinsufficiency of microRNA-processing gene Dgcr823, which is mapped to a proximal part of the microdeletion. Depletion of microRNAs miR-185 and miR-25 leads to presynaptic Ca2+ dysregulation and abnormal STP through the abnormal elevation of (sacro)endoplasmic reticulum ATPase type 2 (Serca2), the Ca2+ pump that extrudes Ca2+ from the cytoplasm into the endoplasmic reticulum23. SERCA2 is also elevated in the hippocampus of schizophrenic patients23, and the most comprehensive genome-wide association study to date linked the ATP2A2 gene, which encodes SERCA2, with SCZ24. Other genes that affect STP remain unknown. Here we report results of our STP screening of the distal region of the 22q11DS microdeletion, which encompasses six genes: Cldn5, Cdc45l, Ufd1l, 2510002D24Rik, Mrpl40, and Hira. Using mutant mice carrying hemizygous deletions of individual genes, we discovered that haploinsufficiency of Mrpl40 (mitochondrial large ribosomal subunit protein 40, also known as Nlvcf) causes abnormal STP and short-term memory deficits via Serca2-independent deregulation of presynaptic Ca2+. Using two-photon Ca2+ imaging of the genetically encoded Ca2+ indicator GCaMP625, expressed either in the presynaptic cytosol or mitochondria, we showed that Mrpl40 haploinsufficiency hindered the extrusion of Ca2+ from the mitochondrial matrix through impaired mitochondrial permeability-transition pore (mPTP). This leads to abnormally high levels of Ca2+ in the presynaptic cytosol and elevated STP. Our data implicate Mrpl40 as a 22q11DS gene, the haploinsufficiency of which contributes to cognitive deficits in microdeletion-related SCZ. Materials and Methods Animals Mature (16–20 weeks) mice of both sexes were used. Production and genotyping of Df(16)5+/−, Dgcr8+/−, Cldn5+/−, and Hira+/− mice were previously described23, 26–28. To generate Cdc45/+/− and Ufd1l+/− mice, we obtained SIGTR embryonic stem (ES) cell clones containing gene-trap disruptions of the pGT01Lxr vector for Cdc45l (cell line AJ0425) and a Ufd1l allele (cell line AW0532) from the Mutant Mouse Regional Resource Center (University of California, Davis). The AJ0425 ES cell line carries a Genetrap insertion in exon 3 of the Cdc45l gene. Offspring were genotyped using the following primers: Cdc45lF: GCTGGGTACCTGAGTGTCATTG, Cdc45lR: CGAGACTGGTATGTGTGTGTGTG, and the betageo primer 2: ATTCAGGCTGCGCAACTGTTGGG, producing a 353-bp wild-type (WT) amplicon and a 309-bp mutant amplicon. The AW0532 ES cell line disrupts the Ufd1l gene in the first intron. This line was genotyped with the following primers: Ufd1lF: GTTGACGCTAACGTCCAGTCAC, Ufd1lR: GAAGCAGCGGTACTGCGTGGAG, and the Betageo primer: ATTCAGGCTGCGCAACTGTTGG, producing a 612-bp WT amplicon and a 304-bp mutant amplicon. To generate the Mrpl40+/− mice, we obtained sperm from the C57BL/6 strain carrying the Mrpl40tm1(KOMP)vlcg allele of the Mrpl40 gene from the trans-NIH Knock-Out Mouse Project (www.komp.org). Sperm was used to produce Mrpl40+/− offspring via in vitro fertilization. Mrpl40+/− mice were genotyped using the following primers: Mrpl40F: CAGGCACACGTCAGACACA, Mrpl40R: GAGATCCCAGAAGGCCAGTAAG, and LacZR: CCCACCAGAGAGCTTC, producing a 102-bp WT amplicon and a 387-bp mutant amplicon. To expand the colony, we bred 2510002D24Rik+/−, Mrpl40+/−, Ufd1l+/−, and Cdc45l+/− mice harboring the disrupted alleles to C57BL/6J mice in the St. Jude Animal Resource Center. All mouse strains in this study were back-crossed onto the C57BL/6J genetic background for at least 5 generations. Individual experiments were conducted in age-matched and sex-matched animals. The care and use of animals were reviewed and approved by the Institutional Animal Care and Use Committee at St. Jude Children’s Research Hospital. Brain slice preparation Mouse brains were quickly removed and placed in cold (4°C) dissecting artificial cerebrospinal fluid (ACSF) containing 125 mM choline-Cl, 2.5 mM KCl, 0.4 mM CaCl2, 6 mM MgCl2, 1.25 mM NaH2PO4, 26 mM NaHCO3, and 20 mM glucose (295–300 mOsm), under 95% O2/5% CO2, and acute transverse hippocampal slices (400-µm thick) were prepared. After dissection, slices were incubated for 1 h in ACSF containing 124 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 1.25 mM NaH2PO4, 26 mM NaHCO3, and 10 mM glucose (285–295 mOsm), under 95% O2/5% CO2 at 32°C to 34°C and then transferred into the submerged recording chamber and superfused (1.5–2 mL/min) with ACSF. Long-term potentiation (LTP) experiments were performed as previously described12. Because the bath temperature is crucial for short-term potentiation (STP) experiments29, all electrophysiology and imaging experiments were conducted in a perfusion chamber maintained at near physiological temperature (33°C-34°C, measured at the center of the bath) by using both inline and bath heaters. Whole-cell electrophysiology Electrophysiological data were acquired using a Multiclamp 700B amplifier, filtered at 2 kHz and digitized at 10/20 kHz using a Digidata 1440 digitizer controlled by Clampex acquisition software. All offline analyses of electrophysiological data were done in Clampfit. To evoke excitatory postsynaptic currents (EPSCs) in CA1 neurons, we stimulated Schaffer collaterals with a concentric bipolar stimulating electrode (125-µm outer diameter; 12.5-µm inner diameter) connected to an Iso-Flex stimulus isolator with 100-µs pulses. Stimulating and recording electrodes were separated by at least 350 to 400 µm. For STP experiments, a cut was placed between the CA3 and CA1 regions to avoid recurrent stimulation. Recording electrodes were borosilicate glass capillaries pulled in a Sutter P-1000 puller and had resistances approximately 3.5 to 5 mΩ. CA1 neurons were whole cell voltage-clamped using electrodes filled with 130 mM K gluconate, 10 mM KCl, 0.5 mM EGTA, 2 mM MgCl2, 5 mM NaCl, 2 mM ATP-Na2, 0.4 mM GTP-Na, 10 mM HEPES, and 10 to 25 µM Alexa 594 (pH 7.35, ~ 290 mOsm). EPSCs were recorded by holding the cells at –74 mV (accounting for a liquid junction potential of ~14 mV). Access resistance was monitored using a –5 mV step before each recording (range, 25–50 mΩ), and cells that displayed unstable access resistance (variations >20%) were excluded from the analyses. All EPSC recordings for STP experiments were done in the presence of the GABAAR antagonist picrotoxin (100 µM) to prevent inhibitory responses and the NMDA receptor antagonist D-2-amino-5-phosphonovalerate (D-AP5, 50 µM) to avoid induction of long-term synaptic plasticity. STP components were separated using an adapted procedure29. A baseline of stable EPSCs was established to evoke 150- to 200-pA EPSCs. Stimulation intensities ranged from 30 to 60 µA and were held constant for each cell throughout the experiment. Stimulation intensity did not differ significantly between slices from WT and mutant littermates (P > 0.05). The stimulation protocol consisted of the following: 5 pre-train pulses (0.2 Hz), a high-frequency train (100 pulses, 80-Hz), 24 pulses at 5 Hz (to measure recovery from depression), and 24 pulses at 0.2 Hz (to measure augmentation). The 80-Hz (100 pulses) train was chosen because it induced both strong depression and augmentation on the basis of STP induced by trains of different frequencies (data not shown). EPSCs during and after the train were normalized to the average of the 5 pre-train EPSCs (shown as a single baseline point), and the amplitudes are reported as normalized EPSC peak amplitudes. For each cell, we averaged the data from 3 to 4 trials by using this stimulation protocol. To measure the EPSC amplitudes during the train, we resorted to a binning procedure, instead of the template waveform-based subtraction protocol, as previously described29. The measured EPSC train of 1,250 ms (80 Hz × 100 pulses) was divided into 100 equal bins of 12.5 ms, and the peak amplitude within each bin was graphed as the normalized EPSC amplitude during the train. This procedure was sufficient to separate EPSCs in the train. Augmentation, the slowest component of STP, has negligible contamination from depression and facilitation and is reported as measured without corrections. Statistical comparisons of augmentation were made using the measured peak augmentation at 5 s after the 80-Hz train, instead of the extrapolated augmentation at 0 s. Recovery from depression (200 to 4,800 ms) after the 80-Hz train was contaminated by augmentation and some residual facilitation. Because the EPSCs measured during the recovery phase had an interstimulus interval of 200 ms, the contamination by facilitation was minimal and corrected only for overlying augmentation. To that end, the measured augmentation decay curve (5–120 s after the 80-Hz train) was extrapolated from 200 to 4,800 ms after the train by using a standard exponential fit in Clampfit. The normalized EPSCs during the recovery phase were then corrected for the overlying extrapolated augmentation for each time point. Facilitation was measured in separate experiments by using paired pulses at interstimulus intervals of 20 to 1,000 ms. To measure excitability, we held CA3 neurons in current clamp mode and added D-AP5 (50 µM) and picrotoxin (100 µM) to the external ACSF bath to avoid possible long-term effects. Action potential (AP) parameters were estimated by clamping the cell at –65 mV using automatic slow-current injection. Input resistance, AP threshold, rheobase (intensity of current reached at threshold), number of APs (evoked by holding at the minimum current needed to reach threshold) were measured by injecting 1-s steps of 25-pA current (12 steps from –25 pA to 250 pA). To measure AP widths, we evoked APs by injecting 5 short current steps (1,500–2,000 pA, 1-ms duration, 0.2 Hz). AP durations were calculated as the time interval between the up-stroke and down-stroke of the AP waveform at –10 mV. AP duration during the 80-Hz train was normalized to the AP duration at baseline, as previously described21. Electrophysiology experiments were done without blinding. Generation of plasmids and viruses To generate adeno-associated viruses (AAVs) expressing cytoplasmic GCaMP6f (GCaMP6) and mitochondrial-targeted GCaMP6f (mitoGCaMP6), we used PCR to amplify the human synapsin promoter (hSyn) from pAAV-6P-SEWB30. The pAAV-GFP (Addgene plasmid 32395) was cut with SnaB1 and Sac1 to replace CMV with hsyn (pAAV-hsyn-GFP). EcoRI and BamHI were used to replace GFP with mCherry to generate pAAV-hsyn-mCherry. To generate pAAV-hSyn-mCherry-2A-GCaMP6, oligonucleotides containing the coding sequence for the 2A peptide were used for PCR amplification of GCaMP6 from pGP-CMV-GCaMP6 (Addgene plasmid 40755). To generate pAAV-hSyn-mCherry-2A-mitoGCaMP6, oligonucleotides containing the coding sequence for the 2A peptide and the mitochondrial-targeting sequence (MSVLTPLLLRGLTGSARRLPVPRAKIHSL) were used for PCR amplification of GCaMP6 from pGP-CMV-GCaMP6 (Addgene plasmid 40755). To generate AAV-Slc25a4 OE, we used PCR to amplify the open reading frame of Slc25a4. Drd2 open reading frame from AAV-CamKII-Drd2 OE31 was replaced with the Slc25a4 open reading frame using HindIII. DNA sequencing was used to verify the absence of PCR-induced mutations. Lentivirus vector siRNA plasmids (control shRNA, 5′-TACGTCCAAGGTCGGGCAGGAAGA-3′; Slc25a4 shRNA1, 5′- GCAAGGGATCTTCCCAGCGAGAATTCAAT-3′; Slc25a4 shRNA2, 5′-CGTTTGACACTGTTCGTCGTAGGATGATG-3′; Slc25a4 shRNA3, 5′- GCACATTATCGTGAGCTGGATGATTGCCC-3′) were generated by Applied Biological Materials (Richmond, BC, Canada). Viruses (1.8 × 108 to 1 × 109 particles/ml) were produced by either the St. Jude or University of Tennessee Health Sciences Center Viral Vector Cores. Two-photon imaging of presynaptic calcium and in vivo injections Two-photon laser-scanning microscopy was performed using an Ultima imaging system (Prairie Technologies, Middleton, WI), a Ti:sapphire Chameleon Ultra femtosecond-pulsed laser (Coherent Inc., Santa Clara, CA), and 60× (0.9 NA) water-immersion IR objectives (Olympus, Center Valley, PA). Calcium (Ca2+) transients in presynaptic terminals were recorded using GCaMP6 expressed in the CA3 hippocampal neurons. To express GCaMP6, mice were anaesthetized using isoflurane (2% for induction and 1.5% for maintenance) in 100% oxygen, and their heads were restrained on a stereotaxic apparatus. An approximately 1-cm midline incision was made centered about 0.25 cm behind bregma. Viruses were injected into 3 locations within the CA3 region, in 1 or both hemispheres. The stereotaxic coordinates for the 3 injections were as follows, in relation to the bregma: (1) –1.5-mm anteroposterior, 1.8-mm lateral, and 1.7-mm deep; (2) 2.2-mm anteroposterior, 2.3-mm lateral, and 1.8-mm deep; (3) 2.5-mm anteroposterior, 2.8-mm lateral, and 2.2-mm deep. Craniotomy holes were drilled at these locations, and 200 nL of AAVs was slowly (20 nL/min) injected via a 33G cannula. After each injection, the cannula was left in place for 2 to 3 min before being retracted. Following injections, the skin was sutured, and the mice were allowed to recover before returning to the holding cages. Imaging experiments were performed 4 to 7 weeks after AAV injections. During each experiment, care was taken to limit the differences in post-injection durations to a maximum of 2 to 3 days across experimental groups to avoid substantial differences in the levels of AAV expression. To visualize GCaMP6 or mitoGCaMP6, we used brain slices prepared from AAV-injected mice. Schaffer collaterals were stimulated via field stimulation using bipolar stimulating electrodes, as in STP experiments. In addition to D-AP5 and picrotoxin, the AMPA receptor antagonist NBQX (3 µM) was added to the bath ACSF to prevent excitation of postsynaptic neurons. GCaMP6 was visualized at 940 nm, and mCherry was visualized at 1,040 nm by two Ti:Si lasers. Presynaptic boutons were identified in a 34 µm × 34 µm region of interest by activity-dependent increase in GCaMP6 fluorescence during time-series scans. Line-scans through identified boutons were then used for experiments. Boutons from 4 to 8 regions of interests in the stratum radiatum of the CA1 area were imaged for each mouse. In some experiments, we used whole-cell recordings from CA3 neurons and filled the cells with Alexa Fluor 594 (30 µM) and Fluo 5F (300 µM) at 820 nm to visualize presynaptic Ca2+. Axons emanating from the cell bodies were identified based on their morphology and lack of spines. Presynaptic terminals were identified as boutons in secondary and tertiary axonal branches. Those axons could not be tracked beyond 200 µm from the CA3 cell bodies due to the limitation of the approach. APs were evoked by holding the cells at –70 mV (current clamp mode) and injecting depolarizing current (3.5–4.0 nA, 500 µs). We recorded Ca2+ transients in line-scan mode from 3 to 7 boutons per cell. For GCaMP6 or mitoGCaMP6 experiments, the baseline fluorescence (F0) was used to calculate the change in signal (ΔF/F0). For Fluo5F experiments, fluorescence changes (ΔG/R) were quantified as an increase in Fluo 5F fluorescence (ΔG) normalized to the respective Alexa 594 fluorescence (R). Imaging experiments were done without blinding. Two-photon glutamate uncaging For two-photon glutamate uncaging (TGU), MNI-glutamate (2.5 mM) was added to the recording ACSF. The timing and intensity of glutamate uncaging were controlled by TriggerSync (Prairie Technologies). In typical experiments, 0.2-ms pulses from a second Ti:sapphire Chameleon Ultra laser (720 nm) were delivered to the vicinity of a targeted dendritic spine, and TGU-evoked EPSCs (uEPSCs) were recorded. The duration and intensity of illumination of the uncaging laser were then adjusted to induce responses that mimicked spontaneous miniature EPSCs, which were recorded in CA1 neurons and averaged 10 to 15 pA. After the uncaging parameters (i.e., site, laser duration, and laser intensity) were adjusted for a single spine, the parameters remained constant for the STP experiments on the particular spine. An 80-Hz train of 100 TGU pulses was delivered to a single dendritic spine to measure TGU-induced STP. Electron microscopy Mice were anesthetized with ethyl carbamate (1.5 g/kg, 25% solution, intraperitoneal) and perfused transcardially with phosphate-buffered saline (PBS) for 1 to 2 min and then a fixative (2.5% gluteraldehyde and 2% paraformaldehyde in 0.2 M sodium cacodylate). Brains were isolated, stored at 4°C overnight in the same fixative, and sagittal sections (100-µm thick) were prepared on a Leica vibratome. Smaller regions (~ 500 µm × 500 µm) containing the stratum radiatum of the CA1 hippocampal region were processed for 3-dimensional (3D) scanning electron microscopy (SEM). The samples were stained with a modified heavy-metal–staining method, processed through a graded series of alcohol and propylene oxide, and then embedded in Epon hard resin32. Sections (0.5-µm thick) were cut to determine the correct area and then coated with iridium in a Denton Desk II sputter coater. The 3D EM images were collected on a Helios Nanolab 660 Dualbeam system. From the 3D stacks of electron micrographs (10 × 10 x 10 nm voxel size, 250 to 260 sections of 10 nm thickness and approximately 30 × 20 µm area, synapses were identified based on the presence of postsynaptic densities and presynaptic vesicles. The mitochondria in presynaptic terminals were identified and counted manually. For transmission electron microscopy (TEM), 100-µm-thick vibratome sections containing the CA1 stratum radiatum region were prepared as described above and fixed in 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4). Sections were post fixed in 2% osmium tetroxide in 0.1 M sodium cacodylate buffer with 0.3% potassium ferrocyanide for 1.5 h. After rinsing in the same buffer, the sections were dehydrated through a series of graded ethanol and propylene oxide solutions, infiltrated and embedded in epoxy resin, and polymerized at 70°C overnight. Thin sections (0.5 µm) were stained with toluidine blue for examination by light microscopy. Ultrathin sections (80 nm) were cut and imaged using an FEI Tecnai F 20 FEG Electron Microscope with A AT XR41 Camera. Mitochondrial DNA quantification RNA was isolated from the hippocampus of 4-month-old mice by using the mirVana RNA isolation kit (Ambion, Life Technologies) The SuperScript III reverse transcriptase kit (Invitrogen, Life Technologies) was used to synthesize cDNA from 1 µg RNA. The following primers were used in the qPCR experiments: COI (5'-GCCCCAGATATAGCATTCCC-3'and 5'-GTTCATCCTGTTCCTGCTCC-3'); ND2 (5'-CCCATTCCACTTCTGATTACC-3' and 5'-ATGATAGTAGAGTTGAGTAGCG-3'); 18S (5'-TAGAGGGACAAGTGGCGTTC-3' and 5'-CGCTGAGCCAGTCAGTGT-3'). The ratio of mitochondrial (CO1 and ND2) transcripts to the nuclear 18S transcript was used to quantify relative mitochondrial DNA. Mitochondrial respiration Mice were decapitated, and their brains were dissected immediately in cold (4°C) dissecting ACSF. Hippocampi were removed and washed in a mitochondrial isolation buffer (MIB) containing 250 mM sucrose, 1 mg/mL bovine serum albumin, 5 mM EDTA, and 10 mM Tris-HCl at pH 7.4. A crude mitochondrial fraction was then isolated by differential centrifugation. Briefly, the hippocampal tissue was finely minced on a pre-chilled glass dish, washed several times, resuspended in 1 mL MIB and homogenized in a 2-mL Dounce homogenizer (glass/glass, 5–7 runs in an ice bath). The homogenized tissue was centrifuged at 700 ×g for 5 min; the supernatant was then transferred to a new 1.5-mL tube and centrifuged at 7,000 ×g for 10 min. The resulting pellet was washed in 1 mL MIB and resuspended in 30 µL MIB. The protein concentration in the pellet was typically 25 to 35 mg protein/mL. The mitochondrial integrity was tested by measuring the glutamate/malate-dependent respiratory control ratio (i.e., State III/State IV of respiration). The resulting mitochondrial samples were used immediately to measure respiration (mitochondrial oxygen consumption) and Ca2+-retention capacity. Mitochondrial oxygen consumption was measured using a Clark-type electrode (Hansatech Instruments, Ltd, Norfolk, U.K.) in a thermostatically controlled chamber equipped with a magnetic stirring device and a gas-tight stopper fitted with a narrow port for additions via a Hamilton microsyringe. Isolated mitochondria were placed in the respiration chamber at 37°C in 0.4 mL respiration buffer (250 mM sucrose, 1 g/L bovine serum albumin, 10 mM KH2PO4, 2.7 mM KCl, 3 mM MgCl2, 40 mM Hepes, 0.5 mM EGTA, pH 7.1) to yield a final concentration of 0.5 mg/mL. State-III respiration was stimulated by the addition of 2 mM ADP. Respiratory control ratios were obtained by dividing the rate of oxygen consumption in the presence of ADP (state III) by that in the absence of ADP (state IV). The measurement protocol involved sequential addition (with 5-min intervals) of 5 mM glutamate, 2.5 mM malate (activating the CI-CIII-CIV span), 1 µM rotenone, 10 mM succinate (activating the CII-CIII-CIV span), 1 µM antymicin A, 10 mM ascorbate, 0.4 mM TMPD (activating the CIV span), and 5 mM KCN. The rates of oxygen consumption were calculated online, as first derivatives of the dioxygen-content changes by manufacturer-provided software. Quantitative real-time PCR The cDNA was generated from 1 µg hippocampal RNA by using Superscript III (Life Technologies, Waltham, MA). Primers for qPCR were as follows: Hira: TCCGCCATCCATCAATTC and CTATCCTTCACCAGCCTAG, Cldn5: CGCAGACGACTTGGAAGG and GCCAGCACAGATTCATACAC, Mrpl40: CTGGTAGTTAGAGATAGGTGGTG and GAGGAGCTGAAACTTGAATCTG, Ufd1l: TCAAGCATGTATTCATTCTGC and TTTATTTACAGTGACTCAGAAGG, 2510002D24Rik: GTGTTCCAGGTCAAGTAA and AGAAGGACAAGTGATAAGC, Cdc45l: GATTTCCGCAAGGAGTTCTACG and TACTGGACGTGGTCACACTGA. Co1: GCCCCAGATATAGCATTCCC and GTTCATCCTGTTCCTGCTCC, Nd2: CCCATTCCACTTCTGATTACC and ATGATAGTAGAGTTGAGTAGCG, 18s: TAGAGGGACAAGTGGCGTTC and CGCTGAGCCAGTCAGTGT. We performed qPCR using SYBR green in an Applied Biosystems 7900HT Fast Real-time PCR system and the standard protocol. A serial dilution of cDNA was used to generate a standard curve for each primer set, and this curve was used to calculate gene concentrations for each sample. All samples were run in triplicate. Mouse behavior Mature animals (16- to 20-weeks) were used for all behavior experiments. Morris water maze One hour prior to testing, animals were brought into the testing room and allowed to habituate. Testing was performed during the animal’s inactive phase under dim-light conditions. Mice were allowed to navigate in the maze, and swim patterns were recorded with a video camera tracking system (HVS Image, Co., Buckingham, UK) mounted above the pool. Animals learned to find a hidden, clear platform by using the standard spatial version of the Morris water maze task for 4 successive days. Each day, animals were given four 1-min trials from each starting position with an inter-trial latency of 60 s. The order of the starting locations was counterbalanced each day by using a Latin-square design. A spatial learning (probe) trial was administered 1 h after the completion of spatial training. A spatial memory (probe) trial was administered 48 h after completion of the spatial learning. During both probe trials, the platform was removed, and the mice received a single 1-min trial in which the animal tried to find the escape platform. These trials originated from the starting location that was the farthest from the platform’s location throughout training. Mice also completed a nonspatial learning task at least 7 days after completion of the spatial protocol. In that task, mice were trained to find a black visible platform for 2 successive days. During Day 1, the escape platform was located in the same position used during spatial training. The next day, the escape platform was moved to a new quadrant. Each day, the mouse was given four 1-min trials in the same manner that occurred during spatial training. To avoid hypothermia, immediately after each round of training and testing trials, animals were dried with paper towels and placed in warmed holding cages. Delayed non–matched-to-position task To motivate mice to complete the delayed non–matched-to-position task, they were subjected to water restriction for 2 days prior to testing. Specifically, mice were allowed 2 h of free access to water per day. Mice were weighed daily to ensure that weight decrease during deprivation did not surpass the recommended 20% loss. One hour prior to testing, animals were brought into the testing room and allowed to habituate. Testing was performed during the animal’s inactive phase under well-lit lighting conditions. The testing apparatus consisted of a Y-maze (Cleversys Inc., Reston, VA) with a start arm (20 cm × 16 cm × 7 cm; l × w × h) leading to 2 goal arms (20 cm × 16 cm × 7 cm). Mice were allowed to habituate to the maze and were given a positive reinforcer (i.e., Chocolate Yoohoo) before behavioral testing. To achieve this, mice were allowed to investigate the Y-maze baited with Yoohoo for 15 min. Maze habituation was performed for 2 consecutive days. After maze habituation was complete and the animal had consumed the food rewards, we conducted a test of spatial working memory. First, the mouse was constrained in the start arm with a guillotine door. Next, a sample arm was determined at random and the choice arm was closed off with a guillotine door (there was a limit of 2 consecutive same-side sample arms in a 10-trial test). Both arms were then baited with 20 µL Yoohoo. The mouse was released from the start chamber and allowed to run to the sample arm and consume the reward. The mouse was then returned to the start arm and the guillotine door to the nonsample arm was removed. All efforts were made to keep the intratrial interval at 5 s. The mouse was again released from the start chamber and allowed to run to either the previously entered sample arm or new choice arm to consume the reward. A return to the sample arm was counted as an incorrect response. Incorrect responses resulted in no reward and return to the start arm. A total of 10 trials were given. The Y-maze spatial recognition The maze was shaped like a “Y”, with 3 equally spaced arms (20 cm × 7 cm × 16 cm) radiating from a triangular center section. The Y-maze was constructed from blue opaque plastic to aid in video detection. The maze was located in a lit room with abundant extra-maze cues. The procedure consisted of an acquisition and a recognition session. During the acquisition session, the mouse was placed facing the distal end of an arm (start arm; determined semirandomly) and allowed to freely explore the maze for 15 min. During acquisition, the mouse was allowed to freely explore 2 of the 3 arms (determined semirandomly). After completing the acquisition session, the mouse was returned to its home cage for 1 h. Next, the mouse was given a 2-min recognition session, where all 3 arms were available for the mouse to freely explore. Time spent in each arm and the triangular center section and the total distance traveled during each trial were automatically recorded using TopScan software (CleverSys Inc.). Mice prefer novelty; therefore, if a mouse recognized and remembered which arms were familiar during the recognition session, the mouse would spend more time in the novel arm (expressed as a percent of total arm time) than would be expected by chance. Acoustic startle and prepulse inhibition Each day before testing, the mice were allowed a 1-h habituation in the testing room after being transported from the animal housing room. Before experiments were initiated, the mice were allowed to acclimate to the Plexiglas restraint chamber (6 cm × 6 cm × 4.8 cm) for 20 min. Acoustic-startle and prepulse inhibition (PPI) tests were performed in ventilated, sound-attenuated chambers (Med Associates, St. Albans City, VT). For acoustic-startle experiments, the mice had a 5-min acclimation period to a 65-dB background white noise, which played throughout the session. Three startle pulses (8 kHz, 120 dB, 40 ms) were then delivered at 15-s intertrial intervals. For PPI experiments (conducted on different days than acoustic-startle experiments), mice had a 5-min acclimation period to a 65-dB background white noise, which played throughout the session. Three acoustic startles (broadband white noise click, 120 dB, 40 ms) were then delivered separated by a 15-s intertrial interval. The testing session consisted of 39 trials of 5 trial types: pulse alone, in which the startle pulse was presented; the combination of a 40-ms prepulse of 74 dB, 82 dB, or 90 dB preceding the startle pulse by 100 ms; and no stimuli. Trials were separated by 15 s and presented in a pseudo-random order. PPI was calculated as follows: 100 × (pulse-alone response – prepulse + pulse response)/pulse-alone response. All mouse behavior experiments were performed in a blind manner in respect to mouse genotypes. Western blotting AAV5-hSyn-mCherry-2A-mitoGCaMP6f or AAV5-hSyn-mCherry-2A-GCaMP6f viruses were injected in vivo into the mouse hippocampus as described above. Four weeks after injections, mice were euthanized and dorsal hippocampi were isolated at 4°C for fractionation into nuclear/cell debris, cytoplasm, and crude mitochondrial fractions. Freshly extracted hippocampi from each mouse were homogenized in isolation buffer (250 mM sucrose, 75 mM mannitol, 1 mM EGTA, and 5 mM Hepes, at pH 7.4) by using a glass-teflon tissue homogenizer on ice. Homogenized hippocampi were centrifuged twice at 1,400 ×g for 3 min at 4°C to separate nuclei and cell debris. The supernatant was then centrifuged at 17,200 ×g for 10 min at 4°C to separate cytosol and crude mitochondria. Ice-cold RIPA buffer (Santa Cruz Biotechnology, Dallas TX) with protease inhibitors (Roche, Basel, Switzerland) was added to the mitochondria but not the cytoplasmic fraction, and both were briefly sonicated. Samples were centrifuged at 17,200 ×g for 20 min at 4°C, and the pellet was kept at –80°C for protein quantification. Protein quantification was performed using a Pierce BCA Protein Assay Kit (Thermo Scientific). For loading-sample preparation, we used NuPAGE LDS sample buffer (Life Technologies). The Western-blot experiments were performed similar to a previously described method12. Briefly, NuPAGE 10% Bis-Tris gels and MES-SDS running buffer (Life Technologies) were used to load mitochondria and cytoplasmic fractions (10 µg total protein/well) and run the gels. The primary antibody used to detect GCaMP6 was rabbit–anti-GFP (1:1000, Abcam, 6556). We also used rabbit–anti-Prohibitin 1 (1:1000, Thermo Scientific, PA5-12274) and mouse–anti–β-actin (1:10,000, Sigma, 5316) primary antibodies. Secondary antibodies were Odyssey goat–anti-rabbit IRDye-680LT (1:40,000, LI-COR Biosciences, Lincoln, NE, 926-68021) or Odyssey donkey–anti-mouse IRDye-800CW (1:15,000, LI-COR Biosciences, 926-32212). Membranes were imaged using an Odyssey Infrared Imager (LI-COR Biosciences). Images were analyzed using Odyssey V3.0 software. Subcellular localization of mitoGCaMP6 and GCaMP6 Neuro-2a (N2a) mouse neuroblastoma cells (ATCC, CCL-131) were plated in 4-well chamber slides (Thermo-Scientific Lab-Tek 177399) at 1.25 × 105 cells/well and maintained in culture using Eagle’s Medium Essential Media (EMEM) (ATCC) plus 10% fetal bovine serum (heat inactivated, Life Technologies) and 1× PenStrep (Life Technologies) in an incubator at 37°C and 95% O2/5% CO2. Following a 24-h incubation, cells were transfected with hSyn-mCherry-2A-mitoGCaMP6f or hSyn-mCherry-2A-GCaMP6f plasmids (0.5 µg DNA/well) using Fugene HD transfection reagent (Promega) (Ration Fugene HD/DNA 1:4) in serum-free EMEM. Twenty-four hours later, cells were supplemented with additional fresh complete media, and 48 h later, they were incubated in complete media with Mitotracker DeepRed (1:2,000, Life Technologies) for 15 min at 37°C. Next, Mitotracker was washed out and replaced with fresh complete media, and cells were placed back to the incubator for 15 to 20 min. Cells were then washed with sterile PBS and fixed in 4% paraformaldehyde for 15 min at room temperature. Following fixation, cells were washed 3 times for 5 min with PBS and blocked with 10% normal goat serum and 0.1% TritonX-100 in PBS for 1 h at room temperature. We used the following antibodies: chicken–anti-GFP to detect GCaMP6 or mitoGCaMP6 (1:1000, Abcam, 13970) and goat–anti-chicken–Alexa-488 (1:1000, Life Technologies, A11039). Cell imaging was performed using an LSM-780 confocal microscope (Zeiss). Other drugs and chemicals Bongkrekic acid solution was purchased from Sigma-Aldrich. D-AP5, picrotoxin, tetrodotoxin, NBQX, CGP 37157, MNI glutamate were from Tocris. Stock solutions of these drugs were prepared in manufacturer recommended solvents and stored at −20°C. Ru360 was from Calbiochem. Statistical analyses Data are presented as the mean ± SEM. Statistical analyses were performed using SigmaPlot software. Parametric or nonparametric tests were chosen based on the normality and variance of data distribution. Independent or paired two-tailed t-tests (t value), Mann-Whitney Rank Sum test (U value), one-way analysis of variance (ANOVA) / Kruskal Wallis one-way analysis of variance on ranks followed by a multiple-comparison procedure (Dunn’s method), two-way ANOVA / two way repeated measures ANOVA with one factor repetition followed by Holm-Sidak multiple comparison procedure were the statistical tests used. F values were reported for ANOVA and Q values from multiple comparison procedure were reported for ANOVA on ranks. Differences with p < 0.05 were considered significant. Results Abnormal presynaptic augmentation underlies aberrant STP in Df(16)5+/− mice To determine whether haploinsufficiency of distal genes in the 22q11DS region contributes to hippocampal pathophysiology, we used Df(16)5+/− mice33 carrying a hemizygous deletion of six genes: Cldn5, Cdc45l, Ufd1l, 2510002D24Rik, Mrpl40, and Hira (Fig. 1a). Similar to the late onset of SCZ symptoms, synaptic plasticity abnormalities in mouse models of 22q11DS do not appear until later in life23; thus, we used 4- to 5-month-old mice for these experiments. Df(16)5+/− mice developed normally and had no visible gross morphologic abnormalities (data not shown). Using the whole-cell voltage-clamp technique we recorded excitatory postsynaptic currents (EPSCs) at glutamatergic synapses between CA3 and CA1 pyramidal neurons in the hippocampus (CA3–CA1 synapses) by electrically stimulating Schaffer collaterals. Basal synaptic transmission measured as the input–output relation between stimulation intensity and EPSCs in Df(16)5+/− mice was comparable to wild-type (WT) littermates [F(1,12) = 1.435, p = 0.231] (Fig. 1b). EPSC kinetics, such as rise time [t(19) = 0.442, p = 0.663], half-width (U = 51, p = 0.805), and decay time [t(19) = 0.0377, p = 0.970] were also normal between mutants and controls (Supplementary Fig. 1). However, an 80-Hz train (100 stimuli) applied to Schaffer collaterals evoked a substantially larger STP [t(16) = 2.196; p = 0.010] in Df(16)5+/− mutants than in WT mice (Fig. 1c). Because STP is not a unitary process but rather consists of several temporally and mechanistically distinct components (e.g., facilitation, augmentation, and depression of presynaptic transmission20), we measured STP components individually. The STP increase in Df(16)5+/− mutants could arise from the increased facilitation/augmentation, decreased short-term depression, or a combination thereof21. To differentiate among these possibilities, we used an established protocol to separate STP components29 and assess their individual contributions to the STP increase in Df(16)5+/− mice. First, using the paired-pulse ratios of two consecutive EPSCs, we found no significant difference [F(1,5) = 2.142, p = 0.146] in facilitation between Df(16)5+/− and WT mice (Fig. 1d). We then examined whether the increased STP in Df(16)5+/− mice resulted from reduced short-term depression. We examined the recovery from depression by using a 5-Hz stimulus train applied during the first 5 s after the 80-Hz train of Schaffer collateral stimuli. Because recovery from depression overlaps with the decay of augmentation during those 5 s, the actual EPSC peak amplitude reflects the net effect of the two processes21. To isolate the depression component, we corrected the synaptic responses for contribution from augmentation21, 29 (see Online Methods). This analysis revealed no significant difference [F(1,23) = 1.947; p = 0.182] in recovery from short-term depression between Df(16)5+/− and WT littermates (Fig. 1e). Next, we assessed the role of augmentation, which is the longest-lasting component of STP and operates on a time scale of tens of seconds29. Using a previously reported approach29, we isolated augmentation by applying a single stimulus to Schaffer collaterals every 5 s for 2 min, starting 5 s after the 80-Hz train (Fig. 1f). Augmentation was significantly increased [t(16) = 4.758, p = 0.0002] in Df(16)5+/− mice compared to WT controls (Fig. 1f). This increased augmentation in Df(16)5+/− mice was maximal at the onset and decayed to normal values after 20 s. Elevated augmentation in Df(16)5+/− mutants was not sensitive to the Serca inhibitor thapsigargin (4 µM) [without thapsigargin: t(16) = 4.404, p < 0.001; with thapsigargin: t(10) = 1.954, p = 0.0396] (Supplementary Fig. 2a). Further, Serca2 protein levels were normal (U = 107, p = 0.836) in Df(16)5+/− mice (Supplementary Fig. 2b), suggesting that haploinsufficiency of genes within the Df(16)5 genomic region resulted in abnormal STP through Serca2-independent mechanisms. To ensure that the STP increase in Df(16)5+/− mice originated from the presynaptic CA3 neurons, we examined the role of the postsynaptic component by performing two-photon glutamate uncaging (TGU) (100 TGU pulses, 80 Hz) to activate individual dendritic spines on CA1 neurons, the postsynaptic sites of CA3 inputs. Because TGU focally releases exogenous glutamate from inactive (caged) glutamate (MNI-glutamate), thereby bypassing the release of endogenous neurotransmitter from CA3 terminals, this method tests only postsynaptic mechanisms of synaptic transmission and plasticity. TGU-induced STP did not differ [t(25) = −0.405, p = 0.689] between Df(16)5+/− and WT littermates (Supplementary Fig. 3a), suggesting that the causative locus of the abnormal STP increase in Df(16)5 mutants resides in the presynaptic CA3 neurons. CA3 pyramidal neuron recordings revealed that presynaptic neurons had normal resting membrane potential [t(18) = −1.621, p = 0.122], input resistance [t(18) = 0.978, p = 0.341], and depolarization-induced action potentials (APs) [rheobase: t(18) = 0.451, p = 0.658 ; AP width: t(18) = 0.384, p = 0.706) (Supplementary Fig. 3b). An 80-Hz train of depolarization pulses delivered to the CA3 soma progressively increased the AP width, but to the same extent [100th AP width: t(12) = 0.845, p = 0.414] in both Df(16)5+/− and WT littermates (Supplementary Fig. 3c, d), thereby ruling out CA3 excitability as the culprit mechanism and implicating abnormal presynaptic glutamate release as a possible underlying mechanism of elevated STP in Df(16)5+/− mice. Mrpl40 haploinsufficiency causes abnormal STP in Df(16)5+/− mice To identify the culprit gene(s) whose haploinsufficiency causes abnormal augmentation and STP in Df(16)5+/− mice, we tested STP parameters in Cldn5+/− and Hira+/− mice27, 28 and the following new mutants that we generated: Cdc45l+/−, Ufd1l+/−, 2510002D24Rik+/−, and Mrpl40+/− mice (Supplementary Fig. 4). All transcripts were reduced by approximately 50% in the Df(16)5+/− mice [Cldn5+/−: p < 0.001; Cdc45l+/−: p = 0.03; Ufd1l+/−: p = 0.004; 2510002D24Rik+/−: p = 0.005; Mrpl40+/− : p = 0.010; Hira+/−: p = 0.003] and the respective individual mutants [Cldn5+/−: p < 0.001; Cdc45l+/: p < 0.001; Ufd1l+/−: p < 0.001; 2510002D24Rik+/−: p = 0.004; Mrpl40+/− : p < 0.001; Hira+/−: p = 0.03] (Supplementary Fig. 4). Testing STP in all six mouse mutants revealed that Mrpl40+/− mice had elevated STP and augmentation compared to their WT littermates [STP: t(29) = 2.368, p = 0.025; augmentation: t(29) = 3.150, p = 0.0037], whereas Cldn5+/−, Cdc45l+/−, Ufd1l+/−, 2510002D24Rik+/−, and Hira+/− mutants had normal STP and augmentation [STP and augmentation, respectively: Cldn5+/−: t(12) = 0.349, p = 0.733 and t(12) = −1.205, p = 0.250; Cdc45l+/−: t(18) = 0.566, p = 0.579 and t(18) = −0.637, p = 0.532; Ufd1l+/−: U = 43, p = 0.903 and U = 35, p = 0.438; 2510002D24Rik+/−: t(35) = 0.789, p = 0.436 and t(35) =1.773, p = 0.085; Hira+/−: t(11) = −0.644, p = 0.532 and t(11) = −0.687, p = 0.507] (Fig. 1g, h, Supplementary Fig. 5). Mrpl40 haploinsufficiency did not affect mRNA expression of other genes in the Df(16)5 microdeletion [Gapdh: U = 28, p = 0.871; Mrpl40: t(14) = 4.225, p < 0.001; Cldn5: t(14) = −0.0174, p = 0.986; Ufd1l: t(14) = −0.368, p = 0.719) (Supplementary Fig. 6). The full complement of Mrpl40 gene appeared to be essential for prenatal development as its homozygous deletion was embryonically lethal. The enhanced STP and augmentation in Mrpl40+/− mice were comparable to that in Df(16)5+/− mice, suggesting that hemizygous deletion of Mrpl40 underlies the abnormal synaptic plasticity in Df(16)5+/− mice. Df(16)5 deletion does not disrupt mitochondrial structure or oxidative phosphorylation Because Mrpl40 is thought to be one of the proteins of the mitoribosome complex34, 35, we investigated whether mitochondrial numbers, structure, or function were affected in Df(16)5+/− mice. We found no significant difference in mitochondrial ultrastructure imaged with TEM in the CA1 area of the hippocampus (Fig. 2a), in total mitochondrial DNA (Co1: U = 89, p = 0.147; Nd2: U = 101, p = 0.318) or in oxidative phosphorylation [F(1,8) = 0.108, p = 0.745] in isolated mitochondria from the hippocampus between Df(16)5+/− and WT littermates, suggesting normal energy production in mice with a Df(16)5-hemizygous deletion (Supplementary Fig. 7). Furthermore, 3D scanning electron microscopy (SEM) imaging of the hippocampal CA1 area revealed a normal distribution of mitochondria in presynaptic terminals of Df(16)5+/− mice compared to WT littermates (U = 34.5, p = 0.093) (Supplementary Fig. 8), suggesting normal trafficking of mitochondria to presynaptic terminals in Df(16)5+/− mice. Dysregulation of activity-dependent presynaptic and mitochondrial Ca2+ in Mrpl40+/− mice Because mitochondria also regulate presynaptic Ca2+ levels, we measured activity-dependent Ca2+ changes in response to the 80-Hz stimulation of Schaffer collaterals in presynaptic CA3 terminals in the hippocampal stratum radiatum (CA1 area). To this end, we took advantage of the highly sensitive genetically encoded Ca2+ indicator GCaMP6f (GCaMP6)25. After infecting the CA3 area of the hippocampus in vivo with recombinant adeno-associated viruses (AAVs) encoding mCherry and GCaMP6, we observed high expression of fluorescent proteins in neuronal cell bodies in the CA3 area but not in the CA1 area (Fig. 2b). In the stratum radiatum we observed fluorescent boutons, which responded to 80-Hz electrical stimulation (10 pulses) of Schaffer collaterals with GCaMP6 fluorescence transients with fast rise and decay (rise time20%-80%, 100.46 ± 4.03 ms; decay time (τ), 222.71 ± 9.40 ms) (Fig. 2c). These activity-dependent kinetics of GCaMP6 fluorescence were similar to those observed in CA3 presynaptic terminals in which the inorganic Ca2+ indicator Fluo 5F was used (data not shown), suggesting that GCaMP6 reliably measures cytosolic Ca2+ in presynaptic terminals. The activity-dependent increase in GCaMP6 fluorescence during the 80-Hz train of stimulations was substantially higher (U = 537, p < 0.001) in Df(16)5+/− mice than in their WT littermates (Fig. 2d), which is consistent with the notion that higher presynaptic Ca2+ levels lead to elevated augmentation and STP20. To directly measure Ca2+ in the CA1 stratum radiatum mitochondria localized to the presynaptic terminals originating from CA3, we expressed GCaMP6 with a mitochondrial-localization signal (mitoGCaMP6) in the CA3 area using recombinant AAVs (Fig. 2b). We verified the specific localization of mitoGCaMP6 to mitochondria using subcellular fractionation followed by Western blotting and co-immunolocalization with mitochondrial markers (Supplementary Fig. 9). We also verified that the activity-dependent increase in mitoGCaMP6 fluorescence was sensitive to an inhibitor of the mitochondrial Ca2+ uniporter Ru360 (10 µM) (U = 15, p < 0.001) (Supplementary Fig. 10). An 80-Hz train (10 pulses) applied to the Schaffer collaterals induced an activity-dependent increase in mitoGCaMP6 but with substantially slower kinetics compared to cytosolic GCaMP6 (mitoGCaMP6: rise time20%-80%, 210.21 ± 27.03 ms, decay time > 1 s; GCaMP6: rise time20%-80%, 100.46 ± 4.03 ms; decay time (τ), 222.71 ± 9.40 ms; U = 540, p < 0.001) (Fig. 2e). Similar to cytosolic Ca2+, mitochondrial Ca2+ was elevated (U = 889, p = 0.020) in Df(16)5+/− mice in response to the 80-Hz synaptic stimulation (Fig. 2e). The increases in cytosolic and mitochondrial Ca2+ were also observed in Mrpl40+/− mice to a similar degree as in Df(16)5+/− mice (GCaMP6: U = 482, p < 0.001; mitoGCaMP6: U = 618, p < 0.001) (Fig. 2f–i). These data suggest that Mrpl40 is the gene in the Df(16)5 genomic region that is responsible for the STP phenotype by deregulating activity-dependent mitochondrial and cytoplasmic presynaptic Ca2+ dynamics. Mrpl40+/− mice are deficient in short-term but not long-term spatial memory or long-term synaptic plasticity To test if the hemizygous Mrpl40 deletion affects cognitive function, we compared the performance of Mrpl40+/− and WT mice in several behavioral tests. Mrpl40+/− mice behaved normally in the acoustic startle [F(1,7) = 0.0711, p = 0.79] and pre-pulse inhibition [F(1,2) = 1.314, p = 0.274] of acoustic startle tests (Fig. 3a, b), a measure of sensorimotor gating that is believed to be associated with positive symptoms of SCZ31. To test spatial working memory, we used a delayed, non-matched-to-position task (DNMPT), in which timing between runs ranged from 0 to 5 s. In this test Mrpl40+/− mice performed significantly worse [t(28) = 3.3, p = 0.003] than WT littermates (Fig. 3c). However, Mrpl40+/− mice performed normally in the tasks that assessed long-term memory (e.g., Morris Water Maze task). In this task, mutant mice learned to find the invisible escape platform (Fig. 3d) and retained this spatial memory for 48 h similar to WT controls [F(1,3) = 0.0425, p = 0.837, Fig. 3e]. Mrpl40+/− mice also performed comparably to WT controls [day 1: t(33) = −0.457, p = 0.651; day 2: U = 149.5, p = 0.921] when the escape platform was visible (Fig. 3f). Mrpl40+/− mutants also showed no memory deficits in the Y-maze [t(14) = 0.160, p = 0.873], where we measured the amount of time a mouse spends in a novel arm 1 h after exploring the other two arms of the maze (Fig. 3g). These data suggest that Mrpl40 haploinsufficiency affects short-term (working) memory but not long-term memory. Consistent with this notion, STP was abnormal in Mrpl40+/− mice (Fig. 1g, h), but long-term potentiation (LTP) of excitatory synaptic transmission, a major form of long-term synaptic plasticity at CA3–CA1 synapses, did not differ (t(53) = 0.161, p = 0.873) between Mrpl40+/− mice and WT littermates (Fig. 3h). Mitochondrial Ca2+–extrusion deficit underlies STP and Ca2+ phenotypes in Mrpl40+/− mice The abnormally high increase in cytoplasmic Ca2+ induced by the 80-Hz synaptic stimulation in Mrpl40+/− mice coincided with the enhanced mitochondrial Ca2+ increase. A role for slow mitochondrial Ca2+ extrusion in STP has been implicated in crayfish neuromuscular junction36 and led us to hypothesize that our results in the hippocampus could also be explained by impaired Ca2+ extrusion from mitochondria. Two major mechanisms extrude Ca2+ from the mitochondrial matrix to the cytoplasm: mitochondrial Ca2+ exchangers and the mPTP37, 38. The selective antagonist of the mitochondrial Na+–Ca2+ exchanger, CGP 37157 (5 µM), had no effect on STP or augmentation in WT mice [STP: t(18) = 0.999, p = 0.333; augmentation: t(18) = 1.007, p = 0.330], suggesting that this Ca2+ extrusion mechanism is not required for Ca2+ handling by mitochondria during 80-Hz–induced synaptic plasticity (Supplementary Fig. 11). However, bongkrekic acid (BKA, 2 µM), a non-selective inhibitor of the adenine nucleotide (ADP/ATP) translocases (ANTs)39, which are required for sensitivity of mPTP to calcium40, significantly increased STP and augmentation [STP: U = 14, p = 0.005; augmentation: t(19) = −3.169, p = 0.005] in WT mice (Fig. 4a, Supplementary Fig. 12a,c). This increase mimicked the STP and augmentation enhancement in Df(16)5+/− and Mrpl40+/− mice. Interestingly, BKA did not increase STP or augmentation further [STP: t(19) = −0.107, p = 0.916; augmentation: t(19) = −0.928, p = 0.365] in Df(16)5+/− mice (Fig. 4a, Supplementary Fig. 12b,d). BKA also did not affect basal synaptic transmission [F(3,12) = 1.179, p = 0.318], paired-pulse facilitation [F(3,5) = 0.402, p = 0.752], or recovery from depression [F(3,23) = 1.114, p = 0.358] in either WT or Df(16)5+/− mice (Supplementary Fig. 13), indicating that the BKA effect is specific for the augmentation component of STP. Furthermore, BKA significantly enhanced the magnitudes of cytosolic and mitochondrial Ca2+ transients evoked by the 80-Hz train in WT mice (GCaMP6: Q = 2.821, p < 0.05; mitoGCaMP6: Q = 4.908, p < 0.05) (Fig. 4b,c, Supplementary Fig. 12e,g) but failed to increase Ca2+ transients in Df(16)5+/− mice (GCaMP6: Q = 0.246, p > 0.05; mitoGCaMP6: Q = 1.767, p > 0.05) (Fig. 4b,c, Supplementary Fig. 12f,h), suggesting that the STP enhancement in Df(16)5+/− mutants acts through the same mechanisms as BKA. Like Df(16)5+/− mice, Mrpl40+/− mice showed no effect of BKA on STP, augmentation, or Ca2+ transients [STP: U = 52, p = 0.371; augmentation: t(23) = −1.091, p = 0.287; GCaMP6: Q = 0.418, p > 0.05; mitoGCaMP6: Q = 0.052, p > 0.05] in presynaptic terminals or mitochondria, whereas BKA enhanced these parameters in WT littermates [STP: t(13) = −3.685, p = 0.0028; augmentation: t(13) = −0.2670, p = 0.02; GCaMP6: Q = 5.038, p < 0.05; mitoGCaMP6: Q = 2.645, p < 0.05] (Fig. 4d–f, Supplementary Fig. 14). These results suggest that haploinsufficiency of the Df(16)5 gene Mrpl40 impairs Ca2+ handling by mPTP. Therefore, we sought to rescue this deficit by enhancing mPTP function through overexpression of ANTs. To that end, we designed the AAV-Slc25a4 OE (Fig. 5), which overexpresses Ant1 (also known as Slc25a4), a gene encoding a regulator of mPTP activity40, 41. Only three isoforms of Ant (Ant1, 2, and 4) have been found in mice42, 43, and we identified Ant1 as the highest expressed isoform in the mouse hippocampus (data not shown). When expressed in the hippocampus, AAV-Slc25a4 OE increased the level of Slc25a4 protein [t(7) = 7.868, p < 0.001] (Supplementary Fig. 15a, b). We verified that AAV-Slc25a4 OE did not change the localization of Slc25a4 protein to mitochondria. Thus, Slc25a4 was co-localized with the mitochondrial fluorescent marker mitotracker (Supplementary Fig. 15c), and AAV-Slc25a4 OE did not change this pattern (Supplemental Fig. 15d). AAV-Slc25a4 OE expressed in the CA3 area of the hippocampus (Fig. 5a) rescued STP and augmentation in Mrpl40+/− mice [STP: U = 15, p = 0.005; augmentation: t(22) = 2.543, p = 0.019] but did not affect WT littermates [STP: t(16) = 0.566, p = 0.578 ; augmentation: t(16) = −0.654, p = 0.522] (Fig 5b, Supplementary Fig. 16). To confirm that the BKA-induced reduction in mPTP function mimics the Df(16)5+/− phenotype, we downregulated Slc25a4 by using the shRNA approach. To that end, we injected lentiviruses encoding three different shRNAs against Slc25a4 into the CA3 area of the hippocampus. STP and augmentation of synaptic transmission at CA3–CA1 synapses increased in WT mice infected with all three Slc25a4 shRNAs compared to the control shRNA [shRNA1 STP: t(13) = −1.882, p = 0.041; augmentation: U = 9, p = 0.029; shRNA2 STP: t(15) = −1.910, p = 0.04; augmentation: t(17) = −2.757, p = 0.014; shRNA3 STP:U = 9, p = 0.02; augmentation: t(13) = −2.475, p = 0.028] (Fig. 5c, Supplementary Fig. 17). However, Slc25a4 shRNAs did not further affect STP or augmentation in Mrpl40+/− mice [shRNA1 STP: U = 23, p = 0.382; augmentation: U = 28, p = 0.721; shRNA2 STP: U = 23, p = 0.091; augmentation: t(19) = 1.038, p = 0.312; shRNA3 STP U = 27, p = 0.267; augmentation: U = 24, p = 0.107] (Fig. 5c, Supplementary Fig. 17), suggesting that Mrpl40 haploinsufficiency altered STP by affecting mPTP function. Discussion Approximately 30% of patients with 22q11DS meet the diagnostic criteria for SCZ4. Cognitive symptoms of SCZ include deficits in working memory, attention, executive function, and learning and memory; these symptoms have a more prognostic value than do the positive or negative symptoms of the disease, and they contribute more to the patients’ functional disability44. The mechanisms underlying the cognitive deficits in SCZ and 22q11DS are still debated. Here we presented evidence that 22q11DS affects working memory and STP through a novel pathogenic mechanism––abnormal Ca2+ handling by mitochondria. We also elucidated several features of this pathogenic mechanism: (1) By screening mice carrying hemizygous deletions of individual genes mapped within the distal part of the 22q11 microdeletion, we identified Mrpl40 as a culprit gene that causes abnormal STP. (2) We identified that the augmentation component of STP is specifically affected. (3) Mrpl40 haploinsufficiency led to abnormal STP through mitochondria-mediated deregulation of presynaptic Ca2+ levels. (4) We pinpointed a functional abnormality in mitochondrial Ca2+ extrusion through the mPTP as a pathogenic mechanism caused by Mrpl40 haploinsufficiency. (5) Mrpl40 haploinsufficiency led to deficits in working memory but not in long-term plasticity or long-term memory. The hippocampus, prefrontal cortex, and interactions between these brain regions all contribute to working memory45–48. Deficits in working memory in patients with SCZ and/or 22q11DS are well documented49, 50. STP, an associative, short-lived synaptic strengthening, is considered a cellular correlate of short-term memory, a term often used synonymously with working memory20, 51. Consistent with this notion, deficits in STP are well established in animal models of SCZ and 22q11DS12, 48, 52–54. Because symptoms of SCZ typically appear during late adolescence or early adulthood, age appears to be an important variable in synaptic plasticity phenotypes associated with 22q11DS mice. In our experiments, we used mice that were older than 4 months. At this age, STP and LTP at CA3–CA1 synapses are substantially increased in Df(16)1+/− mouse models of 22q11DS, whereas at younger ages, both forms of synaptic plasticity are normal23. These experiments led to the identification of the microRNA-processing gene Dgcr8 as the culprit gene residing in the proximal part of the microdeletion affecting both short- and long-term synaptic plasticity in the hippocampus. Deletion of one allele of Dgcr8 causes depletion of miR-185 and miR-25 and posttranscriptional upregulation of Serca2 in the forebrain of older but not younger mice23. Because 22q11DS is a multigene syndrome, it is extremely likely that more than one gene is involved in STP abnormalities. Our present work showed that deletion of one allele of Mrpl40 residing in the distal part of the microdeletion, which is outside of the Df(16)1 genomic region caused independent deficits in STP and working memory in 22q11DS. First, we identified the STP increase in Df(16)5+/− mice, which were hemizygous for genes that mapped distally to Dgcr8. The STP screen revealed that only Mrpl40+/− mice replicated the Df(16)5+/− phenotype. Furthermore, the Mrpl40-related STP increase was mechanistically distinct from the STP increase identified in Dgcr8+/− mice. Dgcr8 haploinsufficiency elevates Serca2 protein12, but in the Df(16)5+/− mice, Serca2 levels were normal, and the Serca inhibitor thapsigragin did not rescue the STP defect. Mrpl40 was originally identified as a nuclear gene involved in mitochondrial function55. Mitochondrial dysfunction has been strongly implicated in SCZ pathophysiology56, though the exact connection between SCZ pathogenesis and mitochondria has not been established. For instance, the gene encoding DISC1 (disrupted-in-schizophrenia 1) has been localized to mitochondria and is involved in maintaining mitochondrial morphology and regulating mitochondrial transport57. Several mitochondrial genes, including Mrpl40, are mapped within the 22q11.2 locus and expressed in the brain throughout development, thus implicating mitochondria in the pathogenesis of 22q11DS58, 59. Mrpl40 haploinsufficiency does not affect the mitochondrial ultrastructure and does not reduce mitochondrial numbers in presynaptic terminals or total mitochondrial DNA, which is consistent with the view that SCZ is not a neurodegenerative disease. This is also consistent with our observations that Df(16)5+/− or Mrpl40+/− mice develop normally and have normal gross brain morphology. Mitochondria provide energy through oxidative phosphorylation and regulate Ca2+ dynamics during synaptic transmission in neurons. Because of the mitochondrion’s high capacity for Ca2+ uptake, it acts as a rapid buffering system during periods of intense synaptic activity, then slowly releases Ca2+ when activity subsides. This rapid uptake of Ca2+ into mitochondria occurs through channels (i.e. voltage-dependent anion channel VDAC1) in the outer membrane and the mitochondrial calcium uniporter in the inner membrane. Ca2+ is slowly released from the mitochondria through the calcium exchangers and via the mPTP. mPTP has been shown to open during high frequency stimulation in mammalian neurons, and that transient activation of the mPTP may play a role in the normal contribution of mitochondria to STP60. Oxidative phosphorylation appeared to be normal, whereas Ca2+ dynamics in presynaptic terminals was substantially altered in Df(16)5+/− and Mrpl40+/− mice. Our two-photon Ca2+ imaging in presynaptic terminals revealed that synaptic high frequency stimulation that induces STP evokes enhanced Ca2+ transients in both the presynaptic mitochondrial matrix and presynaptic cytosol of mutant mice. This data argues for a problem with mitochondrial Ca2+ extrusion rather than with mitochondrial Ca2+ uptake. Indeed, if Mrpl40 haploinsufficiency reduced the Ca2+ uptake into mitochondria, we would expect to see increased amplitudes of cytoplasmic Ca2+ transients but decreased Ca2+ transients within mitochondria. Our data is more consistent with a model for a reduced Ca2+ extrusion from mitochondria. Impaired Ca2+ extrusion from mitochondria will result in Ca2+ accumulation in the mitochondrial matrix and this will contribute to the fast rise in mitochondrial Ca2+ that we observed with mitoGCaMP6. This will then reduce the effective mitochondrial buffering capacity and lead to enhanced cytoplasmic Ca2+ transients that we observed with the cytoplasmic GCaMP6 (Fig. 5d). Moreover, slow Ca2+ extrusion from mitochondria (evident from extremely slow decay times of mitoGCaMP6) is an unlikely contributor to fast cytosol Ca2+ transients that occur during high frequency synaptic activity. Because inhibition of mPTP but not the mitochondrial Na+–Ca2+ exchanger increased presynaptic Ca2+ transients in WT mice, we concluded that the main route of Ca2+ extrusion from mitochondria during STP is mPTP. In WT mice, the pharmacological or molecular inhibition of mPTP with BKA or Slc25a4 shRNA, respectively, mimicked the STP and Ca2+-transient phenotypes we observed in Df(16)5+/− and Mrpl40+/− mice. Moreover, AAV-Slc25a4 OE rescued the STP abnormality in Mrpl40+/− mice, indicating that deletion of one allele of Mrpl40 causes mPTP deficiency and STP increase. Although our experiments demonstrated that Mrpl40 haploinsufficiency leads to abnormal Ca2+ handling by mitochondria during STP, the exact connection between Mrpl40 and mPTP function remains unclear. Some pharmacological agents that affect mPTP function61 can be beneficial adjuvants to antipsychotics to alleviate SCZ symptoms56. However, because mPTP8 is involved not only in Ca2+ extrusion from mitochondria but also in ATP transport, we cannot rule out the possibility that impaired mPTP regulation leads to the abnormal production of ATP in presynaptic terminals, which in turn could result in abnormal presynaptic function. This theory could be addressed using a novel ATP probe that visualizes ATP dynamics in the presynaptic terminals of neurons in culture, but those experiments have not yet been tailored for acute brain slices62. In summary, we report that haploinsufficiency of Mrpl40, a gene mapped within the 22q11.2 microdeletion associated with SCZ, causes deficient working memory. This behavioral abnormality, which typically manifests in patients with SCZ, is associated with abnormal STP changes in synaptic transmission in the hippocampus and caused by elevated presynaptic Ca2+ and impaired Ca2+ extrusion from mitochondria. Supplementary Material 1 This work was supported, in part, by National Institutes of Health grants R01 MH095810 and R01 MH097742 and by ALSAC. We thank Drs. Elizabeth Illingworth, Peter Scambler, and Mikio Furuse for providing Df(16)5+/−, Hira+/−, and Cldn5+/− mice; Sharon Frase, Randall Wakefield and staff in the Electron Microscopy Division of the St. Jude Cell and Tissue Imaging Center for help with electron microscopy; John Swift, Amber Braden, Lisa Emmons, and staff in the St. Jude Transgenic Core for help in producing mouse mutants; and staff in the St. Jude and University of Tennessee Vector Cores for producing the AAVs and lentiviruses. We also thank Jay Blundon for valuable comments and Angela McArthur and Vani Shanker for editing the manuscript. Figure 1 Abnormal synaptic plasticity in Df(16)5+/− mice is caused by Mrpl40 haploinsufficiency (a) Diagram depicting genes in the 22.q11.2 genomic region of the human chromosome 22 and the syntenic region of mouse chromosome 16. Red horizontal bar represents genomic regions hemizygously deleted in Df(16)5+/− mice and grey horizontal bar represents genomic regions hemizygously deleted in Df(16)1+/− mice. Note that 2510002D24Rik, Mrpl40, and Hira genes are mapped outside the Df(16)1 microdeletion. (b) Input–output relations in Df(16)5+/− and WT littermates. (c) STP (comprising facilitation, depression, and augmentation) induced by the high-frequency (80-Hz) train. The first time point represents an average of 5 baseline EPSCs delivered at low frequency. The top inset shows the protocol for measuring STP, recovery from depression, and augmentation in the same experiment. (d-f) Average facilitation tested by paired-pulse ratio in separate experiments (d), recovery from depression tested 5 s after the 80-Hz train (e), and augmentation tested 5 to 120 s after the 80-Hz train in Df(16)5+/− and WT mice (f). Insets show representative EPSC traces. (g,h) Mean STP of EPSCs induced by the 80-Hz train of synaptic stimulation of Schaffer collaterals (g) and augmentation (h) in Mrpl40+/− mice and their WT littermates. Numbers of neurons are shown in parentheses or inside columns. Data are represented as mean ± SEM. *p <0.05. Figure 2 Normal mitochondrial structure but abnormal presynaptic cytosolic and mitochondrial calcium regulation in Df(16)5+/− and Mrpl40+/− mice (a) Three representative TEM images of mitochondrial ultrastructure in the CA1 area of the hippocampus of WT and Df(16)5+/− mice. (b) Representative fluorescent image of mCherry after infection of the CA3 area with AAVs encoding either GCaMP6 or mitoGCaMP6. (c) Line scan of mCherry and GCaMP6 fluorescence in a CA3 presynaptic terminal before and after the 80-Hz stimulation of Schaffer collaterals (arrow). (d-g) Mean normalized cytosolic GCaMP6 (d, f) and mitoGCaMP6 (e, g) fluorescence in CA3 presynaptic terminals imaged in the CA1 area of the hippocampus, before and after 80-Hz stimulation in Df(16)5+/− and WT littermates (d, e) and Mrpl40+/− and WT littermates (f, g). (h, i) Normalized mean peak amplitudes of GCaMP6 (h) or mitoGCaMP6 (i) in Df(16)5+/− and WT littermates and Mrpl40+/− and WT littermates. Numbers of fluorescent puncta are shown inside columns. *P <0.05. Figure 3 Deficient working memory and normal long-term spatial memory, long-term synaptic plasticity, and acoustic startle in Mrpl40+/− mice (a, b) Mean maximal startle responses as a function of sound intensity (a) and prepulse inhibition (PPI) (b) in Mrpl40+/− and WT littermates. (c) The mean numbers of correct choices made in the delayed non–matched-to-position task by Mrpl40+/− and WT littermates. (d-f) Morris water maze tasks. Average time to find a submerged platform during the learning phase (d), time spent in quadrants during the probe test performed 24 h after the last learning session (e), and time to travel to a visible platform (f) in Mrpl40+/− and WT littermates. Abbreviations: AdL, adjacent left quadrant; AdR, adjacent right quadrant; Opp, opposite quadrant; Tg, target quadrant. (g) Average percentage of time spent in the novel arm in the novel-recognition version of the Y-maze task by Mrpl40+/− and WT mice. (h) Long-term potentiation at CA3–CA1 synapses measured as mean field EPSP (fEPSP) as a function of time before and after tetanization of Schaffer collaterals with 200-Hz trains (arrows) in Mrpl40+/− and WT mice. Dashed lines in c and g indicate the level of performance expected by chance. Numbers of mice or slices are shown in parentheses or inside columns. *P <0.05. Figure 4 The mPTP inhibitor BKA mimics the STP and calcium transient phenotypes of Df(16)5+/− and Mrpl40+/− mice (a-f) BKA effect on augmentation (a,d), peak GCaMP6 (b,e), and peak mitoGCaMP6 fluorescence intensities (c,f) in WT and Df(16)5+/− littermates (a-c) or WT and Mrpl40+/− littermates (normalized to WT levels) (d-f). Numbers of neurons or fluorescent puncta are shown inside columns. *P <0.05. Figure 5 The mPTP gain- or loss-of-function molecular manipulations rescue or mimic, respectively, the STP phenotypes of Mrpl40+/− mice (a) Overexpression of Slc25a4 and GFP in the CA3 area of the hippocampus. (b) Mean augmentation measured in sham- or Slc25a4-OE–injected WT and Mrpl40+/− mice. (c) Mean augmentation measured in control or Slc25a4 shRNA–injected WT and Mrpl40+/− mice. The data are shown for three different Slc25a4 shRNAs (shRNA1, shRNA2, shRNA3). Normalized to respective WT control levels. Numbers of neurons are shown inside columns. *P <0.05. (d) Model of mPTP-dependent mechanisms of STP dysfunction in 22q11DS. MRPL40 haploinsufficiency in 22q11DS reduces mitochondrial Ca2+ extrusion through impaired mPTP. This leads to a Ca2+ build-up in the mitochondrial matrix and enhanced Ca2+ transients in the mitochondrial matrix and cytosol during high-frequency activity, which in turn leads to enhanced synaptic vesicle release in presynaptic terminals. VGCC, voltage-gated calcium channels, NCX, Na+/Ca2+ exchanger, MCU, mitochondrial Ca2+ uniporter, VDAC1, voltage-dependent anion channel 1. Upper traces represent cytosolic Ca2+ transients (GCaMP6) and lower traces represent mitochondrial Ca2+ transients (mitoGCaMP6). Conflict of Interest The authors declare no conflict of interest. References 1 Carpenter WT Jr Buchanan RW Schizophrenia N. Engl. J. Med 1994 330 681 690 8107719 2 Nestler EJ Hyman SE Animal models of neuropsychiatric disorders Nat. Neurosci 2010 13 1161 1169 20877280 3 Scambler PJ The 22q11 deletion syndromes Hum. Mol. Genet 2000 9 2421 2426 11005797 4 Pulver AE Nestadt G Goldberg R Shprintzen RJ Lamacz M Wolyniec PS Psychotic illness in patients diagnosed with velo-cardio-facial syndrome and their relatives J. Nerv. Ment. Dis 1994 182 476 478 8040660 5 Chow EW Watson M Young DA Bassett AS Neurocognitive profile in 22q11 deletion syndrome and schizophrenia Schizophr. Res 2006 87 270 278 16753283 6 Lewis DA Cortical circuit dysfunction and cognitive deficits in schizophrenia--implications for preemptive interventions Eur. J. Neurosci 2012 35 1871 1878 22708598 7 Vorstman JA Breetvelt EJ Duijff SN Eliez S Schneider M Jalbrzikowski M Cognitive decline preceding the onset of psychosis in patients with 22q11.2 deletion syndrome JAMA Psychiatry 2015 72 377 385 25715178 8 Heckers S Rauch SL Goff D Savage CR Schacter DL Fischman AJ Impaired recruitment of the hippocampus during conscious recollection in schizophrenia Nat. Neurosci 1998 1 318 323 10195166 9 Tamminga CA Stan AD Wagner AD The hippocampal formation in schizophrenia Am. J Psychiatry 2010 167 1178 1193 20810471 10 Kates WR Krauss BR Abdulsabur N Colgan D Antshel KM Higgins AM The neural correlates of non-spatial working memory in velocardiofacial syndrome (22q11.2 deletion syndrome) Neuropsychologia 2007 45 2863 2873 17618656 11 Sobin C Kiley-Brabeck K Daniels S Blundell M Anyane-Yeboa K Karayiorgou M Networks of attention in children with the 22q11 deletion syndrome Dev. Neuropsychol 2004 26 611 626 15456687 12 Earls LR Bayazitov IT Fricke RG Berry RB Illingworth E Mittleman G Dysregulation of presynaptic calcium and synaptic plasticity in a mouse model of 22q11 deletion syndrome J. Neurosci 2010 30 15843 15855 21106823 13 Stark KL Xu B Bagchi A Lai WS Liu H Hsu R Altered brain microRNA biogenesis contributes to phenotypic deficits in a 22q11-deletion mouse model Nat. Genet 2008 40 751 760 18469815 14 Abbott LF Regehr WG Synaptic computation Nature 2004 431 796 803 15483601 15 Klyachko VA Stevens CF Excitatory and feed-forward inhibitory hippocampal synapses work synergistically as an adaptive filter of natural spike trains PLoS. Biol 2006 4 e207 16774451 16 Kandaswamy U Deng PY Stevens CF Klyachko VA The role of presynaptic dynamics in processing of natural spike trains in hippocampal synapses J. Neurosci 2010 30 15904 15914 21106829 17 Rotman Z Deng PY Klyachko VA Short-term plasticity optimizes synaptic information transmission J. Neurosci 2011 31 14800 14809 21994397 18 Mongillo G Barak O Tsodyks M Synaptic theory of working memory Science 2008 319 1543 1546 18339943 19 Deco G Rolls ET Romo R Synaptic dynamics and decision making Proc. Natl. Acad. Sci. U. S. A 2010 107 7545 7549 20360555 20 Zucker RS Regehr WG Short-term synaptic plasticity Annu. Rev. Physiol 2002 64 355 405 11826273 21 Deng PY Sojka D Klyachko VA Abnormal presynaptic short-term plasticity and information processing in a mouse model of fragile X syndrome J Neurosci 2011 31 10971 10982 21795546 22 Tu H Nelson O Bezprozvanny A Wang Z Lee SF Hao YH Presenilins form ER Ca2+ leak channels, a function disrupted by familial Alzheimer’s disease-linked mutations Cell 2006 126 981 993 16959576 23 Earls LR Fricke RG Yu J Berry RB Baldwin LT Zakharenko SS Age-dependent microRNA control of synaptic plasticity in 22q11 deletion syndrome and schizophrenia J. Neurosci 2012 32 14132 14144 23055483 24 Schizophrenia Working Group of the Psychiatric Genomics Consortium Biological insights from 108 schizophrenia-associated genetic loci Nature 2014 511 421 427 25056061 25 Chen TW Wardill TJ Sun Y Pulver SR Renninger SL Baohan A Ultrasensitive fluorescent proteins for imaging neuronal activity Nature 2013 499 295 300 23868258 26 Lindsay EA Vitelli F Su H Morishima M Huynh T Pramparo T Tbx1 haploinsufficieny in the DiGeorge syndrome region causes aortic arch defects in mice Nature 2001 410 97 101 11242049 27 Nitta T Hata M Gotoh S Seo Y Sasaki H Hashimoto N Size-selective loosening of the blood-brain barrier in claudin-5-deficient mice J Cell Biol 2003 161 653 660 12743111 28 Roberts C Sutherland HF Farmer H Kimber W Halford S Carey A Targeted mutagenesis of the Hira gene results in gastrulation defects and patterning abnormalities of mesoendodermal derivatives prior to early embryonic lethality Mol. Cell Biol 2002 22 2318 2328 11884616 29 Klyachko VA Stevens CF Temperature-dependent shift of balance among the components of short-term plasticity in hippocampal synapses J. Neurosci 2006 26 6945 6957 16807324 30 Christensen M Larsen LA Kauppinen S Schratt G Recombinant Adeno-Associated Virus-Mediated microRNA Delivery into the Postnatal Mouse Brain Reveals a Role for miR-134 in Dendritogenesis in Vivo Front Neural Circuits 2010 3 16 20126250 31 Chun S Westmoreland JJ Bayazitov IT Eddins D Pani AK Smeyne RJ Specific disruption of thalamic inputs to the auditory cortex in schizophrenia models Science 2014 344 1178 1182 24904170 32 Denk W Horstmann H Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure PLoS. Biol 2004 2 e329 15514700 33 Paylor R Glaser B Mupo A Ataliotis P Spencer C Sobotka A Tbx1 haploinsufficiency is linked to behavioral disorders in mice and humans: implications for 22q11 deletion syndrome Proc. Natl. Acad. Sci. U. S. A 2006 103 7729 7734 16684884 34 Greber BJ Boehringer D Leibundgut M Bieri P Leitner A Schmitz N The complete structure of the large subunit of the mammalian mitochondrial ribosome Nature 2014 515 283 286 25271403 35 Greber BJ Boehringer D Leitner A Bieri P Voigts-Hoffmann F Erzberger JP Architecture of the large subunit of the mammalian mitochondrial ribosome Nature 2014 505 515 519 24362565 36 Tang Y Zucker RS Mitochondrial involvement in post-tetanic potentiation of synaptic transmission Neuron 1997 18 483 491 9115741 37 Bernardi P von SS The permeability transition pore as a Ca(2+) release channel: new answers to an old question Cell Calcium 2012 52 22 27 22513364 38 Rizzuto R De SD Raffaello A Mammucari C Mitochondria as sensors and regulators of calcium signalling Nat. Rev. Mol. Cell Biol 2012 13 566 578 22850819 39 Halestrap AP What is the mitochondrial permeability transition pore? J. Mol. Cell Cardiol 2009 46 821 831 19265700 40 Kokoszka JE Waymire KG Levy SE Sligh JE Cai J Jones DP The ADP/ATP translocator is not essential for the mitochondrial permeability transition pore Nature 2004 427 461 465 14749836 41 Bonora M Pinton P The mitochondrial permeability transition pore and cancer: molecular mechanisms involved in cell death Front Oncol 2014 4 302 25478322 42 Brower JV Rodic N Seki T Jorgensen M Fliess N Yachnis AT Evolutionarily conserved mammalian adenine nucleotide translocase 4 is essential for spermatogenesis J. Biol. Chem 2007 282 29658 29666 17681941 43 Levy SE Chen YS Graham BH Wallace DC Expression and sequence analysis of the mouse adenine nucleotide translocase 1 and 2 genes Gene 2000 254 57 66 10974536 44 Carter CS Barch DM Buchanan RW Bullmore E Krystal JH Cohen J Identifying cognitive mechanisms targeted for treatment development in schizophrenia: an overview of the first meeting of the Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia Initiative Biol. Psychiatry 2008 64 4 10 18466880 45 Goldman-Rakic PS Architecture of the prefrontal cortex and the central executive Ann. N. Y. Acad. Sci 1995 769 71 83 8595045 46 Sanderson DJ Good MA Seeburg PH Sprengel R Rawlins JN Bannerman DM The role of the GluR-A (GluR1) AMPA receptor subunit in learning and memory Prog. Brain Res 2008 169 159 178 18394473 47 Floresco SB Seamans JK Phillips AG Selective roles for hippocampal, prefrontal cortical, and ventral striatal circuits in radial-arm maze tasks with or without a delay J. Neurosci 1997 17 1880 1890 9030646 48 Sigurdsson T Stark KL Karayiorgou M Gogos JA Gordon JA Impaired hippocampal-prefrontal synchrony in a genetic mouse model of schizophrenia Nature 2010 464 763 767 20360742 49 Wong LM Riggins T Harvey D Cabaral M Simon TJ Children with chromosome 22q11.2 deletion syndrome exhibit impaired spatial working memory Am. J. Intellect. Dev. Disabil 2014 119 115 132 24679349 50 Bearden CE Woodin MF Wang PP Moss E Donald-McGinn D Zackai E The neurocognitive phenotype of the 22q11.2 deletion syndrome: selective deficit in visual-spatial memory J. Clin. Exp. Neuropsychol 2001 23 447 464 11780945 51 Lisman J The challenge of understanding the brain: where we stand in 2015 Neuron 2015 86 864 882 25996132 52 Crabtree GW Gogos JA Synaptic plasticity, neural circuits, and the emerging role of altered short-term information processing in schizophrenia Front Synaptic. Neurosci 2014 6 28 25505409 53 Kvajo M McKellar H Arguello PA Drew LJ Moore H MacDermott AB A mutation in mouse Disc1 that models a schizophrenia risk allele leads to specific alterations in neuronal architecture and cognition Proc. Natl. Acad. Sci. U. S. A 2008 105 7076 7081 18458327 54 Fenelon K Mukai J Xu B Hsu PK Drew LJ Karayiorgou M Deficiency of Dgcr8, a gene disrupted by the 22q11.2 microdeletion, results in altered short-term plasticity in the prefrontal cortex Proc. Natl. Acad. Sci. U. S. A 2011 108 4447 4452 21368174 55 Funke B Puech A Saint-Jore B Pandita R Skoultchi A Morrow B Isolation and characterization of a human gene containing a nuclear localization signal from the critical region for velo-cardio-facial syndrome on 22q11 Genomics 1998 53 146 154 9790763 56 Manji H Kato T Di Prospero NA Ness S Beal MF Krams M Impaired mitochondrial function in psychiatric disorders Nat. Rev. Neurosci 2012 13 293 307 22510887 57 Atkin TA MacAskill AF Brandon NJ Kittler JT Disrupted in Schizophrenia-1 regulates intracellular trafficking of mitochondria in neurons Mol. Psychiatry 2011 16 122 124 121 21079610 58 Napoli E Tassone F Wong S Angkustsiri K Simon TJ Song G Mitochondrial Citrate Transporter-dependent Metabolic Signature in the 22q11.2 Deletion Syndrome J. Biol. Chem 2015 290 23240 23253 26221035 59 Maynard TM Meechan DW Dudevoir ML Gopalakrishna D Peters AZ Heindel CC Mitochondrial localization and function of a subset of 22q11 deletion syndrome candidate genes Mol. Cell Neurosci 2008 39 439 451 18775783 60 Barsukova A Komarov A Hajnoczky G Bernardi P Bourdette D Forte M Activation of the mitochondrial permeability transition pore modulates Ca2+ responses to physiological stimuli in adult neurons Eur. J. Neurosci 2011 33 831 842 21255127 61 Gieseler A Schultze AT Kupsch K Haroon MF Wolf G Siemen D Inhibitory modulation of the mitochondrial permeability transition by minocycline Biochem. Pharmacol 2009 77 888 896 19041852 62 Rangaraju V Calloway N Ryan TA Activity-driven local ATP synthesis is required for synaptic function Cell 2014 156 825 835 24529383

PMC005xxxxxx/PMC5114849.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0410462 6011 Nature Nature Nature 0028-0836 1476-4687 27383983 5114849 10.1038/nature18849 NIHMS822517 Article Interactions between the microbiota and pathogenic bacteria in the gut Bäumler Andreas J. 1 Sperandio Vanessa 23 1 Department of Medical Microbiology and Immunology, University of California, Davis, School of Medicine, Davis, California 95616, USA 2 Department of Microbiology, University of Texas Southwestern Medical Center, Dallas, Texas 75390-9048, USA 3 Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas 75390-9038, USA Correspondence should be addressed to V.S. (vanessa.sperandio@utsouthwestern.edu) 13 10 2016 06 7 2016 06 7 2016 18 11 2016 535 7610 8593 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The microbiome has an important role in human health. Changes in the microbiota can confer resistance to or promote infection by pathogenic bacteria. Antibiotics have a profound impact on the microbiota that alters the nutritional landscape of the gut and can lead to the expansion of pathogenic populations. Pathogenic bacteria exploit microbiota-derived sources of carbon and nitrogen as nutrients and regulatory signals to promote their own growth and virulence. By eliciting inflammation, these bacteria alter the intestinal environment and use unique systems for respiration and metal acquisition to drive their expansion. Unravelling the interactions between the microbiota, the host and pathogenic bacteria will produce strategies for manipulating the microbiota against infectious diseases. Appreciation of the important role of the microbiota in human health and nutrition has grown steadily in the past decade. Initial studies focused on cataloguing the microbial species that comprise the microbiota and correlating the composition of the microbiota with the health or disease state of the host. The present period of renaissance has resulted in technologies and interdisciplinary research that are conducive to mechanistic studies and, in particular, those that focus on associations between the microbiota, the host and pathogenic bacteria. Exciting research is now starting to unravel how the composition of the microbiota can offer either resistance or assistance to invading pathogenic species. The majority of these studies were conducted in the gastrointestinal tract, in which associations between the host and microbes are of paramount importance. The gut microbiota of each individual is unique at the genus and species levels; however, it is more generally conserved at the phylum level, which is populated most prominently by Bacteroidetes and Firmicutes, followed by Proteobacteria and Actinobacteria. Host genetics, diet and environmental insults such as treatment with antibiotics alter the microbiota1–4, which can lead to varying susceptibility to infectious diseases between individuals5. The microbiota can promote resistance to colonization by pathogenic species6–9. For instance, mice that are treated with antibiotics or that are bred in sterile environments (known as germ-free mice) are more susceptible to enteric pathogenic bacteria such as Shigella flexneri, Citrobacter rodentium, Listeria monocytogenes and Salmonella enterica serovar Typhimurium10–13. And some microbiotas can lead to the expansion or enhanced virulence of pathogenic populations7. A notable example concerns how differences in the composition of microbiotas determine the susceptibility of the mice to infection with C. rodentium: the transplantation of microbiotas from strains of mice that are susceptible to infection induced similar susceptibility in animals that were previously insusceptible, and the transplantation of microbiotas from resistant animals led to resistance to infection in previously susceptible animals14,15. Epidemiological surveys reinforce this idea. For example, differential susceptibility to infection with Campylobacter jejuni was shown to depend on the species composition of the microbiotas in a study of Swedish adults16. Individuals with a higher diversity within their microbiotas, and with an abundance of bacteria from the genera Dorea and Coprococcus, were significantly recalcitrant to C. jejuni infection compared with people who had low-diversity microbiotas and non-abundance of Dorea and Coprococcus. The host’s diet profoundly affects the composition of the microbiota, with repercussions for the physiology, immunity and susceptibility to infectious diseases of the host17. Dietary choices have been shown to affect colonization by enterohaemorrhagic Escherichia coli (EHEC) serotype O157:H7 and the severity and length of its resulting disease18, and supplementation of the diet with phytonutrients promotes the expansion of beneficial Clostridia species that protect mice from colonization by C. rodentium19. The use of innovative technologies, in combination with more conventional approaches, is driving our understanding of the interactions between the microbiota, the host and pathogenic bacteria. The genetic tractability of several species of bacteria, as well as of their mammalian hosts (such as mice), allows for the mechanistic investigation of these relationships. The investigation of changes in the composition of microbiotas has been driven by next-generation sequencing, which also facilitated the analysis of transcriptomes. The growing power and finesse of metabolomics studies are quickly expanding our knowledge of the impact of both the microbiota and of pathogenic bacteria on the metabolic landscape of the gut. Here, we review advances in our understanding of the complex relationships that determine the severity and outcome of gastrointestinal infections. The majority of the mechanistic studies that investigate these interactions have been conducted in S. Typhimurium, EHEC and Clostridium difficile: therefore, these pathogenic organisms are covered more extensively than others in this Review. Antibiotics Antibiotics revolutionized medicine and were justifiably dubbed ‘magic bullets’ against bacterial infections. However, conventional antibiotics are generally bacteriostatic or bactericidal, which means that they indiscriminately kill or prevent the growth of both pathogenic and beneficial microbes. Antibiotics can alter the taxonomic, genomic and functional features of the microbiota, and their effects can be rapid and sometimes everlasting20. They can decrease the diversity of the microbiota, which compromises resistance to colonization by incoming pathogenic bacteria20 — most notably leading to an expansion of C. difficile that can cause diarrhoea that leads to potentially fatal colitis21. C. difficile is a spore-forming bacterium that, on germination, colonizes the large intestine and causes colitis through the action of two toxins: TcdA and TcdB. The majority of C. difficile infections are nosocomial, but there has also been an increase in community-acquired infections, mainly due to the ubiquitous presence of C. difficile spores. C. difficile can colonize the mammalian intestine without causing disease, but one of the most important risk factors for colitis that is mediated by C. difficile is the use of antibiotics21. The antibiotics-mediated loss of resistance to colonization also allows colonization by S. Typhimurium and the development of disease22. Both C. difficile and S. Typhimurium catabolize sialic acid as a source of carbon in the lumen to promote their expansion23. They rely on saccharolytic members of the microbiota, such as Bacteroides thetaiotaomicron, to make this sugar freely available in the intestinal lumen. Treatment with antibiotics increases the abundance of host-derived free sialic acid as well as enhancing its release into the lumen by B. thetaiotaomicron, which promotes the expansion of the two pathogenic bacteria23. Antibiotic use also triggers production of the organic acid succinate, another microbiota-derived nutrient that confers a growth advantage to C. difficile. It is often present at a low concentration in the microbiotas of conventional mice, but its presence increases on treatment with antibiotics, which promotes a bloom of C. difficile24 (Fig. 1). Knowledge of how microbiota disruption affects the ability of bona fide or opportunistic pathogenic organisms to infect hosts is still in its infancy. However, two underlying themes converge: microbiota-induced changes in the metabolite landscape of the gut and inflammation. Utilization of nutrients Simple dietary sugars are absorbed in the small intestine, which means that they are unavailable as sources of carbon for the microbiota and pathogenic bacteria in the colon. The most abundant members of the microbiota are those that are able to utilize the undigested plant polysaccharides and host glycans that are present in the colon25. The gut epithelium is protected by a layer of mucus that is composed of proteins known as mucins that are rich in fucose, galactose, sialic acid, N-acetylgalactosamine, N-acetylglucosamine and mannose. These sugars are harvested by saccharolytic members of the microbiota, such as Bacteroidales in the gut, which makes them available to species within the microbiota that lack this capability26. However, pathogenic bacteria in the gut can also exploit the availability of these sugars to promote their own expansion. Several studies have used B. thetaiotaomicron as a model Bacteroides in which to investigate these syntrophic links. Sialic acid is a terminal sugar of some mucosal glycans, and B. thetaiotaomicron has sialidase activity but lacks the catabolic pathway for sialic-acid utilization. The bacterium therefore releases sialic acid to gain access to underlying glycans that it can use as a source of carbon. The sialic acid that B. thetaiotaomicron releases from the mucus can be catabolized by both C. difficile and S. Typhimurium, which provides them with a growth advantage23. The ability of the microbiota to use sialic acid therefore depends on the action of B. thetaiotaomicron, and mutants that lack sialidase fail to enhance the growth of these two pathogenic bacteria23. B. thetaiotaomicron also releases fucose from the mucus. It harbours multiple enzymes that can cleave fucose from host glycans, so its presence results in the high availability of fucose in the lumen of the gut27–30. This free fucose can also be used as a source of carbon by S. Typhimurium23. Importantly, B. thetaiotaomicron can promote the fucosylation of mucosal glycans when introduced into monoassociated germ-free mice31,32. The microbiota resides in the lumen and the outer mucus layer of the intestine. EHEC, however, aims to achieve a unique niche by closely adhering to the enterocytes of the intestinal epithelium. To achieve its goal, EHEC must successfully compete with the microbiota for nutrients. B. thetaiotaomicron does not need to compete with EHEC, however, because it can utilize polysaccharides; EHEC can only utilize monosaccharides and disaccharides13,33. EHEC’s main competitors are commensal E. coli, which preferentially utilizes fucose as a source of carbon when growing in the mammalian intestine13,33. To circumvent this competition, EHEC utilizes other sources of sugar, such as galactose, the hexuranates, mannose and ribose, which commensal E. coli cannot catabolize optimally33,34 (Fig. 2). EHEC uses fucose as a signalling molecule with which to adjust its metabolism and to regulate the expression of its virulence repertoire in the lumen and the outer mucus layer of the colon35. It horizontally acquired a pathogenicity island of genes that encode a fucose-sensing signalling-transduction system35. This system is unique to EHEC and to C. rodentium35 (which is used extensively in mouse models as a surrogate for the human pathogen EHEC36). It is composed of the membrane-bound histidine sensor kinase FusK, which specifically autophosphorylates in response to fucose. FusK then transfers its phosphate to a response regulator called FusR, which is a transcription factor. Phosphorylation activates FusR, which represses the expression of the fucose utilization genes in EHEC, and helps EHEC to avoid the need to compete for this nutrient with commensal E. coli35. To prevent the unnecessary expenditure of energy by EHEC, FusR represses the genes that encode the EHEC virulence machinery, a syringe-like apparatus known as a type III secretion system (T3SS), which the bacterium uses to adhere itself to enterocytes and highjack the function of these host cells35. EHEC therefore uses fucose, a host-derived signal that is made available by the microbiota, to sense the environment of the intestinal lumen and to modulate its own metabolism and virulence. To reach the lining of the epithelium, EHEC and C. rodentium produce mucinases37, which cleave the protein backbone of mucin-type glycoproteins. Expression of these enzymes is increased by metabolites that are produced by B. thetaiotaomicron38. Because mucus is one of the main sources of sugar in the colon, where EHEC and C. rodentium colonize, obliteration of the mucus layer creates a nutrient-poor environment near the epithelium that is referred to as gluconeogenic. The colonization of mice by B. thetaiotaomicron therefore profoundly changes the metabolic landscape of the mouse gut because it raises the levels of organic acids such as succinate24,38,39. Moreover, several metabolites that indicate a gluconeogenic environment, such as lactate and glycerate, are also elevated38. EHEC and C. rodentium sense this gluconeogenic and succinate-rich environment through the transcriptional regulator Cra. On receiving the cue that they have reached the lining of the gut epithelium, these bacteria activate the expression of their T3SSs38. EHEC therefore exploits metabolic cues from B. thetaiotaomicron, and probably other members of the microbiota, to precisely programme its metabolism and virulence (Fig. 2). Other pathogenic bacteria can also adjust their gene expression in the presence of microbiota-produced succinate. C. difficile induces a pathway that converts succinate to butyrate, which confers a growth advantage in vivo24. Populations of C. difficile mutants that are unable to convert succinate fail to expand in the gut in the presence of B. thetaiotaomicron24. Several short-chain fatty acids that are produced by the microbiota, are important determinants of interactions between the microbiota and pathogenic bacteria in the gut. The abundance and composition of short-chain fatty acids is distinct in each compartment of the intestine, and the ability to sense these differences might help pathogenic bacteria in niche recognition. The most abundant short-chain fatty acids in the gut are acetate, propionate and butyrate. S. Typhimurium preferably colonizes the ileum40, which generally contains acetate at a concentration of 30 mM. This acetate concentration enhances the expression of the S. Typhimurium Salmonella pathogenicity island 1 (SPI-1)-encoded T3SS (T3SS-1), which is involved in the bacterium’s invasion of the host. Conversely, 70 mM propionate and 20 mM butyrate, concentrations typical of the colon, suppress the expression of the T3SS-1 (ref. 41). Propionate and butyrate seem to affect the T3SS-1 regulatory cascade at various levels. However, the detailed mechanism of this regulation is yet to be unravelled. In EHEC, exposure to the levels of butyrate found in the colon increases the expression of the EHEC T3SS through post-transcriptional activation of the transcriptional regulator Lrp42. Exposure to the concentrations of acetate and propionate that are found in the small intestine does not significantly affect the virulence of EHEC. Diet has a profound effect on the composition of the microbiota and the concentration of short-chain fatty acids in the gut17. A diet that is high in fibre results in the enhanced production of butyrate by the gut microbiota. That increases the host’s expression of globotriaosylceramide, which is a receptor for the Shiga toxin that is produced by EHEC18. Shiga toxin can lead to the development of haemolytic uraemic syndrome (HUS) and is the cause of the morbidity and mortality associated with outbreaks of EHEC43. Consequently, animals that are fed a high-fibre diet are more susceptible to Shiga toxin than are those on a low-fibre diet and develop more severe disease18. Conversely, increased levels of microbiota-derived acetate protect animals from disease that is caused by the toxin. Certain species of Bifidobacteria contribute to higher levels of acetate in the gut, which helps to improve the barrier function of the intestinal epithelium and to prevent Shiga toxin from reaching the bloodstream44. Enteric pathogenic bacteria also use other nutrients to successfully overcome the microbiota’s resistance to their colonization. Ethanolamine is abundant in the mammalian intestine45. It can be used as a source of carbon and of nitrogen by a number of pathogenic species46, and food-borne bacteria are particularly adept at using it. However, it cannot be metabolized by the majority of commensal species47. S. Typhimurium, EHEC and L. monocytogenes gain a growth advantage in the intestine through their ability to use this compound45,48,49. Ethanolamine is also used as a signal by EHEC and S. Typhimurium to activate the expression of virulence genes50,51. And S. Typhimurium uses hydrogen produced by the microbiota as an energy source to enhance its growth during the initial stage of infection52. The exploitation of microbiota-derived molecules as both nutrients and signals is crucial for the successful infection of the host by pathogenic bacteria. Although such organisms have clearly developed many strategies through which to circumvent the microbiota’s resistance to colonization, and in many cases even employ its help, the microbiota pushes back, which creates an intense competition for resources. The ability of EHEC to colonize the intestine stems from differences in the sources of sugar that are used by EHEC and by commensal E. coli. For example, the presence of multiple strains of commensal E. coli with overlapping nutritional requirements interferes with the colonization of the mouse intestine by EHEC53. This study uses a streptomycin-treated mouse model of EHEC and three distinct commensal strains of E. coli to assess differential sugar requirements for the successful colonization of the intestines53. EHEC could colonize mice that were pre-colonized with any one of the commensal strains, but it could not colonize mice that were pre-colonized with all three strains53. EHEC has evolved to exploit distinct sources of sugar during colonization of the gut. It utilizes catabolic pathways for the hexuronates glucuronate and galacturonate and for sucrose that are not employed by commensal E. coli within the gut33,53. It can also metabolize several sugars simultaneously. The loss of multiple catabolic pathways has an additive effect on colonization. This phenomenon is not observed in commensal E. coli, however, which suggests that E. coli uses available sugars in a stepwise fashion54. EHEC therefore differs from commensal E. coli in metabolic strategy and the use of nutrients for the colonization of the mammalian intestine. C. rodentium is outcompeted and then cleared from the mouse gut through a bloom in the population of commensal E. coli, which competes with C. rodentium for monosaccharides for nutrition13. By contrast, C. rodentium is not cleared by B. thetaiotaomicron in germ-free mice that are fed a diet that contains both monosaccharides, which can be used by Enterobacteriacae such as C. rodentium, and polysaccharides, which can be used by Bacteroides. However, when the mice are switched to a diet that consists only of monosaccharides, B. thetaiotaomicron and C. rodentium are forced to compete for sugars, and B. thetaiotaomicron outcompetes C. rodentium13. The ability of pathogenic bacteria to successfully compete with commensal species for nutrients is therefore important for their establishment in the gut. Interception of signals from the microbiota and the host The microbiota affects the risks and courses of enteric diseases. Vibrio cholerae is a major cause of explosive diarrhoea in which there is extensive disruption of the intestinal population of microbes. Metagenomic studies of the faecal microbiota of people with cholera in Bangladesh show that recovery is characterized by a certain microbiota signature. Reconstitution of this microbiota in germ-free mice restricts the infectivity of V. cholerae. Specifically, the presence of Ruminococcus obeum can hamper the colonization of the intestines by V. cholerae through the production of the furanone signal autoinducer-2, which causes the repression of several V. cholerae colonization factors55. Another example of the effect of microbiota-derived signals on host colonization is their use by EHEC in the colonization of its ruminal reservoir. EHEC exclusively colonizes the recto–anal junction of adult cattle. Through the sensor protein SdiA, EHEC detects acyl-homoserine lactone signals from the rumen microbiota, which it uses to reprogram itself to survive the acidic pH of the animal’s stomachs and to successfully colonize the rectoanal junction56. As well as being able to directly detect signals that are derived from the microbiota, pathogenic bacteria can detect host-derived signals that have been modified by the microbiota to modulate their virulence. V. cholerae has a type VI secretion system (T6SS), which it uses to kill other bacteria. During its colonization of the intestine, V. cholerae comes in contact with the mucosal microbiota, which can affect the composition of bile acids in the intestine. For example, Bifidobacterium bifidum negatively regulates the T6SS activity of V. cholerae through the metabolic conversion of three bile acids (glycodeoxycholic acid, taurodeoxycholic acid and cholic acid) into the bile acid deoxycholic acid. Deoxycholic acid, but not its unmodified salts, decreases the expression of T6SS genes. This leads to a decrease in the killing of E. coli by V. cholerae owing to bile-acid conversion by other commensals, which decreases the activity of the T6SS57. Another microbiota-modified host signal that is detected by pathogenic bacteria is the neurotransmitter noradrenaline. The gut is highly innervated, and neurotransmitters are important signals in the gastrointestinal tract, where they modulate peristalsis, the flow of blood and the secretion of ions58. The microbiota affects the availability of neurotransmitters in the intestinal lumen, as well as their biosynthesis. For example, the microbiota induces biosynthesis of serotonin59, and microbiota-derived enzymatic activities increase the levels of active noradrenaline in the gut lumen60. Noradrenaline is synthesized by the adrenergic neurons of the enteric nervous system61 and it is inactivated by the host through conjugation with glucuronic acid (to produce a glucuronide). Microbiota-produced enzymes known as glucuronidases then deconjugate glucuronic acid from noradrenaline, which increases the amount of active noradrenaline in the lumen of the intestine60. Several pathogenic bacteria of the gut, including EHEC, S. Typhimurium and V. parahaemolyticus, sense noradrenaline to activate the expression of virulence genes62–65. Two adrenergic sensors have been identified in bacteria: the membrane-bound histidine kinases QseC and QseE66,67. QseC also detects the microbiota-produced signal autoinducer-3 (refs 64 and 66), so the sensing of signals from both the host and the microbiota converge at the level of a single receptor, a process known as inter-kingdom signalling. Inflammation Although diet and the composition of the microbiota heavily influence the availability of nutrients in the gut, the host also has an important part to play. A crucial driver of changes in the gut environment is the inflammatory response of the host. Intestinal inflammation in people is associated with an imbalance in the microbiota, known as dysbiosis, and is characterized by a reduced diversity of microbes, a reduced abundance of obligate anaerobic bacteria and an expansion of facultative anaerobic bacteria in the phylum Proteobacteria, mostly members of the family Enterobacteriaceae68–73. Similar changes in the composition of the gut microbiota are observed in mice with chemically induced colitis74 and genetically induced colitis75. These changes in the structure of the microbiota probably reflect an altered nutritional environment that is created by the inflammatory response of the host. The availability of nutrients in the large intestine is altered during inflammation through changes in the composition of mucous carbohydrates. Interleukin (IL)-22, a cytokine that is prominently induced in the intestinal mucosa when mice and rhesus macaques are infected with S. Typhimurium76,77, stimulates the epithelial expression of galactoside 2-α-L-fucosyltransferase 2 and enhances the α(1,2)-fucosylation of mucus carbohydrates78,79. The gut microbiota can liberate fucose from mucus carbohydrates23,80, which leads to the induction of genes for fucose utilization in E. coli78. Similarly, increased fucosylation of glycans is observed during S. Typhimurium-induced colitis in mice, which correlates with elevated synthesis of the proteins involved in fucose utilization81. Mucus fucosylation that is induced during infection with C. rodentium causes changes in the composition of the gut microbiota that help to protect the host from the expansion and epithelial translocation of the pathobiont Enterococcus faecalis79. Another driver of changes in the nutritional environment of the gut is the generation of reactive oxygen species and reactive nitrogen species during inflammation. Pro-inflammatory cytokines such as interferon-γ (IFN-γ) activate dual oxidase 2 in the intestinal epithelium, which produces hydrogen peroxide82. Increased expression of DUOX2, the gene that encodes dual oxidase 2, in the intestinal mucosa of patients with Crohn’s disease and ulcerative colitis correlates with an expansion of Proteobacteria in the gut microbiota83. IFN-γ also induces epithelial expression of the gene Nos2 (ref. 84), which encodes inducible nitric oxide synthase, the enzyme that catalyses the production of nitric oxide from L-arginine85. As a result, the concentration of nitric oxide is elevated in gases from the colons of people with inflammatory bowel disease86–88. Although reactive oxygen and nitrogen species have antimicrobial activity, these radicals quickly form non-toxic compounds in the lumen of the gut as they diffuse away from the epithelium. For example, when they are generated during inflammation by host enzymes in the intestinal epithelium, these species react to form nitrate89. This by-product of inflammation is present at elevated concentrations in the intestines of mice with chemically induced colitis90 (Fig. 3). Nitrate reductases, enzymes that are broadly conserved among the Enterobacteriaceae, couple the reduction of nitrate to energy-conserving electron transport systems for respiration, a process termed nitrate respiration. However, the genes that encode them are absent from the genomes of obligate anaerobic Clostridia or Bacteroidia91. Nitrate respiration drives the Nos2-dependent expansion of commensal E. coli in mice with chemically or genetically induced colitis, but not in animals without signs of intestinal inflammation91. Respiratory electron acceptors that are generated as a by-product of the host inflammatory response therefore create a niche in the lumen of the intestines that supports the uncontrolled expansion of commensal Enterobacteriaceae rather than of obligate anaerobic bacteria91. The resulting bloom in the inflamed intestine is one of the most consistent and robust ecological patterns that has been observed in the gut microbiota92. The creation of a niche for respiratory nutrients during inflammation is also an important driver of the strategies that pathogenic bacteria from the family Enterobacteriaceae use to invade the gut ecosystem. In the absence of inflammation or treatment with antibiotics, members of the gut microbiota occupy all available nutrient niches, which makes it very challenging for pathogenic Enterobacteriaceae to enter the community. One solution is for these bacteria to trigger intestinal inflammation, which would coerce the host into creating a fresh niche of respiratory nutrients that is suitable for its expansion — an approach that is used by S. Typhimurium93. On ingestion, S. Typhimurium uses T3SS-1 to invade the intestinal epithelium94 and T3SS-2 to survive in the tissue of the host95. Both of these processes trigger acute intestinal inflammation in cattle and in mouse models of gastroenteritis96–98 (Fig. 3). The inflammatory response of the host drives the expansion of S. Typhimurium in the lumen of the gut99, which is required for the transmission of this pathogenic species to a new host through the faecal–oral route100. Although such expansion allows S. Typhimurium to side-step competition with obligate anaerobic Clostridia and Bacteroidia, this strategy forces the bacterium into battle with commensal Enterobacteriaceae over limited resources. For example, S. Typhimurium expands in the inflamed gut through nitrate respiration101,102, which results in rivalry with commensal Enterobacteriaceae that pursue a similar strategy91. S. Typhimurium can gain an edge in this competition through its ability to utilize a broader range of inflammation-derived electron acceptors than its rivals. A source of one such electron acceptor is sulfate-reducing species of Desulfovibrio from the microbiota, which release hydrogen sulfide, a compound that is converted to thiosulfate by the epithelium of the colon to avoid toxicity103. Deployment of the virulence factors of pathogenic bacteria leads to the recruitment of neutrophils to the intestinal mucosa, which is the histopathological hallmark of S. Typhimurium-induced gastroenteritis96. A fraction of these recruited neutrophils migrate into the lumen of the intestine — a diagnostic marker of inflammatory diarrhoea104. In the lumen, neutrophils help to protect the mucosa by engulfing bacteria in the vicinity of the epithelium105, but reactive oxygen species that are generated by the phagocyte-produced NADPH oxidase 2 (also known as cytochrome b-245 heavy chain) convert thiosulfate into tetrathionate, a respiratory electron acceptor that supports the expansion of S. Typhimurium in the lumen of the inflamed gut106 (Fig. 3). Although tetrathionate respiration is a characteristic of Salmonella serovars and has been used empirically in their isolation in clinical microbiology laboratories since 1923 (ref. 107), insights into the respiratory nutrient niche that Salmonella occupies suggest that this property is part of a strategy to edge out competing commensal Enterobacteriaceae in the inflamed gut106. The inflammatory response of the host also ignites competition between commensal and pathogenic Enterobacteriaceae over trace elements such as iron, which is less available during inflammation. IL-22 induces the release of the antimicrobial protein lipocalin-2 (also known as neutrophil gelatinase-associated lipocalin) from the epithelium in mice and rhesus macaques108,109. Lipocalin-2 reduces iron availability by binding to enterobactin, a low-molecular-weight iron chelator (or siderophore) that is produced by Enterobacteriaceae110,111. To overcome this, S. Typhimurium and some commensal E. coli secrete a glycosylated derivative of enterobactin, termed salmochelin, which is not bound by lipocalin-2 (ref. 108). By producing salmochelin as well as two further siderophores that are not bound by lipocalin-2, yersiniobactin and aerobactin, the probiotic E. coli strain Nissle 1917 can limit the expansion of S. Typhimurium in the lumen of the inflamed gut112. Conversely, lipocalin-2 secretion by the epithelium generates an environment that enables S. Typhimurium to edge out commensal Enterobacteriaceae that depend solely on enterobactin for the acquisition of iron109 (Fig. 3). Through its limitation of iron availability, intestinal inflammation also sets the stage for battles between Enterobacteriaceae that use protein-based toxins known as colicins113 that affect a narrow range of hosts. Iron limitation induces the synthesis of siderophore receptor proteins for the bacterial outer membrane113, which also commonly serve as receptors for colicins114–116. Expression of a siderophore receptor protein termed the colicin I receptor (CirA) confers commensal E. coli with sensitivity to colicin Ib produced by S. Typhimurium113. The respiratory nutrient niche that is generated by the inflammatory response of the host is therefore a battleground on which commensal and pathogenic Enterobacteriaceae struggle for dominance using a diverse arsenal of nutritional and antimicrobial strategies. Perspective and the future The study of the microbiome began more than a century ago. equencing of 16S rRNA genes provided the first insights into the taxonomic composition of microbial communities. Later, sequencing of the complete metagenome of microbial communities provided a more detailed insight into the full genetic capacity of such a community. The use of germ-free animals, either alone or in combination with emerging technologies such as laser-capture microdissection and transcriptomics, enabled mechanistic studies of the associations between the microbiota, the host and pathogenic bacteria117. Multi-taxon insertion sequencing now allows researchers to investigate both the assembly and the shared and strain-specific dietary requirements of communities of microbes, and it has also facilitated the informed manipulation of such communities through diet118. The development of quantitative imaging technologies has provided insight into the localization of microbes within the gastrointestinal tract, and it has also enabled studies on the proximity of and interactions between microbes119. The increasing refinement and power of metabolomics, imaging mass spectrometry and three-dimensional mapping of mass-spectrometry data provide a high-resolution image of the complex chemistry landscape of the interactions between microbes and the host, which sets the stage for manipulating this chemistry to prevent or treat infectious diseases24,38,120–127. A marriage of metagenomics and mathematical modelling promises to enhance the precision of microbiome reconstitution, which has proven successful for tackling C. difficile infections in mice128. In these exciting times, the expansion of multidisciplinary research is rapidly generating new technologies and mechanistic insights into interactions between the microbiota, the host and pathogenic bacteria (Box 1). BOX 1 Microbiota interventions as therapeutic strategies to limit pathogen expansion An imbalance in the gut microbiota might underlie many human diseases but, in most cases, the development of treatment options is still in its infancy. This could be in part because the mechanisms that lead to adverse effects in the host differ for each disease, which means that intervention strategies must be developed for each. The treatment options for antibiotic-induced dysbiosis are perhaps the most advanced, mainly because faecal microbiota transplantation can reverse this imbalance in the gut microbiota129. Nonetheless, the mechanisms through which treatment with antibiotics encourages an uncontrolled expansion of the obligate anaerobe C. difficile differ markedly from those that stimulate the growth of the facultative anaerobes Enterobacteriaceae, which has implications for the development of precision microbiome interventions. Mice that are treated with streptomycin have a reduced abundance of members of the class Clostridia130, which are credited with producing the lion’s share of the short-chain fatty acid butyrate in the large intestine131. The resulting depletion of short-chain fatty acids drives an expansion of Enterobacteriaceae through mechanisms that are not fully resolved44,132. Depletion of Clostridia-derived butyrate affects the metabolism of enterocytes in the colon, which derive most of their energy by butyrate respiration133. The depletion of short-chain fatty acids also leads to a contraction in the pool of regulatory T cells in the colonic mucosa134–136. These changes in the host physiology increase the inflammatory tone of the mucosa, as indicated by the elevated expression of Nos2, the gene that encodes inducible nitric oxide synthase, and contributes to the expansion of commensal E. coli through nitrate generation137. Although other mechanisms probably contribute to the post-antibiotic expansion of certain populations of bacteria in the gut126, the transfer of Clostridia, with their capacity for producing short-chain fatty acids, represents the most effective treatment for limiting the growth of E. coli in streptomycin-treated mice138. By contrast, the post-antibiotic expansion of the C. difficile population is driven by a depletion of secondary bile salts. The liver produces the primary bile salts cholate and chenodeoxycholate, which are conjugated to the amino acids taurine (to produce taurocholate and taurochenodeoxycholate) or glycine (to produce glycocholate and glycochenodeoxycholate) and then secreted into the gut. Bile salt hydrolases, enzymes that are produced by many members of the gut microbiota, remove the conjugated amino acid from the primary bile salt. C. scindens is one of a limited number of species of bacteria that can actively transport cholate and chenodeoxycholate into its cytosol, where these unconjugated primary bile salts are converted into the secondary bile salts deoxycholate and lithocholate, which are subsequently secreted into the extracellular environment139 (Box Fig.). Although both primary and secondary bile salts induce the germination of C. difficile spores, only secondary bile salts efficiently prevent the growth of vegetative C. difficile cells140. By significantly reducing the abundance of species that are capable of producing deoxycholate and lithocholate, treatment with antibiotics causes a depletion of these secondary bile salts and promotes the expansion of vegetative C. difficile cells in the large intestine141,142. Faecal microbiota transplantation restores the production of secondary bile salts and therefore prevents the expansion of C. difficile143. Direct supplementation of the diet with secondary bile salts warrants caution because increased concentrations of bile salts have been linked to gastrointestinal cancers144. However, inoculation with only the secondary-bile-salt-producing C. scindens confers mice with resistance to C. difficile expansion following treatment with antibiotics128. This remarkable observation opens the door to novel precision microbiome interventions that aim to prevent or treat the colitis that is associated with C. difficile infection after antibiotic therapy. Work in the V.S. laboratory is supported by US National Institutes of Health (NIH) grants AI053067, AI077613, AI05135 and AI114511. Work in the A.J.B. laboratory is supported by US Department of Agriculture grant 2015-67015-22930 and NIH grants AI044170, AI096528, AI112445, AI114922 and AI117940. The contents of this Review are solely the responsibility of the authors and do not necessarily represent the official views of the NIH National Institute of Allergy and Infectious Diseases. Figure 1 The impact of antibiotics on the microbiota and the expansion of enteric pathogens a, A diverse and non-disturbed microbiota confers resistance to colonization by enteric pathogens in the intestinal epithelium. b, Treatment with antibiotics decreases the diversity of the microbiota and leads to expansion of the C. difficile population. Toxins that are released from C. difficile (TcdA and TcdB) enter and damage the cells of the epithelium, which leads to inflammation (colitis) and cell death. c, Treatment with antibiotics also leads to an increase in the levels of free sialic acid (from the host) and succinate (from the microbiota) in the lumen of the intestine. Elevated sialic acid promotes the expansion of the S. Typhimurium population, which can lead to inflammation (gastroenteritis) if the bacterium invades the cells of the intestinal epithelium. Elevated levels of sialic acid and succinate further promote the expansion of the C. difficile population and the development of colitis and cell death. Figure 2 Modulation of enterohaemorrhagic E. coli virulence through nutrients provided by the microbiota a, The microbiota resides in the lumen and outer mucus layer of the intestine. The saccharolytic bacterium Bacteroides thetaiotaomicron is a prominent member of the microbiota. It can release fucose from the mucus and makes the sugar available to other bacteria. When EHEC senses fucose through the FusKR signalling system, it represses both its use of the sugar and the expression of genes that encode the T3SS, a protein-translocation apparatus that enables the bacterium to secrete effector proteins into host cells. This repression prevents EHEC from competing for fucose with commensal E. coli and from expending energy unnecessarily on T3SS expression. b, Metabolites that are provided by B. thetaiotaomicron, such as succinate, lead to an increase in the expression by EHEC of the enzyme mucinase, which obliterates the mucus layers of the intestine. EHEC is then able to reach the intestinal epithelium. B. thetaiotaomicron then begins to secrete succinate and other metabolites that are required for gluconeogenesis into the now nutrient-poor environment. The compounds are sensed by EHEC, which upregulates its expression of the T3SS to enable the bacterium to attach to the epithelial cells of the host intestine and form lesions that cause diarrhoea. Figure 3 The effect of intestinal inflammation on nutrient availability S. Typhimurium uses its virulence factors (T3SS-1 and T3SS-2) to trigger intestinal inflammation. Cytokines that are released during inflammation, such as IL-22 and IFN-γ, trigger the release of antimicrobial molecules lipocalin-2, reactive oxygen species (ROS) and reactive nitrogen species (RNS) from the intestinal epithelium. Lipocalin-2 can block the growth of commensal Enterobacteriaceae that rely on the siderophore enterobactin for the acquisition of iron (Fe3+). It does not bind to the S. Typhimurium siderophone salmochelin, however, which confers the bacterium with resistance to its effects on growth. RNS and ROS react to form nitrate, which drives the growth of Enterobacteriaceae through nitrate respiration. Microbiota-derived hydrogen sulfide is converted to thiosulfate by colonic epithelial cells. Neutrophils that migrate into the lumen of the intestine during inflammation generate ROS that convert endogenous sulfur compounds (thiosulfate) into an electron acceptor (tetrathionate) that further boosts the growth of S. Typhimurium through tetrathionate respiration. The authors declare no competing financial interests. Readers are welcome to comment on the online version of this paper at go.nature.com/28ix4vg. 1 Eckburg PB Diversity of the human intestinal microbial flora Science 308 1635 1638 2005 15831718 2 Frank DN Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases Proc Natl Acad Sci USA 104 13780 13785 2007 17699621 3 The Human Microbiome Project Consortium Structure function and diversity of the healthy human microbiome Nature 486 207 214 2012 22699609 4 Ley RE Peterson DA Gordon JI Ecological and evolutionary forces shaping microbial diversity in the human intestine Cell 124 837 848 2006 16497592 5 Yurist-Doutsch S Arrieta MC Vogt SL Finlay BB Gastrointestinal microbiota-mediated control of enteric pathogens Annu Rev Genet 48 361 382 2014 25251855 6 Sassone-Corsi M Raffatellu M No vacancy: how beneficial microbes cooperate with immunity to provide colonization resistance to pathogens J Immunol 194 4081 4087 2015 25888704 7 Cameron EA Sperandio V Frenemies: signaling and nutritional integration in pathogen–microbiota–host interactions Cell Host Microbe 18 275 284 2015 26355214 8 Pacheco AR Sperandio V Enteric pathogens exploit the microbiota-generated nutritional environment of the gut Microbiol Spectr 3 MBP-0001–2014 2015 9 Bohnhoff M Drake BL Miller CP Effect of streptomycin on susceptibility of intestinal tract to experimental Salmonella infection Proc Soc Exp Biol Med 86 132 137 1954 13177610 10 Ferreira RB The intestinal microbiota plays a role in Salmonella-induced colitis independent of pathogen colonization PLoS ONE 6 e20338 2011 21633507 11 Sprinz H The response of the germfree guinea pig to oral bacterial challenge with Escherichia coli and Shigella flexneri Am J Pathol 39 681 695 1961 13915950 12 Zachar Z Savage DC Microbial interference and colonization of the murine gastrointestinal tract by Listeria monocytogenes Infect Immun 23 168 174 1979 106003 13 Kamada N Regulated virulence controls the ability of a pathogen to compete with the gut microbiota Science 336 1325 1329 2012 This study showed that enteric pathogens use virulence genes to compete with the microbiota for nutrients in the gut 22582016 14 Ghosh S Colonic microbiota alters host susceptibility to infectious colitis by modulating inflammation, redox status, and ion transporter gene expression Am J Physiol Gastrointest Liver Physiol 301 G39 G49 2011 21454446 15 Willing BP Vacharaksa A Croxen M Thanachayanont T Finlay BB Altering host resistance to infections through microbial transplantation PLoS ONE 6 e26988 2011 22046427 16 Kampmann C Dicksved J Engstrand L Rautelin H Composition of human faecal microbiota in resistance to Campylobacter infection Clin Microbiol Infect 22 61.e1 61.e8 2016 26369602 17 Kau AL Ahern PP Griffin NW Goodman AL Gordon JI Human nutrition, the gut microbiome and the immune system Nature 474 327 336 2011 21677749 18 Zumbrun SD Dietary choice affects Shiga toxin-producing Escherichia coli (STEC) O157:H7 colonization and disease Proc Natl Acad Sci USA 110 E2126 E2133 2013 23690602 19 Wlodarska M Willing BP Bravo DM Finlay BB Phytonutrient diet supplementation promotes beneficial Clostridia species and intestinal mucus secretion resulting in protection against enteric infection Sci Rep 5 9253 2015 25787310 20 Modi SR Collins JJ Relman DA Antibiotics and the gut microbiota J Clin Invest 124 4212 4218 2014 25271726 21 Leffler DA Lamont JT Clostridium difficile infection N Engl J Med 373 287 288 2015 22 Pavia AT Epidemiologic evidence that prior antimicrobial exposure decreases resistance to infection by antimicrobial-sensitive Salmonella J Infect Dis 161 255 260 1990 2299207 23 Ng KM Microbiota-liberated host sugars facilitate post-antibiotic expansion of enteric pathogens Nature 502 96 99 2013 This study showed that treatment with antibiotics enhances the abundance of host sialic acid that can be harvested by the microbiota, which promotes the expansion of enteric pathogens 23995682 24 Ferreyra JA Gut microbiota-produced succinate promotes C. difficile infection after antibiotic treatment or motility disturbance Cell Host Microbe 16 770 777 2014 25498344 25 Ferreyra JA Ng KM Sonnenburg JL The enteric two-step: nutritional strategies of bacterial pathogens within the gut Cell Microbiol 16 993 1003 2014 24720567 26 Rakoff-Nahoum S Coyne MJ Comstock LE An ecological network of polysaccharide utilization among human intestinal symbionts Curr Biol 24 40 49 2014 24332541 27 Alverdy J Chi HS Sheldon GF The effect of parenteral nutrition on gastrointestinal immunity. The importance of enteral stimulation Ann Surg 202 681 684 1985 3935061 28 Fischbach MA Sonnenburg JL Eating for two: how metabolism establishes interspecies interactions in the gut Cell Host Microbe 10 336 347 2011 22018234 29 Chow WL Lee YK Free fucose is a danger signal to human intestinal epithelial cells Br J Nutr 99 449 454 2008 17697405 30 Bourlioux P Koletzko B Guarner F Braesco V The intestine and its microflora are partners for the protection of the host: report on the Danone Symposium “The Intelligent Intestine,” held in Paris, June 14, 2002 Am J Clin Nutr 78 675 683 2003 14522724 31 Bry L Falk PG Midtvedt T Gordon JI A model of host–microbial interactions in an open mammalian ecosystem Science 273 1380 1383 1996 8703071 32 Hooper LV Xu J Falk PG Midtvedt T Gordon JI A molecular sensor that allows a gut commensal to control its nutrient foundation in a competitive ecosystem Proc Natl Acad Sci USA 96 9833 9838 1999 10449780 33 Fabich AJ Comparison of carbon nutrition for pathogenic and commensal Escherichia coli strains in the mouse intestine Infect Immun 76 1143 1152 2008 18180286 34 Autieri SM L-fucose stimulates utilization of D-ribose by Escherichia coli MG1655 ΔfucAO and E. coli Nissle 1917 ΔfucAO mutants in the mouse intestine and in M9 minimal medium Infect Immun 75 5465 5475 2007 17709419 35 Pacheco AR Fucose sensing regulates bacterial intestinal colonization Nature 492 113 117 2012 This paper showed that sugars from the mucus that are released by the microbiota can be perceived as signals by enteric pathogens to regulate the expression of virulence genes 23160491 36 Schauer DB Falkow S Attaching and effacing locus of a Citrobacter freundii biotype that causes transmissible murine colonic hyperplasia Infect Immun 61 2486 2492 1993 8500884 37 Szabady RL Lokuta MA Walters KB Huttenlocher A Welch RA Modulation of neutrophil function by a secreted mucinase of Escherichia coli O157:H7 PLoS Pathog 5 e1000320 2009 19247439 38 Curtis MM The gut commensal Bacteroides thetaiotaomicron exacerbates enteric infection through modification of the metabolic landscape Cell Host Microbe 16 759 769 2014 This study revealed that metabolites that are produced by the microbiota can be exploited as cues to increase the virulence of enteric pathogens 25498343 39 Macy JM Ljungdahl LG Gottschalk G Pathway of succinate and propionate formation in Bacteroides fragilis J Bacteriol 134 84 91 1978 148460 40 Carter PB Collins FM The route of enteric infection in normal mice J Exp Med 139 1189 1203 1974 4596512 41 Lawhon SD Maurer R Suyemoto M Altier C Intestinal short-chain fatty acids alter Salmonella typhimurium invasion gene expression and virulence through BarA/SirA Mol Microbiol 46 1451 1464 2002 12453229 42 Takao M Yen H Tobe T LeuO enhances butyrate-induced virulence expression through a positive regulatory loop in enterohaemorrhagic Escherichia coli Mol Microbiol 93 1302 1313 2014 25069663 43 Karmali MA Petric M Lim C Fleming PC Steele BT Escherichia coli cytotoxin, haemolytic-uraemic syndrome, and haemorrhagic colitis Lancet 2 1299 1300 1983 44 Fukuda S Bifidobacteria can protect from enteropathogenic infection through production of acetate Nature 469 543 547 2011 This study showed that the acetate that is produced by certain probiotic bacteria can enhance the barrier function of the gut epithelium to protect the host from enteric infections 21270894 45 Bertin Y Enterohaemorrhagic Escherichia coli gains a competitive advantage by using ethanolamine as a nitrogen source in the bovine intestinal content Environ Microbiol 13 365 377 2011 20849446 46 Garsin DA Ethanolamine utilization in bacterial pathogens: roles and regulation Nature Rev Microbiol 8 290 295 2010 20234377 47 Korbel JO Systematic association of genes to phenotypes by genome and literature mining PLoS Biol 3 e134 2005 15799710 48 Thiennimitr P Intestinal inflammation allows Salmonella to use ethanolamine to compete with the microbiota Proc Natl Acad Sci USA 108 17480 17485 2011 21969563 49 Joseph B Identification of Listeria monocytogenes genes contributing to intracellular replication by expression profiling and mutant screening J Bacteriol 188 556 568 2006 16385046 50 Kendall MM Gruber CC Parker CT Sperandio V Ethanolamine controls expression of genes encoding components involved in interkingdom signaling and virulence in enterohemorrhagic Escherichia coli O157:H7 mBio 3 00050 12 2012 51 Anderson CJ Clark DE Adli M Kendall MM Ethanolamine signaling promotes Salmonella niche recognition and adaptation during infection PLoS Pathog 11 e1005278 2015 26565973 52 Maier L Microbiota-derived hydrogen fuels Salmonella Typhimurium invasion of the gut ecosystem Cell Host Microbe 14 641 651 2013 24331462 53 Maltby R Leatham-Jensen MP Gibson T Cohen PS Conway T Nutritional basis for colonization resistance by human commensal Escherichia coli strains HS and Nissle 1917 against E. coli O157:H7 in the mouse intestine PLoS ONE 8 e53957 2013 23349773 54 Fabich AJ Comparison of carbon nutrition for pathogenic and commensal Escherichia coli strains in the mouse intestine Infect Immun 76 1143 1152 2008 18180286 55 Hsiao A Members of the human gut microbiota involved in recovery from Vibrio cholerae infection Nature 515 423 426 2014 25231861 56 Hughes DT Chemical sensing in mammalian host–bacterial commensal associations Proc Natl Acad Sci USA 107 9831 9836 2010 20457895 57 Bachmann V Bile salts modulate the mucin-activated type VI secretion system of pandemic Vibrio cholerae PLoS Negl Trop Dis 9 e0004031 2015 26317760 58 Hörger S Schultheiss G Diener M Segment-specific effects of epinephrine on ion transport in the colon of the rat Am J Physiol 275 G1367 G1376 1998 9843774 59 Yano JM Indigenous bacteria from the gut microbiota regulate host serotonin biosynthesis Cell 161 264 276 2015 25860609 60 Asano Y Critical role of gut microbiota in the production of biologically active, free catecholamines in the gut lumen of mice Am J Physiol Gastrointest Liver Physiol 303 G1288 G1295 2012 23064760 61 Furness JB Types of neurons in the enteric nervous system J Auton Nerv Syst 81 87 96 2000 10869706 62 Curtis MM Sperandio V A complex relationship: the interaction among symbiotic microbes, invading pathogens, and their mammalian host Mucosal Immunol 4 133 138 2011 21248724 63 Moreira CG Weinshenker D Sperandio V QseC mediates Salmonella enterica serovar Typhimurium virulence in vitro and in vivo Infect Immun 78 914 926 2010 20028809 64 Sperandio V Torres AG Jarvis B Nataro JP Kaper JB Bacteria–host communication: the language of hormones Proc Natl Acad Sci USA 100 8951 8956 2003 This paper describes how signals from the microbiota and host converge to enhance virulence in enteric pathogens 12847292 65 Nakano M Takahashi A Sakai Y Nakaya Y Modulation of pathogenicity with norepinephrine related to the type III secretion system of Vibrio parahaemolyticus J Infect Dis 195 1353 1360 2007 17397007 66 Clarke MB Hughes DT Zhu C Boedeker EC Sperandio V The QseC sensor kinase: a bacterial adrenergic receptor Proc Natl Acad Sci USA 103 10420 10425 2006 16803956 67 Reading NC Rasko DA Torres AG Sperandio V The two-component system QseEF and the membrane protein QseG link adrenergic and stress sensing to bacterial pathogenesis Proc Natl Acad Sci USA 106 5889 5894 2009 19289831 68 Seksik P Alterations of the dominant faecal bacterial groups in patients with Crohn’s disease of the colon Gut 52 237 242 2003 12524406 69 Gophna U Sommerfeld K Gophna S Doolittle WF Veldhuyzen van Zanten SJ Differences between tissue-associated intestinal microfloras of patients with Crohn’s disease and ulcerative colitis J Clin Microbiol 44 4136 4141 2006 16988016 70 Baumgart M Culture independent analysis of ileal mucosa reveals a selective increase in invasive Escherichia coli of novel phylogeny relative to depletion of Clostridiales in Crohn’s disease involving the ileum ISME J 1 403 418 2007 18043660 71 Walker AW High-throughput clone library analysis of the mucosaassociated microbiota reveals dysbiosis and differences between inflamed and non-inflamed regions of the intestine in inflammatory bowel disease BMC Microbiol 11 7 2011 21219646 72 Gevers D The treatment-naive microbiome in new-onset Crohn’s disease Cell Host Microbe 15 382 392 2014 24629344 73 Chiodini RJ Microbial population differentials between mucosal and submucosal intestinal tissues in advanced Crohn’s disease of the ileum PLoS ONE 10 e0134382 2015 26222621 74 Lupp C Host-mediated inflammation disrupts the intestinal microbiota and promotes the overgrowth of Enterobacteriaceae Cell Host Microbe 2 119 129 corrigendum 2, 204 (2007). This paper reported that the induction of inflammation in the host disrupts the microbiota and promotes a bloom of Enterobacteriaceae 75 Garrett WS Enterobacteriaceae act in concert with the gut microbiota to induce spontaneous and maternally transmitted colitis Cell Host Microbe 8 292 300 2010 20833380 76 Raffatellu M Simian immunodeficiency virus-induced mucosal interleukin-17 deficiency promotes Salmonella dissemination from the gut Nature Med 14 421 428 2008 18376406 77 Godinez I T cells help to amplify inflammatory responses induced by Salmonella enterica serotype Typhimurium in the intestinal mucosa Infect Immun 76 2008 2017 2008 18347048 78 Pickard JM Rapid fucosylation of intestinal epithelium sustains host–commensal symbiosis in sickness Nature 514 638 641 2014 This paper showed that cytokines that are induced by enteric infection promote fucosylation in the host 25274297 79 Pham TA Epithelial IL-22RA1-mediated fucosylation promotes intestinal colonization resistance to an opportunistic pathogen Cell Host Microbe 16 504 516 2014 25263220 80 Sonnenburg JL Glycan foraging in vivo by an intestine-adapted bacterial symbiont Science 307 1955 1959 2005 15790854 81 Ansong C Studying Salmonellae and Yersiniae host–pathogen interactions using integrated ’omics and modeling Curr Top Microbiol Immunol 363 21 41 2013 22886542 82 Harper RW Differential regulation of dual NADPH oxidases/peroxidases, Duox1 and Duox2, by Th1 and Th2 cytokines in respiratory tract epithelium FEBS Lett 579 4911 4917 2005 16111680 83 Haberman Y Pediatric Crohn disease patients exhibit specific ileal transcriptome and microbiome signature J Clin Invest 124 3617 3633 2014 25003194 84 Salzman A Induction and activity of nitric oxide synthase in cultured human intestinal epithelial monolayers Am J Physiol 270 G565 G573 1996 8928785 85 Palmer RM Rees DD Ashton DS Moncada S L-arginine is the physiological precursor for the formation of nitric oxide in endothelium-dependent relaxation Biochem Biophys Res Commun 153 1251 1256 1988 3390182 86 Lundberg JO Weitzberg E Lundberg JM Alving K Intragastric nitric oxide production in humans: measurements in expelled air Gut 35 1543 1546 1994 7828969 87 Singer II Expression of inducible nitric oxide synthase and nitrotyrosine in colonic epithelium in inflammatory bowel disease Gastroenterology 111 871 885 1996 8831582 88 Enocksson A Lundberg J Weitzberg E Norrby-Teglund A Svenungsson B Rectal nitric oxide gas and stool cytokine levels during the course of infectious gastroenteritis Clin Diagn Lab Immunol 11 250 254 2004 15013971 89 Bai P Protein tyrosine nitration and poly(ADP-ribose) polymerase activation in N-methyl-N-nitro-N-nitrosoguanidine-treated thymocytes: implication for cytotoxicity Toxicol Lett 170 203 213 2007 17428624 90 Dudhgaonkar SP Tandan SK Kumar D Raviprakash V Kataria M Influence of simultaneous inhibition of cyclooxygenase-2 and inducible nitric oxide synthase in experimental colitis in rats Inflammopharmacology 15 188 195 2007 17943250 91 Winter SE Host-derived nitrate boosts growth of E. coli in the inflamed gut Science 339 708 711 2013 This study revealed that nitrate respiration drives the expansion of commensal E. coli during colitis 23393266 92 Winter SE Baumler AJ Why related bacterial species bloom simultaneously in the gut: principles underlying the ‘like will to like’ concept Cell Microbiol 16 179 184 2014 24286560 93 Rivera-Chávez F Baumler AJ The pyromaniac inside you: Salmonella metabolism in the host gut Annu Rev Microbiol 69 31 48 2015 26002180 94 Galán JE Curtiss R III Cloning and molecular characterization of genes whose products allow Salmonella typhimurium to penetrate tissue culture cells Proc Natl Acad Sci USA 86 6383 6387 1989 2548211 95 Hensel M Simultaneous identification of bacterial virulence genes by negative selection Science 269 400 403 1995 7618105 96 Tsolis RM Adams LG Ficht TA Baumler AJ Contribution of Salmonella typhimurium virulence factors to diarrheal disease in calves Infect Immun 67 4879 4885 1999 10456944 97 Barthel M Pretreatment of mice with streptomycin provides a Salmonella enterica serovar Typhimurium colitis model that allows analysis of both pathogen and host Infect Immun 71 2839 2858 2003 12704158 98 Coburn B Li Y Owen D Vallance BA Finlay BB Salmonella enterica serovar Typhimurium pathogenicity island 2 is necessary for complete virulence in a mouse model of infectious enterocolitis Infect Immun 73 3219 3227 2005 15908346 99 Stecher B Salmonella enterica serovar Typhimurium exploits inflammation to compete with the intestinal microbiota PLoS Biol 5 2177 2189 2007 17760501 100 Lawley TD Host transmission of Salmonella enterica serovar Typhimurium is controlled by virulence factors and indigenous intestinal microbiota Infect Immun 76 403 416 2008 17967858 101 Lopez CA Phage-mediated acquisition of a type III secreted effector protein boosts growth of Salmonella by nitrate respiration mBio 3 e00143 12 2012 22691391 102 Lopez CA Rivera-Chavez F Byndloss MX Baumler AJ The periplasmic nitrate reductase NapABC supports luminal growth of Salmonella enterica serovar Typhimurium during colitis Infect Immun 83 3470 3478 2015 26099579 103 Levitt MD Furne J Springfield J Suarez F DeMaster E Detoxification of hydrogen sulfide and methanethiol in the cecal mucosa J Clin Invest 104 1107 1114 1999 10525049 104 Harris JC Dupont HL Hornick RB Fecal leukocytes in diarrheal illness Ann Intern Med 76 697 703 1972 4554412 105 Loetscher Y Salmonella transiently reside in luminal neutrophils in the inflamed gut PLoS ONE 7 e34812 2012 22493718 106 Winter SE Gut inflammation provides a respiratory electron acceptor for Salmonella Nature 467 426 429 2010 This paper showed that tetrathionate is a unique electron acceptor that is induced through inflammation that promotes expansion of S. Typhimurium 20864996 107 Muller HJ Partial list of biological institutes and biologists doing experimental work in Russia at the present time Science 57 472 473 1923 17783219 108 Raffatellu M Lipocalin-2 resistance confers an advantage to Salmonella enterica serotype Typhimurium for growth and survival in the inflamed intestine Cell Host Microbe 5 476 486 2009 19454351 109 Behnsen J The cytokine IL-22 promotes pathogen colonization by suppressing related commensal bacteria Immunity 40 262 273 2014 24508234 110 Goetz DH The neutrophil lipocalin NGAL is a bacteriostatic agent that interferes with siderophore-mediated iron acquisition Mol Cell 10 1033 1043 2002 12453412 111 Flo TH Lipocalin 2 mediates an innate immune response to bacterial infection by sequestrating iron Nature 432 917 921 2004 15531878 112 Deriu E Probiotic bacteria reduce Salmonella typhimurium intestinal colonization by competing for iron Cell Host Microbe 14 26 37 2013 This paper revealed that probiotic strains of E. coli restrict infection of the gut with S. Typhimurium by competing for sources of iron 23870311 113 Cascales E Colicin biology Microbiol Mol Biol Rev 71 158 229 2007 17347522 114 Guterman SK Colicin B: mode of action and inhibition by enterochelin J Bacteriol 114 1217 1224 1973 4576402 115 Cardelli J Konisky J Isolation and characterization of an Escherichia coli mutant tolerant to colicins Ia and Ib J Bacteriol 119 379 385 1974 4604639 116 Patzer SI Baquero MR Bravo D Moreno F Hantke K The colicin G, H and X determinants encode microcins M and H47, which might utilize the catecholate siderophore receptors FepA, Cir, Fiu and IroN Microbiology 149 2557 2570 2003 12949180 117 Hooper LV Molecular analysis of commensal host–microbial relationships in the intestine Science 291 881 884 2001 This was the first report to show that the microbiota can change the expression of mammalian genes to modulate the immune system of the host 11157169 118 Wu M Genetic determinants of in vivo fitness and diet responsiveness in multiple human gut Bacteroides Science 350 aac5992 2015 26430127 119 Earle KA Quantitative imaging of gut microbiota spatial organization Cell Host Microbe 18 478 488 2015 26439864 120 Bouslimani A Molecular cartography of the human skin surface in 3D Proc Natl Acad Sci USA 112 E2120 E2129 2015 25825778 121 Marcobal A A metabolomic view of how the human gut microbiota impacts the host metabolome using humanized and gnotobiotic mice ISME J 7 1933 1943 2013 23739052 122 Lau W Fischbach MA Osbourn A Sattely ES Key applications of plant metabolic engineering PLoS Biol 12 e1001879 2014 24915445 123 Marcobal A Metabolome progression during early gut microbial colonization of gnotobiotic mice Sci Rep 5 11589 2015 26118551 124 Rath CM Molecular analysis of model gut microbiotas by imaging mass spectrometry and nanodesorption electrospray ionization reveals dietary metabolite transformations Anal Chem 84 9259 9267 2012 This study used imaging mass spectrometry to map differences in the gut metabolic landscape that are promoted by the microbiota 23009651 125 Dorrestein PC Mazmanian SK Knight R Finding the missing links among metabolites, microbes, and the host Immunity 40 824 832 2014 24950202 126 Antunes LC Antivirulence activity of the human gut metabolome mBio 5 e01183 14 2014 25073640 127 Antunes LC Impact of Salmonella infection on host hormone metabolism revealed by metabolomics Infect Immun 79 1759 1769 2011 21321075 128 Buffie CG Precision microbiome reconstitution restores bile acid mediated resistance to Clostridium difficile Nature 517 205 208 2015 This paper showed that the introduction of a single species of bacteria that can produce secondary bile salts confers the host with resistance to the expansion of C. difficile after treatment with antibiotics 25337874 129 Koenigsknecht MJ Young VB Faecal microbiota transplantation for the treatment of recurrent Clostridium difficile infection: current promise and future needs Curr Opin Gastroenterol 29 628 632 2013 24100717 130 Sekirov I Antibiotic-induced perturbations of the intestinal microbiota alter host susceptibility to enteric infection Infect Immun 76 4726 4736 2008 18678663 131 Louis E Libioulle C Reenaers C Belaiche J Georges M Genomics of inflammatory bowel diseases: basis for a new molecular classification and new therapeutic strategies of these diseases [in French] Rev Med Liege 64 S1 24 28 2009 132 Meynell GG Antibacterial mechanisms of the mouse gut. II. The role of Eh and volatile fatty acids in the normal gut Br J Exp Pathol 44 209 219 1963 13935361 133 Donohoe DR Wali A Brylawski BP Bultman SJ Microbial regulation of glucose metabolism and cell-cycle progression in mammalian colonocytes PLoS ONE 7 e46589 2012 23029553 134 Atarashi K Umesaki Y Honda K Microbiotal influence on T cell subset development Semin Immunol 23 146 153 2011 21292500 135 Atarashi K Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota Nature 500 232 236 2013 23842501 136 Furusawa Y Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells Nature 504 446 450 2013 24226770 137 Spees AM Streptomycin-induced inflammation enhances Escherichia coli gut colonization through nitrate respiration mBio 4 e00430 13 2013 138 Itoh K Freter R Control of Escherichia coli populations by a combination of indigenous clostridia and lactobacilli in gnotobiotic mice and continuous-flow cultures Infect Immun 57 559 565 1989 2643576 139 Wells JE Hylemon PB Identification and characterization of a bile acid 7α-dehydroxylation operon in Clostridium sp. strain TO-931, a highly active 7α-dehydroxylating strain isolated from human feces Appl Environ Microbiol 66 1107 1113 2000 10698778 140 Wilson KH Efficiency of various bile salt preparations for stimulation of Clostridium difficile spore germination J Clin Microbiol 18 1017 1019 1983 6630458 141 Sorg JA Sonenshein AL Bile salts and glycine as cogerminants for Clostridium difficile spores J Bacteriol 190 2505 2512 2008 18245298 142 Theriot CM Antibiotic-induced shifts in the mouse gut microbiome and metabolome increase susceptibility to Clostridium difficile infection Nature Commun 5 3114 2014 24445449 143 Weingarden AR Microbiota transplantation restores normal fecal bile acid composition in recurrent Clostridium difficile infection Am J Physiol Gastrointest Liver Physiol 306 G310 G319 2014 24284963 144 Bernstein H Bernstein C Payne CM Dvorakova K Garewal H Bile acids as carcinogens in human gastrointestinal cancers Mutat Res 589 47 65 2005 15652226

PMC005xxxxxx/PMC5114884.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985117R 4816 J Immunol J. Immunol. Journal of immunology (Baltimore, Md. : 1950) 0022-1767 1550-6606 27815424 5114884 10.4049/jimmunol.1601279 EMS70109 Article Activating KIR2DS4 is expressed by uterine NK cells and contributes to successful pregnancy1 Kennedy Philippa R. *#¶ Chazara Olympe *# Gardner Lucy *# Ivarsson Martin A. *# Farrell Lydia E. *# Xiong Shiqiu *#║ Hiby Susan E. *# Colucci Francesco #§ Sharkey Andrew M. *# Moffett Ashley *# * Department of Pathology, University of Cambridge, Cambridge CB2 1QP, United Kingdom # Centre for Trophoblast Research, University of Cambridge, Downing Street, Cambridge CB2 3EG, United Kingdom ¶ Manchester Collaborative Centre for Inflammation Research, University of Manchester, 46 Grafton Street, Manchester M13 9NT, United Kingdom ║ Department of Molecular and Cell Biology, University of Leicester, Leicester LE1 7RH, United Kingdom § Department of Obstetrics and Gynaecology, University of Cambridge School of Clinical Medicine, NIHR Cambridge Biomedical Research Centre, Addenbrooke’s Hospital, Box 111, Hills Road, Cambridge CB2 0SP, United Kingdom. Phone Number +441612751536, Fax Number +441612755082, philippa.kennedy@manchester.ac.uk, am485@cam.ac.uk 16 11 2016 4 11 2016 1 12 2016 01 6 2017 197 11 42924300 This file is available to download for the purposes of text mining, consistent with the principles of UK copyright law. Tissue-specific Natural Killer (NK) cells are abundant in the pregnant uterus and interact with invading placental trophoblast cells that transform the maternal arteries to increase the feto-placental blood supply. Genetic case-control studies have implicated Killer-cell Immunoglobulin-like Receptor (KIR)2 genes and their Human Leukocyte Antigen (HLA) ligands in pregnancy disorders characterized by failure of trophoblast arterial transformation. Activating KIR2DS1 or KIR2DS5 (when located in the centromeric region as in Africans) lower the risk of disorders when there is a fetal HLA-C allele carrying a C2 epitope. Here, we investigate another activating KIR, KIR2DS4, and provide genetic evidence for a similar effect when carried with KIR2DS1. KIR2DS4 is expressed by ~45% of uterine NK cells (uNK)3. Similarly to KIR2DS1, triggering of KIR2DS4 on uNK led to secretion of GM-CSF and other chemokines, known to promote placental trophoblast invasion. In addition, XCL1 and CCL1, identified in a screen of 120 different cytokines, were consistently secreted upon activation of KIR2DS4 on uNK. Inhibitory KIR2DL5A, carried in linkage disequilibrium with KIR2DS1, is expressed by peripheral blood NK cells (pbNK)4 but not by uNK cells, highlighting the unique phenotype of uNK compared to pbNK. That KIR2DS4, KIR2DS1 and some alleles of KIR2DS5 contribute to successful pregnancy suggests that activation of uNK by KIR binding to HLA-C is a generic mechanism promoting trophoblast invasion into the decidua. Introduction Natural Killer (NK) cells use a combination of activating and inhibitory receptors to recognise viruses and cancerous cells (1). That the same receptors are also used to recognise fetal cells by tissue-specific uterine NK cells (2) indicates two strong contrasting evolutionary pressures: disease resistance and successful reproduction, both showing evidence of balancing selection (3, 4). NK cells in the pregnant uterus, decidual NK cells (dNK)5, are different phenotypically and functionally from peripheral blood NK cells (pbNK)6 (5–10). Evidence from genetic and functional studies suggests that dNK regulate trophoblast transformation of the uterine spiral arteries necessary for increasing the blood supply to the feto-placental unit until the end of gestation (11–14). The NK cell receptors particularly implicated in reproductive health are the highly polymorphic Killer-Cell Immunoglobulin-like Receptor family (KIR)7 (15). A KIR genotype is made up of two KIR haplotypes that can differ by both gene content and allelic variation. The genes in these haplotypes are so densely clustered on chromosome 19 that they are generally inherited as haplotypic centromeric and telomeric blocks (16, 17) (Figure 1A). The dominant ligands for KIR are HLA-C allotypes. All individuals have KIR that will bind to HLA-C allotypes as two groups depending on the C1 or C2 epitope that they bear. There is an increased risk of pregnancy disorders with certain inhibitory maternal KIR and fetal HLA-C combinations. Case-control genetic studies of Europeans have shown that pregnancy disorders that result from defective placentation with inadequate trophoblast arterial transformation (eg pre-eclampsia, fetal growth restriction, FGR and recurrent miscarriage, RM) are linked to an absence of the telomeric-B (Tel-B)8 KIR region in the mother (Figure 1A) and the presence of paternal C2 in the fetus (13, 18, 19). In contrast, pregnancies resulting in babies with increased birth weights are also associated with the presence of a paternal C2 allele in the fetus, but with a maternal Tel-B KIR region (20). The tight linkage disequilibrium (LD)9 of KIR makes it difficult to determine through genetic studies alone which gene is responsible, so functional studies are required to complement this work. Of the KIR in the Tel-B region, activating KIR2DS1 is the most likely candidate for enhancing placentation, because it can bind to C2 allotypes. The inhibitory counterpart, KIR2DL1, also binds strongly to C2 allotypes, is present in the centromeric-A (Cen-A)10 and some centromeric-B (Cen-B)11 regions and is carried by ~98% of individuals. Therefore, in the absence of KIR2DS1 (55-60% of Europeans), the dominant effect of paternal trophoblast C2 allotypes interacting with dNK is inhibition. Ligation of KIR2DS1 on dNK induces production of cytokines and chemokines, such as GM-CSF, which can induce trophoblast migration (12). Thus, our current model of pregnancy indicates that when C2 allotypes derived from the father are expressed by trophoblast, KIR2DS1 activates dNK to secrete cytokines that encourage deeper invasion of the uterus by trophoblast, promote spiral artery remodelling and a better blood supply for the fetus (2). In the absence of KIR2DS1, insufficient activation of dNK results in poor trophoblast invasion, placental stress, growth restriction of the fetus and pre-eclampsia. In a similar Ugandan case-control study, we found no protective effect for pre-eclampsia of the Tel-B region including KIR2DS1 (carried by ~20% of control women). Instead, certain alleles of an activating KIR, KIR2DS5, present in Cen-B were more frequent in controls compared to pre-eclamptic pregnancies (21). KIR2DS5 is always located in the Tel-B region in non-African populations and is carried in tight LD with KIR2DS1. It thus could contribute to the protective effect of Tel-B in Europeans but whether it is expressed or binds C2 allotypes is still controversial. In addition KIR2DS1 and KIR2DS5, KIR2DL5A is also present in Tel-B and remains an enigmatic KIR in terms of ligands and functions (22). Other activating KIR that might recognise ligands on trophoblast and influence pregnancy outcome include KIR3DS1 and KIR2DS2-4. KIR3DS1, in LD with KIR2DS1, binds HLA-B allotypes carrying the Bw4 motif (23), but HLA-B molecules are never expressed by trophoblast (24, 25). KIR2DS2 is predicted to bind the C1 motif through homology with KIR2DL2/3; the presence of fetal C1 alone is always neutral in our genetic case control studies. KIR2DS3 is not expressed at the cell surface (26). This leaves KIR2DS4, present in the Tel-A region, that occurs either as a truncated (KIR2DS4del) (alleles *003/004/006 are carried by ~80% of Europeans) or full length (KIR2DS4wt) form (allele *001 is carried by ~35% of Europeans). KIR2DS4del has a 22bp deletion which introduces a frameshift mutation that results in a soluble protein with only one intact Ig-like domain (27). While KIR2DS4wt has been reported to bind some HLA-C alleles carrying both the C1 and C2 epitopes, soluble KIR2DS4del does not bind HLA class I molecules (28). We previously found a negative association of KIR2DS4del with pregnancy outcome, but no positive effect of KIR2DS4wt (13). Here, in order to investigate the role of KIR other than KIR2DS1 in successful pregnancy, we have studied the expression and function of KIR2DS4 and KIR2DL5 on dNK cells. From this we demonstrate that activation of dNK is a general mechanism that is beneficial to pregnancy. Materials and Methods Primary tissue Tissue and matched peripheral blood samples were obtained from women undergoing elective terminations in the first trimester of pregnancy; blood was also obtained from healthy volunteers. Both sets of patients gave informed consent. Ethical approval for the use of these tissues was obtained from the Cambridge Local Research Ethics Committee (REC 04/Q0108/23). Leukocytes and placental samples were isolated as previously described (29). Cell lines Cell lines transfected with cDNA for single KIR were used to test antibody specificities. KIR2DL1+, KIR2DL3+, KIR2DS1+, KIR2DS2+, KIR2DS4+ (30) or KIR3DS1+ (31) BWZ cells were the gift of Eric Vivier. KIR2DL2+, KIR2DS5+, KIR3DL1+ (31) or KIR3DL3+ (32) BA/F3 cells were the gift of Chiwen Chang as was cDNA for KIR2DL5 used to transiently transfect HEK293T cells. KIR2DL4+ Jurkat cells were the gift of Kerry Campbell. Paul Norman supplied cDNA of KIR3DL2 for transient transfection into HEK293T cells. Flow cytometry dNK were gated on as live, CD9+ CD56+ cells. pbNK were gated on as live CD56+ CD3- cells. The following antibodies were used: LIVE/DEAD discriminator (LifeTechnologies), CD9 (SN4 or M-L13 from eBioscience or Becton Dickinson, respectively), CD56 (HCD56 from Biolegend) and CD3 (SK7) from Becton Dickinson. Fibroblasts and macrophages were identified using CD10 (HI10a from Biolegend) and CD14 antibodies (MφP9 and HCD14 from BD Pharmingen and Biolegend), respectively. The following antibodies were used to stain KIR: UPR1 (KIR2DL5) from Biolegend and Carlos Vilches (33); 179315 (KIR2DS4), 143211 (KIR2DL1) and 181703 (KIR2DL4) from R&D Systems; FES172 (KIR2DS4) and EB6 (KIR2DL1/S1) from Beckman Coulter; CHL (KIR2DL2/3/S2) from BD Pharmingen, DX9 (KIR3DL1) from Biolegend; NKVFS1 (KIR2DL1/2/3/S1/2/4) from Abcam; 5.133 (KIR3DL2) from Marco Colonna (34); and FLAG antibodies (Sigma Aldrich). Intracellular staining was performed according to manufacturers’ instructions with antibodies against Ki647, (BD Pharmingen), CCL3 (R&D Systems) and GM-CSF (BD Biosciences). Functional assays Purified NK cells (CD56-positive selection using magnetic beads, Miltenyi Biotec) or mixed decidual mononuclear cells were stimulated with plate-bound anti-KIR2DS4 (179315) antibodies or an isotype control for 4-48 hours. After this time supernatants were removed (spun at 500g for 5min to remove cellular contaminants) or stimulated cells were mechanically dislodged. Supernatants were analysed using a chip-based fluorescence-linked immunosorbent assay (Human Cytokine Antibody Array G Series 1000, RayBioTech) or a standard ELISA for CCL1 and XCL1 (DuoSets, R&D Systems). Cells activated cells in the presence of monensin and brefeldin-A for 5h were analysed for surface expression of CD107a (H4A3, BD Pharmingen) or the intracellular cytokines listed above. Immunohistochemistry Paraffin sections of decidual implantation sites were heat-treated in 0.1M citrate buffer for 20min at 99.5°C. Slides were left in hot buffer for a further 20min for antigen unmasking. Anti-XCR1 (191704 from R&D Systems) was stained in Tris-buffered saline with 0.1% Tween for 45min. The staining was detected with goat anti-rabbit IgG-biotin and avidin-biotin-HRP complexes (Vector Laboratories). Genetic typing The case-control cohort analysed here has previously been described (13). KIR and HLA-C1/2 genetic typing of new patient samples was performed as in this previous study. Two-digit HLA-C typing was performed by the Tissue Typing facility at Addenbrookes Hospital, Cambridge. Statistics Statistical tests were carried out using the computational site http://vassarstats.net/; the statistical packages within GraphPad Prism v6 (GraphPad Software, California); the Real Statistics Resource Pack for Excel 2010 http://www.real-statistics.com/ ; and PLINK (version 1.07) http://pngu.mgh.harvard.edu/purcell/plink/ (35). The product rule was calculated by multiplying the observed frequency of individual receptors to generate the expected frequency of double positive receptors: pExp(AB) = pObs(A).pObs(B). The following genetics tests were performed: Chi-squared test and Fisher Exact test with two-tailed mid p adjustment; Breslow-Day test; Cochran-Mantel-Haenszel test. Results KIR2DS4wt in epistasis with KIR2DS1 is associated with a lower risk of pre-eclampsia KIR2DS4wt, the full length form of activating KIR2DS4, is potentially important in pregnancy as it can bind to some HLA-C allotypes (28, 36). Indeed, we previously found in a case-control cohort of European women that KIR2DS4del associates with increased risk of pregnancy disorders (13). The presence of KIR2DS4wt was neutral in this analysis. However, we only considered presence/absence of this gene and did not consider the effect of both KIR telomeric regions that make up the women’s genotypes. There are three possible regions: Tel-A containing KIR2DS4wt, Tel-A containing KIR2DS4del and Tel-B containing KIR2DS1 (Figure 1A) that provides a strong protective effect (13). Here, therefore, we re-analysed this dataset for the effect of KIR2DS4wt, now controlling for the clear protective effect of KIR2DS1. Indeed, the presence/absence of KIR2DS1 does alter the effect of KIR2DS4wt, indicative of epistasis (Breslow-Day test, p=0.003). KIR2DS4wt is protective compared to KIR2DS4del in KIR2DS1+ women (p=5.7x10-4, OR=0.59)(Figure 1B). This effect is not found in the absence of KIR2DS1 (p=0.83, OR=1.0). This indicates that women who carry both KIR2DS4wt and KIR2DS1 are further protected against disorders of pregnancy affecting placentation (p=6.8x10-5, OR=0.45)(Figure 1C). Because of the similar functions and overlapping ligands of KIR2DS1 and KIR2DS4, it is likely that the epistasis detected at the statistical level reflects a biological interaction. KIR2DS4 is expressed by a large proportion of both pbNK and dNK Two mAbs (FES172 and 179315) were tested to confirm specificity against KIR2DS4 on cell lines expressing single KIR (Supplemental Figure 1). The frequency of KIR2DS4+ CD56+ cells is high in both dNK and pbNK populations (Figure 2A-C). In contrast, both KIR2DS1 and KIR2DL1 have increased frequency of expression in dNK compared to pbNK (12, 37, 38) and so, in accordance with the product rule, there is a higher frequency of dNK co-expressing these KIR than for pbNK (12). This means that the proportion of cells co-expressing KIR2DS4 and other KIR is probably different for dNK and pbNK. We chose to look at the distribution of KIR2DS4 relative to KIR2DL1, because KIR2DL1 is carried by almost all donors allowing us to analyse KIR co-expression with sufficient statistical power. KIR2DL1 is also critical to our model of pregnancy disorders as it is strongly inhibitory for HLA-C allotypes bearing C2 epitopes. Our findings (Figure 2D-E) show that in pbNK, most KIR2DS4+ cells lack KIR2DL1 (Figure 2E, mid grey segment), but in dNK, most KIR2DS4+ cells co-express KIR2DL1 (Figure 2E, dark grey segment). This increased co-expression obeys the product rule (Supplemental Figure 2A), suggesting it reflects the combined frequency of KIR2DL1 and KIR2DS4. In line with this, Ki-67 staining shows that KIR+ dNK proliferate more than KIR- cells, but there is no preferential proliferation by KIR2DS4+ KIR2DL1+ cells compared to single positive cells (Supplemental Figure 2B). Therefore, in donors carrying KIR2DS4wt, a large proportion of dNK co-express KIR2DS4 with other KIR that have the potential to modulate its function. Using KIR Fc-fusion proteins, KIR2DS4 binds and responds to certain HLA-C alleles carrying both C1 and C2 epitopes (28, 36). Binding of KIR2DS4 on dNK to trophoblast HLA-C ligands might affect the frequency of KIR2DS4+ cells, but we find no difference in the proportion of dNK expressing KIR2DS4 when the mother or fetus carries its ligands (Supplemental Figure 3A-B). There is a suggestion that allogeneic ligands affect KIR2DS4 expression as there are fewer dNK expressing KIR2DS4 when the fetus alone carries a ligand, compared to the mother alone (Supplemental Figure 3C). Given that KIR2DS4wt is protective in genetic case-control studies only in the presence of KIR2DS1, and that both are mutually exclusive on a KIR haplotype, protected individuals must have one copy of each gene. Therefore, we analysed the effect of KIR2DS4wt copy number on frequency of expression: as two copies; as one copy in the presence of KIR2DS4del; or as one copy in the presence of KIR2DS1. KIR2DS4 frequency on dNK is similar in these different genetic backgrounds, suggesting that an altered frequency of KIR2DS4+ dNK in KIR2DS1+ KIR2DS4+ individuals is not the mechanism by which KIR2DS4 provides protection against pregnancy disorders (Supplemental Figure 3D). KIR2DS4 activation on dNK induces cytokine responses HLA-C ligands for KIR2DS4 are shared with other NK receptors on dNK. To investigate the functional consequences specific to activation of KIR2DS4 alone we used cross-linking with a specific mAb. Decidual NK cells are poor killers, as measured by chromium-release assays (6, 9, 39), but CD107a degranulation does occur in the presence of low dose IL-15 (40) and offers a reproducible assay to quantify dNK activation. We find that degranulation of both pbNK and dNK occurs in response to increasing concentrations of anti-KIR2DS4 (Scheirer-Ray-Hare modification of Kruskal-Wallis test, effect of mAb concentration p=4.1x10-10)(Figure 3). Since cytokine responses are more physiologically relevant to human pregnancy than degranulation (12, 41), we next analysed the cytokines produced following KIR2DS4 stimulation of dNK using a semi-quantitative screen of 120 cytokines (Supplemental Table I). Mixed decidual mononuclear cells were co-cultured in wells coated with anti-KIR2DS4 or control IgG antibody so that contact with stromal cells is maintained, as this improves viability. We identified 8 candidates that were upregulated >1.25 fold in at least 1 out of 4 donors tested (Figure 4A-B). After cross-linking with anti-KIR2DS4, flow cytometry (GM-CSF and CCL3) or ELISA (XCL1 and CCL1) assays were used to validate 4 of these 8 cytokines (Figure 4C-F). The percentage of dNK positive for intracellular GM-CSF and the median fluorescence intensity for CCL3 increases (p<0.05)(Figure 4C-E) and secretion assayed by ELISA for both XCL1(p<0.01) and CCL1 (p<0.05) is augmented (Figure 4F). In summary, stimulation of KIR2DS4 on dNK cells triggers the release of cytokines, many of which are related to the cytokines upregulated at the mRNA level by dNK cells upon KIR2DS1 activation (XCL2, CCL3L3, GM-CSF, IFNG) (12), although this is the first time XCL1 and CCL1 have been identified as secreted by dNK cells in response to activating signals. Trophoblast and maternal decidual cells express receptors for newly-identified cytokines produced by activated dNK Recently we have shown that GM-CSF induces migration of human primary trophoblast cells (12). CCL3 production by decidual and trophoblast cells may attract NK cells, as well as monocytes and T cells, which all bear receptors for this cytokine (42, 43). Receptors for chemokines XCL1 and CCL1 have not been described on cells at the site of placentation. We therefore stained sections of decidua and placenta and cell isolates by flow cytometry for these receptors. Several cell types within the placenta, including fetal endothelial cells, villous trophoblast and invasive EVT express XCR1, the receptor for XCL1 (Figure 5). Within the dNK-rich decidua, XCR1 is found on cells with branching processes, (Figure 5B) identified by flow cytometry as a small proportion of the CD14+ macrophage population (Figure 5D-E). CCR8, the receptor for CCL1, is expressed by all decidual macrophages and a small proportion of dNK (Figure 5F). KIR2DL5 – the only inhibitory receptor in the Tel-B region – is not expressed on the surface of dNK In the Tel-B region, KIR2DS1 is in LD with KIR2DL5A, which codes for an orphan inhibitory receptor. To determine whether KIR2DL5A affects dNK activation and pregnancy outcome, we first looked for expression of KIR2DL5A in dNK. KIR2DL5 alleles are also found in the Cen-B region, where they are known as KIR2DL5B (Figure 1A). To distinguish between these alleles we used antibody UPR1, which binds the most common KIR2DL5A allele in Europeans, KIR2DL5A*001 (~30% Europeans) (33), but not KIR2DL5A*005 (~8% Europeans) or KIR2DL5B (~20-40% Europeans) (22). UPR1 binds KIR2DL5 and not other KIR using KIR-negative cell lines transfected with single KIR genes (Supplemental Figure 1). In donors where there were detectable KIR2DL5+ pbNK, there was no surface expression of KIR2DL5 on dNK (Figure 6A-B). The donors who expressed KIR2DL5 in blood, always also carry KIR2DS5 (Figure 6C), which is in LD with KIR2DS1 and KIR2DL5*001 in the Tel-B haplotype in Europeans, suggesting we might only detect Tel-B KIR2DL5*001 and not other KIR2DL5 allotypes. The absence of KIR2DL5 surface expression on dNK means it is unlikely to be affecting the activity of dNK and so the protective effect of the Tel-B region is due to activating KIR alone. Discussion We have shown that KIR2DS4wt is associated with lower risk of pregnancy disorders in the presence of KIR2DS1 – a synergistic interaction. KIR2DS4wt has been linked to higher viral load and increased transmission of HIV infection (44–47), and with clinical outcomes in arthritis (48–50), cancer (51) and allogeneic cell transplantation (52). Because KIR are in tight LD and there are confounding effects of genes on alternative haplotypes, it can be difficult to determine which particular KIR has a role in disease (53). Here we have ruled out the effect of alternative haplotypes by stratifying the cohort according to the presence of Tel-B. A clear biological rationale for a particular KIR’s involvement can also help distinguish the effect of KIR in tight LD. KIR2DS4wt is in LD with KIR3DL1 alleles, but in the context of pregnancy, the ligand for KIR3DL1, HLA-Bw4, is not expressed by trophoblast. HLA-Bw4 can be expressed by stromal cells, so it is possible it modulates uNK activity. Nevertheless, KIR2DS4, known to bind to certain HLA-C allotypes expressed on trophoblast, is the likely candidate for the protective effect. To explain how a particular KIR could affect trophoblast migration, the candidate KIR needs to be expressed by NK cells in contact with EVT in the decidua. KIR2DS4 is expressed by a large proportion of dNK and its frequency of expression follows the product rule of co-expression with other KIR. Co-expression of KIR is relevant because the balance of activating and inhibitory signals within the cell determines activation of NK cells. Like KIR2DL1, KIR2DS1 has increased frequency of expression in dNK compared to pbNK. KIR2DS4wt could swing the balance in favour of activation when co-expressed with KIR2DS1 or in the context of activating cytokines, but may fail to activate dNK in the absence of another activating receptor. Indeed, KIR2DS1 may also require the presence of another activating receptor to have measurable effects on population genetics, but unlike KIR2DS4wt, KIR2DS1 is in LD with two other activating receptors in Europeans. There is precedence for co-operation of activating KIR from pregnancy and allogeneic hematopoietic stem cell transplantation, where cumulative Cen-B and Tel-B haplotypes that carry multiple activating KIR, contribute to increasing beneficial effects (18, 54). Indeed, our finding that certain centromeric alleles of another different activating KIR, KIR2DS5, is protective against pre-eclampsia in Ugandans (21) supports this model. There is still limited evidence that KIR2DS4 responds to HLA-C molecules, but our preliminary findings suggest that the size of the KIR2DS4+ dNK subset is smaller when KIR2DS4 ligands are present in the fetus, but not the mother. This observation supports the hypothesis that KIR2DS4 does bind these HLA-C ligands on trophoblast. Why should KIR2DS4 act as a co-receptor in this way, requiring the presence of another activating receptor? The mechanism of this co-operation remains unclear, but we can exclude some factors. Firstly, we have shown that the frequency of expression of KIR2DS4wt on dNK is unaffected by the presence of Tel-B or Tel-A on the women’s other haplotype. Therefore higher frequency of expression of KIR2DS4wt in the presence of KIR2DS1 cannot be the mechanism by which epistasis is achieved. Similarly, there is only one prevalent allele of KIR2DS1 and functional KIR2DS4 amongst Europeans (KIR2DS1*002 and KIR2DS4*001), so allelic variation on particular haplotypes is unlikely to affect the association in our European case-control cohorts. One reason for the dependence of KIR2DS4wt on the presence of KIR2DS1 could be the nature of its interaction with HLA-C molecules. Whilst there are functional responses of KIR2DS1+ NK cells upon interaction with HLA-C alleles carrying C2 epitopes ex vivo (12, 55, 56), similar responses of KIR2DS4+ NK cells have only been demonstrated for HLA-C*0401 (36) and HLA-A*1102 (28). The interaction of KIR2DS4 with HLA-C could be of lower avidity than that of KIR2DS1; KIR2DS4 recognition of HLA-C allotypes might be peptide dependent, as has been shown for KIR3DS1 (23); or KIR2DS4 may be interacting with open conformers of HLA molecules (57) expressed by trophoblast. All these factors could affect the way KIR2DS4 binds to HLA-C molecules on trophoblast. Specialised functions for pbNK and dNK are likely to have arisen because of the conflicting demands of disease resistance and reproductive success (3). When trying to assess the impact of KIR in health and disease, it is necessary, therefore, to study these receptors in the species and tissue of interest. Upon triggering of KIR2DS4 with specific antibodies, dNK degranulate and secrete cytokines, such as GM-CSF, that are known to have direct effects on trophoblast migration, and other cytokines (XCL1, CCL1 and CCL3) that have the potential to directly impact trophoblast and other cells in the decidua, including decidual macrophages. Recently, KIR2DS4 has been highlighted for promoting trogocytosis (58), a process that has been implicated in dNK acquisition of HLA-G from trophoblast (59). There may be several mechanisms, therefore by which triggering of dNK could aid placentation. The view that immune cells must be suppressed for successful pregnancy – both locally in the uterus and systemically - originated with Medawar and the birth of transplant biology (60). There is now mounting evidence that for uterine NK cells this is not correct. We show that activation of dNK cells through KIR2DS4wt provides help to trophoblast migration and the establishment of pregnancy. Perhaps KIR2DS5 in the Cen-B may protect Africans from pre-eclampsia in the same way (21). Moreover, we find here that inhibitory receptor KIR2DL5A in the protective Tel-B region is not expressed by dNK, suggesting it does not affect pregnancy outcome. Together these data support a model of generic activation of dNK cells counter-acting strong inhibition by KIR2DL1 and benefitting pregnancy. Supplementary Material Supplementary material Acknowledgements We would like to thank all the donors and research nurses for providing samples. We thank Carlos Vilches, Hugo Hilton, Paul Norman, Peter Parham, Eric Vivier, Chiwen Chang, John Trowsdale, Kerry Campbell, Michela Falco and Massimo Vitale for reagents relating to this study, as well as Sarah Peacock and Craig Taylor of the Tissue Typing facility at Addenbrookes Hospital Cambridge. We thank Daniel Davis for helpful comments on the manuscript. Figure 1 KIR2DS4wt in epistasis with KIR2DS1 is associated with a lower risk of pregnancy disorders. A. The LD blocks that make up more than 94% of Caucasian KIR genotypes (17). An individual’s KIR genotype contains two haplotypes, each with one centromeric (left) and one telomeric (right) block. These blocks contain activating (white) and inhibitory (black) genes in LD. Framework genes (grey) are found in all haplotypes. The three most common telomeric blocks contain either KIR2DS4wt, KIR2DS4del or KIR2DS1. B. Women were stratified according to the presence or absence of the protective gene KIR2DS1 as a Breslow-Day test indicated epistasis between KIR2DS1 and KIR2DS4wt. The carrier frequency of KIR2DS4wt was then compared between women with affected pregnancies and healthy control pregnancies within each subgroup. The presence of KIR2DS4wt was protective, Cochran-Mantel-Haenszel test p = 5.7x10-4, OR 0.59. C. Then, within the women carrying KIR2DS1, it is the double positive KIR2DS1+ KIR2DS4wt+ that are the most protected, p = 6.78x10-5, OR = 0.45. Figure 2 KIR2DS4 is expressed by a large proportion of both pbNK and dNK. A. Flow cytometry plots from a typical donor showing the gating strategy for pbNK and dNK. B. Flow cytometry plots showing KIR2DS4 and KIR2DL1 staining on pbNK and dNK from a representative donor. The percentage of cells in each quadrant is shown. C. The proportion of KIR2DS4+ cells was compared for pbNK and dNK in matched donors (n=22). The proportion of KIR2DL1+ (n=41) and KIR2DS1+ (n=11) NK cells is shown for comparison only. Data points for KIR2DS1 and some KIR2DL1 (shown in grey) are already published (12) and are reproduced with permission from the Journal of Clinical Investigation. D-E. The proportion of NK cells from each KIR+ subset (single positive (sp), double positive (dp) or double negative (dn) for the receptors) was compared for pbNK and dNK cells. D. Values for individual donors. Black lines represent donors, red lines represent the median, *** p<0.001 Wilcoxon Signed Rank Test. E. The mean values for each subset displayed as a pie chart. Figure 3 KIR2DS4 is functional on dNK cells. pbNK and dNK cells from KIR2DS4+ donors were incubated in wells coated with anti-KIR2DS4 or an isotype control for 4 hours in the presence of monensin. A. dNK cells from a represenative donor, gated as in Figure 2, are shown stained for KIR2DS4 and CD107a following activation with plate-bound antibody (anti-KIR2DS4 or an isotype control). B. The percentage of KIR2DS4+ NK cells positive for CD107a upon activation was calculated by subtracting the %CD107a+ when cells were cross-linked with IgG. The extent of degranulation for a range of antibody concentrations is shown. pbNK and dNK cells were not from the same donor. Scheirer-Ray-Hare modification of Kruskal-Wallis test, effect of concentration p=4.1x10-10; effect of cell type p=0.24; effect of interaction p=0.87. Bars represent medians and interquartile ranges. Figure 4 Cytokine secretion by dNK cells in response to KIR2DS4 activation. A-B. A semi-quantitative fluorescent chip-based sandwich ELISA was used to screen for 120 cytokines in supernatants taken from mixed decidual leukocytes of KIR2DS4+ donors (see Supplemental Table 1). Leukocytes were cultured on antibody-coated plastic for 12-24 hours, where the only cells to express KIR2DS4 were the dNK cells. Fluorescent spots for cytokines of interest are highlighted in A. The cropped regions of interest are taken from different chips and different donors. They are grouped according to whether they show greater than 1.5-fold increase in secretion on average across all donors (‘Increase’); secretion was already high within the isotype control stimulation, so the screen was insensitive (‘Ambiguous’); and control spots (‘Control’). B. The cytokines that were upregulated more than 1.25-fold upon KIR2DS4 activation in at least one of four donors tested are listed in the table. The mean fold change across all four donors in shown to the right. C-E. Mixed decidual leukocytes were cultured on plastic coated with either anti-KIR2DS4 antibody or an isotype control (IgG2a) in the presence of monensin and brefeldin A. After 5h, cells were fixed and live CD56+ CD9+ KIR2DS4+ dNK cells were identified by flow cytometry (C). Although KIR2DS4 expression reduced upon cross-linking (C) the majority retained KIR2DS4 expression. These KIR2DS4+ dNK were assessed for intracellular cytokines: (D) GM-CSF (n=7) and (E) CCL3 (n=7). F. Where antibodies for flow cytometry were not available, purified dNK cells were cultured on antibody-coated plastic for 12-48 hours and the production of CCL1 and XCL1 (n=8) were detected in supernatants by commercial sandwich ELISA. Colour-coded according to donor. Wilcoxon signed rank test, * p<0.05, ** p<0.01. Figure 5 Receptors for XCL1 and CCL1 on placenta and in the pregnant uterus. A. XCR1, the receptor for XCL1, was identified on trophoblast through immunohistochemical staining of sections of placenta invading the pregnant uterus. VT villous trophoblast, EVT extravillous trophoblast. B. Within the maternal compartment, XCR1 was identified on individual cells with branching processes often adjacent to vessels. C. Isotype control staining of trophoblast (left panel) invading the uterus (right panel). D. Chemokine receptor expression on live CD14+ decidual macrophages was confirmed by flow cytometry for (E) XCR1 (n=6) and (F) CCR8 (n=5). A population of cells that did not express the receptor (dNK for XCR1 and fibroblasts for CCR8, since some dNK express low amounts of CCR8) are shown for comparison. Figure 6 KIR2DL5 is detected by flow cytometry on the surface of pbNK cells, but not dNK cells. A. Flow cytometry plots from a typical donor showing matched pbNK and dNK, gated as in Figure 2, stained for KIR2DL5 (mAb UPR1) or an isotype control (IgG1). B. The frequency of the KIR2DL5+ population was defined as (%UPR1+) - (%IgG1+). The frequency of this population was measured for all pbNK and dNK cells from donors where there was UPR1+ staining in blood (n=8). For comparison, staining of KIR2DL5- donors is shown alongside (n=7). Each line represents one donor. Wilcoxon signed rank test. * p<0.05 C. All donors that carried KIR2DL5 according to SSP-PCR were assessed for KIR2DL5 staining, but a positive KIR2DL5+ subset was only seen as part of the genotype that carried KIR2DS5 alongside KIR2DL5. Lines show the median. Each dot represents one donor (n=22). 1 This research was supported by the Wellcome Trust, The Centre for Trophoblast Research, The British Heart Foundation and the Cambridge Philosophical Society. 2 Killer-cell Immunoglobulin-like Receptor 3 Uterine NK cell 4 Peripheral blood NK cell 5 Decidual NK cell 6 Peripheral blood NK cell 7 Killer-Cell Immunoglobulin-like Receptor 8 Telomeric-B 9 Linkage disequilibrium 10 Centromeric-A 11 Centromeric-B 1 Lanier LL NK cell recognition Annu Rev Immunol 2005 23 225 74 15771571 2 Moffett-King A Natural killer cells and pregnancy Nat Rev Immunol 2002 2 656 663 12209134 3 Parham P Moffett A Variable NK cell receptors and their MHC class I ligands in immunity, reproduction and human evolution Nat Rev Immunol 2013 13 133 44 23334245 4 Moffett A Colucci F Co-evolution of NK receptors and HLA ligands in humans is driven by reproduction Immunol Rev 2015 267 283 297 26284484 5 Koopman La Kopcow HD Rybalov B Boyson JE Orange JS Schatz F Masch R Lockwood CJ Schachter AD Park PJ Strominger JL Human decidual natural killer cells are a unique NK cell subset with immunomodulatory potential J Exp Med 2003 198 1201 12 14568979 6 Kopcow HD Allan DSJ Chen X Rybalov B Andzelm MM Ge B Strominger JL Human decidual NK cells form immature activating synapses and are not cytotoxic Proc Natl Acad Sci U S A 2005 102 15563 15568 16230631 7 Sharkey AM Xiong S Kennedy PR Gardner L Farrell LE Chazara O Ivarsson Ma Hiby SE Colucci F Moffett A Tissue-Specific Education of Decidual NK Cells J Immunol 2015 195 3026 32 26320253 8 Ferry BL Starkey PM Sargent IL Watt GM Jackson M Redman CW Cell populations in the human early pregnancy decidua: natural killer activity and response to interleukin-2 of CD56-positive large granular lymphocytes Immunology 1990 70 446 452 1697563 9 King A Birkby C Loke YW Early human decidual cells exhibit NK activity against the K562 cell line but not against first trimester trophoblast Cell Immunol 1989 118 337 44 2910502 10 Ritson A Bulmer JN Isolation and functional studies of granulated lymphocytes in first trimester human decidua Clin Exp Immunol 1989 77 263 8 2776361 11 Ashkar AA Di Santo JP Croy BA Interferon gamma contributes to initiation of uterine vascular modification, decidual integrity, and uterine natural killer cell maturation during normal murine pregnancy J Exp Med 2000 192 259 70 10899912 12 Xiong S Sharkey AM Kennedy PR Gardner L Farrell LE Chazara O Bauer J Hiby SE Colucci F Moffett A Maternal uterine NK cell-activating receptor KIR2DS1 enhances placentation J Clin Invest 2013 123 4264 4272 24091323 13 Hiby SE Apps R Sharkey AM Farrell LE Gardner L Mulder A Claas FH Walker JJ Redman CC Morgan L Tower C Maternal activating KIRs protect against human reproductive failure mediated by fetal HLA-C2 J Clin Invest 2010 120 4102 4110 20972337 14 Kieckbusch J Gaynor LM Moffett A Colucci F MHC-dependent inhibition of uterine NK cells impedes fetal growth and decidual vascular remodelling Nat Commun 2014 5 3359 24577131 15 Khakoo SI Carrington M KIR and disease: A model system or system of models? Immunol Rev 2006 214 186 201 17100885 16 Robinson J Mistry K Mcwilliam H Lopez R Marsh SGE Ipd-the immuno polymorphism database Nucleic Acids Res 2009 38 863 869 17 Jiang W Johnson C Jayaraman J Simecek N Noble J Moffatt MF Cookson WO Trowsdale J Traherne Ja Copy number variation leads to considerable diversity for B but not A haplotypes of the human KIR genes encoding NK cell receptors Genome Res 2012 22 1845 54 22948769 18 Hiby SE Walker JJ O’shaughnessy KM Redman CWG Carrington M Trowsdale J Moffett A Combinations of maternal KIR and fetal HLA-C genes influence the risk of preeclampsia and reproductive success J Exp Med 2004 200 957 965 15477349 19 Hiby SE Regan L Lo W Farrell L Carrington M Moffett a Association of maternal killer-cell immunoglobulin-like receptors and parental HLA-C genotypes with recurrent miscarriage Hum Reprod 2008 23 972 976 18263639 20 Hiby SE Apps R Chazara O Farrell LE Magnus P Trogstad L Gjessing HE Carrington M Moffett A Maternal KIR in combination with paternal HLA-C2 regulate human birth weight J Immunol 2014 192 5069 73 24778445 21 Nakimuli A Chazara O Hiby SE Farrell L Tukwasibwe S Jayaraman J Traherne Ja Trowsdale J Colucci F Lougee E Vaughan RW A KIR B centromeric region present in Africans but not Europeans protects pregnant women from pre-eclampsia Proc Natl Acad Sci U S A 2015 112 845 50 25561558 22 Cisneros E Moraru M Gomez-Lozano N Lopez-Botet M Vilches C KIR2DL5: An orphan inhibitory receptor displaying complex patterns of polymorphism and expression Front Immunol 2012 3 1 8 22679445 23 O’Connor GM Vivian JP Gostick E Pymm P Lafont BaP Price Da Rossjohn J Brooks AG McVicar DW Peptide-Dependent Recognition of HLA-B*57: 01 by KIR3DS1 J Virol 2015 89 5213 21 25740999 24 King A Burrows TD Hiby SE Bowe JM Joseph S Verma S Lim PB Gardner L Le Bouteiller P Ziegler A Uchanska-Ziegler B Loke YW Surface expression of HLA-C antigen by human extravillous trophoblast Placenta 2000 21 376 387 10833373 25 Apps R Murphy SP Fernando R Gardner L Ahad T Moffett A Human leucocyte antigen (HLA) expression of primary trophoblast cells and placental cell lines, determined using single antigen beads to characterize allotype specificities of anti-HLA antibodies Immunology 2009 127 26 39 19368562 26 VandenBussche CJ Mulrooney TJ Frazier WR Dakshanamurthy S Hurley CK Dramatically reduced surface expression of NK cell receptor KIR2DS3 is attributed to multiple residues throughout the molecule Genes Immun 2009 10 162 173 19005473 27 Hsu KC Liu X-R Selvakumar A Mickelson E O’Reilly RJ Dupont B Killer Ig-like receptor haplotype analysis by gene content: evidence for genomic diversity with a minimum of six basic framework haplotypes, each with multiple subsets J Immunol 2002 169 5118 5129 12391228 28 Graef T Moesta AK Norman PJ Abi-Rached L Vago L Older Aguilar AM Gleimer M Hammond JA Guethlein LA Bushnell DA Robinson PJ KIR2DS4 is a product of gene conversion with KIR3DL2 that introduced specificity for HLA-A*11 while diminishing avidity for HLA-C J Exp Med 2009 206 2557 2572 19858347 29 Male V Trundley A Gardner L Northfield J Chang C Apps R Moffett A Natural Killer Cells in Human Pregnancy Campbell KS Natural Killer Cell Protocols. Methods in Molecular Biology 2010 612 Humana Press Totowa, NJ 447 463 30 Schleinitz N Cognet C Guia S Laugier-Anfossi F Baratin M Pouget J Pelissier J-F Harle J-R Vivier E Figarella-Branger D Expression of the CD85j (leukocyte Ig-like receptor 1, Ig-like transcript 2) receptor for class I major histocompatibility complex molecules in idiopathic inflammatory myopathies Arthritis Rheum 2008 58 3216 23 18821690 31 Trundley A Frebel H Jones D Chang C Trowsdale J Allelic expression patterns of KIR3DS1 and 3DL1 using the Z27 and DX9 antibodies Eur J Immunol 2007 37 780 7 17301953 32 Trundley AE Hiby SE Chang C Sharkey AM Santourlidis S Uhrberg M Trowsdale J Moffett A Molecular characterization of KIR3DL3 Immunogenetics 2006 57 904 16 16391939 33 Estefanía E Flores R Gómez-Lozano N Aguilar H López-Botet M Vilches C Human KIR2DL5 is an inhibitory receptor expressed on the surface of NK and T lymphocyte subsets J Immunol 2007 178 4402 4410 17371997 34 Dohring C Scheidegger D A human killer inhibitory receptor specific for HLA-A1, 2 J Immunol 1996 3 6 35 Purcell S Neale B Todd-Brown K Thomas L Ferreira MAR Bender D Maller J Sklar P de Bakker PIW Daly MJ Sham PC PLINK: A tool set for whole-genome association and population-based linkage analyses Am J Hum Genet 2007 81 559 575 17701901 36 Katz G Markel G Mizrahi S Arnon TI Mandelboim O Recognition of HLA-Cw4 but not HLA-Cw6 by the NK cell receptor killer cell Ig-like receptor two-domain short tail number 4 J Immunol 2001 166 7260 7 11390475 37 Sharkey A Gardner L Hiby S Killer Ig-like receptor expression in uterine NK cells is biased toward recognition of HLA-C and alters with gestational age J 2008 181 39 46 38 Male V Sharkey A Masters L Kennedy PR Farrell LE Moffett A The effect of pregnancy on the uterine NK cell KIR repertoire Eur J Immunol 2011 41 3017 3027 21739430 39 Verma S Hiby SE Loke YW King a Human decidual natural killer cells express the receptor for and respond to the cytokine interleukin 15 Biol Reprod 2000 62 959 68 10727265 40 Apps R Sharkey A Gardner L Male V Kennedy P Masters L Farrell L Jones D Thomas R Moffett A Ex vivo functional responses to HLA-G differ between blood and decidual NK cells Mol Hum Reprod 2011 17 577 86 21471023 41 Hanna J Goldman-Wohl D Hamani Y Avraham I Greenfield C Natanson-Yaron S Prus D Cohen-Daniel L Arnon TI Manaster I Gazit R Decidual NK cells regulate key developmental processes at the human fetal-maternal interface Nat Med 2006 12 1065 1074 16892062 42 Red-Horse K Drake PM Gunn MD Fisher SJ Chemokine Ligand and Receptor Expression in the Pregnant Uterus Am J Pathol 2001 159 2199 2213 11733370 43 Drake PM Gunn MD Charo IF Tsou CL Zhou Y Huang L Fisher SJ Human placental cytotrophoblasts attract monocytes and CD56(bright) natural killer cells via the actions of monocyte inflammatory protein 1alpha J Exp Med 2001 193 1199 212 11369791 44 Hong HA Paximadis M Gray GE Kuhn L Tiemessen CT KIR2DS4 allelic variants: Differential effects on in utero and intrapartum HIV-1 mother-to-child transmission Clin Immunol 2013 149 498 508 24239756 45 Merino A Malhotra R Morton M Mulenga J Allen S Hunter E Tang J Kaslow RA Impact of a functional KIR2DS4 allele on heterosexual HIV-1 transmission among discordant Zambian couples J Infect Dis 2011 203 487 495 21216870 46 Merino AM Dugast AS Wilson CM Goepfert PA Alter G Kaslow RA Tang J KIR2DS4 promotes HIV-1 pathogenesis: New evidence from analyses of immunogenetic data and natural killer cell function PLoS One 2014 9 47 Olvera A Pérez-Álvarez S Ibarrondo J Ganoza C Lama J Lucchetti A Cate S Hildebrand W Bernard N Gomez G Sanchez J Brander C The HLA-C*04:01/KIR2DS4 gene combination and 1 HLA alleles with high population frequency drive rate of HIV disease progression AIDS 2015 29 507 17 25715101 48 Zhou J Tang X Ding Y An Y Zhao X Natural killer cell activity and frequency of killer cell immunoglobulin-like receptors in children with different forms of juvenile idiopathic arthritis Pediatr Allergy Immunol 2013 24 691 6 24112428 49 Yen J-H Lin C-H Tsai W-C Wu C-C Ou T-T Hu C-J Liu H-W Killer cell immunoglobulin-like receptor gene’s repertoire in rheumatoid arthritis Scand J Rheumatol 2006 35 124 7 16641046 50 Majorczyk E Pawlik A Gendosz D Kuśnierczyk P Presence of the full-length KIR2DS4 gene reduces the chance of rheumatoid arthritis patients to respond to methotrexate treatment BMC Musculoskelet Disord 2014 15 256 25069714 51 Giebel S Boratyn-Nowicka A Karabon L Jedynak A Pamula-Pilat J Tecza K Kula D Kowal M Frydecka I Grzybowska E Associations between genes for killer immunoglobulin-like receptors and their ligands in patients with epithelial ovarian cancer Hum Immunol 2014 75 508 513 24755350 52 Bao XJ Hou LH Sun aN Qiu QC Yuan XN Chen MH Chen ZX He J The impact of KIR2DS4 alleles and the expression of KIR in the development of acute GVHD after unrelated allogeneic hematopoietic SCT Bone Marrow Transplant 2010 45 1435 41 20062104 53 Hammond JA Carrington M Khakoo SI A vision of KIR variation at super resolution Immunology 2016 1 1 4 54 Cooley S Weisdorf DJ Guethlein La Klein JP Wang T Le CT Marsh SGE Geraghty D Spellman S Haagenson MD Ladner M Donor selection for natural killer cell receptor genes leads to superior survival after unrelated transplantation for acute myelogenous leukemia Blood 2010 116 2411 2419 20581313 55 Pittari G Liu X-R Selvakumar A Zhao Z Merino E Huse M Chewning JH Hsu KC Dupont B NK cell tolerance of self-specific activating receptor KIR2DS1 in individuals with cognate HLA-C2 ligand J Immunol 2013 190 4650 60 23554313 56 Stewart CA Laugier-Anfossi F Vély F Saulquin X Riedmuller J Tisserant A Gauthier L Romagné F Ferracci G Arosa Fa Moretta A Recognition of peptide-MHC class I complexes by activating killer immunoglobulin-like receptors Proc Natl Acad Sci U S A 2005 102 13224 9 16141329 57 Goodridge JP Burian A Lee N Geraghty DE HLA-F and MHC class I open conformers are ligands for NK cell Ig-like receptors J Immunol 2013 191 3553 62 24018270 58 Pesce S Carlomagno S Moretta A Sivori S Marcenaro E Uptake of CCR7 by KIR2DS4 + NK Cells Is Induced upon Recognition of Certain HLA-C Alleles J Immunol Res 2015 2015 1 10 59 Tilburgs T Evans JH Crespo ÂC Strominger JL The HLA-G cycle provides for both NK tolerance and immunity at the maternal–fetal interface Proc Natl Acad Sci 2015 201517724 60 Medawar P Some immunological and endocrinological problems raised by the evolution of viviparity in vertebrates Symp Soc Exp Biol 1953 7 320 38

PMC005xxxxxx/PMC5115177.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9110829 2209 J Biomol NMR J. Biomol. NMR Journal of biomolecular NMR 0925-2738 1573-5001 26968894 5115177 10.1007/s10858-016-0028-y NIHMS768266 Article Insights into Furanose Solution Conformations: Beyond the Two-State Model Wang Xiaocong Woods Robert J. * Complex Carbohydrate Research Center, University of Georgia, Athens, GA, 30605 * To whom all correspondence should be addressed, rwoods@ccrc.uga.edu, Phone: +1 (706) 542-4454, Fax: +1 (706) 542-4412 5 4 2016 12 3 2016 4 2016 01 4 2017 64 4 291305 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. A two-state model is commonly used for interpreting ring conformations of furanoses based on NMR scalar 3J-coupling constants, with the ring populating relatively narrow distributions in the North and the South of the pseudorotation itinerary. The validity of this simple approach has been questioned, and is examined here in detail employing molecular dynamics (MD) simulations with a new GLYCAM force field parameter set for furanoses. Theoretical 3J-coupling constants derived from unrestrained MD simulations with the new furanose-specific parameters agreed with the experimental coupling constants to within 1 Hz on average. The results confirm that a two state model is a reasonable description for the ring conformation in the majority of methyl furanosides. However, in the case of methyl α-D-arabinofuranoside the ring populates a continuum of states from North to South via the eastern side of the pseudorotational itinerary. Two key properties are responsible for these differences. Firstly, East and West regions in β- and α-anomers, respectively, are destabilized by the absence of the anomeric effect. And, secondly, East or West conformations can be further destabilized by repulsive interactions among vicinal hydroxyl groups and ring oxygen atoms when the vicinal hydroxyl groups are in syn-configurations (such as in ribose and lyxose) more so than when in anti (arabinose, xylose). Pseudorotation Anomeric effect Potential energy surface Molecular dynamics simulation GLYCAM AMBER Introduction Furanoses are essential components in the backbone of nucleic acids (RNA and DNA) and are also frequent components of complex polysaccharides found in organisms ranging from bacteria to protozoa, fungi and plants (Taha et al. 2013). Their conformational properties affect the structure and recognition of the polymers in which they are found. In marked contrast to six-membered sugar rings (pyranoses), the five-membered forms (furanoses) exhibit a high level of internal ring flexing (Bartenev et al. 1987; Levitt and Warshel 1978; Seo et al. 2008). As a result, furanoses can interconvert between multiple ring conformations (Houseknecht et al. 2003a), whereas pyranoses are usually found in single, energy-favorable chair conformation (Angyal 1984). This feature of furanoses greatly complicates the development and interpretation of structure-function relationships for these small but crucial molecules. The conformational analysis of ring conformations of carbohydrates molecules in solution relies heavily on deconvoluting NMR scalar 3J-coupling constants into ring puckering geometries and populations (De Leeuw and Altona 1983; Padrta and Sklenar 2002; Sun et al. 2004). In the case of furanoses, interpreting their conformational populations from 3J-coupling constants typically requires assumptions regarding to the number of states that are present. In the two-state model introduced by Altona et al. (1972) to interpret the ring conformations of the ribofuranosyl ring in nucleic acids, it is assumed that the rings populate only two states (Altona and Sundaral.M 1972; Altona and Sundaralingam 1973), referred to as North and South, with respect to their location on the pseudorotational itinerary (Fig. 1). However, the validity of a two-state model has been questioned when applied to other furanoses, particularly in the case of arabinofuranose (Hendrickx et al. 2010; Seo et al. 2008; Taha et al. 2010; Taha et al. 2011). An alternative approach that has shown promise, is to perform MD simulations of the furanose (Hatcher et al. 2009), and then back-calculate the 3J-values directly from the MD data (Hendrickx et al. 2010; Raman et al. 2010; Taha et al. 2010). This approach eliminates the need for assumptions regarding states, but requires an accurate force field, adequate conformational sampling, and a reliable method for deriving scalar couplings from the MD trajectory. In general, the calculation of NMR 3J-values from MD data can be performed in one of two ways, most commonly by employing a Karplus-type relationship to convert torsion angles between coupled spins into 3J-values (Altona et al. 1994; Bose et al. 1998; Cloran et al. 1999; Haasnoot et al. 1980; Houseknecht et al. 2003b; Stenutz et al. 2002; Taha et al. 2010; Taha et al. 2011; Zhao et al. 2007), or by computing the 3J-values quantum mechanically (QM) for each state (Gonzalez-Outeirino et al. 2006), The latter approach eliminates the approximations implicit in Karplus-relationships, but is far more computationally demanding, and has not yet been applied to furanoses, although it has shown benefits when applied to analyses of pyranose conformational properties (Gonzalez-Outeirino et al. 2006). Previously, the GLYCAM06 force field (Kirschner et al. 2008) was created to be applicable broadly to a range of carbohydrates, including pyranoses and furanoses, however, MD simulations of furanoses have not resulted in consistently acceptable reproduction of solution NMR properties (Seo et al. 2008). Therefore, developing a reliable force field for furanoses that is capable of describing their conformational properties in solution, is essential for both theoretical and experimental studies of these molecules. As might be expected, the fluxional properties of furanosyl rings leads to unique challenges in developing a force field for these molecules, requiring long MD simulation times to ensure converged conformational sampling. In addition, the five-membered ring has more strain than found in pyranoses, as evidenced by the compression of bond angles, and this feature places unique demands on the derivation of the torsion terms relevant to the ring. To address the question of the apparent discrepancy between MD-derived and NMR-derived ring populations in furanoses (Seo et al. 2008), we begin by first developing a set of furanose-specific parameters that are compatible with the GLYCAM06 force field (Kirschner et al. 2008), which are able to reproduce the quantum-computed relative energies of the ring as it traverses the pseudorotational itinerary. Having derived a suitable set of parameters, MD simulations were performed in explicit water for both α and β anomers of the four methyl D-aldofuranosides: methyl D-arabinofuranoside (1), methyl D-lyxofuranoside (2), methyl D-ribofuranoside (3), methyl D-xylofuranoside (4), as well as methyl 2-deoxy-β-D-ribofuranoside (5β) (Fig. 2). 3J-coupling constants derived from the MD simulations of the examined furanosides with this new set of parameters were found to be in agreement with NMR 3J-values to within an average error of less than 0.9 Hz. More importantly, ring conformations from the MD simulations have shown the two-state model is a reasonable description of the ring properties in the majority of the furanosides, but not in the case of the arabino configuration (1α). Further analysis indicated that, for the two-state model to be applicable, barriers in the East and West must exist to confine the conformations to low-energy regions in the northern and southern quadrants of the pseudorotational surface. The relative stability of northern and southern conformations has been rationalized in terms of the numbers of stabilizing gauche effects between vicinal oxygen atoms (Callam and Lowary 2001; D'Souza et al. 2000; Plavec et al. 1993; Thibaudeau and Chattopadhyaya 1997) as well as the presence of the anomeric effect (Callam and Lowary 2001; Cosse-Barbi and Dubois 1987; Cossé-Barbi et al. 1989; D'Souza et al. 2000; Ellervik and Magnusson 1994; Houseknecht et al. 2003a; Plavec et al. 1994; Plavec et al. 1996; Plavec et al. 1993; Thibaudeau and Chattopadhyaya 1997; Thibaudeau et al. 1994). Here, we find that whether or not there is a barrier in the eastern or western regions depends on the absence of an anomeric effect, as well as the presence of unfavorable syn-orientations of vicinal hydroxyl groups. In the unique case of 1α, there is no barrier in the eastern quadrant due to the absence of any conformation in which hydroxyl groups can adopt syn-orientations, enabling the furanose to populate a continuum of states from North to South. The present analysis provides an independent method for interpreting furanose conformational properties that does not rely on experimental NMR data as constraints, nor on assumptions of preferred states. In addition, QM calculations have been employed here to quantify for the first time the strength of the anomeric and exo-anomeric effects in furanosides. Results & Discussion Parameters New atom type In order to develop the parameters specific for furanoses, which are orthogonal with the current parameters for pyranoses, two new atom types were introduced: “Cf”, representing ring carbon atoms, and “Of”, representing the ring oxygen atom (Fig. S1). The “upper case lower case” naming of atom types in AMBER is reserved for use with carbohydrates in the GLYCAM force field Van der Waals terms for Cf and Of were assumed to be the same as for the corresponding atoms in pyranoses (Kirschner et al. 2008). Ensemble-averaged partial atomic charges for the new atom types, as well as all other atoms (Table S1 and Table S2), were computed according to the standard GLCYAM protocol (Kirschner et al. 2008), discussed in METHODS. Bond length and bond angle parameters The average bond lengths for C-C and C-O bonds in furanose ring are nearly identical with those in pyranose ring (Ishii et al. 2014; Ishii et al. 2015; Longchambon et al. 1976; McDonald and Beevers 1952; Ohanessian et al. 1977; Ohanessian and Gillierpandraud 1976). Therefore, the bond length parameters for furanose ring were imported from GLYCAM06, and simulations with these parameters demonstrated good agreement with the corresponding crystallographic values (Table S3 and Table S4). In earlier bond angle parameter development for pyranoses in GLYCAM06 (Kirschner et al. 2008), acyclic molecules were distorted from their equilibrium geometries in order to determine the force constants and equilibrium angle values. Similarly, bond lengths were stretched or compressed to determine appropriate bond parameters. However, to determine these values for furanoses, an alternative procedure was adopted, in which the ring was distorted by twisting the relevant torsion angles over a range of +/− 50 degrees. QM energy profiles for valence bond and angle parameter fitting were generated by optimizing tetrahydrofuran (THF) structures with ring torsion angles changing from −50° to 50° in 5° increments. Due to the symmetry of THF, only three energy profiles were necessary, namely for the C1-C2-C3-C4, C2-C3-C4-O4 and C3-C4-O4-C1 torsion angles. Parameters for the Cf-Cf-Cf, Cf-Cf-Of and Cf-Of-Cf angles were derived simultaneously by minimizing the differences between QM and molecular mechanics (MM) ring distortion energies. To assess the performance of the new parameters in reproducing the QM energy profiles, the errors between QM and MM energies were computed over the entire range of the distortion curves (<|Error|>all), as well as for the low energy conformations (<|Error|>low). The new parameters reproduced the gas-phase relative energies for THF markedly better than GLYCAM06 (Fig. S2). The <|Error|>all for the new parameters for the three energy profiles in Fig. S2 is 0.1 kcal/mol, and in contrast to GLYCAM06, the new parameters better reproduced the QM energies in high energy regions. Average values for the C-C-C, C-C-O, and C-O-C angles were computed from 300 ns MD simulations and compared to relevant crystallographic values (Table S3 and Table S4). Although both C-C-O and C-O-C angles showed larger differences from the crystallographic values, they are within the standard deviation. Torsion angle parameters Torsion terms associated with the five-membered ring were optimized based on their ability to reproduce the QM pseudorotational energies for discrete conformations of the ring generated at 1° increments in the ring phase angle (P). In this procedure several contributing torsion terms could be simultaneously optimized. Simultaneous parameter fitting has been applied recently in the generation of a set of force field parameters for protein modeling (Cerutti et al. 2014). Model structures were selected that nevertheless enabled the fitting of as few simultaneous torsion terms as possible, in order to minimize the number of new torsion terms, maximize parameter transferability, and provide insight into the underlying structural preferences (Table S5). All other torsion terms were generated directly by fitting to internal rotational energies. In order to reduce energy variations originating from interactions involving exocyclic moieties, the conformation of each of the exocyclic moieties was restrained throughout the MM and QM pseudorotation energy minimizations (Table S5). To assess the performance of the torsion angle parameters, the average errors between QM and MM energies were computed for the entire energy curves, <|Error|>pseudo, as well as for the minima, <|Error|>minima. The parameters resulting from the development protocol discussed below are presented in Table S6. THF The torsion terms (Cf-Cf-Cf-Cf, Cf-Cf-Cf-Of and Cf-Cf-Of-Cf) derived simultaneously from fitting to the pseudorotational energies for THF (6) are fundamental to all furanoses. The agreement between the MM and QM energies (Fig. 3) for the pseudorotational energy of THF computed with the optimized torsion parameters was within 0.1 kcal/mol for both <|Error|>pseudo and <|Error|>minima. It is notable that the energy minima for the THF pseudorotation was observed to lie in the East/West quadrants, rather than North/South, as typically observed for furanoses. This observation suggests that the exocyclic groups in furanoses alter the conformational preferences of the five-membered ring, although the parries to interconversion are less than approximately 0.5 kcal/mol. As can be seen in Fig. 3, the energy minima for the five-membered ring in THF occur at P = 90 and 270°, corresponding to East and West pseudorotational quadrants, respectively. In these conformations the C1-C2-C3-C4 torsion angle is 0°. However, in this conformation hydroxyl groups at C2 and C3 can be eclipsed (as in lyxose and ribose), disfavoring East/West conformations. Models for mono- and dihydroxy substituents in furanoses The ring conformation of furanoses is greatly affected by the configurations of the hydroxyl groups (Church et al. 1997; Houseknecht et al. 2003b; Kline and Serianni 1990; Serianni and Barker 1984; Taha et al. 2010; Taha et al. 2011), and presumably arises from a combination of steric and electrostatic interactions. These interactions may be between hydroxyl groups, as well as between the ring atoms and the hydroxyl groups. To reproduce the effect of these interactions on the ring psudorotational energies, torsion terms were developed for the Oh/Of-Cf-Cf-Oh, and Cf/Hc-Cf-Cf-Oh sequences, employing the model compounds cis- (7) and trans-tetrahydrofuran-3,4-diol (8) and tetrahydrofuran-3-ol (9). As shown in Fig. S3, the new parameters reproduced the shapes of the QM pseudorotational energy curves for both molecules. In the case of 9, the errors in the fits were similarly small. With the introduction of the vicinal hydroxyl groups, the minima now occur in the North/South regime, as expected in general for furanoses. Models for pentofuranoses To extend the model to pentofuranoses, a hydroxymethyl group was introduced into THF, which corresponds to the C4-substituent in pentofuranoses, employing 2-(hydroxymethyl)tetrahydrofuran (10) as the model compound. This molecule enabled us to develop the Cf-Cf-Cf-Cg and Cf-Of-Cf-Cg torsion terms (pertinent to the C2-C3-C4-C5 and C1-O4-C4-C5 atomic sequences). The fitting to the pseudorotational energies for 10 was performed for each of the three rotamers of the hydroxymethyl substituent (Fig. S4). While the average errors were all less than 0.5 kcal/mol, not all low-energy states were well reproduced, however in the worst case the error was less than 1 kcal/mol. Similar agreements were obtained for cis- and trans-2-(hydroxymethyl)tetrahydrofuran-3-ol (11 and 12, respectively), which tested the ability of the parameters to reproduce the relative energies of the C4-C5 rotamers in the presence of a vicinal hydroxyl group at C3 (Fig. S5 and Fig. S6). Models for methyl furanosides The simplest model for the effect of anomeric configuration on five-membered ring conformational energy is (S)-2-methoxytetrahydrofuran (13), and, by analogy to the preceding section, the effect of interactions between the anomeric substituent and the adjacent hydroxyl group at C2 can be inferred from an analysis of cis- and trans-2-methoxytetrahydrofuran-3-ol (14 and 15, respectively). Torsion terms derived from fitting to 13 (Cf-Of-Cf-Os and Cf-Cf-Cf-Os), implicitly include any contributions from the anomeric effect (Callam and Lowary 2001; Cosse-Barbi and Dubois 1987; Cossé-Barbi et al. 1989; D'Souza et al. 2000; Ellervik and Magnusson 1994; Plavec et al. 1993; Thibaudeau and Chattopadhyaya 1997). In GLYCAM06, both the endo- and exo-anomeric torsion terms were parameterized by fitting to small acyclic molecular fragments, leading to the same set of terms for both applications. Here the torsion terms were derived for intact five-membered rings, with furanose-specific atom types for the ring atoms (Cf and Of), leading to unique terms for the endo- (Cf-Of-Cf-Os, and Cf-Cf-Cf-Os) and exo-anomeric (Of-Cf-Os-Cf/Cg and Cf-Cf-Os-Cf/Cg) sequences. As shown in Fig. S7, the torsion angle terms for endo-anomeric sequence reproduced the lowest energy conformation, as well as the overall shape of pseudorotational energy curve. It should be noted that the new parameters lead to a local energy minimum at P = 99°, while a similar minimum appears at P = 71° in the QM energy curve. The torsion angle parameters derived from 14 and 15 gave rise to slightly larger errors, reflecting in part the challenge of fitting highly asymmetric energy curves using a sum of symmetric cosine functions. The torsion angle terms for exo-anomeric sequence reproduced the rotational energy profiles of 13 in both northern (P = 0°) and southern (P = 180°) ring conformations (Fig. S8). Features of the endo- and exo-anomeric effects in furanoses are discussed in the following sections. Endo-anomeric effect In carbohydrates, the anomeric effect relates to the preference of the endocyclic C-O-C-O torsion angle to adopt the gauche orientation over the anti conformation (Cossé-Barbi et al. 1989; Jeffrey et al. 1978; Juaristi and Cuevas 1992), and has been estimated from quantum mechanical calculations for pyranoses (Lii et al. 2003; Senderowitz et al. 1996) and related acyclic fragments (Jeffrey et al. 1978; Lii et al. 2005) to be 1–2 kcal/mol. Its presence in furanoses has been invoked as contributing to ring conformational preferences (Callam and Lowary 2001; Cosse-Barbi and Dubois 1987; Cossé-Barbi et al. 1989; D'Souza et al. 2000; Ellervik and Magnusson 1994; Houseknecht et al. 2003a; Plavec et al. 1994; Plavec et al. 1996; Plavec et al. 1993; Thibaudeau and Chattopadhyaya 1997; Thibaudeau et al. 1994), and is consistent with the observation that electronegative substituents at C1 in furanoses prefers a pseudoaxial over a pseudoequatorial configuration,29–34 but it has not been quantified computationally. We address this here by noting that, over the course of the pseudorotational itinerary for 13, the endocyclic C4-O4-C1-O torsion angle spans a range of 80 to 160°, or pseudo-gauche to pseudo-anti. Thus the strength of the endo-anomeric effect in furanoses can be estimated to be 3.2 kcal/mol at the B3LYP/6-31G* level. In contrast, methoxycyclopentane (16) shows no such effect (Fig. 4). The endo-anomeric effect stabilizes eastern ring conformations (C4-O4-C1-O angle is less than 100°) but not western conformations (C4-O4-C1-O angle is greater than 140°). Therefore, western conformations in α-anomers (or eastern conformation in β-anomers) display high potential energies. Exo-anomeric effect The preference of the exocyclic C1-O bond in pyranoses to adopt gauche orientations is a direct corollary to the endo-anomeric effect, and this exo-anomeric effect has been extensively studied in pyranoses and acyclic molecules (Jeffrey et al. 1978; Kirschner et al. 2008; Lemieux and Koto 1974; Tvaroska and Bleha 1989). To determine the strength of the exo-anomeric effect in furanoses, the rotational energies of the O4-C1-O-CH3 torsion angle in 13 were computed for both North (P = 0°) and South (P = 180°) conformations. The exo-anomeric energy stabilizes the gauche orientations of the C1-O bond in 13, relative to the cyclopentane analog 16, by approximately 4 kcal/mol, which is comparable to that reported for pyranoses (Kirschner et al. 2008) (Fig. 5). This contribution is particularly important for defining the 3D structure of polymers of furanoses, as it reduces the conformational freedom of the C1-O bond to effectively a single state. Assessments of solution populations The performance of the new parameters was assessed by comparing NMR 3J-coupling constants for conformations observed in MD simulations (300 ns) of furanosides with their experimental values. Based on monitoring the P-values (Fig. S9) convergence in ring interconversion was obtained within this timescale. By employing quantum mechanical 3J-coupling constants corresponding to the states observed in the MD simulation, it was possible to avoid employing empirical Karplus equations (Altona et al. 1994; Bose et al. 1998; Cloran et al. 1999; Haasnoot et al. 1980; Houseknecht et al. 2003b; Stenutz et al. 2002; Taha et al. 2010; Taha et al. 2011; Zhao et al. 2007) or invoking approximations associated with the two-state model. GLYCAM06 was developed for use with partial atomic charges derived from electrostatic fitting to QM values computed at the HF/6-31G* level, however, we wished to examine what impact might arise from fitting to potentials computed at a higher level of theory. Recently, it has been shown that dipole moments computed at the B3LYP/cc-pVTZ level have approximately one half of the average error of those computed at the HF/6-31G* level (Hickey and Rowley 2014), and so we compared data for monosaccharide charge sets computed at both levels. Both charge sets led to very similar distributions of solution conformations, as indicated by the resulting J-couplings, which differ by on average 0.1 Hz (Table 1 and Table S7). Thus the following discussion focuses only on the data generated with the B3LYP/cc-pVTZ charges. The average error between the computed and experimental 3J-coupling constants for 1–4 (α and β) and 5β was within 0.9 Hz. It is notable that, while the overall agreement with experiment is good, in the case of 2α, the theoretical value for the 3JC1,H4 coupling constant (0.6 Hz) is significantly below that reported experimentally (3.7 Hz). However, this coupling constant is less than 0.5 Hz in all other α-anomers, suggesting a potential error in the experimental data. In the case of 5β there is also a significant difference between the theoretical (9.2 Hz) and experimental (~5.7 Hz) values for the 3JH2R,H3 coupling constant, however, the experimental report noted uncertainty in that particular measurement (Church et al. 1997). The rotamer distributions for the C4-C5 were also generated, and the average error between theoretical and experimental 3J-values associated with this bond for 1 (α and β), 3 (α and β) and 5β was 0.4 Hz; while that for 2 (α and β) and 4 (α and β) was 1.4 Hz (Table 2 and Table S8). The origin of the larger average error in these latter cases is uncertain, but may relate to the fact that the parameters for the C4-C5 bond were imported from GLYCAM06 and not re-derived herein. The overall correlation (R2) between the theoretical and experimental 3J-values for coupling constants related to ring conformation was computed to be 0.76 (see Fig. S10). Two-state equilibrium model The ring conformation distributions of 1–4 (α and β) and 5β in explicit solvent MD simulations are shown in Fig. 6. The majority of the furanoses populated conformations predominantly in the northern and southern regions, and could therefore be described as approximately satisfying a two-state distribution, regardless of the partial charge model. A notable exception to this was seen for 1α, which populated a continuum of states from North to South via the eastern side of the pseudorotational itinerary. This is in contrast both to the other furanoses, and to earlier studies of 1α, performed with GLYCAM06 that indicated a preponderant population of states in the North (Seo et al. 2008). The overall agreement between the theoretical and experimental NMR data from the present simulations provides compelling support for the present simulation results. By plotting conformational energies onto the pseudorotational surface (Fig. 7 and Fig. S11), it is possible to conveniently visualize the presence or absence of energy barriers between northern and southern states. In the case of a furanoside such as 3β, that prefers conformations in the North and South, both quantum- and classically-computed conformational energies confirm the presence of barriers in the East and West, resulting in a clear division of the conformational space into northern and southern quadrants (Fig. 7). In contrast, in the case of 1α, both quantum and classical mechanical methods indicate a barrier only in the West, enabling the ring to populate a continuous distribution of conformations from North to South via the East. Previous NMR (Angyal 1979; Serianni and Barker 1984; Taha et al. 2010), as well as QM (D'Souza et al. 2000; Gordon et al. 1999) studies, have also been shown to be consistent with conformations in the North, East, and South and it has been proposed that the anomeric effect (Cosse-Barbi and Dubois 1987; Cossé-Barbi et al. 1989; D'Souza et al. 2000; Ellervik and Magnusson 1994; Juaristi and Cuevas 1992; Plavec et al. 1993; Richards et al. 2013; Thibaudeau and Chattopadhyaya 1997) and the gauche effect (Callam and Lowary 2001; D'Souza et al. 2000; Houseknecht et al. 2003a; Plavec et al. 1994; Plavec et al. 1996; Plavec et al. 1993; Thibaudeau and Chattopadhyaya 1997; Thibaudeau et al. 1994) among the vicinal hydroxyl groups and the ring oxygen atom were the key influences on ring conformation in furanoses. In the case of 1α, it has been argued that the anomeric and gauche effects both stabilize the South conformation (Callam and Lowary 2001; D'Souza et al. 2000; Gordon et al. 1999; Houseknecht et al. 2003a). From the analysis of the anomeric effect presented in Fig. 4, we can conclude that the barrier in the western quadrant in 1α is due in part to the absence of stabilizing contributions from the anomeric effect. In the case of other furanoses, for example 3β, in which there are syn-interactions between hydroxyl groups, an examination of the interaction energies between the hydroxyl groups and the ring oxygen atom (Fig. 8) leads to the conclusion that repulsions between these groups introduce a barrier in the eastern quadrant. Thus in furanoses other than 1α (that has no cis-hydroxyl groups), either the lack of an anomeric effect, or the presence of repulsive oxygen-oxygen interactions results in barriers in both the East and West, effectively establishing a preference for the populations in the North and South. Conclusions The present study confirms that the two-state model, commonly employed in interpreting ring conformations of furanoses in solution, is not applicable to all furanoses. Results from MD simulation indicated that 1α exhibited a continuum distribution from North to South via East of pseudorotation, which agreed with the low energy pathway found in its pseudorotational potential energy surface, and gave rise to experimentally-consistent scalar coupling values for the ring protons. Nevertheless, the two-state model is a reasonable description for other furanoses, to the extent that their ring conformations broadly populate the northern and southern quadrants. The division (or not) of the populations into two states can be explained by the presence (or absence) energy penalties in East and West regions of the pseudorotation itinerary. These barriers are found to arise from the absence of a stabilizing anomeric effect and unfavorable interactions among vicinal hydroxyl groups and the ring oxygen atom. Notably, the new force field parameters permit accurate simulations of furanoses to be performed (average difference between theoretical and experimental NMR 3J-values was within 1 Hz), making the need to adopt an assumption about state preferences obsolete. This is likely to be particularly valuable when examining complex oligofuranosides that may contain additional chemical modifications, as are typical in bacterial (Besra et al. 1995; Bhamidi et al. 2008; Brennan and Nikaido 1995; Crick et al. 2001) or fungal pathogen (Beauvais et al. 2007; Costachel et al. 2005; Fontaine et al. 2000; Lamarre et al. 2009; Latge et al. 1994; Leitao et al. 2003) surfaces. The ability to characterize the conformational properties of such structures is an important component in understanding the mechanisms of disease infection (Carapito et al. 2009; Golgher et al. 1993), as well as immune response (Cooper and Petrovsky 2011; Predy et al. 2005). Supplementary Material 10858_2016_28_MOESM1_ESM The authors thank the NIH for support (R01 GM100058, P41 GM103390) and wish to thank Dr. Anthony S. Serianni at the University of Notre Dame for helpful discussions. Fig. 1 Pseudorotational itinerary of furanoses depicting different Envelope (E) and Twist (T) ring conformations with associated conformational phase angle (P) values in degrees. Fig. 2 Methyl D-furanosides examined in present study. Fig. 3 Pseudorotational energy curves for 6. Solid lines: energies computed at the B3LYP/6-31G* level to be consistent with the GLYCAM06 parameters (Kirschner et al. 2008); dashed lines: energies computed with new parameters. Fig. 4 Upper: pseudorotational energy curves for 13 with each quadrant of pseudorotation color coded and 16 (grey) computed at the B3LYP/6-31G* level, with the orientation of the exocyclic O4-C1-O-CH3 torsion angle restrained at 60°. Lower: the relative energy of 13 and 16 as functions of the C4-O4-C1-O and C4-C5-C1-O torsions, respectively. Fig. 5 Rotational energy curves of O4-C1-O-CH3 angle in 13 (black) and C5-C1-O-CH3 angle in 16 (grey) computed at B3LYP/6-31G* level, while maintaining the ring conformation at P = 0 (a) and 180° (b); rotational energy curves of O5-C1-O-CH3 angle in (S)-2-methoxytetrahydropyran (17) (black) and C6-C1-O-CH3 angle in methoxycyclohexane (18) (grey) computed at B3LYP/6-31G* level, while maintaining the ring conformation at 1C4 (c) and 4C1 (d). Regions stabilized by the exo-anomeric effect are indicated by vertical arrows. Fig. 6 Ring conformation distribution for 1–4 (α and β) and 5β from MD simulations (300 ns each), red/blue lines correspond to data from the HF/6-31G* and B3LYP/cc-pVTZ-derived partial charge models, respectively. a-d: α anomers of 1–4, e-i: β anomers of 1–5. Fig. 7 Pseudorotational potential energy surfaces for 1α (left) and 3β (right), generated with partial atomic charges computed at the B3LYP/cc-pVTZ level. a, b: energies computed at the B3LYP/6-31G* level; c, d: energies computed after energy minimization in the gas phase; e, f: energies computed for conformations observed in explicitly solvated MD simulations, without energy minimization. Details provided in METHODS. Fig. 8 Pseudorotational energy curves for 7 (upper) and 8 (bottom) computed at the B3LYP/6-31G* level, with the orientation of the vicinal hydroxyl groups (C2-C3-O3-H3O and C3-C2-O2-H2O) restrained at 180°. Curves in each quadrant of pseudorotation are color coded. Table 1 MD-deriveda, Karplus-equation-derived, and experimental 1H-1H and 13C-1H coupling constants for 1–4 (α and β) and 5β. 1H- 1H coupling constants (Hz) 3JH1,H2 3JH2,H3 3JH3,H4 3JC1,H4 MD/ QMb MD/ Karplusc,d Expt. MD/ QM MD/ Karplusd Expt. MD/ QM MD/ Karplusd Expt. MD/ QM MD/ Karplus f Expt. 1α 1.7 2.2 1.7e 3.6 2.9 3.4e 5.5 4.6 5.8e 0.8 1.0 <0.5f,g 1β 5.4 5.6 4.5h 5.9 4.7 7.9h 5.3 4.8 6.7h 3.9 1.8 3.2f 2α 3.4 4.5 3.6i 5.4 4.9 4.8i 5.4 4.9 4.3i 0.7 0.6 3.7f 2β 6.1 5.7 4.8i 5.6 4.9 5.0i 4.7 4.5 4.6i 1.8 0.3 2.2f 3α 4.7 5.5 4.3j 6.2 5.3 6.2j 5.9 5.4 3.4j 0.7 1.1 <0.5f,g 3β 0.6 1.2 1.2j 5.7 4.9 4.6j 8.2 7.0 6.9j 2.8 0.9 2.7f 4α 4.4 5.0 4.5i 2.9 2.3 5.5i 5.3 4.6 6.1i 0.1 0.3 <0.5f,g 4β 0.4 1.3 -i 1.5 1.3 1.7i 4.6 4.7 5.1i 1.4 0.4 2.4f 5β 1.7/6.2k 2.3/6.5k 2.6/5.4k,l 7.9/9.2k 7.7/6.9k 6.7/~5. 7k,l 5.2 5.2 4.2l a Computed with partial atomic charges fit to the B3LYP/cc-pVTZ QM potentials. b The standard derivations of all MD derived 3J-values are within approximately 0.3 Hz. c The standard derivations of all Karplus-equation-derived 3J-values are within approximately 0.1 Hz. d Altona et al. 1994. e Taha et al. 2010. f Houseknecht et al. 2003b. g A value of 0 Hz was employed in calculations of average error between the computed and experimental 3J-values. h Taha et al. 2011. i Serianni and Barker 1984. j Kline and Serianni 1990. k Reference are reported for couplings involving H2S and H2R (S/R). l Church et al. 1997. Table 2 MD-deriveda, Karplus-equation-derived, and experimental 3JH4,H5S and 3JH4,H5R values for 1–4 (α and β) and 5β. 1H-1H coupling constants (Hz) 3JH4,H5S 3JH4,H5R MD/QMb MD/Karplusc,d Expt. MD/QM MD/Karplus Expt. 1α 3.9 3.3 3.3e 6.1 5.3 5.8e 1β 4.0 3.4 3.4f 6.5 5.8 6.7f 2α 3.5 3.1 4.4g 8.7 7.8 6.7g 2β 3.1 3.0 4.5g 9.2 8.1 7.6g 3α 3.5 3.0 3.3h 5.1 4.8 4.6h 3β 3.5 3.1 3.1h 6.8 6.0 6.6h 4α 3.6 3.3 3.8g 8.8 7.7 6.0g 4β 3.3 3.1 4.4g 9.1 7.9 7.6g 5β 3.4 3.2 4.6i 6.8 6.0 7.0i a Computed with partial atomic charges fit to the B3LYP/cc-pVTZ QM potentials. b The standard derivations of all MD derived 3J-values are within approximately 0.5 Hz. c The standard derivations of all Karplus equation derived 3J-values are within approximately 0.4 Hz. d Altona et al. 1994. e Taha et al. 2010. f Taha et al. 2011. g Serianni and Barker 1984. h Kline and Serianni 1990. i Church et al. 1997. Altona C Francke R Dehaan R Ippel JH Daalmans GJ Hoekzema A Vanwijk J 1994 EMPIRICAL GROUP ELECTRONEGATIVITIES FOR VICINAL NMR PROTON-PROTON COUPLINGS ALONG A C-C BOND - SOLVENT EFFECTS AND REPARAMETERIZATION OF THE HAASNOOT EQUATION Magn Reson Chem 32 670 678 10.1002/mrc.1260321107 Altona C Sundaral M 1972 CONFORMATIONAL-ANALYSIS OF SUGAR RING IN NUCLEOSIDES AND NUCLEOTIDES - NEW DESCRIPTION USING CONCEPT OF PSEUDOROTATION J Am Chem Soc 94 8205 10.1021/ja00778a043 5079964 Altona C Sundaralingam M 1973 Conformational analysis of the sugar ring in nucleosides and nucleotides. Improved method for the interpretation of proton magnetic resonance coupling constants J Am Chem Soc 95 2333 2344 10.1021/ja00788a038 4709237 Angyal SJ 1979 Hudson's rules of isorotation as applied to furanosides, and the conformations of methyl aldofuranosides Carbohydr Res 77 37 50 doi:http://dx.doi.org/10.1016/S0008-6215(00)83791-X Angyal SJ 1984 THE COMPOSITION OF REDUCING SUGARS IN SOLUTION AdvCarbohydrChemBiochem 42 15 68 10.1016/s0065-2318(08)60122-5 Bartenev VN Kameneva NG Lipanov AA 1987 A STATISTICAL STEREOCHEMICAL MODEL OF THE FLEXIBLE FURANOSE RING Acta Crystallogr Sect B-Struct Sci 43 275 280 10.1107/s010876818709788x Beauvais A 2007 An extracellular matrix glues together the aerial-grown hyphae of Aspergillus fumigatus Cellular Microbiology 9 1588 1600 10.1111/j.1462-5822.2007.00895.x 17371405 Besra GS Khoo KH McNeil MR Dell A Morris HR Brennan PJ 1995 A NEW INTERPRETATION OF THE STRUCTURE OF THE MYCOLYL ARABINOGALACTAN COMPLEX OF MYCOBACTERIUM-TUBERCULOSIS AS REVEALED THROUGH CHARACTERIZATION OF OLIGOGLYCOSYLALDITOL FRAGMENTS BY FAST-ATOM-BOMBARDMENT MASS-SPECTROMETRY AND H-1 NUCLEAR-MAGNETIC-RESONANCE SPECTROSCOPY Biochemistry 34 4257 4266 10.1021/bi00013a015 7703239 Bhamidi S Scherman MS Rithner CD Prenni JE Chatterjee D Khoo KH McNeil MR 2008 The identification and location of succinyl residues and the characterization of the interior arabinan region allow for a model of the complete primary structure of Mycobacterium tuberculosis mycolyl arabinogalactan Journal of Biological Chemistry 283 12992 13000 10.1074/jbc.M800222200 18303028 Bose B 1998 Three-Bond C O C C Spin-Coupling Constants in Carbohydrates: Development of a Karplus Relationship J Am Chem Soc 120 11158 11173 10.1021/ja981127f Brennan PJ Nikaido H 1995 THE ENVELOPE OF MYCOBACTERIA Annual Review of Biochemistry 64 29 63 10.1146/annurev.bi.64.070195.000333 Callam CS Lowary TL 2001 Synthesis and conformational investigation of methyl 4a-carba-D-arabinofuranosides J Org Chem 66 8961 8972 10.1021/jo010827r 11749629 Carapito R 2009 Molecular Basis of Arabinobio-hydrolase Activity in Phytopathogenic Fungi CRYSTAL STRUCTURE AND CATALYTIC MECHANISM OF FUSARIUM GRAMINEARUM GH93 EXO-alpha-L-ARABINANASE Journal of Biological Chemistry 284 12285 12296 10.1074/jbc.M900439200 19269961 Cerutti DS Swope WC Rice JE Case DA 2014 ff14ipq: A Self-Consistent Force Field for Condensed-Phase Simulations of Proteins J Chem Theory Comput 10 4515 4534 10.1021/ct500643c 25328495 Church TJ Carmichael I Serianni AS 1997 C-13-H-1 and C-13-C-13 spin-coupling constants in methyl beta-D-ribofuranoside and methyl 2-deoxy-beta-D-erythro-pentofuranoside: Correlations with molecular structure and conformation J Am Chem Soc 119 8946 8964 10.1021/ja970231e Cloran F Carmichael I Serianni AS 1999 Density Functional Calculations on Disaccharide Mimics: Studies of Molecular Geometries and Trans-O-glycosidic 3JCOCH and 3JCOCC Spin-Couplings J Am Chem Soc 121 9843 9851 10.1021/ja984384t Cooper PD Petrovsky N 2011 Delta inulin: a novel, immunologically active, stable packing structure comprising beta-D- 2 -> 1 poly(fructo-furanosyl) alpha-D-glucose polymers Glycobiology 21 595 606 10.1093/glycob/cwq201 21147758 Cosse-Barbi A Dubois JE 1987 Anomeric orbital and steric control in static conformations and systems dynamics: rotations of methoxy groups in 2,2-dimethoxypropane and similar crystallographic COCOC fragments J Am Chem Soc 109 1503 1511 10.1021/ja00239a033 Cossé-Barbi A Watson DG Dubois JE 1989 Anomeric effects in carbohydrates: non equivalence of endocyclic oxygen lone pairs Tetrahedron Letters 30 163 166 doi:http://dx.doi.org/10.1016/S0040-4039(00)95147-5 Costachel C Coddeville B Latge JP Fontaine T 2005 Glycosylphosphatidylinositol-anchored fungal polysaccharide in Aspergillus fumigatus Journal of Biological Chemistry 280 39835 39842 10.1074/jbc.M510163200 16204227 Crick DC Mahapatra S Brennan PJ 2001 Biosynthesis of the arabinogalactan-peptidoglycan complex of Mycobacterium tuberculosis Glycobiology 11 107R 118R 10.1093/glycob/11.9.107R D'Souza FW Ayers JD McCarren PR Lowary TL 2000 Arabinofuranosyl oligosaccharides from mycobacteria: Synthesis and effect of glycosylation on ring conformation and hydroxymethyl group rotamer populations J Am Chem Soc 122 1251 1260 10.1021/ja993543l De Leeuw FAAM Altona C 1983 Computer-assisted pseudorotation analysis of five-membered rings by means of proton spin–spin coupling constants: Program PSEUROT Journal of Computational Chemistry 4 428 437 10.1002/jcc.540040319 Ellervik U Magnusson G 1994 Anomeric Effect in Furanosides. Experimental Evidence from Conformationally Restricted Compounds J Am Chem Soc 116 2340 2347 10.1021/ja00085a013 Fontaine T 2000 Molecular organization of the alkali-insoluble fraction of Aspergillus fumigatus cell wall Journal of Biological Chemistry 275 27594 27607 10869365 Golgher DB Colli W Soutopadron T Zingales B 1993 GALACTOFURANOSE-CONTAINING GLYCOCONJUGATES OF EPIMASTIGOTE AND TRYPOMASTIGOTE FORMS OF TRYPANOSOMA-CRUZI Molecular and Biochemical Parasitology 60 249 264 10.1016/0166-6851(93)90136-l 8232416 Gonzalez-Outeirino J Kirschner KN Thobhani S Woods RJ 2006 Reconciling solvent effects on rotamer populations in carbohydrates - A joint MD and NMR analysis Can J Chem-Rev Can Chim 84 569 579 10.1139/v06-036 Gordon MT Lowary TL Hadad CM 1999 A computational study of methyl alpha-D-arabinofuranoside: Effect of ring conformation on structural parameters and energy-profile J Am Chem Soc 121 9682 9692 10.1021/ja9915091 Haasnoot CAG de Leeuw FAAM Altona C 1980 The relationship between proton-proton NMR coupling constants and substituent electronegativities—I : An empirical generalization of the karplus equation Tetrahedron 36 2783 2792 doi:http://dx.doi.org/10.1016/0040-4020(80)80155-4 Hatcher E Guvench O MacKerell AD 2009 CHARMM Additive All-Atom Force Field for Aldopentofuranoses, Methyl-aldopentofuranosides, and Fructofuranose J Phys Chem B 113 12466 12476 10.1021/jp905496e 19694450 Hendrickx PMS Corzana F Depraetere S Tourwe DA Augustyns K Martins JC 2010 The Use of Time-Averaged (3)J(HH) Restrained Molecular Dynamics (tar-MD) Simulations for the Conformational Analysis of Five-Membered Ring Systems: Methodology and Applications Journal of Computational Chemistry 31 561 572 10.1002/jcc.21345 19530112 Hickey AL Rowley CN 2014 Benchmarking Quantum Chemical Methods for the Calculation of Molecular Dipole Moments and Polarizabilities The Journal of Physical Chemistry A 118 3678 3687 10.1021/jp502475e 24796376 Houseknecht JB Lowary TL Hadad CM 2003a Gas- and solution-phase energetics of the methyl alpha- and beta-D-aldopentofuranosides J Phys Chem A 107 5763 5777 10.1021/jp027716w Houseknecht JB Lowary TL Hadad CM 2003b Improved Karplus equations for (3)J(C1,H4) in aldopentofuranosides: Application to the conformational preferences of the methyl aldopentofuranosides J Phys Chem A 107 372 378 10.1021/jp026610y Ishii T Ohga S Fukada K Morimoto K Sakane G 2014 beta-d-Gulose Acta crystallographica Section E, Structure reports online 70 o569 10.1107/s1600536814008046 Ishii T Senoo T Kozakai T Fukada K Sakane G 2015 Crystal structure of beta-d,l-allose Acta crystallographica Section E, Crystallographic communications 71 o139 10.1107/s2056989015000353 Jeffrey GA Pople JA Binkley JS Vishveshwara S 1978 APPLICATION OF ABINITIO MOLECULAR-ORBITAL CALCULATIONS TO STRUCTURAL MOIETIES OF CARBOHYDRATES. 3 J Am Chem Soc 100 373 379 10.1021/ja00470a003 Juaristi E Cuevas G 1992 RECENT STUDIES OF THE ANOMERIC EFFECT Tetrahedron 48 5019 5087 10.1016/s0040-4020(01)90118-8 Kirschner KN Yongye AB Tschampel SM Gonzalez-Outeirino J Daniels CR Foley BL Woods RJ 2008 GLYCAM06: A generalizable Biomolecular force field Carbohydrates Journal of Computational Chemistry 29 622 655 10.1002/jcc.20820 17849372 Kline PC Serianni AS 1990 C-13-ENRICHED RIBONUCLEOSIDES - SYNTHESIS AND APPLICATION OF C-13-H-1 AND C-13-C-13 SPIN-COUPLING CONSTANTS TO ASSESS FURANOSE AND N-GLYCOSIDE BOND CONFORMATIONS J Am Chem Soc 112 7373 7381 10.1021/ja00176a043 Lamarre C 2009 Galactofuranose attenuates cellular adhesion of Aspergillus fumigatus Cellular Microbiology 11 1612 1623 10.1111/j.1462-5822.2009.01352.x 19563461 Latge JP 1994 CHEMICAL AND IMMUNOLOGICAL CHARACTERIZATION OF THE EXTRACELLULAR GALACTOMANNAN OF ASPERGILLUS-FUMIGATUS Infection and Immunity 62 5424 5433 7960122 Leitao EA 2003 beta-Galactofuranose-containing O-linked oligosaccharides present in the cell wall peptidogalactomannan of Aspergillus fumigatus contain immunodominant epitopes Glycobiology 13 681 692 10.1093/glycob/cwg089 12851285 Lemieux RU Koto S 1974 The conformational properties of glycosidic linkages Tetrahedron 30 1933 1944 doi:http://dx.doi.org/10.1016/S0040-4020(01)97324-7 Levitt M Warshel A 1978 Extreme conformational flexibility of the furanose ring in DNA and RNA J Am Chem Soc 100 2607 2613 10.1021/ja00477a004 Lii J-H Chen K-H Allinger NL 2003 Alcohols, ethers, carbohydrates, and related compounds. IV. carbohydrates Journal of Computational Chemistry 24 1504 1513 10.1002/jcc.10271 12868113 Lii JH Chen KH Johnson GP French AD Allinger NL 2005 The external-anomeric torsional effect Carbohydr Res 340 853 862 10.1016/j.carres.2005.01.032 15780251 Longchambon F Avenel D Neuman A 1976 CRYSTAL-STRUCTURE OF ALPHA-D-MANNOPYRANOSE Acta Crystallogr Sect B-Struct Sci 32 1822 1826 10.1107/s0567740876006468 McDonald TRR Beevers CA 1952 THE CRYSTAL AND MOLECULAR STRUCTURE OF ALPHA-GLUCOSE Acta Crystallographica 5 654 659 10.1107/s0365110x52001787 Ohanessian J Avenel D Kanters JA Smits D 1977 2 INDEPENDENT DETERMINATIONS OF CRYSTAL-STRUCTURE OF ALPHA-D-TALOPYRANOSE Acta Crystallogr Sect B-Struct Sci 33 1063 1066 10.1107/s0567740877005378 Ohanessian J Gillierpandraud H 1976 CRYSTAL-STRUCTURE OF ALPHA-D-GALACTOSE Acta Crystallogr Sect B-Struct Sci 32 2810 2813 10.1107/s0567740876008947 Padrta P Sklenar V 2002 Program MULDER - A tool for extracting torsion angles from NMR data Journal of Biomolecular Nmr 24 339 349 10.1023/a:1021656808607 12522298 Plavec J Thibaudeau C Chattopadhyaya J 1994 How Does the 2'-Hydroxy Group Drive the Pseudorotational Equilibrium in Nucleoside and Nucleotide by the Tuning of the 3'-Gauche Effect? J Am Chem Soc 116 6558 6560 10.1021/ja00094a009 Plavec J Thibaudeau C Chattopadhyaya J 1996 How do the energetics of the stereoelectronic gauche and anomeric effects modulate the conformation of nucleos(t)ides? Pure Appl Chem 68 2137 2144 10.1351/pac199668112137 Plavec J Tong W Chattopadhyaya J 1993 How do the gauche and anomeric effects drive the pseudorotational equilibrium of the pentofuranose moiety of nucleosides? J Am Chem Soc 115 9734 9746 10.1021/ja00074a046 Predy GN Goel V Lovlin R Donner A Stitt L Basu TK 2005 Efficacy of an extract of North American ginseng containing poly-furanosyl-pyranosyl-saccharides for preventing upper respiratory tract infections: a randomized controlled trial Canadian Medical Association Journal 173 1043 1048 10.1503/cmaj.1041470 16247099 Raman EP Guvench O MacKerell AD 2010 CHARMM Additive All-Atom Force Field for Glycosidic Linkages in Carbohydrates Involving Furanoses The Journal of Physical Chemistry B 114 12981 12994 10.1021/jp105758h 20845956 Richards MR Bai Y Lowary TL 2013 Comparison between DFT- and NMR-based conformational analysis of methyl galactofuranosides Carbohydr Res 374 103 114 10.1016/j.carres.2013.03.030 23660004 Senderowitz H Parish C Still WC 1996 Carbohydrates: United Atom AMBER*Parameterization of Pyranoses and Simulations Yielding Anomeric Free Energies J Am Chem Soc 118 2078 2086 10.1021/ja9529652 Seo M Castillo N Ganzynkowicz R Daniels CR Woods RJ Lowary TL Roy PN 2008 Approach for the simulation and modeling of flexible rings: Application to the alpha-D-arabinofuranoside ring, a key constituent of polysaccharides from Mycobacterium tuberculosis J Chem Theory Comput 4 184 191 10.1021/ct700284r 25339852 Serianni AS Barker R 1984 C-13 -ENRICHED TETROSES AND TETROFURANOSIDES - AN EVALUATION OF THE RELATIONSHIP BETWEEN NMR PARAMETERS AND FURANOSYL RING CONFORMATION J Org Chem 49 3292 3300 10.1021/jo00192a009 Stenutz R Carmichael I Widmalm G Serianni AS 2002 Hydroxymethyl Group Conformation in Saccharides: Structural Dependencies of 2JHH, 3JHH, and 1JCH Spin Spin Coupling Constants The Journal of Organic Chemistry 67 949 958 10.1021/jo010985i 11856043 Sun G Voigt JH Filippov IV Marquez VE Nicklaus MC 2004 PROSIT: Pseudo-Rotational Online Service and Interactive Tool, Applied to a Conformational Survey of Nucleosides and Nucleotides Journal of Chemical Information and Computer Sciences 44 1752 1762 10.1021/ci049881+ 15446834 Taha HA Castillo N Sears DN Wasylishen RE Lowary TL Roy PN 2010 Conformational Analysis of Arabinofuranosides: Prediction of (3)J(H,H) Using MD Simulations with DFT-Derived Spin-Spin Coupling Profiles J Chem Theory Comput 6 212 222 10.1021/ct900477x 26614332 Taha HA Richards MR Lowary TL 2013 Conformational Analysis of Furanoside-Containing Mono- and Oligosaccharides Chem Rev 113 1851 1876 10.1021/cr300249c 23072490 Taha HA Roy PN Lowary TL 2011 Theoretical Investigations on the Conformation of the beta-D-Arabinofuranoside Ring J Chem Theory Comput 7 420 432 10.1021/ct100450s 26596163 Thibaudeau C Chattopadhyaya J 1997 The discovery of intramolecular stereoelectronic forces that drive the sugar conformation in nucleosides and nucleotides Nucleosides & Nucleotides 16 523 529 10.1080/07328319708002912 Thibaudeau C Plavec J Chattopadhyaya J 1994 Quantitation of the Anomeric Effect in Adenosine and Guanosine by Comparison of the Thermodynamics of the Pseudorotational Equilibrium of the Pentofuranose Moiety in N- and C-Nucleosides J Am Chem Soc 116 8033 8037 10.1021/ja00097a010 Tvaroska I Bleha T 1989 ANOMERIC AND EXO-ANOMERIC EFFECTS IN CARBOHYDRATE CHEMISTRY Tipson RS Horton D Advances in Carbohydrate Chemistry and Biochemistry 45 124 Zhao H Pan Q Zhang W Carmichael I Serianni AS 2007 DFT and NMR Studies of 2JCOH, 3JHCOH, and 3JCCOH Spin-Couplings in Saccharides: C O Torsional Bias and H-Bonding in Aqueous Solution The Journal of Organic Chemistry 72 7071 7082 10.1021/jo0619884 17316047

PMC005xxxxxx/PMC5115637.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101652055 43566 J Nat Sci J Nat Sci Journal of nature and science 2377-2700 27868087 5115637 NIHMS827866 Article Diabetic Wound Healing and Activation of Nrf2 by Herbal Medicine Senger Donald R. 1 Cao Shugeng 2* 1 Department of Pathology and Center for Vascular Biology Research, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, Boston, MA 02215, USA 2 Department of Pharmaceutical Sciences, Daniel K Inouye College of Pharmacy, University of Hawaii at Hilo, 200 W. Kawili Street, Hilo, HI 96720, USA * Corresponding Author. scao@hawaii.edu 6 11 2016 2016 18 11 2016 2 11 e247This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Nrf2 defense is a very important cellular mechanism to control oxidative stress, which is implicated in wound healing. Nrf2 can induce many cytoprotective genes, including HO-1, NQO1 and G6PD. Among many natural products that have been reported as Nrf2 activators, sulforaphane and curcumin have been studied more widely than any others, and both are in clinical trials for non-cancerous disorders. Recently, we reported 4-ethyl catechol and 4-vinyl catechol as Nrf2 co-factors that can induce Nrf2 as potently as sulforaphane and curcumin. These new Nrf2 co-factors were identified in hot aqueous extract of an herbal medicine Barleria lupulina, and fermented Noni (Morinda citrifolia) juice, which are used traditionally for diabetic wound healing. Nrf2 diabetic wound healing herbal medicine Barleria lupulina Morinda citrifolia alkyl catechols The Nrf2 defense pathway In mammals, the master regulator of antioxidant defense is the Nrf2 pathway1,2. The Nrf2 transcription factor controls expression of anti-oxidant and detoxifying enzymes that maintain a healthy cellular redox state and protect against toxic foreign chemical substances through phase II modification3, which neutralizes dangerous species, generating less reactive and more soluble substances that are readily eliminated4. The network of cytoprotective genes regulated by Nrf2 is more than two hundred5, accounting for more than 1% of the genome6. In the absence of stress, Nrf2 with a half-life of about 20 minutes7 is sequestered within the cytosol by the actin-binding protein Kelch-like ECH-associated protein 1 (KEAP1) and Cullin 3, which degrade Nrf2 through ubiquitination8. Under conditions of oxidative stress, Nrf2 is released from Keap1 and rapidly moves to the nucleus to induce transcription of anti-oxidant and detoxifying enzymes. The “redox sensor” mechanism that releases Nrf2 from Keap1, resulting in Nrf2 transport to the nucleus, involves oxidation-sensitive sulfhydryl groups in cysteine residues of Keap19. Nrf2 induction of cytoprotective genes and diabetic wound healing Compelling evidence for the importance of the Nrf2 pathway comes from numerous studies with mice lacking the Nrf2 gene. These Nrf2 null mice exhibit increased sensitivity to a multitude of chemical toxins, resulting in increased inflammation and damage to brain, lung, and kidney2,10. Similarly, Nrf2 null mice are considerably more sensitive to chemical carcinogens, with increased incidence of cancers demonstrated in skin, stomach, colon, and bladder2,11,12. Also, mice engineered with a dominant negative Nrf2 mutant transgene develop skin cancer at three times the frequency of control mice in a classical two-stage model of chemical carcinogenesis13. Additional problems for Nrf2 null mice include impaired liver regeneration14, accelerated UVB-induced photo-ageing of skin15, increased rheumatoid arthritis16, development of lupus-like autoimmune nephritis17, and development of age-related retinopathy18. Thus, the protective importance of the Nrf2 pathway is well established. Diabetes often causes slow-healing wounds that can worsen rapidly. The Keap1-Nrf2 system is a critical target for preventing the onset of diabetes mellitus19. Nrf2 transcription factor, a novel target of keratinocyte growth factor action, regulates gene expression and inflammation in the healing skin wound20. In a streptozotocin-induced diabetes mouse model, Nrf2−/− mice have delayed wound closure rates compared with Nrf2+/+ mice21. Hence, Nrf2 can be activated to potentially heal diabetic wounds and improve overall health. Nrf2 small molecule co-factors were first discovered as “cancer-protective” compounds, which could be utilized for diabetic wound healing Long before the discoveries of Nrf2 and the antioxidant response element (ARE) to which Nrf2 binds, a variety of small chemical compounds were observed to protect rodents from chemically induced carcinogenesis22. Remarkably, these “cancer-protective” compounds were from distinctly different chemical classes, but they all shared the critical property of high susceptibility to oxidation-reduction reactions23. The simplest of these active cancer-protective compounds were identified as 1,4-diphenol (hydroquinone) and 1,2-diphenol (catechol), and more complex examples include the isothiocyanate sulforaphane isolated from broccoli24 and curcumin from the turmeric plant25 (see Figure 1). With the discovery of Nrf2, it became clear that these previously identified redox-sensitive, cancer-protective compounds worked as co-factors for Nrf2 activation26. Collectively, these findings suggested that support of Nrf2 activation by redox-sensitive co-factors, particularly dietary factors such as sulforaphane and curcumin, could be employed as an effective anti-cancer strategy27. This provided impetus for clinical trials28,29, which are continuing. Sulforaphane and curcumin are also currently in clinical trials for non-cancer disorders in which the Nrf2 pathway has been implicated. Current challenges with the application of sulforaphane and curcumin for clinical benefit appear to involve bioavailability of these compounds28–32. Acrolein (CHO-CH=CH) induces Nrf2 translocation and ARE-luciferase reporter activity33. Cinnamaldehyde (Ph-CH=CH-CHO, trans double bond) inhibits thioredoxin reductase and induce Nrf234. Recent studies have demonstrated the protective role of Nrf2 and the potential therapeutic effect of Nrf2 activators, sulforaphane and cinnamaldehyde in a diabetic nephropathy animal model21. The alkyl catechols, 4-ethyl catechol and 4-vinyl catechol, potent Nrf2 co-factors, from Barleria lupulina and Morinda citrifolia, both of which are used traditionally for diabetic wound healing Barleria lupulina (BL): Recently, our work with a traditional Vietnamese medicine (Barleria lupulina, BL) has identified 4-ethyl catechol and 4-vinyl catechol as potently active, natural Nrf2 co-factors35 (Figure 2) in the hot water extract of Barleria lupulina. Although these compounds had been reported, they had not been recognized as important Nrf2 co-factors. Nonetheless, these compounds each satisfy the well-defined structural criteria for “oxidation-reduction lability” that is required for a compound to induce protective enzymes23,36. Interestingly, catechol (Figure 2) was among the first compounds recognized as an Nrf2 co-factor23, but 4-ethyl catechol and 4-vinyl catechol had not received attention. Importantly, we found that 4-ethyl catechol and 4-vinyl catechol are much more potent Nrf2 co-factors than catechol and that they are comparably potent to sulforaphane and curcumin37. Thus, apart from sulforaphane and curcumin, we believe that 4-ethyl catechol and 4-vinyl catechol may be the most potent of the naturally occurring Nrf2 co-factors. Moreover, the relatively small size and simple structure of these compounds, suggests the likelihood of better bioavailability than sulforaphane and curcumin. We have reported the activities of these alkyl catechols recently37. Besides inducing Nrf2, BL, 4-EC and 4-VC significantly improved the organization of the endothelial cell actin cyto-skeleton, reduced actin stress fibers, organized cell-cell junctions, and induced expression of mRNA encoding claudin-5 that is important for formation of endothelial tight junctions and reducing vascular leak35. Morinda citrifolia (Noni): Encouraged by the discovery of new Nrf2 co-factors, 4-MC, 4-EC, and 4-VC from the hot water extract of Barleria lupulina, we screened many herbs for their activity of Nrf2 activation. Only fermented Noni juice (Order Number 809979, Virgin Noni juice, http://www.virginnonijuice.com) showed Nrf2 activation, which was as strong as BL. The traditional medicinal plant Morinda citrifolia L. (Rubiaceace) is believed to have the origin in Southeast Asia and later on distributed to Polynesia. It is called Indian mulberry in India, ba ji tian () in China, nono in Tahiti, and noni in Hawaii38. Noni is now widely cultivated in tropical areas of the South Pacific, including Hawaii39. Traditionally, noni bark and roots were used as dye or clothing, while medicinal usage of all plant parts, including leaves and fruits, were mostly restricted to treat wounds, infections, menstrual cramps, bowel irregularities, diabetes, high blood pressure or as a purgative40. Pacific Islanders and Native Hawaiians consume fresh fruits or noni juice prepared by fermenting the fruits39,41. Claims of its ‘healing powers’ fuelled much of the commercial interest in noni and promoting a worldwide market for noni-based dietary supplements including fruit juice, in North America, Mexico, Australia, and Asia. Young Noni fruits are green, and will turn yellow before ripe, but harvested Noni fruits are mainly whitish (Figure 3), and will turn dark after fermentation. We tested FNJ from ripe Noni fruits, and juices from both green and whitish Noni fruits for their activity of Nrf2 activation. Results showed that juice from young green Noni fruits was inactive, and both juice from ripe whitish Noni fruits and FNJ exhibited Nrf2 activation (Figure 4, Figure 5). Using the same assay-guided separation method as described in our previous publication35, we have identified 4-MC, 4-EC, and 4-VC (Figure 2, Figure 6) in FNJ, which partially accounted for the Nrf2 activation. We are in progress of identifying other Nrf2 activators. At the same time, questions remain why juice from ripe whitish Noni fruits with a fouling smell was active while juice from young green Noni fruits was not, and why people prefer FNJ to juice from ripe whitish Noni fruits. Discussion Nrf2 is a master regulator that can modulate many cytoprotective genes. Many small molecules, including natural products, were reported as Nrf2 activators, but only a few showed strong Nrf2 induction. Sulforaphane and curcumin are in clinical trials for non-cancer disorders, and dimethyl fumarate (CH3O-CO-CH= CH-OC-OCH3, trans), an Nrf2 activator, is also in phase III clinical trials for multiple sclerosis42. Many traditional herbal medicines have been used for diabetic wound healing, but most of the active compounds and mechanisms of actions are unknown. We have studied BL and more recently Noni, and we have identified 4-MC, 4-EC, and 4-VC as Nrf2 co-factors. 4-EC and 4-VC demonstrated Nrf2 induction, which were as potent as sulforaphane and curcumin. Since Nrf2 activation by Noni has not been fully characterized, it is worthy of further investigation. This publication was made possible by grant number R01AT007022 (to D.R.S. and S.C.) from the National Center for Complementary and Integrative Health (NCCIH) at the National Institutes of Health. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NCCIH. This study was also supported by faculty start-up funds from the Daniel K. Inouye College of Pharmacy, University of Hawaii at Hilo (to S.C). Abbreviation BL Barleria lupulina Noni Morinda citrifolia FNJ fermented Noni juice 4-MC 4-methyl catechol 4-EC 4-ethyl catechol 4-VC 4-vinyl catechol Figure 1 Known natural Nrf2 co-factors Figure 2 New Nrf2 co-factors identified in Barleria lupulina and Morinda citrifolia. Figure 3 A Noni tree on campus of University of Hawaii at Hilo and fruits (Young: bottom left, green & hard; Ripe: bottom right, whitish & soft). Figure 4 FNJ induces nuclear translocation of Nrf2 in MVECs MVECs were incubated with FNJ at 1:25 dilution. Staining of Nrf2 in human MVECs incubated with FNJ (right), in comparison with control (left). Green = Nrf2, red = F-actin. Note increased nuclear staining for Nrf2 in cells incubated with FNJ. Overall intensity of Nrf2 staining is also increased, because Nrf2 activation typically involves Nrf2 stabilization in combination with nuclear translocation. Figure 5 Induction of Nrf2 target gene RNAs by FNJ in MVECs, as measured with RT-PCR Y-axis = (mRNA copies)/(106 18S rRNA copies). Nrf2 target genes = heme oxygenase-1 (HO-1), NAD(P)H:quinone oxidoreductase 1 (NQO1), glucose 6-phosphate dehydrogenase (G6PD). Control, non-NRF2 target mRNAs = CD31 (PECAM-1) and VE-cadherin (cadherin-5). FNJ was added to a final dilution of 1:25, and cells were harvested at 24 hours. For all panels, error bars = ± standard deviation (S.D.); n ≥ 4 for each data point. Statistical significance: For HO-1, NQO1, and G6PD panels: FNJ vs. vehicle Ctrl = all extremely significant (p<0.001); for CD31 and VE-cadherin panels: no significant differences. Figure 6 Agilent prep-HPLC HPLC chromatogram of the aqueous acetonitrile eluent of FNJ Sample: FNJ (2 mL/injection, about 150 mg); Flow-rate: 10 mL/min; Solvents: 100% water (0.1% formic acid) for 10 min (0–10 min), then to 100% acetonitrile (0.1% formic acid) in 20 min (10–30 min), finally 100% acetonitrile for 10 min (30–40 min). Y-axis = absorbance at 254nm (mAU), X-axis = minutes. Fraction 11 (tR = 29–31 min) was active and Nrf2 activators 4-MC, 4-EC and 4-VC (minor compounds) were identified. Conflict of Interest: no conflicts declared. 1 Motohashi H Yamamoto M Nrf2-Keap1 defines a physiologically important stress response mechanism Trends in molecular medicine 2004 10 11 549 557 15519281 2 Kensler TW Wakabayashi N Biswal S Cell survival responses to environmental stresses via the Keap1-Nrf2-ARE pathway Annual review of pharmacology and toxicology 2007 47 89 116 3 Hayes JD Dinkova-Kostova AT The Nrf2 regulatory network provides an interface between redox and intermediary metabolism Trends in biochemical sciences 2014 39 4 199 218 24647116 4 Kwak MK Wakabayashi N Kensler TW Chemoprevention through the Keap1-Nrf2 signaling pathway by phase 2 enzyme inducers Mutat Res 2004 555 1–2 133 148 15476857 5 Carmona-Aparicio L Perez-Cruz C Zavala-Tecuapetla C Overview of Nrf2 as Therapeutic Target in Epilepsy International journal of molecular sciences 2015 16 8 18348 18367 26262608 6 Holmstrom KM Baird L Zhang Y Nrf2 impacts cellular bioenergetics by controlling substrate availability for mitochondrial respiration Biology open 2013 2 8 761 770 23951401 7 Kobayashi A Kang MI Okawa H Oxidative stress sensor Keap1 functions as an adaptor for Cul3-based E3 ligase to regulate proteasomal degradation of Nrf2 Molecular and cellular biology 2004 24 16 7130 7139 15282312 8 Itoh K Mimura J Yamamoto M Discovery of the negative regulator of Nrf2, Keap1: a historical overview Antioxidants & redox signaling 2010 13 11 1665 1678 20446768 9 Holland R Fishbein JC Chemistry of the cysteine sensors in Kelch-like ECH-associated protein 1 Antioxidants & redox signaling 2010 13 11 1749 1761 20486763 10 Liu M Grigoryev DN Crow MT Transcription factor Nrf2 is protective during ischemic and nephrotoxic acute kidney injury in mice Kidney international 2009 76 3 277 285 19436334 11 Khor TO Huang MT Prawan A Increased susceptibility of Nrf2 knockout mice to colitis-associated colorectal cancer Cancer prevention research 2008 1 3 187 191 19138955 12 Ramos-Gomez M Kwak MK Dolan PM Sensitivity to carcinogenesis is increased and chemoprotective efficacy of enzyme inducers is lost in nrf2 transcription factor-deficient mice Proceedings of the National Academy of Sciences of the United States of America 2001 98 6 3410 3415 11248092 13 auf dem Keller U Huber M Beyer TA Nrf transcription factors in keratinocytes are essential for skin tumor prevention but not for wound healing Molecular and cellular biology 2006 26 10 3773 3784 16648473 14 Beyer TA Xu W Teupser D Impaired liver regeneration in Nrf2 knockout mice: role of ROS-mediated insulin/IGF-1 resistance The EMBO journal 2008 27 1 212 223 18059474 15 Hirota A Kawachi Y Yamamoto M Koga T Hamada K Otsuka F Acceleration of UVB-induced photoageing in nrf2 gene-deficient mice Experimental dermatology 2011 20 8 664 668 21569103 16 Wruck CJ Fragoulis A Gurzynski A Role of oxidative stress in rheumatoid arthritis: insights from the Nrf2-knockout mice Annals of the rheumatic diseases 2011 70 5 844 850 21173018 17 Yoh K Itoh K Enomoto A Nrf2-deficient female mice develop lupus-like autoimmune nephritis Kidney international 2001 60 4 1343 1353 11576348 18 Zhao Z Chen Y Wang J Age-related retinopathy in NRF2-deficient mice PloS one 2011 6 4 e19456 21559389 19 Uruno A Furusawa Y Yagishita Y The Keap1-Nrf2 system prevents onset of diabetes mellitus Molecular and cellular biology 2013 33 15 2996 3010 23716596 20 Braun S Hanselmann C Gassmann MG Nrf2 transcription factor, a novel target of keratinocyte growth factor action which regulates gene expression and inflammation in the healing skin wound Molecular and cellular biology 2002 22 15 5492 5505 12101242 21 Long M Rojo de la Vega M Wen Q An Essential Role of NRF2 in Diabetic Wound Healing Diabetes 2016 65 3 780 793 26718502 22 Wattenberg LW Coccia JB Lam LK Inhibitory effects of phenolic compounds on benzo(a)pyrene-induced neoplasia Cancer research 1980 40 8 Pt 1 2820 2823 7388831 23 Prochaska HJ De Long MJ Talalay P On the mechanisms of induction of cancer-protective enzymes: a unifying proposal Proceedings of the National Academy of Sciences of the United States of America 1985 82 23 8232 8236 3934671 24 Zhang Y Talalay P Cho CG Posner GH A major inducer of anticarcinogenic protective enzymes from broccoli: isolation and elucidation of structure Proceedings of the National Academy of Sciences of the United States of America 1992 89 6 2399 2403 1549603 25 Rao CV Rivenson A Simi B Reddy BS Chemoprevention of colon carcinogenesis by dietary curcumin, a naturally occurring plant phenolic compound Cancer research 1995 55 2 259 266 7812955 26 Xu C Huang MT Shen G Inhibition of 7,12-dimethylbenz(a)anthracene-induced skin tumorigenesis in C57BL/6 mice by sulforaphane is mediated by nuclear factor E2-related factor 2 Cancer research 2006 66 16 8293 8296 16912211 27 Cornblatt BS Ye L Dinkova-Kostova AT Preclinical and clinical evaluation of sulforaphane for chemoprevention in the breast Carcinogenesis 2007 28 7 1485 1490 17347138 28 Egner PA Chen JG Wang JB Bioavailability of Sulforaphane from two broccoli sprout beverages: results of a short-term, cross-over clinical trial in Qidong, China Cancer prevention research 2011 4 3 384 395 21372038 29 Kensler TW Egner PA Agyeman AS Keap1-nrf2 signaling: a target for cancer prevention by sulforaphane Topics in current chemistry 2013 329 163 177 22752583 30 Shureiqi I Baron JA Curcumin chemoprevention: the long road to clinical translation Cancer prevention research 2011 4 3 296 298 21372027 31 Carroll RE Benya RV Turgeon DK Phase IIa clinical trial of curcumin for the prevention of colorectal neoplasia Cancer prevention research 2011 4 3 354 364 21372035 32 Prasad S Tyagi AK Aggarwal BB Recent developments in delivery, bioavailability, absorption and metabolism of curcumin: the golden pigment from golden spice Cancer research and treatment: official journal of Korean Cancer Association 2014 46 1 2 18 24520218 33 Wu CC Hsieh CW Lai PH Lin JB Liu YC Wung BS Upregulation of endothelial heme oxygenase-1 expression through the activation of the JNK pathway by sublethal concentrations of acrolein Toxicology and applied pharmacology 2006 214 3 244 252 16480751 34 Wondrak GT Cabello CM Villeneuve NF Cinnamoyl-based Nrf2-activators targeting human skin cell photo-oxidative stress Free radical biology & medicine 2008 45 4 385 395 18482591 35 Senger DR Hoang MV Kim KH Li C Cao S Anti-inflammatory activity of Barleria lupulina: Identification of active compounds that activate the Nrf2 cell defense pathway, organize cortical actin, reduce stress fibers, and improve cell junctions in microvascular endothelial cells J Ethnopharmacol 2016 36 Bensasson RV Zoete V Dinkova-Kostova AT Talalay P Two-step mechanism of induction of the gene expression of a prototypic cancer-protective enzyme by diphenols Chemical research in toxicology 2008 21 4 805 812 18361512 37 Senger DR Li D Jaminet S-C Cao S Activation of the Nrf2 Cell Defense Pathway by Ancient Foods: Disease Prevention by Important Molecules and Microbes Lost from the Modern Western Diet PloS one 2016 11 2 e0148042 26885667 38 Abbott IA Shimazu C The geographic origin of the plants most commonly used for medicine by Hawaiians J Ethnopharmacol 1985 14 2–3 213 222 4094468 39 Wang MY West BJ Jensen CJ Morinda citrifolia (Noni): a literature review and recent advances in Noni research Acta pharmacologica Sinica 2002 23 12 1127 1141 12466051 40 Brown AC Anticancer activity of Morinda citrifolia (Noni) fruit: a review Phytotherapy research: PTR 2012 26 10 1427 1440 22344842 41 McClatchey W From Polynesian healers to health food stores: changing perspectives of Morinda citrifolia (Rubiaceae) Integrative cancer therapies 2002 1 2 110 120 14664736 42 Fox RJ Miller DH Phillips JT Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis N Engl J Med 2012 367 1087 1097 22992072

PMC005xxxxxx/PMC5115934.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7704136 4076 Histopathology Histopathology Histopathology 0309-0167 1365-2559 27374168 5115934 10.1111/his.13027 NIHMS799889 Article A Proportion of Primary Squamous Cell Carcinomas of the Parotid Gland Harbor High Risk Human Papillomavirus Xu Bin 1 Wang Lu 1 Borsu Laetitia 1 Ghossein Ronald 1 Katabi Nora 1 Ganly Ian 2 Dogan Snjezana 1 1 Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, NY 2 Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY Corresponding Author: Snjezana Dogan, Department of Pathology, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, New York, 10065, dogans@mskcc.org 1 9 2016 23 8 2016 12 2016 01 12 2017 69 6 921929 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Aims In the current study, we aimed to examine primary parotid squamous cell carcinoma (ParSCC) for the presence of HR-HPV and associated molecular alterations. Methods and results Eight cases of ParSCC were retrieved after a detailed clinicopathologic review to exclude a possibility of metastasis and/or extension from another primary site. HR-HPV status was determined based on immunohistochemistry (IHC) for p16 protein expression and by chromogenic in situ hybridization (CISH) for HR-HPV. All cases were genotyped by multiplexed mass spectrometry assay interrogating 91 hotspot mutations in 8 cancer-related genes (EGFR, KRAS, NRAS, BRAF, PIK3CA, AKT1, MEK1 and ERBB2), and studied by fluorescence in situ hybridization (FISH) for PTEN copy number alteration. Three of 8 cases (37.5%) were positive for presence of HR-HPV by CISH and p16 IHC. One of three (33%) HR-HPV-positive cases harbored PTEN hemizygous deletion, and one (33%) HR-HPV-positive case harbored PIK3CA E545K somatic mutation. No alteration of PTEN-PI3K pathway was detected in HR-HPV-negative tumors. Over a median follow-up period of 66.2 months, only the patient with HR-HPV-positive PIK3CA-mutated tumor died of his disease, while the remaining 7 patients were disease free. Conclusions Given the established etiologic role of HR-HPV in other head and neck squamous cell carcinoma, it is likely that HR-HPV represents an oncogenic driver in the pathogenesis of more than one third of ParSCC. Presence of HR-HPV in ParSCC may be coupled with alterations in PTEN-PI3K pathway. HR-HPV and molecular characterization of a larger number of ParSCC is needed to determine the clinical significance of these findings. human papillomavirus squamous cell carcinoma parotid gland p16 in situ hybridization Introduction High risk human papillomavirus (HR-HPV) is a well-established oncogenic agent causing about 5% of human cancers worldwide, in particular the vast majority of cervical and oropharyngeal carcinomas, and a subset of anogenital squamous cell carcinoma 1–3. Over the past two decades, the prevalence of HR-HPV in oropharyngeal squamous cell carcinoma (SCC) has increased dramatically, from 21% prior to 1990 to 75% at present 3, 4. More importantly, recent research has shown that HR-HPV is an important prognostic and predictive biomarker in oropharyngeal squamous carcinoma. HPV-positive oropharyngeal SCC tends to affect relatively younger individuals that those usually affected by conventional HPV-negative SCC, and is associated with a favorable overall prognosis, excellent loco-regional control, prolonged disease specific survival, and improved response to radiation therapy 5, 6. The clinical significance of HR-HPV in head and neck SCC outside of the oropharynx, in particular salivary glands, remains unclear. A handful of published studies have investigated the effects of HPV, including HR-HPVs, in epithelial neoplasms, benign or malignant, of salivary glands 7–15. The results were somewhat inconsistent with a documented incidence of HPV ranging from nil to 100%. In addition, the majority of prior published studies have employed polymerase chain reaction (PCR)-based assays. Despite its well-documented high sensitivity, a positive PCR result does not necessarily imply the integration of the viral oncogenes into the host DNA, which is the first initiating step leading to malignant transformation. Primary squamous cell carcinoma of the parotid gland (parSCC) is a rare and aggressive malignant epithelial tumor with a 5-year disease specific survival rate of 33% to 50% 16, 17. To date, only one published study has reported HPV positivity in a single case of squamous cell carcinoma of salivary gland, using a nested two-step-PCR assay and customized primers targeting both high risk and low risk HPVs 8. In the present study, our aim was to examine ParSCC for the presence of HR-HPV and any associated molecular alterations. Additionally, we performed molecular and cytogenetic analyses to explore the molecular alterations of HPV-positive and negative parSCCs. Material and Methods Case selection and characteristics and Confirmation of HPV status using p16 immunohistochemistry and CISH The study was approved by the institutional review board. Eight cases fulfilling the following criteria were retrieved from the pathology database: 1) patients who underwent surgical resection at Memorial Sloan Kettering Cancer Center (MSKCC, New York, NY, US) between 2000 to 2014; 2) a final pathological diagnosis of squamous cell carcinoma of the parotid gland (parSCC); 3) no documented prior history of squamous cell carcinoma in the head and neck region; 4) no squamous cell carcinoma outside of the parotid gland detected on extensive radiological, clinical, and endoscopic work-up; 5) no evidence of direct connection of the tumor to the skin on radiological studies, macroscopic and microscopic examination; and 6) an epicenter of the tumor within the parotid gland by imaging and/or macroscopic examination. All histologic and immunohistochemistry slides were reviewed by four head and neck pathologists. The pathologic and clinical stages were assigned using the American Joint Committee on Cancer (AJCC) staging manual 18. Immunohistochemistry studies were performed to confirm the presence of squamous differentiation and to exclude diagnostic mimics. Antibodies and ISH probes utilized are listed in Table 1. HPV status was confirmed using p16 immunostain and CISH against HPV types 16, 18, 31, 33, and 51. Immunopositivity for p16 was defined as a diffuse and strong cytoplasmic and nuclear stain of p16 in at least 70% of tumor cells (Figure 1G). A carcinoma was considered as positive for HR-HPV only when the integrated pattern was observed (Figure 1H). Additionally, FISH for PLAG1, MAML2 or EWSR-1 loci was performed in three individual cases to exclude diagnostic mimickers. Clinico-pathological features captured included: age, gender, smoking history, duration of clinical follow-up, disease status at the last follow-up, clinical staging, size of the tumor, tumor necrosis, mitotic index,, perineural invasion, lymphovascular invasion, AJCC pT and pN stages 18. Molecular and cytogenetic analyses All cases were genotyped using a multiplexed mass spectrometry assay for hotspot alterations (Sequenom) based on matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). DNA from the tumor samples was used to interrogate presence of single nucleotide variation (SNV) in 91 hot-spots in 8 oncogenes: EGFR, KRAS, BRAF, PIK3CA, AKT1, NRAS, MEK1, and ERBB2 as previously described 19. Immunohistochemistry studies and FISH were performed to investigate the status of PTEN in these tumors. Results A total of 8 patients were identified from the MSKCC pathology database, fulfilling the inclusion criteria stated above. All eight cases harbored poorly differentiated predominantly non-keratinizing squamous cell carcinoma, among which four exhibited focal abrupt keratinization (Figure 1). All eight tumors studied were infiltrative and were devoid of lymphoid cuff. The squamous differentiation was confirmed by diffuse and strong immunopositivity for p63 (7/8, 88%), cytokeratin 34βE12 (5/5, 100%), Cytokeratin 5/6 (4/4, 100%) and/or p40 (8/8, 100%); as well as immuno-negativity for androgen receptor (0/8), calponin (0/3), smooth muscle actin (0/4), and S100 (0/4). A positive HPV status, as confirmed using CISH positivity for HR-HPV, was identified in 3 tumors (38%). All tumors that were positive for HR-HPV on CISH analysis showed strong nuclear and cytoplasmic p16 immunostain in > 70% of cells; while only one of the five HR-HPV negative cases were positive for p16 immunostain. The demographic, clinical and pathological features of HPV-positive and HPV-negative carcinomas are summarized in Table 2. In our series, primary parSCC affected predominantly elderly male patients. The median age of presentation was 68 years and the male to female ratio was 7:1. None of the patients had an identifiable history of previous radiation to the head and neck region. Three patients were ex-smokers (10 to 70 pack-years), while the remaining were non-smokers. No noticeable difference was identified between the HPV-positive and HPV-negative carcinomas in terms of age, gender, and risk factors. HPV-positive carcinomas were associated with a higher risk of lymphovascular invasion and lymph node metastases, compared to their HPV-negative counterparts. The remaining histological features, including tumor size, mitotic index, tumor necrosis, perineural invasion, AJCC pT and pN staging, were indistinguishable between HPV-positive and negative tumors. The median follow-up period was 61 months (range: 8 to 150 months). All patients underwent parotidectomy, selective neck dissection, and adjuvant radiation therapy at initial presentation. Seven patients were disease-free at the time of last follow-up, including 5 HPV-negative and 2 HPV-positive cases. One patient who was diagnosed with pT3 pN2b HPV-positive poorly differentiated SCC developed a neck recurrence and possible metastases in the mediastinum and lungs 5 months post-surgery and died 4 months later. While no mutation was identified using Sequenom assay from the tumors of the 7 patients who were disease-free at last follow-up, a missense somatic mutation of PIK3CA c.1633G>A (p.E545K) on exon 9 was detected in the patient with HPV-positive parSCC who suffered disease-specific death. Additionally, one HPV-positive parSCC demonstrated complete loss of PTEN protein expression, and a somatic hemizygous PTEN deletion on FISH assay (Figure 1E and 1F). The remaining tumors did not show PTEN deletion. Discussion Recent epidemiological data have implicated a key pathogenesis role of HR-HPV in head and neck squamous cell carcinoma, particularly those arising within the oropharynx 2, 20. In addition, HR-HPV has emerged as a prognostic and predictive biomarker in oropharyngeal SCC. A positive HR-HPV status has been associated with a favorable loco-regional control, overall survival, and disease specific survival, as well as an improved response to radiation therapy 5, 6. The roles of HR-HPV in head and neck cancer outside of the oropharynx, in particular within the salivary glands, remain unclear. Emerging yet scanty and controversial evidence has been published demonstrating the presence of HPV in certain but not all types of salivary gland epithelial neoplasms with a reported incidence ranging widely from nil to 100% (reviewed and summarized in Table 3) 7–15. Since the parotid gland directly communicates with the oral cavity through Stensen’s duct, it is possible that HR-HPV could theoretically gain access to the parotid gland via retrograde transportation. Using ISH or PCR techniques, HR-HPV positivity was detected in a variety of benign and malignant salivary gland epithelial lesions, including pleomorphic adenoma, oncocytoma, Warthin’s tumor, squamous cell carcinoma, acinic cell carcinoma, adenoid cystic carcinoma, adenocarcinoma, and mucoepidermoid carcinoma (Table 3) 7–15. However, most of these studies included only a limited number of cases, and the documented frequency of any particular entity was not always consistent across studies. For example, Vagali et al. (2007) 9 and Teymoortash et al. (2013) 14 showed that HPV DNA was absent in Warthin’s tumor, a benign salivary epithelial neoplasm. In contrast, Teng et al. (2014) 15 detected the presence of HR-HPVs, including HPV 16 and 18, in 4 of 12 (33%) Warthin’s tumors. Similarly, the reported incidence of HR-HPV in mucoepidermoid carcinoma, the most common malignant salivary gland epithelial tumor, varied from nil 7, 11 to 47 – 100% 12, 15. With regard to squamous cell carcinoma of the parotid gland, the only case with documented HPV status has been reported by Fischer and Von Winterfeld (2003) 8 who detected the presence of HPV in a single case of squamous cell carcinoma of the parotid gland. The authors employed a nested two step-PCR assay with customized primers targeting HPV 6, 11, 13, 16, 31 and 33. Thus, it was unclear whether this specific tumor harbored HR or low risk-HPV. The marked variation in the reported frequency of HPV in salivary gland tumors might be due to the small number of cases included in these studies, but could also be attributed to the different detection methods used. Although HPV infection is nearly ubiquitous in humans, the vast majority are transient and not carcinogenic. Thus, the ultimate goal of HPV detection is to recognize oncogenic HR-HPV infection. PCR-based assays, in particular nested PCR, are extremely sensitive. Therefore, a positive PCR result using GP5+/6+ primers detects the presence of the virus genome, even at a very low copy number, and may not represent clinically relevant HPV infection. Indeed, HPV is only implicated in tumorigenesis when E6/E7 mRNA is transcriptionally activated, while a significant proportion (14–50%) of carcinomas with detected HR-HPV genomic DNA using PCR do not contain transcribed E6/E7 mRNA 1, 21–23. Moreover, PCR techniques do not distinguish between episomal and integrated HPV DNA, or between non-neoplastic and neoplastic tissue, leading to a further increase in the detection of clinically insignificant HPV infection. In situ hybridization (ISH), on the other hand, is a highly specific method allowing reliable topographical visualization of HPV within the nuclei of tumor cells. Additionally, the ISH technique allows discrimination between the integrated and the episomal state of HPV. Compared with the diffuse nuclear staining for episomal virus, a punctuated nuclear signal indicates integration of the viral DNA into the host DNA, which would then trigger carcinogenesis in the host cell 1, 21–23. P16 protein overexpression, detected by immunohistochemistry, is an indirect consequence of E7 oncogene transcription, and can serve as a cost-effective surrogate marker with 94–100% sensitivity and 79 – 82% of specificity for tumorigenic HR-HPV infection 1, 21–23. The majority of previously published studies have adopted PCR-based assays 7–9, 11, 12, 14, 15. Hence, the detection of HPV in these studies did not necessarily indicate a pathogenic role of HR-HPV in these tumors. Only three groups have studied the status of HPV in salivary gland tumor using ISH techniques 10, 13, 14. While Seethala et al. (2012) 13 did not detect HR-HPV in 7 cases of lymphadenoma Boland et al. (2012) 10 identified HR-HPV in 2 of 25 (8%) cases of adenoid cystic carcinoma; and Teymoortash et al. (2013) 14 reported HR-HPV positivity in 13 of 40 (33%) Warthin’s tumors. Interestingly, the ISH staining pattern in HPV-positive Warthin’s tumors was episomal 14, suggesting that the HR-HPV may not be pathogenic in Warthin’s tumor. The current study is the first study showing that some primary SCCs of the parotid gland (38%) may be driven by HR-HPV as demonstrated by positive CISH studies and confirmed by immunohistochemistry for p16. One potential weakness of the present study was that the CISH probe utilized only detected the most common HR-HPV types, namely 16, 18, 31, 33, and 51, but not other rare types of HR-HPVs. The current study is also the only study with a long-term clinical follow-up allowing evaluation of the clinical relevance and prognostic value of HR-HPV in these tumors. We intentionally excluded cases from our consult service, and included only those patients who received surgical resection and clinical follow-up at our center. This approach minimized selection bias, ensured comparable and consistent results of HPV detection, and allowed collection of reliable long term clinical data. Unlike SCC of the oropharynx which is associated with an improved clinical outcome 2, 20, our study showed that HPV positivity in SCC of the parotid gland had no significant impact on clinical and pathological staging and prognosis. However, the small number of cases is a limiting factor, and additional larger scale studies are needed. The facts that HR-HPV related parotid squamous cell carcinomas was associated with a propensity for lymphovascular invasion and metastases to neck lymph nodes may raise concerns that these cases are parotid metastases rather than primary squamous cell carcinoma of the parotid gland. However, this possibility was excluded on the basis of an extensive clinico-radiological examination and/or multiple benign biopsies of the oropharynx. Additionally, two of three patients with HR HPV-related ParSCC showed no evidence of disease 23 and 66 months after the initial resection, which also argues against the metastatic nature of their disease. Recent genomic evidence has shown that alteration in PTEN-PIK3CA-AKT-mTOR pathway is the most frequent genomic alteration detected in 48 to 53% of HPV-positive head and neck squamous cell carcinomas 24–28. Such alterations include PTEN inactivation by mutation and deletion, and PIK3CA mutation or amplification. Using Sanger sequencing, Chiosea et al. (2013) 24 have detected PIK3CA alteration in 31% of oropharyngeal SCC, the most common one being p.E545K missense mutation. Similarly, using massive parallel high throughout DNA sequencing, three recent studies have shown that PIK3CA and PTEN alterations are present in 37% and 11% of HPV-positive head and neck squamous cell carcinomas, respectively 26, 27, 29. Consistent with what has been reported in the literature, we identified PIK3CA E545K mutation (1/3, 33%) and PTEN hemizygous deletion (1/3, 33%) in HPV-positive parSCCs. Although previous studies reported that the presence of PIK3CA alteration did not alter clinical prognosis of head and neck squamous cell carcinoma 24, the patient with HPV-associated SCC harboring PIK3CA somatic mutation in our series was the only patient who suffered adverse outcomes, including distant recurrence and ultimately disease specific death. As SCC is a rare malignancy of salivary gland, future collaborative studies across different tertiary centers are needed to collect sufficient evidence in order to clarify the prognostic value of PIK3CA somatic mutation in HPV-positive and HPV-negative SCCs of parotid gland. Conclusions To the best of our knowledge, our study is the first to report the presence of high risk human papillomavirus in a proportion of primary squamous cell carcinoma of the parotid gland suggesting that HPV may contribute to the pathogenesis of parSCCs. PTEN hemizygous deletion and PIK3CA somatic mutation were detected in HPV-positive primary SCCs, suggesting that alteration in PTEN-PIK3CA-AKT-mTOR pathway may play a role in HPV-related parSCC. Unlike oropharyngeal squamous cell carcinoma in which HPV is associated with an improved survival, HPV in parotid SCC may not correlate with better outcome. However, future studies with larger number of cases are needed to confirm such findings. Grant numbers and sources of support: Research reported in this publication was supported in part by the Cancer Center Support Grant of the National Institutes of Health/National Cancer Institute under award number P30CA008748. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Figure 1 Table 1 Antibodies and in situ hybridization probes utilized in the current study. Antibody specificity Clone Dilution Source p16 E6H4 RTU Ventana Medical Systems Inc p63 4A4 RTU Ventana Medical Systems Inc p40 PC373 1: 3000 EMD Millipore Cytokeratin 5/6 D5/16B4 RTU Ventana Medical Systems Inc Cytokeratin 34βE12 34BE12 RTU Ventana Medical Systems Inc Calponin EP789Y RTU Ventana Medical Systems Inc Smooth muscle actin asm-1 1:50 Vector laboratory S100 Z0311 1:8000 DAKO PTEN 6H2.1 1:100 DAKO Androgen receptor AR441 1:200 DAKO CISH probe for high risk HPV type 16, 18, 3133, and 51 PATHO-GENE® NA ENZO Life Sciences Inc. FISH probe for PTEN Vysis® LSI PTEN/CEP 10 FISH probe NA Abbott Laboratories RTU: ready to use Table 2 Clinicopathologic characteristics of high risk human papillomavirus (HR-HPV)-negative and positive squamous cell carcinoma in the parotid gland a. Overall HR-HPV (-) HR-HPV (+) N 8 5 3 Clinical characteristics Age, median ( range) 68 (51-87) 67(51 – 87) 70 (58 – 81) Gender Male 7 5 2 Female 1 0 1 Smoking status Non-smoker 5 3 2 Ex-Smoker 3 2 1 Clinical staging II 4 4 0 III 1 1 0 IVa 3 0 3 Follow-up period (months)a 61 (8 – 150) 78 (21 – 150) 31 (8 – 69) Disease status at last F/U No evidence of disease 7 5 2 Dead of disease 1 0 1 Pathological characteristics Tumor Size (cm) 3.7 (2.1 – 5.0) 3.1 (2.1 – 4.5) 4.7 (2.0 – 6.0) Mitotic index (/10 HPFs) 12 (2 – 38) 11 (2 – 38) 13 (9 – 19) Surgical Margin Status Negative 4 3 1 Close (<0.1 cm) 1 1 0 Positive 3 1 2 Lymphovascular invasion Yes 4 1 3 No 4 4 0 Perineural invasion Yes 5 2 3 No 3 3 0 Tumor necrosis Yes 5 2 3 No 3 3 0 AJCC pT stage T2 4 3 1 T3 3 2 1 T4a 1 0 1 AJCC pN stage N0 5 5 0 N2b 3 0 3 PIK3CA E545K mutation Yes 1 0 1 No 7 5 2 Loss of PTEN immunoexpression and PTEN deletion on FISH analysis Yes 1 0 1 No 7 5 2 a Values were expressed as N or mean (range). AJCC: American Joint Committee on Cancer, FISH: fluorescence in situ hybridization, F/U: clinical follow-up, HPFs: high power fields (400X), HPV: human papilloma virus. Table 3 Reported frequency of HPV in salivary gland epithelial neoplasms in English literature Ref. Tumor tested (N) HPV detection methods primers/probe) HPV N (%) HR- HPV N (%) HPV16 N (%) HPV18 N (%) Atula, 1998 [7] N = 34 PA (19) ACC (4) SCC (3) MEC (3) Adenocarcinoma (2) AdCC (1) EMC (1) MC (1) PCR (GP5+/GP6+) 0 N/A N/A N/A Fischer, 2003 [8] SCC of parotid (1) Nested PCR (customized primers) 1 (100%) N/A N/A N/A Vageli, 2007 [9] N = 8 Oncocytoma (1) ACC (1) Warthin’s tumor (1) Adenocarcinoma HG (1) PA (2) Lymphoepithelial cyst (1) PLGA (1) Solution PCR (GP5+/GP6+) Multiplex qRT-PCR (E6/E7 of HPV16, 18, 31, 33, 35, 52, 58 and 67) In situ PCR 6 (75%) 6 (75%) 5 (63%) 4 (50%) Boland, 2012 [10] AdCC (n = 25) ISH for LR- and HR- HPV 2 (8%) 2 (8%) N/A N/A Descamps, 2012 [11] N = 79 PA (40) AdCC (15) MEC (9) CAexPA (9) ACC (6) PCR (GP5+/GP6+) qRT-PCR (type specific sequences) 4 (5%) PA (3) ACC (1) 2 (3%) PA (1) ACC (1) 2 (3%) 0 Isayeva, 2012 [12] MEC (89) Nested RT-PCR (primers N/A) IF (C1P5 antibody) N/A 42 (47%) N/A N/A Seethala 2012 [13] Lymphadenoma (7) ISH (wide spectrum HPV-DNA probe) 0 N/A N/A N/A Teymoortash, 2013 [14] Warthin’s tumor (40) ISH for HR-HPV PCR (GP5+/GP6+ and E1 gene regions) 13 (33%) episomal by FISH 0 by PCR N/A N/A N/A Teng, 2014 [15] N = 59 PA (36) Adenoma (3) Warthin’s tumor (12) Myoepithelioma (1) Adenocarcinoma (3) AdCC (1) MEC (1) Others (2) PCR (37 LR and HR HPV type-specific probes) 34 (58%) N/A 11 (19%) 14 (24%) ACC: acinic cell carcinoma, AdCC: adenoid cystic carcinoma, CAexPA: carcinoma ex-pleomorphic adenoma, EMC: epithelial myoepithelial carcinoma, HPV: human papillomavirus, HR-HPV: high risk-human papillomavirus, IF: immunofluorescence, ISH: in situ hybridization, LR-HPV: low risk-human papillomavirus, MC: myoepithelial carcinoma, MEC: myoepithelial carcinoma, N/A: not available, PA: pleomorphic adenoma, PCR: polymerase chain reaction, PLGA: polymorphous low-grade adenocarcinoma, qRT-PCR: quantitative real-time PCR, Ref. References, RT-PCR: reverse transcriptase PCR, SCC: squamous cell carcinoma. Disclosure/conflict of interest: No competing financial interests exist for all contributory authors. Author contributions: Bin Xu: Organized the database, reviewed the immunohistochemistry and histologic slides, and drafted the manuscript. Lu Wang: Performed the fluorescence in situ hybridization studies. Laetitia Borsu: Performed the sequenom (mass spectrum) studies. Ronald Ghossein: Reviewed the pathology, and provided feedbacks on the study design and manuscript. Nora Katabi: Reviewed the pathology, and provided feedbacks on the study design and manuscript. Ian Ganly: Provide clinical aspect of the study and feedbacks on the manuscript. Snjezana Dogan: Designed the study, reviewed all pathology and molecular aspects of the study, and finalized the manuscript. References 1 Groves IJ Coleman N Pathogenesis of human papillomavirus-associated mucosal disease The Journal of pathology 2015 235 527 538 25604863 2 Woods R Sr. O'Regan EM Kennedy S Martin C O'Leary JJ Timon C Role of human papillomavirus in oropharyngeal squamous cell carcinoma: A review World journal of clinical cases 2014 2 172 193 24945004 3 Stein AP Saha S Yu M Kimple RJ Lambert PF Prevalence of human papillomavirus in oropharyngeal squamous cell carcinoma in the united states across time Chemical research in toxicology 2014 27 462 469 24641254 4 Young D Xiao CC Murphy B Moore M Fakhry C Day TA Increase in head and neck cancer in younger patients due to human papillomavirus (hpv) Oral oncology 2015 5 Masterson L Moualed D Liu ZW De-escalation treatment protocols for human papillomavirus-associated oropharyngeal squamous cell carcinoma: A systematic review and meta-analysis of current clinical trials European journal of cancer 2014 50 2636 2648 25091798 6 Petrelli F Sarti E Barni S Predictive value of human papillomavirus in oropharyngeal carcinoma treated with radiotherapy: An updated systematic review and meta-analysis of 30 trials Head & neck 2014 36 750 759 23606404 7 Atula T Grenman R Klemi P Syrjanen S Human papillomavirus, epstein-barr virus, human herpesvirus 8 and human cytomegalovirus involvement in salivary gland tumours Oral oncology 1998 34 391 395 9861347 8 Fischer M von Winterfeld F Evaluation and application of a broad-spectrum polymerase chain reaction assay for human papillomaviruses in the screening of squamous cell tumours of the head and neck Acta oto-laryngologica 2003 123 752 758 12953779 9 Vageli D Sourvinos G Ioannou M Koukoulis GK Spandidos DA High-risk human papillomavirus (hpv) in parotid lesions The International journal of biological markers 2007 22 239 244 18161653 10 Boland JM McPhail ED Garcia JJ Lewis JE Schembri-Wismayer DJ Detection of human papilloma virus and p16 expression in high-grade adenoid cystic carcinoma of the head and neck Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 2012 25 529 536 11 Descamps G Duray A Rodriguez A Detection and quantification of human papillomavirus in benign and malignant parotid lesions Anticancer research 2012 32 3929 3932 22993339 12 Isayeva T Li Y Maswahu D Brandwein-Gensler M Human papillomavirus in non-oropharyngeal head and neck cancers: A systematic literature review Head and neck pathology 2012 6 Suppl 1 S104 S120 22782230 13 Seethala RR Thompson LD Gnepp DR Lymphadenoma of the salivary gland: Clinicopathological and immunohistochemical analysis of 33 tumors Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 2012 25 26 35 14 Teymoortash A Bohne F Jonsdottir T Human papilloma virus (hpv) is not implicated in the etiology of warthin's tumor of the parotid gland Acta oto-laryngologica 2013 133 972 976 23944949 15 Teng WQ Chen XP Xue XC Distribution of 37 human papillomavirus types in parotid gland tumor tissues Oncology letters 2014 7 834 838 24527091 16 Lee S Kim GE Park CS Primary squamous cell carcinoma of the parotid gland American journal of otolaryngology 2001 22 400 406 11713725 17 Wang L Li H Yang Z Chen W Zhang Q Outcomes of primary squamous cell carcinoma of major salivary glands treated by surgery with or without postoperative radiotherapy Journal of oral and maxillofacial surgery : official journal of the American Association of Oral and Maxillofacial Surgeons 2015 18 Edge SB Byrd DR Compton CC Fritz AG Greene FL Trotti A Ajcc cancer staging manual 2009 Springer 19 Arcila M Lau C Nafa K Ladanyi M Detection of kras and braf mutations in colorectal carcinoma roles for high-sensitivity locked nucleic acid-pcr sequencing and broad-spectrum mass spectrometry genotyping The Journal of molecular diagnostics : JMD 2011 13 64 73 21227396 20 El-Mofty SK Hpv-related squamous cell carcinoma variants in the head and neck Head and neck pathology 2012 6 Suppl 1 S55 S62 22782224 21 Mirghani H Amen F Moreau F Human papilloma virus testing in oropharyngeal squamous cell carcinoma: What the clinician should know Oral oncology 2014 50 1 9 24169585 22 Mirghani H Amen F Moreau F Lacau St Guily J Do high-risk human papillomaviruses cause oral cavity squamous cell carcinoma? Oral oncology 2014 23 Venuti A Paolini F Hpv detection methods in head and neck cancer Head and neck pathology 2012 6 Suppl 1 S63 S74 22782225 24 Chiosea SI Grandis JR Lui VW Pik3ca, hras and pten in human papillomavirus positive oropharyngeal squamous cell carcinoma BMC cancer 2013 13 602 24341335 25 Chun SH Jung CK Won HS Kang JH Kim YS Kim MS Divergence of p53, pten, pi3k, akt and mtor expression in tonsillar cancer Head & neck 2014 26 Cancer Genome Atlas N Comprehensive genomic characterization of head and neck squamous cell carcinomas Nature 2015 517 576 582 25631445 27 Chung CH Guthrie VB Masica DL Genomic alterations in head and neck squamous cell carcinoma determined by cancer gene-targeted sequencing Ann Oncol 2015 26 1216 1223 25712460 28 Lechner M Frampton GM Fenton T Targeted next-generation sequencing of head and neck squamous cell carcinoma identifies novel genetic alterations in hpv+ and hpv− tumors Genome Med 2013 5 49 23718828 29 Seiwert TY Zuo Z Keck MK Integrative and comparative genomic analysis of hpv-positive and hpv-negative head and neck squamous cell carcinomas Clinical cancer research : an official journal of the American Association for Cancer Research 2014

PMC005xxxxxx/PMC5115953.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8302946 4093 Hepatology Hepatology Hepatology (Baltimore, Md.) 0270-9139 1527-3350 27628766 5115953 10.1002/hep.28811 NIHMS817609 Article IRAKM-Mincle axis links cell death to inflammation: Pathophysiological implications for chronic alcoholic liver disease Zhou Hao 1 Yu Minjia 1 Zhao Junjie 1 Martin Bradley N. 1 Roychowdhury Sanjoy 2 McMullen Megan R. 2 Wang Emily 1 Fox Paul L. 3 Yamasaki Sho 4 Nagy Laura E. 256 Li Xiaoxia 16 1 Department of Immunology, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA 2 Center for Liver Disease Research, Department of Pathobiology, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA 3 Department of Cellular and Molecular Medicine, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA 4 Division of Molecular Immunology, Research Center for Infectious Diseases, Medical Institute of Bioregulation, Kyushu University, 3-1-1 Maidashi Higashiku, Fukuoka, Japan 5 Department of Gastroenterology, Cleveland Clinic, Cleveland, Ohio, USA Correspondence should be addressed to L. E. N. (nagyL3@ccf.org) or X. L. (lix@ccf.org) 6 These authors contributed equally to this work. 22 9 2016 25 10 2016 12 2016 01 12 2017 64 6 19781993 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Lipopolysaccharide (LPS)-mediated activation of Toll-like receptors (TLRs) in hepatic macrophages and injury to hepatocytes are major contributors to the pathogenesis of alcoholic liver disease (ALD). However, the mechanisms by which TLR-dependent inflammatory responses and alcohol-induced hepatocellular damage coordinately lead to ALD are not completely understood. In this study, we found that mice deficient in IRAKM, a proximal Toll-like receptor pathway molecule typically associated with inhibition of TLR signaling, were actually protected from chronic ethanol-induced liver injury. In bone marrow derived macrophages challenged with low concentrations of LPS, which reflect the relevant pathophysiological levels of LPS in both alcoholic patients and ethanol-fed mice, the IRAKM Myddosome was preferentially formed. Further, the IRAKM Myddosome mediated the up-regulation of Mincle, a sensor for cell death. Mincle-deficient mice were also protected from ethanol-induced liver injury. The endogenous Mincle ligand, SAP130 is a danger signal released by damaged cells; culture of hepatocytes with ethanol increased the release of SAP130. Ex vivo studies in bone marrow derived macrophages suggested that the endogenous Mincle ligand, SAP130, and LPS synergistically activated inflammatory responses, including inflammasome activation. Conclusion: This study reveals a novel IRAKM-Mincle axis that contributes to the pathogenesis of ethanol-induced liver injury. Toll-like receptor C-type lectin receptor innate immune response Introduction Alcoholic liver disease (ALD) ranges from simple steatosis to alcoholic hepatitis, fibrosis, cirrhosis and hepatocellular carcinoma. Alcohol exposure induces endoplasmic reticulum (ER) stress and mitochondrial dysfunction in hepatocytes, which leads to hepatocyte apoptosis, necrosis, necroptosis and inflammation (1). Alcohol disrupts the balance of gut microflora associated with an increased intestinal permeability, resulting in increased translocation of bacterial products into the circulation (2, 3). Increased levels of lipopolysaccharide (LPS) are indeed detected in the serum of alcoholic patients (4, 5) and alcohol-treated experimental animals, ranging from 80–130 pg/ml (5–9). The activation of hepatic macrophages (Kupffer cells) in response to portal endotoxin/LPS plays a key role in the early pathogenesis of alcohol-induced liver injury. LPS recognition by Toll-like receptor 4 (TLR4) on hepatic macrophages results in the release of inflammatory cytokines, such as TNF and IL-1, that can in turn impact hepatocyte function. Mice deficient in TLR4 or interlukin-1 receptor (IL-1R) are resistant to ethanol-induced liver disease (9, 10). Conversely, previous studies have shown that signals released from hepatocytes injured by alcohol are also critical to the activation of hepatic macrophages. Thus, one important question is how TLR-dependent inflammatory responses and alcohol-induced cellular damage coordinately lead to the pathogenesis of ALD. TLRs transduce signals through the adaptor molecule MyD88 and IL-1R-associated kinase (IRAK) family members, which include: IRAK1, IRAK2, IRAKM (also known as IRAK3) and IRAK4 (11). We recently reported the co-existence of two parallel TLR-mediated NFκB activation pathways: TAK1-dependent and MEKK3-dependent, respectively (12–14). IRAK4 kinase activity and its substrates, IRAK1 and IRAK2, are necessary for TAK1-dependent NFκB activation and mRNA stabilization of chemokines and cytokines, but not for MEKK3-dependent NFκB activation (15–17). IRAKM is able to interact with MyD88-IRAK4 to form IRAKM Myddosome, allowing it to control TAK1-independent MEKK3-dependent NFκB activation (18). This IRAKM-dependent pathway is necessary for the second wave of TLR-induced NFκB activation in the company of IRAK1/IRAK2. Notably, the IRAKM-dependent pathway only induces expression of the inhibitory molecules SOCS1, SHIP1, A20 and IκBα. Thus, IRAKM exerts an overall inhibitory effect on inflammatory response. In this study, we were surprised to discover that IRAKM-deficient mice were protected from ethanol-induced liver injury and inflammation. Mechanistically, IRAKM-dependent NFκB activation was the dominant pathway in hepatic macrophages challenged with low dose of LPS (100 pg/ml). Analysis of TLR4-induced IRAKM-dependent genes revealed one gene of particular interest: Mincle, a C-type lectin receptor, that senses non-homeostatic cell death. Induction of Mincle in hepatic macrophages was ablated in IRAKM-deficient mice and Mincle-deficient mice were protected from ethanol-induced liver injury. Spliceosome-associated Protein 130 (SAP130, also known as SF3B3), the endogenous ligand of Mincle, was released from hepatocytes after ethanol exposure, while recombinant SAP130 synergized with low dose LPS to robustly induce inflammatory gene expression and activate the inflammasome in macrophages. Since Mincle is a sensor for cell death and its expression is IRAKM-dependent, our results suggest that TLR-induced IRAKM-dependent Mincle up-regulation in hepatic macrophages critically links alcohol-induced cell death to subsequent inflammatory responses, contributing to the pathogenesis of ALD. Materials and methods Details regarding materials, details of the animal procedures and basic biochemical methods are given in the Supporting Information. Mouse models IRAKM deficient and Mincle deficient mice were previously described (19, 20). All procedures using animals were approved by the Cleveland Clinic Institutional Animal Care and Use Committee. KO and wild-type (WT) mice in the ethanol-fed groups were allowed free access to an ethanol containing diet: 1% (vol/vol) ethanol for 2d followed by 2% ethanol for 2d, 4% ethanol for 1w, 5% ethanol for 1w followed by 6% ethanol (32% of total calories) for the final week. Control mice were pair-fed a control diet which iso-calorically substituted maltose dextrins for ethanol over the entire feeding period. Inflammasome activation Bone marrow derived macrophages (BMDMs) were plated in 6-well plates at a concentration of 2.0 × 106 cells per well the day before the experiment. The day of the experiment, cells were stimulated with low concentration of LPS (100 pg/ml), recombinant SAP130 (5 µg/ml) or TDB (10 µg/ml) for the indicated time. Cell free supernatants were then prepared as described in Supporting information. Immunoblotting Cells and tissues were harvested and lysed in a Triton-containing lysis buffer (0.5% Triton X-100, 20 mM HEPES (pH 7.4), 150 mM NaCl, 12.5 mM β-glycerophosphate, 1.5 mM MgCl2, 10 mM NaF, 2 mM dithiothreitol, 1 mM sodium orthovanadate, 2mM EGTA, 1 mM phenylmethylsulfonyl fluoride and complete protease inhibitor cocktail from Roche). Cell lysates were then separated by 10% SDS-PAGE, transferred to Immobilon-P membranes (Millipore), and subjected to immunoblotting. Data analysis and Statistics Date were normally distributed, if not, non-parametric statistical analysis was applied to all data sets. Data are expressed as mean ± SEM. Differences were analyzed by Student t test and one-way ANOVA were used to compare normally distributed data set. Mann-Whitney U test and Kruskal-Wallis test were used for non-parametric analysis. P < 0.05 was considered significant. Results IRAKM is required for low dose LPS-mediated NFκB activation Recently, we reported a novel role for IRAKM in TLR signaling: IRAKM interacts with MyD88-IRAK4 to mediate a MEKK3-dependent second wave NFκB activation (18). In response to typical doses of LPS (10 ng/ml-1 µg/ml), MEKK3 modification and second wave IκBα phosphorylation are ablated in IRAKM-deficient macrophages, whereas LPS-induced IRAK1 modification/degradation, TAK1 activation and IKKα/β phosphorylation are not affected ((18), Fig. 1A and Suppl. Fig. 1). This IRAKM-dependent second wave of NFκB activation in response to LPS (10 ng/ml–1 µg/ml) induces expression of genes that are not regulated at the posttranscriptional level (including inhibitory molecules SOCS-1, SHIP-1, A20 and IκBα). Surprisingly, here we find that IRAKM-MEKK3-dependent NFκB activation was actually the dominant pro-inflammatory pathway in response to stimulation with low dose of LPS (100 pg/ml). In BMDMs (Fig. 1A) or primary Kupffer cells (Fig. 1B) treated with low dose LPS (100 pg/ml), IκBα phosphorylation – without IκBα degradation was observed, while IRAK1 modification/degradation and IKKα/β phosphorylation were absent, together indicating MEKK3-dependent NFκB activation. Importantly, IκBα phosphorylation and MEKK3 modification induced by low dose LPS were completely abolished in IRAKM-deficient cells (Fig. 1A & B). Based on these results, we hypothesized that IRAKM is dominantly activated to mediate MEKK3-dependent NFκB activation at low concentrations of LPS, whereas the IRAK1-TAK1-dependent pathway is not activated by low dose LPS. In support of this, we found that mRNA induction of the inflammatory genes CXCL1, TNF-α and IL-6 in response to low dose LPS was normal in IRAK1/2-deficient macrophages, but was greatly reduced in IRAKM-deficient cells (Fig. 1C). Similar results were also observed in cultures of primary Kupffer cells isolated from liver of WT and IRAKM-deficient mice (Fig. 1 D). Together, these data suggest that low dose LPS preferentially induces formation of the IRAKM Myddosome, leading to MEKK3-dependent NFκB activation. Low-dose LPS preferentially induces the formation of an IRAKM Myddosome via the death domain of IRAKM One important question is how low dose LPS activate the IRAKM-MEKK3 pathway. We previously reported that upon high dose ligand stimulation, the kinase activity of IRAK4 is required for TLR-induced IRAK1-TAK1-, but not MEKK3-, dependent NFκB activation (12–14). Further, IRAK1 and its phosphorylation are required for TLR-induced TAK1- but not MEKK3-, dependent NFκB activation in response to high doses of ligand (15, 16). Thus, we hypothesized that the TLR4 complex induced by low dose LPS fails to trigger the aggregation of IRAK4, leading to recruitment of IRAKM, but not IRAK1, and formation of the IRAKM Myddosome, with subsequent MEKK3-dependent NFκB activation. We indeed found that, upon low dose LPS stimulation (100 pg/ml), IRAKM, but not IRAK1, was recruited to the IRAK4-MyD88 complex, whereas both IRAK1 and IRAKM interacted with IRAK4 upon high dose LPS stimulation (1 µg/ml) (Fig. 2A). Furthermore, low dose LPS-induced MEKK3 modification and NFκB activation were unaffected in IRAK4 kinase inactive knock-in (KI) BMDMs (Suppl. Fig. 2). Taken together, these data suggest that IRAKM-dependent MEKK3-mediated NFκB activation in response to low dose LPS does not require IRAK4 kinase activity. To further elucidate the mechanism for the specific activation of the IRAKM-MEKK3-dependent pathway by low dose LPS, we compared IRAKM versus IRAK1 in their differential recruitment to the TLR-MyD88-IRAK4 complex. The crystal structure of the Myddosome complex suggests that assembly of MyD88-IRAK4 with IRAK1, IRAK2 or IRAKM occurs via death domain (DD) interactions (21). We thus hypothesized that differential recruitment of IRAKM versus IRAK1 in response to low versus high dose LPS stimulation is determined by their respective death domains. We generated a chimeric protein by replacing the death domain of IRAK1 with that of IRAKM (DD(M)-IRAK1). IRAK1/2/M-deficient macrophages were then retrovirally infected with Flag-tagged wild-type IRAK1, wild-type IRAKM, and DD(M)-IRAK1, followed by stimulation with low dose LPS. Indeed, we found that interaction of both wild-type IRAKM and DD(M)-IRAK1, but not IRAK1, with MyD88-IRAK4 was induced in response to low dose LPS stimulation (Fig. 2B). The weak constitutive interaction of between IRAK1 and IRAK4 was probably due to the overexpression of IRAK1 (Fig. 2B). Similarly, IRAKM and DD(M)-IRAK1, but not IRAK1, restored low dose LPS-induced IκBα phosphorylation (Fig. 2C) and target gene expression (Fig. 2D). Together, these data suggest that the IRAKM death domain mediates the specific recruitment of IRAKM to the TLR-MyD88-IRAK4 complex in response to low dose LPS and promotes MEKK3-dependent NFκB activation. IRAKM-dependent pathway is required for the development of chronic ethanol-induced liver disease Alcohol intake increases gut permeability, allowing accumulation of low levels of TLR ligands in the circulation. Activation of hepatic macrophages (Kupffer cells) in response to portal endotoxin/LPS plays a key role in the early pathogenesis of ALD (22). Considering the reported low concentrations of LPS in serum of alcoholic patients and ethanol-treated experimental animals (ranging from 80–130 pg/ml), our findings led us to hypothesize that IRAKM-dependent signaling induced at low concentrations of LPS might be the dominant TLR-activated pathway in ALD. To test this hypothesis, we subjected female IRAKM-deficient and littermate control WT mice to chronic ethanol feeding via the Lieber-DeCarli liquid diet to model steatosis and mild inflammation. Compared to pair-fed control mice, chronic ethanol-fed WT mice had increased hepatocyte injury, as assessed by serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels (Fig. 3A) and steatosis, characterized by the presence of lipid droplets in hepatocytes by H&E and Oil Red O staining, (Fig. 3B and 3C) and hepatic triglycerides (Fig. 3D). Compared to WT mice, IRAKM-deficient mice had reduced serum AST and ALT levels (Fig. 3A) and hepatic steatosis (Fig. 3B–D). Moreover, IRAKM deficiency significantly reduced the chronic ethanol-induced upregulation of TNF-α, IL-6, pro-IL-1β and MCP-1 mRNA in liver of ethanol-fed mice (Fig. 3E). Together, these data support the crucial role of IRAKM in the pathogenesis of chronic ethanol-induced inflammatory responses in the liver. IRAKM is required for TLR-mediated Mincle expression Given that low dose of LPS induced only modest levels of inflammatory cytokines and chemokines in wild-type BMDMs (Fig. 1E), we suspected that the impact of IRAKM on ALD might not be solely due to its direct role in induction of canonical inflammatory genes. Interestingly, we found that Mincle, a C-type lectin (CLR) membrane receptor that senses cell death, was strongly induced by low dose of LPS in WT, but not in IRAKM-deficient, BMDMs (Fig. 4A) and primary Kupffer cells (Suppl. Fig. 3A). It is important to note that induction of Mincle was not affected in IRAK1/2-double deficient macrophages (Fig. 4B). We next examined Mincle expression in liver in pair-fed and ethanol-fed WT and IRAKM-deficient mice. Interestingly, immunofluorescent staining showed that Mincle expression was increased in liver of ethanol-fed wild-type mice. Mincle expression co-localized with F4/80, a marker of resident macrophages (Fig. 4C). In contrast, chronic ethanol-induced expression of Mincle was attenuated in IRAKM-deficient mice. To further confirm this finding, we performed flow cytometry-based sorting of macrophages (F4/80+CD11b+) from the livers of pair-fed and ethanol-fed mice. Mincle mRNA expression was induced by ethanol-feeding in resident macrophages of WT mice; this response was attenuated in resident macrophages from IRAKM-deficient mice (Fig. 4D). Consistent with an induction of Mincle, the phosphorylation of Syk, a tyrosine kinase known to be activated CLR-mediated signaling(23, 24), was increased in the liver of ethanol-fed WT mice, but not in IRAKM-deficient mice (Fig. 4E). Taken together, these data implicate the induction of Mincle as a possible effector in mediating the impact of IRAKM in the pathogenesis of ethanol-induced liver injury. Mincle-mediated pathway is required for the development of chronic ethanol-induced liver disease Given that ethanol feeding induced Mincle expression in hepatic macrophages in an IRAKM-dependent manner, we hypothesized that this upregulation of Mincle might contribute to the pathogenesis of ALD. To test this hypothesis, we examined the impact of genetic deletion of Mincle on ethanol-induced liver injury. Compared to WT, Mincle-deficient mice indeed showed reduced chronic ethanol-induced hepatocyte injury and steatosis, demonstrated by lower levels of ALT/AST in the plasma, reduced lipid droplet formation in hepatocytes and reduced hepatic triglyceride concentrations (Fig. 5A–D). Further, phosphorylation of Syk was upregulated in liver of ethanol-fed WT mice, but not Mincle-deficient mice (Fig. 5E). Mincle deficiency also attenuated the upregulation of TNF-α, IL-6, pro-IL-1β and MCP-1 gene expression in the livers of ethanol-fed mice, suggesting a crucial role for Mincle in the pathogenesis of chronic ethanol-induced inflammatory responses in the liver (Fig. 5F). SAP130 activates Mincle and induces inflammasome activity following priming with low dose LPS Mincle was originally discovered based on its strong induction in macrophages by inflammatory stimuli, including TLR ligands (25). Interestingly, a known endogenous ligand for Mincle is SAP130, a component of the small nuclear ribonucloprotein, which diffuses out of dying cells and initiates an inflammatory response (19). Alcohol exposure is known to induce endoplasmic reticulum (ER) stress and mitochondrial dysfunction in hepatocytes, resulting in hepatocyte apoptosis, necroptosis and necrosis. Exposure of primary hepatocyte cultures to ethanol results in both apoptotic and necroptotic cell death (26) (Suppl. Fig. 4). Interestingly, we observed that hepatocytes challenged with ethanol released SAP130 into the culture supernatant (Fig. 6A), suggesting the possibility that Mincle might mediate chronic ethanol-induced inflammatory responses in the liver in response to SAP130. We indeed found that, while individually low dose LPS or SAP130 only induced very low levels of inflammatory gene expression in BMDMs, SAP130 was able to induce much higher levels of inflammatory gene expression in wild-type BMDMs primed by low dose of LPS (Fig. 6B). This synergistic impact of low LPS with SAP130 was abolished by Mincle deficiency, indicating low LPS-induced Mincle expression allows SAP130, a ligand associated with dying cells, to induce a robust inflammatory response (Fig. 6B). In addition to ligands released during necrotic cell death, fungal and mycobacterial ligands for Mincle have also been identified. Mincle is essential for recognition of the mycobacterial cord factor and its synthetic analog Trehalose-Dibehenate (TDB) (27). Interestingly, while TDB has been shown to activate the NLRP3 inflammasome (28, 29), Dectin-1, another member of C-type lectin receptors, activates processing of IL-1β via a noncanonical caspase-8 inflammasome (30). Thus, we tested the possibility that Mincle might also mediate inflammasome activation in response to SAP130. Indeed, recombinant SAP130 induced caspase-1 cleavage and IL-1β production in macrophages primed with low dose LPS (to up-regulate Mincle). This response was abolished in Mincle-, ASC- and NLRP3-deficient macrophages (Fig. 6 C–D & Suppl. Fig. 5). These results suggest that Mincle ligation by SAP130 in macrophages results in activation of ASC-dependent inflammasomes. Recent studies have shown that mice deficient in inflammasome components, such as the adaptor ASC and caspase 1, or IL-1R are protected from ethanol-induced liver injury (10). However, it has remained unclear how the inflammasome is activated in the liver of ethanol-fed mice. Based on the ability of recombinant SAP130 to activate the inflammasome in vitro, we hypothesized that in vivo Mincle senses SAP130 released during chronic ethanol-induced hepatocyte necrosis or necroptosis. SAP130 ligation of Mincle then activates an NLRP3/ASC-dependent inflammasome, leading to caspase1 activation and IL-1β production. Indeed, we detected increased caspase-1 activity (shown as cleaved caspase 1 and processed IL-1β) in the liver of ethanol-fed mice, which was greatly reduced in IRAKM- and Mincle-deficient mice (Fig. 6 E–F). Discussion Alcohol intake increases gut permeability and leads to accumulation of low levels of LPS in the circulation. Here we have identified a novel TLR4-IRAKM-dependent signaling pathway induced by low dose LPS that contributes to the pathogenesis of ALD. We found that low dose LPS preferentially induces the formation of an IRAKM Myddosome resulting in MEKK3-dependent NFκB activation. Importantly, whereas the low dose LPS-activated IRAKM Myddosome induced only very modest induction of inflammatory gene expression, it mediated strong up-regulation of Mincle, a sensor of cell death. SAP130, the endogenous ligand of Mincle, was released from hepatocytes after ethanol exposure, while recombinant SAP130 synergized with low dose LPS to robustly induce inflammatory gene expression and inflammasome activation in macrophages. Consistent with the IRAKM-Mincle synergy observed ex vivo, upregulation of Mincle was ablated in the liver of IRAKM-deficient mice and deficiency of either IRAKM or Mincle was strongly protective from chronic ethanol-induced liver injury (Fig. 7). Further, phosphorylation of Syk, a hallmark of Mincle signaling, was increased in the liver of ethanol-fed WT mice, but not in IRAKM-deficient mice. These results suggest that the IRAKM-Mincle axis may represent a critical and previously missing link between ethanol-induced hepatocellular damage and innate immune activation during the pathogenesis of ALD. Low concentrations of LPS have been detected in alcoholic patients and in ethanol-treated experimental animals. Importantly, mice deficient in TLR4 show a reduced ALD phenotype (9, 31). To our knowledge, this is the first study to show that the MyD88 downstream component IRAKM-Mincle axis is required for the development and pathogenesis of chronic ethanol-induced liver injury. Interestingly, Wang et al reported that IRAKM deficiency worsened liver injury in mice in response to an acute challenge with alcohol; in their study, mice were treated with 10% alcohol in drinking water for 7 day then gavaged with alcohol (6 g/Kg body weight) on day 7 (32). Furthermore, Mandrekar et al, making use of primary human monocytes in culture, found that IRAKM differentially contributes to the acute and chronic effects of alcohol on LPS-induced inflammation (33). Taken together with our study, these results suggest that the IRAKM-Mincle axis may differentially contribute to acute versus chronic effects of ethanol. Another important model to consider is the acute on chronic (Gao-binge) model of ethanol exposure that induces a more severe liver damage (34). In preliminary studies, we found that Mincle expression was not induced in Kupffer cells isolated from mice exposed to the Gao-binge protocol (unpublished data, Li and Nagy), in contrast to the strong induction observed in response to the Lieber-DeCarli chronic ethanol feeding protocol used here (Fig. 4C & D). Future studies are required to carefully compare and contrast the role of IRAKM-Mincle axis in these different models of ethanol-induced liver injury, likely leading to a better understanding of the multiple pathways by which different drinking patterns (acute, chronic and binge) can differentially lead to liver injury. We and others have previously shown that high doses of LPS lead to rapid clustering/aggregation of the TLR-MyD88-IRAK4 complex and activation of IRAK4, which then recruits and phosphorylates IRAK1 to activate TAK1-dependent NFκB activation (14). In support of this, we reported that IRAK4 dimerization is required for both IRAK4 auto-phosphorylation and activation of kinase activity (35). Importantly, the IRAKM Myddosome, via the MEKK3-dependent pathway, is required for the late phase of NFκB activation which upregulates the inhibitory molecules SOCS-1, SHIP-1, A20 and IκBα. Furthermore, IRAKM, through interaction with IRAK2, inhibits TLR-mediated production of cytokines and chemokines at the levels of translational. Thus, IRAKM is known to function as a net negative regulator of overt microbial infection (modeled in vitro by high-doses LPS) (20). However, the function of IRAKM in chronic inflammatory diseases, in which LPS concentrations are much lower than in response to infections, has not been carefully studied. Here, our data suggest that the low dose LPS-induced IRAKM Myddosome-dependent pathway plays a pathogenic pro-inflammatory role in ALD. Mechanistically, low dose LPS-induced IRAKM-dependent MEKK3-mediated NFκB activation does not require IRAK4 kinase activity, but does require IRAK4 as a structural adaptor for recruitment of IRAKM to proximal TLR-MyD88 complexes. Furthermore, the death domain of IRAKM specifically directs IRAKM recruitment to the TLR-MyD88-IRAK4 complex in response to low dose of LPS, leading to MEKK3-dependent NFκB activation. One important question is how the IRAKM-dependent pathway mediates the pathogenesis of ALD. Since low dose of LPS induced only modest levels of inflammatory cytokines and chemokines in wild-type BMDMs, we suspect that the impact of IRAKM on ALD is not solely due to its direct role in TLR-mediated inflammatory gene expression. Through a search for novel IRAKM-dependent genes, we found that Mincle, a CLR membrane receptor that senses microbial products and cell death, is strongly induced by low dose of LPS and this induction was greatly reduced in IRAKM-deficient cells. Mincle-mediated responses are critical for anti-fungal and anti-mycobacterial host defense (36–39). Importantly, Mincle also functions as a sensor of non-homeostatic cell death in which the cellular contents are released (e.g. necroptosis and pyroptosis), thus promoting sterile inflammation. SAP130 was identified as an endogenous ligand recognized by Mincle following necrotic cell death (19). In this study, we found that TLR-induced Mincle expression is dependent on IRAKM in macrophages. In addition, Mincle is upregulated in hepatic macrophages from of ethanol-fed wild-type mice, but not in IRAKM-deficient mice. Both IRAKM- and Mincle-deficient mice were protected from chronic ethanol-induced hepatocyte injury, steatosis and liver inflammation. Since IRAKM’s expression is restricted to myeloid cells (Suppl. Fig. 3A–C), these results suggest that IRAKM-dependent Mincle expression in myeloid cells likely contributes to the pathogenesis of ALD. Mincle is commonly thought to be expressed mainly in myeloid lineage cells, such as macrophages. Notably, some Mincle positive cells in the liver of ethanol-fed wild-type and IRAKM-deficient mice did not co-localize with F4/80+ cells. We indeed found that Mincle expression was induced in primary hepatocytes by IL-1β (Suppl. Fig. 3B–C). Cell-specific deletion will help to determine the impact of Mincle in different cell populations on the pathogenesis of ALD. In the future, we will also test whether restoration of Mincle expression in IRAKM-deficient mice via generation of Mincle-transgenic mice would be sufficient to drive ethanol-induced liver injury. Recent studies have shown that mice with genetic deletion of either inflammasome components or IL-1R are protected from ALD (10). In this study, we found that low dose LPS upregulates Mincle expression in macrophages and subsequent treatment with recombinant SAP130 was able to activate the ASC/NLRP3/caspase-1 inflammasome. Chronic ethanol-feeding induced caspase-1 and IL-1β activation in the liver; this response was dramatically reduced in IRAKM- and Mincle-deficient mice. However, future studies are required to elucidate the precise molecular mechanism by which the SAP130-induced Mincle-dependent pathway activates the inflammasome. Interestingly, Tanaka el al reported that resident macrophages in adipose tissue express Mincle, which is activated by an endogenous ligand released from dying adipocytes. This Mincle signaling was shown to be required for macrophages to form crown-like structures (CLS), as well as for the development of adipose tissue fibrosis (40). Additionally, Mincle was shown to play a pivotal role in a model of ischemic stroke (41). These studies thus underline the potentially wide-ranging importance of the IRAKM-Mincle axis and suggest this novel pathway as a possible therapeutic target for the treatment of diverse disease states. Supplementary Material Supp info This work was supported by grants from NIH (1R01AA023722 to X.L. and L.E.N. (MPI); 2PO1HL029582-26A1 & PO1CA062220-16A1 to X.L; R01AA011975 & 5P20AA017837 to L.E.N). We thank Dr. Richard A Flavell (Department of Immunobiology, Yale University School of Medicine, New Haven, Connecticut) for providing the IRAKM-deficient mice. We thank Dr. Christine A. Wells (The Australian Institute for Bioengineering and Nanotechnology, University of Queensland, Brisbane, Australia) for providing anti-Mincle antibody for immune fluorescence staining. Abbreviations LPS lipopolysaccharide TLRs Toll-like receptors ALD alcoholic liver disease IL-1R Interlukin-1 receptor IRAK IL-1R-associated kinase KO knockout WT wild-type TDB trehalose-dibehenate DD death domain SAP130 Spliceosome-associated protein 130 AST Aspartate transaminase ALT Alanine transaminase Fig. 1 IRAKM is required for low dose LPS-mediated NFκB activation Cell lysates were prepared from A. WT and IRAKM KO BMDMs and B. primary cultures of mouse Kupffer cells after they were untreated or treated with high dose LPS (1 µg/ml) and low dose LPS (100 pg/ml) for the indicated times and analyzed by Western blot analysis. C. Total mRNA from BMDMs of WT, IRAK1/2-DKO and IRAKM KO mice treated with low dose LPS (100 pg/ml) for the indicated times were subjected to RT-PCR analyses. D. Total mRNA from BMDMs of WT and IRAKM KO mice treated with low dose LPS (100 pg/ml) for the indicated times were subjected to RT-PCR analyses. E. BMDMs were treated with high dose LPS (1 µg/ml) or low dose LPS (100 pg/ml) for 24 hours. Cell-free supernatants were collected and cytokine concentrations were measured by ELISA assay. The experiments were repeated for five times with similar results. Data represent mean ± SEM; *, P<0.05. Fig. 2 Low-doses LPS preferentially induces the formation of IRAKM Myddosome via the death domain of IRAKM A. WT BMDMs were treated with high dose LPS (1 µg/ml) and low dose LPS (100 pg/ml) for the indicated times, followed by immunoprecipitation (IP) with anti-IRAK-4 antibody and analyzed by Western blot analysis. B–D. IRAK-1/2/M-triple deficient immortalized BMDMs infected with adenovirus expressing FLAG-tagged IRAK1, FLAG-tagged chimeric IRAKM death domain with IRAK1 kinase domain (DD(M)+IRAK1) and FLAG-tagged IRAKM were treated with low dose LPS (100 pg/ml) for indicated times. B. immunoprecipitates and C. cell lysates were analyzed by Western blot analysis. D. Total mRNA from transfected cells was subjected to RT-PCR analyses. Fig. 3 IRAKM-dependent pathway is required for the development of chronic ethanol-induced liver disease WT and IRAKM KO mice were allowed free access to ethanol (EtOH) (n=6) or pair-fed control diets (n=4). A. AST and ALT activity was determined in plasma. B. Paraffin-embedded liver sections were stained with hematoxylin and eosin. C. Frozen liver sections were subjected to Oil Red O staining. All images were acquired using a 10× objective. D. Hepatic triglyceride content was measured in whole liver homogenates. E. Total mRNAs from livers of WT and IRAKM KO mice (pair-fed and EtOH-fed) were subjected to RT-PCR analysis. Fig. 4 IRAKM is required for TLR and chronic ethanol-induced Mincle upregulation A. Cell lysates from WT and IRAKM KO BMDMs untreated or treated with low dose LPS (100 pg/ml) for the indicated times were analyzed by Western blot analysis B. Cell lysates from WT and IRAK1/2 DKO BMDMs untreated or treated with low dose LPS (100 pg/ml) for the indicated times were analyzed by Western blot analysis. C. Immunostaining of F4/80 (green) and Mincle (red) was performed on frozen sections of liver from of WT and IRAKM KO mice (pair-fed and EtOH-fed). Confocal images are representative of five mice per group. D. Resident macrophages (F4/80+CD11b+) from livers of pair-fed and ethanol-fed mice were isolated by flow cytometry-based sorting. Total RNA was subjected to RT-PCR analysis. E. Tissue lysates from the whole liver of WT and IRAKM KO mice (pair-fed and EtOH-fed) were analyzed by Western blot analysis. Fig. 5 Mincle-dependent pathway is required for the development of chronic ethanol-induced liver disease WT and Mincle KO mice were allowed free access to ethanol (EtOH) (n=6) or pair-fed control diets (n=4). A. AST and ALT activity was determined in plasma. B. Paraffin-embedded liver sections were stained with hematoxylin and eosin. C. Frozen liver sections were subjected to Oil Red O staining. All images were acquired using a 10× objective. D. Hepatic triglyceride content was measured in whole liver homogenates. E. Tissue lysates from the whole liver of WT and Mincle KO mice (pair-fed and EtOH-fed) were analyzed by Western blot analysis. The black arrows point out the specific bands of pSyk. F. Total mRNAs from livers of WT and Mincle KO mice (pair-fed and EtOH-fed) were subjected to RT-PCR analysis. Fig. 6 SAP130-mediated mincle activation is required for low concentration LPS induced inflammation through inflammasome activation A. Primary hepatocytes from WT mice were treated with EtOH (50 mM) for 24 hours. Cell-free supernatants were collected and SAP130 was measured by Western blot analysis. B. BMDMs from WT and Mincle KO mice were treated with PBS, LPS (100 pg/ml) for 24 hours, SAP130 (5 ug/ml) for 2 hours, TDB (100 µg/ml, 2 hours), LPS (100 pg/ml, 24 hours) + SAP130 (5 µg/ml, 2 hours) or LPS (100 pg/ml, 24 hours) + TDB (100 µg/ml, 2 hours). Total mRNA was subjected to RT-PCR analyses. C. BMDMs from WT and Mincle KO mice were treated with PBS, LPS (100 pg/ml) for 24 hours, LPS (100 pg/ml, 24 hours) + SAP130 (5 µg/ml, 6 hours) or LPS (100 pg/ml, 24 hours) + TDB (100 µg/ml, 6 hours). Cell lysates and supernatants were collected together and analyzed by Western blot. D. IL-1β was measured in Cell-free supernatants by ELISA. E. Tissue lysates from the whole liver of WT and IRAKM KO mice (pair-fed and EtOH-fed) were analyzed by Western blot. F. Tissue lysates from the whole liver of WT and Mincle KO mice (pair-fed and EtOH-fed) were analyzed by Western blot analysis. Fig. 7 IRAKM-Mincle axis contributes to the pathogenesis and development of ALD Chronic alcohol consumption results in increased intestinal permeability and changes in bacterial microflora increase levels of bacterial products in alcoholic patients and animal models of ALD. Further, alcohol exposure can induce endoplasmic reticulum (ER) stress and mitochondrial dysfunction in hepatocytes, contributing to hepatocellular injury and death. TLR4- induced IRAKM-mediated MEKK3-dependent NFκB activation is required for the up-regulation of Minlce in hepatic macrophages. Mincle sense the necroptotic hepatocytes-released nuclear protein, SAP130, which in turn activates the inflammasome activation in macrophages. Secreted IL-1β may further act on hepatocytes inducing pyroptosis or on Stellate cells which leads to fibrosis. Reference 1 Gao B Bataller R Alcoholic liver disease: pathogenesis and new therapeutic targets Gastroenterology 2011 141 1572 1585 21920463 2 Rao R Endotoxemia and gut barrier dysfunction in alcoholic liver disease Hepatology 2009 50 638 644 19575462 3 Mandayam S Jamal MM Morgan TR Epidemiology of alcoholic liver disease Semin Liver Dis 2004 24 217 232 15349801 4 Parlesak A Schafer C Schutz T Bode JC Bode C Increased intestinal permeability to macromolecules and endotoxemia in patients with chronic alcohol abuse in different stages of alcohol-induced liver disease J Hepatol 2000 32 742 747 10845660 5 Bala S Marcos M Gattu A Catalano D Szabo G Acute binge drinking increases serum endotoxin and bacterial DNA levels in healthy individuals PLoS One 2014 9 e96864 24828436 6 Michelena J Altamirano J Abraldes JG Affo S Morales-Ibanez O Sancho-Bru P Dominguez M Systemic inflammatory response and serum lipopolysaccharide levels predict multiple organ failure and death in alcoholic hepatitis Hepatology 2015 62 762 772 25761863 7 Zhong W McClain CJ Cave M Kang YJ Zhou Z The role of zinc deficiency in alcohol-induced intestinal barrier dysfunction Am J Physiol Gastrointest Liver Physiol 2010 298 G625 G633 20167873 8 Lambert JC Zhou Z Wang L Song Z McClain CJ Kang YJ Prevention of alterations in intestinal permeability is involved in zinc inhibition of acute ethanol-induced liver damage in mice J Pharmacol Exp Ther 2003 305 880 886 12626662 9 Uesugi T Froh M Arteel GE Bradford BU Thurman RG Toll-like receptor 4 is involved in the mechanism of early alcohol-induced liver injury in mice Hepatology 2001 34 101 108 11431739 10 Petrasek J Bala S Csak T Lippai D Kodys K Menashy V Barrieau M IL-1 receptor antagonist ameliorates inflammasome-dependent alcoholic steatohepatitis in mice J Clin Invest 2012 122 3476 3489 22945633 11 Kawai T Akira S Toll-like receptors and their crosstalk with other innate receptors in infection and immunity Immunity 2011 34 637 650 21616434 12 Fraczek J Kim TW Xiao H Yao J Wen Q Li Y Casanova JL The kinase activity of IL-1 receptor-associated kinase 4 is required for interleukin-1 receptor/toll-like receptor-induced TAK1-dependent NFkappaB activation J Biol Chem 2008 283 31697 31705 18794297 13 Kim TW Staschke K Bulek K Yao J Peters K Oh KH Vandenburg Y A critical role for IRAK4 kinase activity in Toll-like receptor-mediated innate immunity J Exp Med 2007 204 1025 1036 17470642 14 Yao J Kim TW Qin J Jiang Z Qian Y Xiao H Lu Y Interleukin-1 (IL-1)-induced TAK1-dependent Versus MEKK3-dependent NFkappaB activation pathways bifurcate at IL-1 receptor-associated kinase modification J Biol Chem 2007 282 6075 6089 17197697 15 Cui W Xiao N Xiao H Zhou H Yu M Gu J Li X beta-TrCP-mediated IRAK1 degradation releases TAK1-TRAF6 from the membrane to the cytosol for TAK1-dependent NF-kappaB activation Mol Cell Biol 2012 32 3990 4000 22851693 16 Kim TW Yu M Zhou H Cui W Wang J DiCorleto P Fox P Pellino 2 is critical for Toll-like receptor/interleukin-1 receptor (TLR/IL-1R)-mediated post-transcriptional control J Biol Chem 2012 287 25686 25695 22669975 17 Wan Y Xiao H Affolter J Kim TW Bulek K Chaudhuri S Carlson D Interleukin-1 receptor-associated kinase 2 is critical for lipopolysaccharide-mediated post-transcriptional control J Biol Chem 2009 284 10367 10375 19224918 18 Zhou H Yu M Fukuda K Im J Yao P Cui W Bulek K IRAK-M mediates Toll-like receptor/IL-1R-induced NFkappaB activation and cytokine production The EMBO journal 2013 32 583 596 23376919 19 Yamasaki S Ishikawa E Sakuma M Hara H Ogata K Saito T Mincle is an ITAM-coupled activating receptor that senses damaged cells Nat Immunol 2008 9 1179 1188 18776906 20 Kobayashi K Hernandez LD Galan JE Janeway CA Jr Medzhitov R Flavell RA IRAK-M is a negative regulator of Toll-like receptor signaling Cell 2002 110 191 202 12150927 21 Lin SC Lo YC Wu H Helical assembly in the MyD88-IRAK4-IRAK2 complex in TLR/IL-1R signalling Nature 2010 465 885 890 20485341 22 Wang HJ Gao B Zakhari S Nagy LE Inflammation in alcoholic liver disease Annu Rev Nutr 2012 32 343 368 22524187 23 Deng Z Ma S Zhou H Zang A Fang Y Li T Shi H Tyrosine phosphatase SHP-2 mediates C-type lectin receptor-induced activation of the kinase Syk and anti-fungal TH17 responses Nat Immunol 2015 16 642 652 25915733 24 Strasser D Neumann K Bergmann H Marakalala MJ Guler R Rojowska A Hopfner KP Syk kinase-coupled C-type lectin receptors engage protein kinase C-sigma to elicit Card9 adaptor-mediated innate immunity Immunity 2012 36 32 42 22265677 25 Matsumoto M Tanaka T Kaisho T Sanjo H Copeland NG Gilbert DJ Jenkins NA A novel LPS-inducible C-type lectin is a transcriptional target of NF-IL6 in macrophages J Immunol 1999 163 5039 5048 10528209 26 Bakhautdin B Das D Mandal P Roychowdhury S Danner J Bush K Pollard K Protective role of HO-1 and carbon monoxide in ethanol-induced hepatocyte cell death and liver injury in mice J Hepatol 2014 61 1029 1037 24946281 27 Ishikawa E Ishikawa T Morita YS Toyonaga K Yamada H Takeuchi O Kinoshita T Direct recognition of the mycobacterial glycolipid, trehalose dimycolate, by C-type lectin Mincle J Exp Med 2009 206 2879 2888 20008526 28 Schweneker K Gorka O Schweneker M Poeck H Tschopp J Peschel C Ruland J The mycobacterial cord factor adjuvant analogue trehalose-6,6'-dibehenate (TDB) activates the Nlrp3 inflammasome Immunobiology 2013 218 664 673 22921586 29 Shenderov K Barber DL Mayer-Barber KD Gurcha SS Jankovic D Feng CG Oland S Cord factor and peptidoglycan recapitulate the Th17-promoting adjuvant activity of mycobacteria through mincle/CARD9 signaling and the inflammasome J Immunol 2013 190 5722 5730 23630357 30 Gringhuis SI Kaptein TM Wevers BA Theelen B van der Vlist M Boekhout T Geijtenbeek TB Dectin-1 is an extracellular pathogen sensor for the induction and processing of IL-1beta via a noncanonical caspase-8 inflammasome Nat Immunol 2012 13 246 254 22267217 31 Hritz I Mandrekar P Velayudham A Catalano D Dolganiuc A Kodys K Kurt-Jones E The critical role of toll-like receptor (TLR) 4 in alcoholic liver disease is independent of the common TLR adapter MyD88 Hepatology 2008 48 1224 1231 18792393 32 Wang Y Hu Y Chao C Yuksel M Colle I Flavell RA Ma Y Role of IRAK-M in alcohol induced liver injury PLoS One 2013 8 e57085 23437317 33 Mandrekar P Bala S Catalano D Kodys K Szabo G The opposite effects of acute and chronic alcohol on lipopolysaccharide-induced inflammation are linked to IRAK-M in human monocytes J Immunol 2009 183 1320 1327 19561104 34 Bertola A Park O Gao B Chronic plus binge ethanol feeding synergistically induces neutrophil infiltration and liver injury in mice: a critical role for E-selectin Hepatology 2013 58 1814 1823 23532958 35 Ferrao R Zhou H Shan Y Liu Q Li Q Shaw DE Li X IRAK4 dimerization and trans-autophosphorylation are induced by Myddosome assembly Mol Cell 2014 55 891 903 25201411 36 Wells CA Salvage-Jones JA Li X Hitchens K Butcher S Murray RZ Beckhouse AG The macrophage-inducible C-type lectin, mincle, is an essential component of the innate immune response to Candida albicans J Immunol 2008 180 7404 7413 18490740 37 Yamasaki S Matsumoto M Takeuchi O Matsuzawa T Ishikawa E Sakuma M Tateno H C-type lectin Mincle is an activating receptor for pathogenic fungus, Malassezia Proc Natl Acad Sci U S A 2009 106 1897 1902 19171887 38 Feinberg H Jegouzo SA Rowntree TJ Guan Y Brash MA Taylor ME Weis WI Mechanism for recognition of an unusual mycobacterial glycolipid by the macrophage receptor mincle J Biol Chem 2013 288 28457 28465 23960080 39 Ishikawa T Itoh F Yoshida S Saijo S Matsuzawa T Gonoi T Saito T Identification of distinct ligands for the C-type lectin receptors Mincle and Dectin-2 in the pathogenic fungus Malassezia Cell Host Microbe 2013 13 477 488 23601109 40 Tanaka M Ikeda K Suganami T Komiya C Ochi K Shirakawa I Hamaguchi M Macrophage-inducible C-type lectin underlies obesity-induced adipose tissue fibrosis Nat Commun 2014 5 4982 25236782 41 Suzuki Y Nakano Y Mishiro K Takagi T Tsuruma K Nakamura M Yoshimura S Involvement of Mincle and Syk in the changes to innate immunity after ischemic stroke Sci Rep 2013 3 3177 24212132

PMC005xxxxxx/PMC5115960.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7707449 656 Ann Neurol Ann. Neurol. Annals of neurology 0364-5134 1531-8249 27686464 5115960 10.1002/ana.24789 NIHMS820622 Article RNAi prevents and reverses phenotypes induced by mutant human ataxin-1 Keiser Megan S. PhD 1 Monteys Alejandro Mas PhD 1 Corbau Romuald PhD 12 Gonzalez-Alegre Pedro PhD, MD 13 Davidson Beverly L. PhD 14 1 The Raymond G Perelman Center for Cellular and Molecular Therapeutics, The Children's Hospital of Philadelphia, Philadelphia, PA 2 Spark Therapeutics Inc. Philadelphia, PA 3 Department of Neurology, the University of Pennsylvania, Philadelphia, PA 4 Department of Pathology and Laboratory Medicine, the University of Pennsylvania, Philadelphia, PA Correspondence should be addressed to Beverly L. Davidson, Raymond G Perelman Center for Cellular and Molecular Therapeutics, 5060 Colket Translational Research Building, 3501 Civic Center Boulevard, Philadelphia, PA 19104, davidsonbl@email.chop.edu 4 10 2016 2 11 2016 11 2016 02 11 2017 80 5 754765 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objective Spinocerebellar ataxia type 1 is an autosomal dominant fatal neurodegenerative disease caused by a polyglutamine expansion in the coding region of ATXN1. We showed previously that partial suppression of mutant ataxin-1 (ATXN1) expression, using virally-expressed RNAi triggers, could prevent disease symptoms in a transgenic mouse model and a knock in mouse model of the disease, using a single dose of virus. Here, we set out to test if RNAi triggers targeting ATXN1 could not only prevent, but also reverse disease readouts when delivered after symptom onset. Methods We administered recombinant adeno-associated virus (rAAV) expressing miS1, an artificial miRNA targeting human ATXN1 mRNA (rAAV.miS1) to a mouse model of SCA1 (B05 mice). Viruses were delivered prior to or after symptom onset at multiple doses. Control B05 mice were treated with rAAVs expressing a control artificial miRNA, or with saline. Animal behavior, molecular phenotypes, neuropathology and magnetic resonance spectroscopy were done on all groups and data were compared to wildtype littermates. Results We found that SCA1 phenotypes could be reversed by partial suppression of human mutant ATXN1 mRNA by rAAV.miS1 when delivered after symptom onset. We also identified the therapeutic range of rAAV.miS1 that could prevent or reverse disease readouts. Interpretation SCA1 disease may be reversible by RNAi therapy, and the doses required for advancing this therapy to humans are delineated. INTRODUCTION Spinocerebellar ataxia type 1 (SCA1) is an autosomal dominant progressive neurodegenerative disease caused by a polyglutamine repeat expansion in exon 8 of the ataxin-1 (ATXN1) gene.1 Healthy individuals have from 6 – 42 CAG repeats in ATXN1 interspersed by 1 – 3 CAT codons. A pure CAG expansion exceeding 40 glutamines causes disease pathogenesis by a toxic gain-of-function mechanism. Although ataxin-1 protein (ATXN1) is ubiquitously expressed, specific cerebellar Purkinje cells (PCs) and brainstem neurons are more susceptible to expanded ATXN1 expression and carry the bulk of pathology. As a result, patients with SCA1 develop a prominent cerebellar syndrome with brainstem features, including gait and limb ataxia, dysarthria or dysphagia, among others. To date, SCA1 remains a fatal disease with no known disease modifying therapies. The initial mouse model for SCA1 is a transgenic line expressing mutant human ATXN1 from a Purkinje cell specific promoter (Pcp2).2,3 This mouse, known as the B05 model, develops progressive disease, with many features similar to SCA1 including measurable gait deficits by 7 weeks of age. Additionally, there are transcriptional changes by postnatal day 25 and PC loss by 24 weeks of age.3 Earlier, a doxycycline-inducible form of the B05 model revealed that pre-existing phenotypes were reversible after many weeks of mutant ATXN1 expression.4 This suggests that a window of opportunity exists not only for halting the disease, but perhaps also to reverse the presence or severity of symptoms once present. We previously assessed the utility of gene silencing strategies as a putative therapy for SCA1 in this and a knock in (KI) model of disease. Delivery of recombinant AAVs (rAAV) expressing RNAi triggers in the form of shRNAs provided clinical benefit in B05 mice after direct injection into the cerebellar cortex.5 More recently, we showed that delivery of rAAVs expressing artificial miRNAs directed at mutant ATXN1 mRNA, to the deep cerebellar nuclei (DCN) for transport to PCs and other brainstem neurons, also improved pathologic and behavioral phenotypes in this model.6 Similarly, rAAVs expressing microRNAs (miRNAs) directed at mutant mouse Atxn1 mRNA was beneficial when applied to the KI SCA1 model.7 Importantly, the DCN-directed approach is scalable from mice to larger mammals; clinically relevant biodistribution and silencing was observed when tested in nonhuman primates.8 While these studies provided the necessary proof-of-principal for advancing RNAi therapy for SCA1, the treatment was applied before symptom onset. Here, we set out to test the clinically-relevant hypothesis stating that viral-expressed RNAi provides therapeutic benefit when delivered after symptom onset. Furthermore, before advancing this treatment to patients it is critical to define the dosing window between the minimal effective doses and the toxicity threshold. We therefore evaluated the effects of increasing doses of a clinical SCA1 gene therapy product directed at silencing mutant ATXN1 mRNA, on the prevention or reversal of clinical and neuropathological signs in B05 mice. Methods and Materials Plasmids and viral vectors The therapeutic miRNA sequence targeting human and rhesus Ataxin-1 (miS1) has been previously described.7 The original therapeutic vector rAAV1.miS1.eGFP was modified to no longer express eGFP and instead contain a “safe stuffer” sequence.9 Recombinant rAAV serotype 2/1 vectors (rAAV1.miS1 and rAAV1.miControl) were generated at The Children’s Hospital of Philadelphia Research Vector Core. AAV vectors were resuspended in Diluent Buffer (The Children’s Hospital of Philadelphia Research Vector Core) and titers (viral genomes/ml) were determined by QPCR. Cell culture and transfection HEK293 cells were transfected (Lipofectamine™ 2000) in quadruplicate in 24-well plates per manufacturer’s instructions with 500ng of plasmid containing pAAV.miS1.eGFP, pAAV.miS1, pAAV, or no plasmid. Total RNA was harvested 24 hours with TRIzol®. Animals All animal protocols were approved by The Children’s Hospital of Philadelphia Animal Care and Use Committee. Wild type FVB mice were obtained from Jackson Laboratories (Bar Harbor, ME). B05 transgenic mice were previously provided by Dr. H.T. Orr and re-derived by Jackson Laboratories. The B05 line was maintained on the FVB background. Mice were genotyped using primers specific for the mutant human ataxin-1 transgene.2 Hemizygous and age-matched wildtype littermates were used for the indicated experiments. Treatment groups comprised of approximately equal numbers of male and female mice. Mice were housed in a controlled temperature environment on a 12-hour light/dark cycle. Food and water were provided ad libitum. AAV injections and brain tissue isolation B05 mice were injected with rAAV1 vectors expressing miS1 or a control scrambled miRNA sequence (miC). Mice were stereotaxically injected bilaterally to the deep cerebellar nuclei (coordinates −6.0 mm caudal to bregma, ± 2.0 mm from midline, and −2.2 mm deep from the cerebellar surface) with 4 μl of rAAV1 virus at doses of 1×107 vg, 1×108 vg, 6×108 vg, 1×109 vg, 6×109 vg or 1×1010 vg/hemisphere or saline (Diluent Buffer). Mice were anesthetized with 4% isoflurane/oxygen mixtures and transcardially perfused with 20 ml of ice cold saline. Mice were decapitated, and for histological analyses, brains were removed and post-fixed overnight in 4% paraformaldehyde. Brains were stored in 30% sucrose/0.05% azide solution at 4°C until cut on a sledge microtome at 40 μm thickness and stored at – 20 °C in a cryoprotectant solution. For RNA and metabolite analyses, brains were removed and cerebellar hemispheres were flash frozen in liquid nitrogen and stored at – 80 °C. RNA was isolated from whole cerebellum using 1 ml of Trizol®, RNA quantity and quality were measured using a NanoDrop® 2000. For metabolite analysis, tissues were subjected to a perchloric acid extraction. Frozen samples were weighed and homogenized using beads in a TissueLyzer LT (Qiagen). Ice cold 3.6% HClO4 was added, further homogenized and centrifuged at 4°C. Supernatant was buffered to a pH of ~7.0 with KOH, centrifuged, lyophilized and again weighed. Immunohistochemical analysis Free-floating sagittal cerebellar sections (40 μm thick) were washed in 1 X TBS with 0.05% Triton®X-100 at room temperature and blocked for 1 hour in 5% serum, 0.05% Triton®X-100, in 1X TBS. Sections were incubated with primary antibody in 3% serum and 0.05% Triton®X-00 in TBS overnight at room temperature. Primary antibodies used were polyclonal rabbit anti-Calbindin (1:2000; Sigma), polyclonal anti-Iba1 (1:1000; WAKO), polyclonal anti-GFAP (1:2000; DAKO), and polyclonal rabbit 12NQ (1:1000; Orr Lab10). For Fluorescent IHC, sections were incubated with goat anti-rabbit Alexa Fluor 488 or 568 (1:200; Life Technologies) in 3% serum and 0.05% Triton®X-100 in 1 X TBS for 1 hour at room temperature. For DAB IHC, sections were incubated in goat anti-rabbit biotin-labeled secondary antibody (1:200; Jackson Immunoresearch) in 3% serum and 0.05% Triton®X-100 in 1 X TBS for 1 hour at room temperature. Tissues were developed with Vectastain® ABC Elite Kit (Vector Laboratories), according to the manufacturer’s instructions. All sections were mounted onto Superfrost Plus slides (Fischer Scientific) and cover-slipped with Fluoro-Gel (Electron Microscopy Sciences) or dehydrated and cover-slipped with DPX. Images were captured on a Leica DM6000B fluorescence microscope using LAS X software. Semi-quantitative PCR Reverse transcription (High Capacity cDNA Reverse Transcription Kit, Applied Biosystems) was performed on 1 μg total RNA collected from cerebellum using a standard stem-loop PCR primer designed to identify miS1 previously described.7 sqRT-cDNA was subjected to RT-PCR with a standard reverse primer and a forward primer specific to miS1. Quantitative PCR Random-primer first-strand cDNA synthesis was performed using 2 μg total RNA (High Capacity cDNA Reverse Transcription Kit, Applied Biosystems) per manufacturer’s instructions. Assays were performed on a BioRad CFX384 Real Time System using TaqMan® (Thermo-Fischer Scientific) primer/probe sets specific for human Ataxin-1, mouse Pcp2, mouse Grm1 or mouse β-Actin (TaqMan® 2X Universal Master Mix by Life Technologies). 1H-magnetic resonance spectroscopy Analysts were blinded to the treatment groups. NMR spectroscopy was performed at 400MHz on a Bruker Avance III 400 wide-bore spectrometer. Each lyophilized tissue extract was dissolved in 0.4 ml of D2O, the pH adjusted to 7.0 and the solution was introduced in a 5 mm NMR tube. An external standard made of a sealed capillary containing a solution of trimethylsilylpropionic acid (TSP) in D2O was centered in the NMR tube and used as chemical shift reference and quantitation standard. Fully relaxed proton spectra were acquired with a 5 mm proton probe. Standard acquisition conditions were as follows: PW 5 μs (45°) TR 8.84s (AQ 4.84s, D1 4s), SW 6775 Hz, TD 64k, 128 scans 4 DS. A soft water saturation pulse was applied during the 4s relaxation delay. Rotarod Analysis Mice were tested by a tester blinded to the treatment groups on an accelerated rotarod apparatus (model 47600; Ugo Basile). For distribution to groups of equal abilities at baseline, mice were first tested at 5 weeks of age prior to treatment. Mice were habituated to the rotarod for 4 min then subjected to three trials per day (with at least 30 min of rest between trials) for four consecutive days. For each trial, acceleration was from 4 to 40 rpm over 5 min, and then speed maintained at 40 rpm. Latency to fall (or if mice hung on for two consecutive rotations without running) was recorded for each mouse per trial. Trials were stopped at 500 seconds, and mice remaining on the rod at that time were scored as 500 seconds. Two-way analysis of variance followed by a Tukey post-hoc analysis was used to assess for significant differences. Variables were time and treatment. Statistical Analysis For all studies, p values were obtained by using one-way analysis of variance followed by Tukey post-hoc analysis to assess for significant differences between individual groups. In all statistical analysis, P < 0.05 was considered significant. Figure Preparation All photographs were formatted with Adobe Photoshop software. All graphs were made with GraphPad Prism® software. All figures were constructed with Adobe Illustrator software. RESULTS AAV1.miS1 prevents the development of rotarod deficits in B05 mice We first modified the former rAAV1.miS17 wherein the eGFP reporter expression cassette was replaced with a stuffer sequence. The stuffer sequence ensured appropriate length of the expression cassette required for optimal AAV packaging (Fig 1A). To assess that this modification did not impact the silencing potency of the miRNA we transfected HEK 293 cells with the shuttle plasmid pAAV.miS1.eGFP, pAAV.miS1 or a control plasmid. Compared to control transduced cells, both pAAV.miS1.eGFP and pAAV.miS1 significantly reduced ATXN1 mRNA expression 24 hours post-transfection (Fig 1B); replacing eGFP with stuffer sequence did not alter the potency of the artificial miRNA. We used the B05 model2 for this work as the RNAi trigger expressed from rAAV1.miS1 targets human ATXN1. B05 transgenic mice show progressive disease, with transcriptional changes evident prior to noted behavioral deficits (Fig 1C). To identify the efficacy and toxicity thresholds to prevent disease progression, mice were injected bilaterally into the deep cerebellar nuclei (DCN) with increasing doses of rAAV1.miS1, rAAV1.miC or saline after baseline behavior testing (Table 1). Twenty-four weeks after injection (30 weeks of age) animals were re-assayed by rotarod and then euthanized and tissues collected for post-necropsy analysis (Fig 1C). As seen in Figure 1D, at 30 weeks of age control treated transgenic mice could not remain on the rotarod apparatus after 98.8 ± 22 seconds. B05 mice treated with rAAV1.miS1 at doses of 8×108 and 8×109 vector genomes (vg) performed significantly better than control treated transgenic animals and were not statistically differently than their wildtype littermates. rAAV1.miS1 reduces ATXN1 mRNA in SCA1 mouse in a dose-dependent manner, preventing changes in cerebellar metabolites Semi-quantitative PCR on whole cerebellar lysates confirmed miS1 expression (Fig 2A), with a clear dose-response. Quantitative RT-PCR for mutant human ATXN1 mRNA (Fig 2B) showed a similar dose response that inversely correlated with miS1 levels. B05 mice treated with rAAV1.miC or rAAV1.miS1 at a dose of 8×107 vg had similar levels of ATXN1 mRNA (98 ± 4% and 100 ± 3% respectively) relative to saline-treated animals. B05 mice administered 8×108 or 8×109 vg of rAAV1.miS1 had increasingly reduced levels of ATXN1 mRNA (77 ± 3 and 47 ± 7%, respectively). B05 mice in the high dose group had almost complete reduction of human ATXN1 mRNA levels (4 ± 1) relative to the control treated animals. High field proton magnetic resonance spectroscopy (1H MRS) allows quantitation of biomarkers in SCA1 patients in a non-invasive manner. In 2010, Oz-G and colleagues showed that SCA1 patients have reduced N-acetylaspartate (NAA) levels and elevated inositol levels.11 The same observations have been made in SCA1 mouse models. We therefore assayed metabolite levels in cerebellar lysates from all groups using nuclear magnetic resonance (NMR). Similar to results on untreated B05 mice12, control treated mice had reduced NAA/inositol ratios compared to wildtype littermates (Fig 2C). However, this difference was normalized to wildtype levels in B05 mice treated with 8×109 or 8×1010 vg of rAAV1.miS1 (Fig 2C). rAAV.miS1 at 8×109 vg prevents cerebellar pathology In SCA1, PC dendrites progressively retract, resulting in cerebellar molecular layer (ML) thinning.13 To investigate the potential protective effect of rAAV1.miS1 on this phenotype, brain sections were evaluated by anti-calbindin staining of sagittal sections, and ML widths quantified. In control treated B05 animals, lobules III-IV/V have marked thinning, as do sections from animals injected with 8×108 vg of rAAV1.miS1 compared to wildtype littermates (Fig 3A). However, there was no significant difference between the ML widths of wildtype animals and B05 mice treated with 8×108, 8×109, or 8×1010 vg of rAAV1.miS1. In lobules IV/V-VI, all groups of B05 treated mice, except those treated with rAAV1.miS1 at 8×109 vg, were significantly reduced relative to their wildtype littermates (Fig 3B). Immunohistochemistry for human ATXN1 in PCs of control treated mice showed, as expected, expression of the transgene in B05 but not wildtype mice (Fig 3C). B05 mice treated with increasing doses of rAAV1.miS1 had progressively less ATXN1-positive PCs. At a dose of 8×1010 vg, no ATXN1-positive PCs were detectable. Histological staining for glial fibrillary acidic protein (Gfap, a marker of astroglial activation), revealed enhanced immunoreactivity at the site of injection (the DCN) in all injected animals, and those treated at 8×1010 vg had robust enhancement (Fig 3D). Histological staining for ionized calcium-binding adapter molecule 1 (Iba1), a marker for microglial activation, did not show differences among any experimental groups except for those receiving the highest dose of rAAV1.miS1 (Fig 3E, F). rAAV2/1.miS1 is effective after disease onset Two doses (8×108 and 8×109 vg) were effective and non-toxic in our pre-onset treatment design. Because most patients with SCA1 present to the clinic with some disease manifestations, we tested the effects of miS1 therapy after disease onset. B05 mice have deficits on the rotarod by 10-11 weeks of age (Fig 4A). We baseline tested B05 mice and wildtype littermates on the rotarod at 11 weeks, to confirm deficits (Fig 4B), and then performed dosing studies as before, except that 5 doses (rather than 4) of rAAV1.miS1, or control were injected at 12 weeks of age. Additionally, the doses stepped up by ½ log and the lowest dose in the pre-disease onset treatment paradigm, which resulted in no silencing, was omitted (Table 2). End-study rotarod was conducted at 20 weeks of age (Fig 4C) and 2 weeks later tissue collected for post-necropsy analysis. Nine weeks after injection, wildtype mice performed significantly better than B05 mice receiving saline, rAAV1.miC, and the low and two high dose groups. In contrast to wildtype mice, treated B05 mice in these groups had poorer performance relative to their baseline. However, B05 mice treated with 2.6×109 vg or 8×109 vg of rAAV1.miS1 performed significantly better than they did at 11 weeks of age, and also significantly better than the other B05 treatment groups. Cumulatively, the data support the hypothesis that rAAV1.miS1 delivery into the DCN of symptomatic ataxic mice improves motor symptoms in the B05 model of SCA1. rAAV1.miS1 reduces ATXN1 and improves molecular readouts in symptomatic B05 mice miS1 expression was detected in B05 mice treated with rAAV1.miS1 (Fig 5A), and there was a clear dose-dependent reduction of human ATXN1 mRNA levels in cerebellar lysates (Fig 5B). ATXN1 mRNA levels were not different between B05 mice treated with saline, rAAV1.miC, and 8×108 vg rAAV1.miS1. B05 mice treated with 2.6×109 vg, 8×109, 2.6×1010 or 8×1010 vg of rAAV1.miS1 had progressively greater levels of knockdown relative to saline-treated B05 mice. Evaluation of transcripts from Purkinje cell protein 2 (Pcp2) and the metabotropic glutamate receptor type 1 (Grm1), two transcripts down-regulated in this model were also done.14,15 B05 mice treated with or rAAV1.miC had significantly lower levels of Pcp2 mRNA than wildtype littermates (Fig 5C). Although B05 mice treated with 8×108 or 8×1010 vg of rAAV1.miS1 had significantly lower levels of Pcp2, B05 mice given 8×109 vg of rAAV1.miS1 had Pcp2 levels that were not different from wildtype. Similar to Pcp2, we detected significantly reduced Grm1 mRNA levels in control treated B05 mice or B05 mice treated with rAAV1.miS1 at 8×108 or 8×1010 vg (Fig 5D). Of note, SCA1 mice treated with 8×109 vg of rAAV1.miS1 expressed Grm1 at levels not significantly different from wildtype littermates. Consistent with the results shown earlier (Fig 2C), cerebellar lysates of control treated B05 mice had abnormal NAA/ inositol ratios as measured by NMR that improved with treatment (Fig 5E). B05 mice treated with 8×109 vg of rAAV1.miS1 had a NAA/inositol ratio that was not significantly different from their wildtype littermates. rAAV1.miS1 improves cerebellar pathology in symptomatic B05 mice Sagittal sections were processed to quantify the molecular layer widths in medial cerebellar regions of lobules IV/V and VI. The data show marked thinning in control-treated B05 mice, and mice treated with 8×108 or 8×1010 vg of rAAV.miS1 relative to those regions in wildtype mice (Fig 6A). However, there was no significant difference between wildtype mice and B05 mice treated with 8×109 vg of rAAV1.miS1. ML widths in the caudal medial cerebellar sections between lobules VIII and IX also show significant thinning in control treated B05 mice and B05 mice administered 8×1010 vg rAAV1.miS1 (Fig 6B). B05 mice treated with 8×108 or 8×109 vg of rAAV1.miS1 had ML widths that were not significantly different from wildtype littermates. Similar to the results observed in the pre-symptomatic dosing experiment, B05 mice treated with saline or rAAV1.miC are immunoreactive for human ATXN1 in most PCs. B05 mice treated post-symptomatically with rAAV1.miS1 show decreasing levels of ATXN1-positive PCs that correlate inversely to the dose injected (Fig 6C). B05 mice treated with rAAV1.miS1 at 8×108 vg had fewer ATXN1-positive PCs than control treated mice; whereas those treated with rAAV1.miS1 at 8×109 or 8×1010 have few to no detectable ATXN1-positive PCs. In the DCN, the site of injection, there were similar amounts of Gfap+ immunoreactive cells in all sections, with the exception of enhanced immunoreactivity in B05 mice treated with 8×1010 vg of rAAV1.miS1 (Fig 6D). B05 mice treated with saline or rAAV1.miC showed slightly higher levels of Iba1 immunoreactivity in the cortex and DCN than those treated with 8×108 or 8×109 vg rAAV1.miS1, or wildtype mice. B05 mice treated with rAAV1.miS1 at 8×1010 vg showed elevated Iba1 immunoreactivity in both the cortex and the DCN (Fig 6E, F). Discussion This work provides solid experimental evidence demonstrating that RNAi-mediated suppression of ATXN1 mRNA alters disease progression, reverses symptoms and normalizes cerebellar pathology and disease biomarkers in a SCA1 model. In addition, it identifies doses within the efficacy-toxicity window to guide clinical development of RNAi for the treatment of SCA1. We identified the least effective dose, the toxicity threshold, and several effective doses that could either prevent or improve SCA1 readouts. Importantly, the doses used in prior work6,7 showing the efficacy of the approach for preventing mutant (human or mouse) ataxin-1-induced symptoms, were recapitulated here. Cumulatively, our data in symptomatic mice corroborate and extend earlier work demonstrating that eliminating mutant human ataxin-1 expression is therapeutic, even after cerebellar pathology and neurological deficits are evident.4,16 In earlier studies by Orr and colleagues, mutant ataxin-1 was completely eliminated and aggregates resolved quickly with recovery of cerebellar pathology.4 Our results are similar but contrasting in important ways. The similarities are that RNAi trigger expression, even when initiated after symptom onset, was therapeutic and improved symptomatology. The major difference is that our suppression was partial rather than complete, and mutant ATXN1 was not suppressed in every Purkinje cell. This result is exciting and suggests that i) partial suppression after disease onset can be beneficial, and ii) that limiting coverage of the RNAi therapy to even a portion of the cerebellum, most notably the medial regions, can improve behavioral outcomes. We found that 2.6×109 vg had 48%, and 8×109 vg had ~71% reduction of mutant ATXN1 mRNA, and both provided benefit. rAAV.miS1 not only prevented further disease progression, but also improved disease readouts (e.g. see rotarod studies). At baseline, B05 animals were significantly impaired (Fig 4B, C). After receiving miS1 at doses of 2.6×109 and 8×109 vg, rotarod performance at 20 weeks of age demonstrated a reversal of pre-existing impairment, and they performed no differently from their wildtype littermates. The difference between a 28% reduction in ATXN1 produced by 8×108 vg and a 48% reduction in ATXN1 produced by 2.6×109 vg of rAAV1.miS1 delineates the threshold for reversal of rotarod performance when delivered to post-symptomatic B05 mice. This is the first time, to our knowledge, that delivery of an RNAi vector has been shown to quantifiably reverse disease pathology in B05 SCA1 mice. Of note, pre-symptomatic B05 mice receiving 8×108 vg at 6 weeks of age failed to develop the phenotypic rotarod deficit by 30 weeks of age, suggesting that earlier treatment with less viral load can be beneficial. PC dysfunction occurs prior to cell loss in SCA1 and in SCA1 mice models. One measure of this is a reduction in molecular layer width due to PC dendritic retraction. The anterior lobe (rostral lobules) of the cerebellum is key to maintain balance, and dysfunction causes truncal ataxia.17 It is also a region affected early in the pathogenesis of SCA1.18,19 The posterior lobe (caudal) plays an important role in motor coordination.20 In mice that received 8×108 vg, molecular layer widths were similar in width to wildtype mice in caudal lobules, with measurable thinning in rostral lobules. However mice treated with 8×109 vg of rAAV.miS1 retained molecular layer widths similar to wildtype in both rostral and caudal lobules. This suggests that this dose, with scaling for human use, may provide preservation of both rostral and caudal aspects of the molecular layers lending improved balance and motor coordination respectively, as well as rescue motor symptoms. It is worth noting that in the B05 model, the human transgene is expressed in the PCs only. However, assays for transduction efficiency, molecular readouts of efficacy, and transgene (miS1) levels were performed on whole cerebellar lysates. These data demonstrate the PC-targeting efficiency of this vector system. Molecular indicators of efficacy include Pcp2, the mRNA of which is trafficked to the dendrites,21 and Grm1, a metabotropic glutamate receptor 1 located in the post-synaptic termini of dendrites.21 The latter is critical for coordinated motor function.22 Both Pcp2 and Grm1 deficits were reversed in mice by 67% reduction in ATXN1. It has been previously shown that B05 mice have elevated Iba1+ immunoreactivity and is reversible in the SCA1 conditional model.23 We also noted enhanced Iba1+ immunoreactivity in control treated B05 mice relative to WT animals, and qualitatively more in the 8×1010 vg rAAV1.miS1 treatment group. In B05 mice treated with 8×108 or 8×109 vg, Iba1+ immunoreactivity was reduced. Thus, even a 28% reduction in ATXN1 (8×108 vg dose), which does not result in behavioral rescue, can abate the phenotypic increase in Iba1+ signal. Non-invasive biomarkers will provide critical tools for assessing efficacy in SCA1. High field proton magnetic resonance spectroscopy (1H MRS) indicates that SCA1 patients have lower levels of N-acetylaspartate (NAA), and elevated levels of inositol.11 These data have been recapitulated in transgenic SCA1 mice12 and were reversed in a conditional mouse model of SCA1.16 NAA reduction usually precedes neuronal loss, and is used as a marker of neuronal dysfunction.24 In this work, NAA levels were modestly reduced in control treated SCA1 mice at 20 weeks of age, a time prior to significant PC loss. Importantly, mice treated with 8×109 vg of rAAV.miS1 had a NAA/inositol similar to their wildtype littermates. This, together with previous studies suggests that the NAA/Inositol may be a sensitive, non-invasive measure of efficacy and a possible biomarker in disease-modifying clinical trials for SCA1. Motor deficits quantified by the Scale for Assessment and Rating of Ataxia (SARA) correlate with altered neurochemical levels quantified by MRS.25 We see a similar relationship in our untreated B05 mice, with improved “scores” upon treatment. These data suggest that rAAV1.miS1 could provide therapeutic benefit and prevention of further pathogenesis in SCA1 patients if administered prior to disease onset. Moreover, rAAV1.miS1 could halt or even reverse pre-existing motor deficits in early, symptomatic SCA1 patients. Thus, SARA scores could stabilize or improve with treatment, along with concomitant improvements in neurochemical levels. While the B05 model was useful for assessing the utility of our clinical product, a more genetically relevant model would be a humanized knock-in that would allow testing of miS1. This would allow efficacy testing in a setting more biologically similar to SCA1 patients, as compared to the transgenic model where the transgene is only expressed in PCs. Of note is that RNAi was tested in mice expressing an expanded CAG repeat in mouse ataxin-1,26 with efficacy at the same E9 dose, (using a RNAi trigger that targeted mouse ataxin-1). Whether the doses that reverse disease are similar to what we found in the B05 model remain to be tested, but are important to determine in future work. Additionally, we cannot assume that miS1 (targets human) and miSCA1 (targets mice) are equivalent in safety, as they are different sequences. Nonetheless it is encouraging that similar doses prevented disease in the current models that are available. Although we noted prevention or reversal of disease at several doses, we also noted toxicity at the highest dose tested. The reasons for this are unclear. It could be due to the near complete absence of ATXN1 in transduced cells, although we think this is not likely due the fact that chronic, 100% loss of Atxn1 in mice causes transcriptional misregulation but no noted neuropathology or ataxia. 27,28 A second possibility is that the high level expression of miS1 somehow causes toxicity due to abnormal processing of endogenous miRNAs. This can be evaluated in future work to examine endogenous miRNA processing and miRNA activity. We think this is improbable, however, as we used miS1 in an older vector platform and reversed aberrant miRNA expression in the B05 model.29 A third possibility is that the high capsid dose is toxic. Testing for this could be done using empty capsid preparations as controls at the doses evaluated here. We are investigating the exact mechanisms underlying the toxicity as part of a follow up study. And while this data informs us of a toxicity ceiling and the maximum tolerable dose, understanding the mechanism for that toxicity is important moving forward. In summary, we show that AAV-mediated delivery of RNAi triggers can reverse neuropathological phenotypes, transcriptional changes, and behavioral phenotypes in a mouse model of SCA1. We also identify the minimal effective and maximally tolerated doses that will guide our clinical application for SCA1 therapy. Together with earlier work demonstrating the scalability of our approach to nonhuman primates,8 our studies will guide the initial application to SCA1 and other cerebellar diseases in which PCs, brainstem neurons and the DCN are important therapeutic targets. Acknowledgements The authors would like to thank The Children’s Hospital of Philadelphia Research Vector Core for the viruses used in this work. The authors would also like to thank Suzanne L. Wehrli and The Children’s Hospital of Philadelphia Small Animal Imaging Facility and Matthew Sowada for assistance with behavioral experiments. This work was funded by the NIH, NINDS (UH2NS094355), the National Ataxia Foundation Pioneer Award, and The Children’s Hospital of Philadelphia Research Institute. FIGURE 1 Experimental design and rotarod analysis. (A) Cartoon of therapeutic viral construct. Inverted terminal repeats (ITR) surround murine U6 promoter driving expression of an artificial miRNA targeting human ATXN1 (miS1) followed by a non-coding stuffer sequence. (B) Comparative silencing efficiency of pAAV.miS1.eGFP and pAAV1.miS1 in HEK293 cells (N=4 replicates; ***p < 0.0001). (C) Timeline of disease progression in B05 mice and study design. (D) Rotarod performance over 4 days at 30 weeks of age. Treatment groups consisted of untreated wildtype mice, B05 mice injected with saline, B05 mice injected at 8 × 108 vector genomes (vg) of rAAV2/1.miControl (denoted as miC), B05 mice injected at 8 × 107 vg (denoted as 8E7) of rAAV2/1.miS1, B05 mice injected at 8 × 108 vg (denoted as 8E8) of rAAV2/1.miS1, B05 mice injected at 8 × 109 vg (denoted as 8E9) of rAAV2/1.miS1, B05 injected at 8 × 1010 vg (denoted as 8E10) of rAAV2/1.miS1 (*** denotes p < 0.0001; * p < 0.05; difference from rAAV1.miC and saline treated B05 animals). FIGURE 2 sqPCR, qRT-PCR, and NMR analyses of cerebellar extracts from treated B05 mice and untreated wildtype littermates. (A) Semi-quantitative analysis of miS1 from whole cerebellar extracts. (B) qRT-PCR for human ATXN1 mRNA levels from whole cerebellar extracts (N ≥ 3; ** p < 0.01; *** p < 0.0001; differences from saline injected SCA1 mice). (C) Ratio of NAA/Inositol levels from whole cerebellar extracts (N=3; *** p < 0.0001; differences from control treated B05 mice). FIGURE 3 Cerebellar pathology. (A) Molecular layer widths of lobules III and IV/V from sagittal cerebellar sections 0.5 mm from midline (N ≥ 3; ** denotes p < 0.01; differences from wildtype). (B) Molecular layer widths of lobules IV/V and VI from sagittal cerebellar sections 0.5 mm from midline (N ≥ 3; * denotes p < 0.05; ** denotes p < 0.01; *** denotes p < 0.0001; differences from wildtype). (C) Representative photomicrographs of sagittal cerebellar sections immunostained for human ataxin-1. Scale Bar = 100 μm. (D) Representative photomicrographs of sections from DCN immunostained for Gfap. Scale Bar = 100 μm. (E) Representative photomicrographs of sagittal cerebellar cortices immunostained for Iba1. Scale Bar = 100 μm. (F) Representative photomicrographs of sagittal cerebellar DCN immunostained for Iba1. Scale Bar = 100 μm. FIGURE 4 Experimental design of reversal study and rotarod data. (A) Timeline of disease progression in B05 mice and study design. (B) 11 week old baseline rotarod. Rotarod performance over 4 days at 11 weeks of age (*p < 0.05; ***p < 0.0001). (C) Rotarod performance over 4 days at 11 and 20 weeks of age (*p < 0.05; *** p < 0.0001; difference from rAAV1.miC and saline treated B05 animals). FIGURE 5 sqPCR, qRT-PCR, and NMR analyses of cerebellar extracts. (A) Semi-quantitative expression of miS1 from whole cerebellar extracts. (B) qRT-PCR for human ATXN1 mRNA levels from whole cerebellar extracts (N ≥ 3; ** p < 0.01; *** p < 0.0001; differences from saline injected B05 mice). (C) qRT-PCR for mouse Pcp2 mRNA levels from whole cerebellar extracts (N=3-4; *p < 0.05; ** denotes p < 0.01; ***p < 0.0001; differences from wildtype mice). (D) qRT-PCR for mouse Grm1 mRNA levels from whole cerebellar extracts (N=3-4; *p < 0.05; **p < 0.01; differences from wildtype mice). (E) Ratio of NAA/Inositol levels from whole cerebellar extracts (***p < 0.0001; differences from wildtype mice). FIGURE 6 Cerebellar pathology in mice treated after disease onset. (A) Molecular layer widths of lobule IV/V and VI from sagittal cerebellar sections 0.5 mm from midline (N ≥ 3; ***p < 0.0001 differences from wildtype). (B) Molecular layer widths layers of lobules VIII and IX from sagittal cerebellar sections 0.5 mm from midline (N ≥ 3; **p < 0.01; ***p < 0.0001; differences from wildtype). (C) Representative photomicrographs of sagittal cerebellar sections immunostained for human ataxin-1. Scale Bar = 100 μm. (D) Representative photomicrographs of sagittal DCN immunostained for Gfap. Scale Bar = 100 μm. (E) Representative photomicrographs of sagittal cerebellar cortex immunostained for Iba1. Scale Bar = 100 μm. (F) Representative photomicrographs of sagittal cerebellar DCN immune stained for Iba1. Scale Bar = 100 μm. TABLE 1 Treatment Groups for Preventative Study Genotype Injectate Dose (vg) B05 rAAV1.miS1 8×107 B05 rAAV1.miS1 8×108 B05 rAAV1.miS1 8×109 B05 rAAV1.miS1 8×1010 B05 rAAV1.miC 8×108 B05 Saline Wildtype TABLE 2 Treatment Groups for Reversal Study Genotype Injectate Dose (total vg) B05 rAAV1.miS1 8×108 B05 rAAV1.miS1 2.6×109 B05 rAAV1.miS1 8×109 B05 rAAV1.miS1 2.6×1010 B05 rAAV1.miS1 8×1010 B05 rAAV1.miC 8×108 B05 rAAV1.miC 8×109 B05 Saline Wildtype Author Contributions Study concept and design: M.S.K., A.M.M., R.C., P.G.A., B.L.D.; data acquisition and analysis: M.S.K. R.C., P.G.A., B.L.D.; drafting the manuscript and figures: M.S.K., A.M.M., R.C., P.G.A., B.L.D. Potential Conflict of Interest B.L.D. is a founder of Spark, Inc, a gene therapy company, that has licensed the technology described in this study. B.L.D. is also on the SAB of Intellia Therapeutics and Serepta Therapeutics. REFERENCES 1 Orr HT Zoghbi HY Trinucleotide repeat disorders Annual review of neuroscience 2007 30 575 621 doi:10.1146/annurev.neuro.29.051605.113042 2 Burright EN SCA1 transgenic mice: a model for neurodegeneration caused by an expanded CAG trinucleotide repeat Cell 1995 82 937 948 7553854 3 Clark HB Purkinje cell expression of a mutant allele of SCA1 in transgenic mice leads to disparate effects on motor behaviors, followed by a progressive cerebellar dysfunction and histological alterations J Neurosci 1997 17 7385 7395 9295384 4 Zu T Recovery from polyglutamine-induced neurodegeneration in conditional SCA1 transgenic mice J Neurosci 2004 24 8853 8861 doi:10.1523/JNEUROSCI.2978-04.2004 15470152 5 Xia H RNAi suppresses polyglutamine-induced neurodegeneration in a model of spinocerebellar ataxia Nat Med 2004 10 816 820 doi:10.1038/nm1076 15235598 6 Keiser MS Boudreau RL Davidson BL Broad therapeutic benefit after RNAi expression vector delivery to deep cerebellar nuclei: implications for spinocerebellar ataxia type 1 therapy Mol Ther 2014 22 588 595 doi:10.1038/mt.2013.279 24419082 7 Keiser MS Geoghegan JC Boudreau RL Lennox KA Davidson BL RNAi or overexpression: alternative therapies for Spinocerebellar Ataxia Type 1 Neurobiol Dis 2013 56 6 13 doi:10.1016/j.nbd.2013.04.003 23583610 8 Keiser MS Kordower JH Gonzalez-Alegre P Davidson BL Broad distribution of ataxin 1 silencing in rhesus cerebella for spinocerebellar ataxia type 1 therapy Brain 2015 138 3555 3566 doi:10.1093/brain/awv292 26490326 9 Monteys AM Single nucleotide seed modification restores in vivo tolerability of a toxic artificial miRNA sequence in the mouse brain Nucleic Acids Res 2014 42 13315 13327 doi:10.1093/nar/gku979 25332397 10 Servadio A Expression analysis of the ataxin-1 protein in tissues from normal and spinocerebellar ataxia type 1 individuals Nat Genet 1995 10 94 98 7647801 11 Oz G Neurochemical alterations in spinocerebellar ataxia type 1 and their correlations with clinical status Mov Disord 2010 25 1253 1261 doi:10.1002/mds.23067 20310029 12 Oz G Noninvasive detection of presymptomatic and progressive neurodegeneration in a mouse model of spinocerebellar ataxia type 1 J Neurosci 2010 30 3831 3838 doi:10.1523/JNEUROSCI.5612-09.2010 20220018 13 Klement IA Ataxin-1 nuclear localization and aggregation: role in polyglutamine-induced disease in SCA1 transgenic mice Cell 1998 95 41 53 9778246 14 Serra HG Gene profiling links SCA1 pathophysiology to glutamate signaling in Purkinje cells of transgenic mice Hum Mol Genet 2004 13 2535 2543 doi:10.1093/hmg/ddh268 15317756 15 Skinner PJ Vierra-Green CA Clark HB Zoghbi HY Orr HT Altered trafficking of membrane proteins in purkinje cells of SCA1 transgenic mice Am J Pathol 2001 159 905 913 doi:10.1016/S0002-9440(10)61766-X 11549583 16 Oz G In vivo monitoring of recovery from neurodegeneration in conditional transgenic SCA1 mice Exp Neurol 2011 232 290 298 doi:10.1016/j.expneurol.2011.09.021 21963649 17 Dale Purves GJA Fitzpatrick David Hall William C. LaMantia Anthony-Samuel McNamara James O. Williams S. Mark Neuroscience 2004 439 Sinauer Associates, Inc Third 18 Robitaille Y Schut L Kish SJ Structural and immunocytochemical features of olivopontocerebellar atrophy caused by the spinocerebellar ataxia type 1 (SCA-1) mutation define a unique phenotype Acta Neuropathol 1995 90 572 581 8615077 19 Robitaille Y Lopes-Cendes I Becher M Rouleau G Clark AW The neuropathology of CAG repeat diseases: review and update of genetic and molecular features Brain Pathol 1997 7 901 926 9217975 20 Cicirata F Angaut P Panto MR Serapide MF Neocerebellar control of the motor activity: experimental analysis in the rat. Comparative aspects Brain Res Brain Res Rev 1989 14 117 141 2752228 21 Vassileva G Smeyne RJ Morgan JI Absence of neuroanatomical and behavioral deficits in L7/pcp-2-null mice Brain Res Mol Brain Res 1997 46 333 337 9191112 22 Knopfel T Grandes P Metabotropic glutamate receptors in the cerebellum with a focus on their function in Purkinje cells Cerebellum 2002 1 19 26 doi:10.1007/BF02941886 12879970 23 Cvetanovic M Ingram M Orr H Opal P Early activation of microglia and astrocytes in mouse models of spinocerebellar ataxia type 1 Neuroscience 2015 289 289 299 doi:10.1016/j.neuroscience.2015.01.003 25595967 24 Demougeot C N-Acetylaspartate, a marker of both cellular dysfunction and neuronal loss: its relevance to studies of acute brain injury J Neurochem 2001 77 408 415 11299303 25 Adanyeguh IM In vivo neurometabolic profiling in patients with spinocerebellar ataxia types 1, 2, 3, and 7 Mov Disord 2015 30 662 670 doi:10.1002/mds.26181 25773989 26 Lorenzetti D Repeat instability and motor incoordination in mice with a targeted expanded CAG repeat in the Sca1 locus Hum Mol Genet 2000 9 779 785 10749985 27 Goold R Down-regulation of the dopamine receptor D2 in mice lacking ataxin 1 Hum Mol Genet 2007 16 2122 2134 doi:10.1093/hmg/ddm162 17599952 28 Matilla A Mice lacking ataxin-1 display learning deficits and decreased hippocampal paired-pulse facilitation J Neurosci 1998 18 5508 5516 9651231 29 Rodriguez-Lebron E Liu G Keiser M Behlke MA Davidson BL Altered Purkinje cell miRNA expression and SCA1 pathogenesis Neurobiol Dis 2013 54 456 463 doi:10.1016/j.nbd.2013.01.019 23376683

PMC005xxxxxx/PMC5115965.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8302946 4093 Hepatology Hepatology Hepatology (Baltimore, Md.) 0270-9139 1527-3350 27629435 5115965 10.1002/hep.28817 NIHMS817614 Article CFTR controls biliary epithelial inflammation and permeability by regulating Src tyrosine kinase activity Fiorotto Romina 14 Villani Ambra 1 Kourtidis Antonis 2 Scirpo Roberto 1 Amenduni Mariangela 1 Geibel Peter J. 3 Cadamuro Massimilano 45 Spirli Carlo 14 Anastasiadis Panos Z. 2 Strazzabosco Mario 145 1 Section of Digestive Diseases, Liver Center, Yale University, New Haven, Connecticut, USA 2 Department of Cancer Biology, Mayo Clinic Comprehensive Cancer Center, Jacksonville, Florida, USA 3 Department of Surgery, Yale University, New Haven, Connecticut, USA 4 International Center for Digestive Health, University of Milan-Bicocca, Milan Italy 5 Section of Digestive Diseases, Department of Medicine and Surgery, University of Milan-Bicocca, Milan, Italy Address correspondence to: Mario Strazzabosco, Section of Digestive Diseases, Liver Center, Yale University School of Medicine, Cedar Street 333, New Haven, CT 06511, USA. Phone: 203-737-1451; Fax: 203-785-7273; mario.strazzabosco@yale.edu. 21 9 2016 27 10 2016 12 2016 01 12 2017 64 6 21182134 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. In the liver, CFTR regulates bile secretion and other functions at the apical membrane of biliary epithelial cells (i.e cholangiocytes). CF-related liver disease (CFLD) is a major cause of death in patients with CF. CFTR dysfunction affects innate immune pathways, generating a para-inflammatory status in the liver, and other epithelia. This study investigates the mechanisms linking CFTR to TLR4 activity. We found that CFTR is associated in a multi-protein complex at the apical membrane of normal mouse cholangiocytes, with proteins that negatively control Src activity. In CFTR-defective cholangiocytes, Src tyrosine kinase self-activates and phosphorylates TLR4, resulting in activation of NF-κB, and increased pro-inflammatory cytokines production in response to endotoxins. This Src/NF-κB-dependent inflammatory process attracts inflammatory cells, but also generates changes in the apical junctional complex and loss of epithelial barrier function. Inhibition of Src decreased the inflammatory response of CF-cholangiocytes to LPS, rescued the junctional defect in-vitro and significantly attenuated endotoxin-induced biliary damage and inflammation in vivo (Cftr-KO mice). Conclusion Our findings reveal a novel function of CFTR as regulator of TLR4 responses and cell polarity in biliary epithelial cells. This mechanism is pathogenetic, as shown by the protective effects of Src inhibition in vivo and maybe a novel therapeutic target in CFLD and other inflammatory cholangiopathies. Cystic Fibrosis TLR4 cytokines cytoskeleton cell polarity Cystic Fibrosis (CF) is a disease of secretory epithelia leading to chronic damage and insufficiency of several organs, including pancreas, lung and liver. CF is caused by mutations in the gene encoding for the cystic fibrosis transmembrane conductance regulator (CFTR), a cAMP-regulated epithelial chloride channel expressed at the apical membrane of most secretory epithelia (1, 2). More than twenty years after its discovery, the understanding of CFTR function(s) is still incomplete. In addition of functioning as a channel, CFTR may also act as a “hub” protein that forms macromolecular complexes with several other proteins at the apical membrane, mostly through PDZ binding motifs (3). Through these interactions, CFTR may be involved in key cellular functions such as regulation of ATP release, membrane protein and vesicle trafficking and inflammation (4, 5). This multi-functional role of CFTR reflects the variety of manifestations associated with its defective function, and the diversity of the pathogenetic mechanisms involved in the target organs of CF, such as lung, pancreas, liver and intestine. Several defective inflammatory responses and altered innate immune pathways have been linked to CFTR deficiency in different organs affected by CF(6-8). In the liver, CFTR is specifically expressed at the apical membrane of the epithelial cells lining the biliary epithelium (cholangiocytes) where it regulates chloride and bicarbonate secretion in response to cAMP/PKA stimulation (9). Defective CFTR function causes impaired biliary secretion of chloride and bicarbonate, and ductal cholestasis (10). A percentage of patients with Cystic Fibrosis develop liver disease (CFLD). CFLD is a chronic inflammatory cholangiopathy that can evolve into sclerosing cholangitis and focal biliary cirrhosis, and is the third leading cause of death in CF (11, 12). Severe CFLD affects only a minority of CF patients, suggesting that ductal cholestasis may be a predisposing factor to liver disease, which progresses only in the presence of a second insult (13, 14). We have previously shown that the CFTR-defective biliary epithelium, when exposed to gut-derived bacterial endotoxins, produces high levels of cytokines/chemokines and attracts inflammatory cells (i.e macrophages and neutrophils) leading to peribiliary inflammation. In CF biliary epithelial cells TLR4/NF-κB-dependent innate immune response to endotoxins is upregulated (14). In CFTR-defective cells, we have also found an increased activity of non-receptor tyrosine kinases belonging to the Src family (SFKs) (14). Src kinases are involved in a variety of signaling pathways including the regulation of inflammatory responses and phosphorylation of the cytoplasmic domain of TLRs (i.e TLR 2, 3, 4 and 5), which is involved in the activation of the downstream signaling cascade (15-17). Members of the SFK family share a common and well-conserved regulatory mechanism. Src is able to self-activate, unless it is negatively regulated by phosphorylation mediated by the C-terminal Src kinase (Csk), a cytosolic protein that anchors to the membrane through the adaptor Cbp (Csk Binding Protein) (18, 19). Cbp is anchored to the cytoskeleton by the PDZ domains of EBP50 (Ezrin-radix-moesin binding protein 50) (20, 21). EBP50 is an adapter protein localized at the apical region of epithelial cells. EBP50 also binds CFTR favoring its association into macromolecular signaling complexes at the plasma membrane (22-24). In this study, we address the hypothesis that CFTR regulates TLR4-dependent inflammatory responses in the biliary epithelium by modulating Src activity. Expression of CFTR at the apical membrane facilitates the assembly of a protein complex able to maintain the kinase inactive. In the absence of CFTR this complex does not assemble, resulting in self-activation of Src and subsequent increase in TLR4 phosphorylation and NF-κB–dependent cytokine production, in response to endotoxins. We also documented that epithelial cell-cell junctions and monolayer barrier function are impaired in CFTR-defective cells as a secondary effect of the NF-κB-dependent inflammatory process. The protective effects of SFKs inhibition in vitro and in vivo demonstrate the pathogenetic relevance of this mechanism and suggest that targeting SFK activity or NF-κB is a potential therapeutic target for CF-liver disease and possibly other cholangiopathies. MATERIALS AND METHODS Reagents and additional methods are detailed in the supplemental materials and methods. Animals and Experimental Protocol All procedures involving animals in this study were performed according to protocols approved by the Yale University Institutional Animal Care and Use Committee. Congenic B6.129P2-Cftrtm1Unc mice, which possess the S489X mutation that blocks transcription of CFTR (25) and ΔF508-CFTR (B6-129-Cftrtm1Kth) (26) mice with targeted mutation corresponding to the ΔF508 mutation in human, were used for in vivo experiments and/or for the isolation of primary cholangiocytes cell lines. Animals were bred in our facility or provided by the CF Core Center Animal Core (Case Western Reserve University, Cleveland, OH) and were maintained as previously described (14, 27, 28). Cftr-KO mice and WT littermates were exposed to DSS (14) alone or in combination with the tyrosine kinase inhibitor PP2 (1mg/Kg body weight/day) by i.p. At the end of the treatment mice were sacrificed and liver tissue was harvested formalin-fixed and paraffin-embedded for histochemical analysis. Cell culture Mouse cholangiocytes were isolated from WT, DeltaF508 and Cftr-KO mice as described (27). After the first passage cells were plated into 25-cm2 tissue culture flasks coated with rat-tail collagen as previously described (14, 27). Before selected experiments cells were cultured in transwell inserts with a 0.4 μm pore semipermeable membrane (Becton, Dickinson, and Co, Franklin Lakes, NJ); in this condition, cells grow as a polarized monolayer that can be accessed from both the apical and basolateral domain distinctly. Establishment of a confluent monolayer was routinely checked, measuring transepithelial resistance and membrane potential difference (Millicell ERS System; Millipore, Billerica, MA). One week after confluence, transepithelial resistance was >1000 Ω·cm2 and cells were ready for analysis. Determination of cytokine secretion Polarized WT and Cftr-KO cholangiocytes were untreated or treated with PP2 (10 μM), LPS (100 ng/ml) or their combination for 12 h. At the end of the treatment, apical and basolateral media were collected and processed for cytokines/chemokines quantification analysis by using the Milliplex Mouse Cytokines/Chemokines Kit EMD Millipore (Billerica,MA) coupled with BioPlex Luminex platform (Bio-Rad Laboratories, Inc, Hercules, CA) following the manufacturer’s instructions. Data were normalized for the cell protein content as previously described (14). Co-immunoprecipitation Polarized WT and Cftr-KO cholangiocytes were lysated in ODG buffer (50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 2% Octyl-D-glucoside, 5mM βmercaptoethanol, 0.25 M NaCl, 1mM Na3VO4, 20 mM NaF, 5% glycerol, 100μl Triton X-100). To immuno-precipitate Csk or Cbp, 3 mg of proteins were incubated with the proper primary antibody for 1 h at 4°C under rotation (see supplemental table 1). Samples were then incubated with 30 μl of Protein A/G PLUS-Agarose for 3 h at 4°C under rotation. After several washing, the beads were resuspended in 50 μl of 4x NuPAGE Lithium Dodecyl Sulphate (LDS) sample buffer. Western blot with antibodies against CFTR, CSK, Cbp or EBP50 was performed on the immune-precipitated lysates (see supplementary table 1). Immunofluorescence and confocal microscopy Polarized WT and Cftr-KO cholangiocytes were processed for immunofluorescence staining. Briefly, cells were fixed with either a) 100% methanol and washed with PBS; b) or 3.7% PFA, washed using PBS/10mM glycine, and permeabilized with PBS/0.2%TritonX-100. Unspecific binding sites were blocked with Normal Horse Serum (1:20) or with 3% non-fat milk for 45 minutes at RT. Cells were incubated overnight at 4°C with specific primary antibody in blocking buffer (supplementary table 1) and for 1 hour at RT with the proper secondary antibody (conjugated with Alexa Fluor 488, 555, 594 or 647 in PBS/BSA/Gly). Cells were then mounted using a Vectashield Kit (Vector Laboratories, Inc, Burlingame, CA) with 4’,6-diamidino-2-phenylindole (DAPI). Confocal analysis was performed using a Zeiss LSM 710 Duo confocal microscope or a Zeiss LSM 510 Meta with x63, 1.4 NA objective. Serial optical sections (0.5 μm thick) were collected for 3D analysis. Proximity ligation assay (PLA) PLA experiments were done using Duolink In Situ reagents (Olink Bioscience). WT and CFTR-KO cholangiocytes seeded on collagen I coated coverslips were fixed with 3.7% PFA, washed using PBS/10mM glycine, and permeabilized with PBS/TritonX-100 0.2%. Cells were incubated with blocking solution (Olink Bioscience) for 30 minutes and then with one (negative control) or two primary antibodies for 1h at RT. Coverslips were then incubated with secondary antibodies linked to PLA oligonucleotide probes PLUS and MINUS (Olink Bioscience) for 1 h at 37 °C. The samples were then incubated with the ligation ligase solution for 30 min at 37 °C to hybridize oligonucleotides tagged on probes. The coverslips were then incubated with the amplification polymerase solution for 100 min at 37 °C to amplify hybridized oligonucleotides and fluorescently label (Alexa-Fluor 488) the amplification products. Cells were then stained with rhodamine conjugate phalloidin for 20 minutes and coverslips were mounted on slides with Duolink In Situ Mounting Medium with DAPI. Imaging was done using a Zeiss LSM 710 Duo confocal microscope. Immunohistochemistry Paraffin-embedded 4 μm liver slides were processed and stained with the cholagiocyte-specific marker K19 to visualize the ductular reaction, with the leukocyte specific marker CD45 and the macrophage marker F4/80 to analyze the inflammatory cell infiltrate. Quantification of the K19, CD45 and F4/80 positive areas by morphometric analysis was performed as described (14). Epithelial permeability assay Epithelial permeability was assessed in polarized monolayers of WT and Cftr-KO cholangiocytes by measuring transepithelial flux of 10-kilodalton FITC-labelled dextrans as previously described (29). Briefly, epithelial monolayers were equilibrated with PBS at 37°C, then the fluorescent marker (400 μl diluted in PBS at a concentration of 200 mg/mL) was added to the apical chamber. Samples were removed at 1-hour intervals for 4 hours. The volume removed (100 μl) was replaced with 37°C PBS. In experiments with SFK or NF-κB inhibitors, PP2 (0.1, 1 or 10 μM) or Bay 11- 7082 (5 μM) were added to the apical chamber for the last 2 hours and epithelial permeability was compared at the end of the 4 hours time point. At the end of the treatment proteins were extracted for Western Blot processing. After sampling, fluorescence intensity (excitation, 485nm; emission, 530nm) was measured on a fluorescent plate reader Sinergy2 (Biotek Instruments, Winooski, VT). Tracer concentrations were determined by linear regression using dilutions of FITC-dextran in PBS. Measurement of trans-epithelial resistance (TER) WT and Cftr-KO cholangiocytes were seeded in transwell inserts with a 0.4 μm pore semipermeable membrane (Becton, Dickinson, and Co, Franklin Lakes, NJ). After four days in culture TER was measured daily for 9 days using a Millicell ERS System (Millipore, Billerica, MA). TER was calculated as ohms/cm2 multiplying it by the surface area of the monolayer (0.33 cm2). Statistics Results are shown as mean ± SD. Statistical comparisons were made using one-way analysis of variance or the Wilcoxon–Mann–Whitney 2-sample rank sum test, where appropriate, using Prism GraphPad. P values less than .05 were considered significant. RESULTS Src activity and TLR4 phosphorylation are increased in CFTR-KO and DeltaF508 mutated cholangiocytes Cholangiocytes isolated from CFTR-null mice (CFTR-KO; Cftrtm1Unc) have higher Src activity and TLR4 phosphorylation at tyrosine 674 (see figure 1A), as previously reported by our group (14). CFLD occurs in patients with severe CFTR mutations (class I-III), ΔF508 being the most frequent (about 80% of CF patients) (30). To understand if Src signaling is enhanced also in the most common mutation, we isolated cholangiocytes from DeltaF508;Cftrtm1Kth mice (corresponding to the ΔF508 mutation in humans) (26) and tested Src activity, measuring the phosphorylation at the tyrosine 419 of the catalytic domain and TLR4 phosphorylation at the tyrosine 674. As shown in figure 1A, similarly to Cftr-KO cholangiocytes, DeltaF508 cells have significant increased Src tyrosine kinase activity (pY419) and increased TLR4 phosphorylation compared to WT cholangiocytes. Increased Src activation in Cftr-KO cholangiocytes is caused by lack of CFTR expression at the membrane To investigate if the increased Src kinase activity depends on the defective CFTR channel function, we treated WT cells for 1h, 6hrs and 24hrs with a specific CFTR blocker (CFTR inh-172, 10 μM), known to alter the channel gating function (31). Abolishment of cAMP-mediated chloride efflux (32) confirmed the effective inhibition of CFTR activity in WT cholangiocytes treated with CFTR inh-172 (Supplementary figure 2). As shown in figure 1B, both basal levels of Src and TLR4 phosphorylation were unaffected, suggesting that the localization of CFTR at the membrane, rather than its channel function is important for the regulation of Src activity. Thus, we hypothesized that the membrane expression of CFTR might be necessary for the assembly and stabilization of a protein complex able to regulate Src. As discussed in the introduction, Csk and Cbp could interact with CFTR through the PDZ domains of the scaffold protein EBP50, to maintain Src in an inactive state. To demonstrate the interaction of the endogenous proteins CFTR-EBP50-Cbp-Csk in a complex, we used three different approaches. Co-immunoprecipitation experiments showed that in WT cholangiocytes, Csk co-immunoprecipitates with Cbp and CFTR, and Cbp co-immunoprecipitates with EBP50 (figure 1C). Interestingly, although by Western blot the expression level of EBP50, Csk and Cbp was comparable in WT and CF cells (figure 1D), we found no co-immunoprecipitation between Cbp, Csk and EBP50 in CFTR-defective cells. By confocal microscopy, we visualized the apical distribution of CFTR, EBP50, Cbp and Csk and their interactions in the polarized monolayer. As shown in panel 2A, in WT cells expressing CFTR, EBP50 staining is restricted to the apical membrane and co-localizes with CFTR, Cbp and Csk at the apical membrane (figure 2B). Csk also appears to be localized in close proximity to TLR4 at the apical membrane (figure 2C). On the contrary, in CFTR-defective cells, EBP50 staining is diffuse and mislocalized in the cytosol (figure 2A). Furthermore, co-localization of Cbp and Csk at the apical membrane is lost (figure 2B). TLR4 on the other hand, appears to be normally expressed at the apical membrane also in Cftr-KO cells (figure 2C), suggesting that the different susceptibility of CF cells to LPS does not depend on an altered membrane distribution of the TLR4 receptor. Using an in-situ proximity ligation assay, based on the formation of fluorescent spots when two proteins of interest are located within a distance of 40nm, we confirmed that EBP50 and Cbp strictly interact in normal cholangiocytes, whereas no signal was detected in Cftr-KO cells (Figure 2D). For negative control, detection was performed in WT and Cftr-KO cells when one of the two antibodies was omitted. Altogether, these results indicate that in WT cells, CFTR physically interacts with Cbp and Csk through EBP50, and that this interaction is missing in CF cells. In CF cells, the negative regulation of Src is lost and the kinase is free to self-activate and to phosphorylate TLR4, increasing its responsiveness to TLR4 agonists. Inhibition of Src activity decreases LPS-induced activation of NF-κB and cytokine secretion in CF cholangiocytes To study if inhibition of Src kinase decreases the response of the CF biliary epithelium to endotoxins, polarized WT and Cftr-KO cholangiocytes were exposed to LPS (100 ng/ml) for 6 hours in the presence or absence of the SFK family inhibitor PP2 (10 μM) (14). PP2 treatment did not alter the PKA signaling as shown by secretin receptor gene expression analysis and intracellular cAMP production (supplementary figure 3). NF-κB activation was assessed by Western blot for the p65 subunit of NF-κB in nuclear fractions and by measuring the levels of phospho-IkBα. As shown in figure 3A, in Cftr-KO cholangiocytes, as compared to WT cells, NF-κB activation was significantly increased, both in basal conditions and after LPS stimulation. In CF cells not exposed to LPS, treatment with PP2 restored NF-κB to levels similar to WT cells. When cells were treated with LPS in the presence of PP2, NF-κB activation was significantly inhibited. Moreover, the levels of phospho-IkBα were decreased after treatment with PP2 correlating with the decreased p65 nuclear expression (supplementary figure 4). A similar experimental approach was used to study the effects of Src inhibition on the secretion of NF-κB-dependent cytokines previously shown to be increased in CF-cholangiocytes (i.e. G-CSF, CXCL1, LIX and CXCL2)(14, 33). The concentration of cytokines was quantified in the apical and basolateral media by Luminex assay in CF-cholangiocytes exposed to LPS and compared with CF-cholangiocytes exposed to LPS and PP2 (10μM). As shown in figure 3B, LPS-stimulated secretion of G-CSF, CXCL1, LIX and CXCL2 was significantly increased in Cftr-KO cholangiocytes as compared to WT. However, secretion of these inflammatory mediators was significantly decreased by treatment with PP2, consistent with the hypothesis that Src controls TLR4/NF-κB response to LPS in CFTR-defective cells. The distribution of apical junction proteins and actin cytoskeleton is altered in Cftr-KO cholangiocytes. While performing the in-situ proximity assay, we noticed significant changes in the actin cytoskeleton in CFTR-defective cholangiocytes that lead us to investigate it further. Phalloidin staining revealed stark differences in the structure of the actin cytoskeleton in CFTR-defective cells both in not polarized (see figure 2D) and polarized culture conditions (figure 4). As shown in figure 4, while cadherin-catenin complexes accumulated normally at areas of cell-cell contact in WT cells, their association with circumferential actin fibers was largely abolished in cells lacking CFTR. Apical actin filaments were still present, but unlike WT cells where the subcortical actin ring co-localized closely with cell-cell junctions, in Cftr-KO cells the actin appeared split in multiple fibers and failed to co-localize with E-cadherin at the adherens junctions. Interestingly, we observed increased actin accumulation at the tricellular junctions (structures where three cells meet). Since actin filaments are directly connected with cell junction structures (34, 35), we investigated if the defective organization of actin filaments reflected a loss of tight junctions (TJs) and adherens junctions (AJs) integrity. Using confocal microscopy, we analyzed the distribution of proteins of the apical junction complex formed by TJs (i.e ZO-1, zona occludens-1) and AJs (i.e afadin). As shown in figure 5A,B while in WT cells ZO-1 and afadin staining are strictly localized at the cellular junctions, in CF cells both proteins appear diffusely distributed in the cytoplasm. The analysis of the fluorescence intensity confirmed a significant increase of the cytoplasmic/junctional ratio in CF as compared to WT cells suggesting a destabilization of the cellular junctions (figure 5C). Barrier function is impaired in Cftr-KO cell monolayer. The apical junction complex is a functional unit formed by tight junctions (TJs), adherens junctions (AJs) and the perijunctional F-actin cytoskeleton that functionally interact to preserve the barrier function of the epithelium. To determine whether the junctional defect affected the barrier function of Cftr-KO cholangiocyte monolayers, we measured the trans-epithelial movement of FITC-labeled dextrans (10 KDa) across the epithelial monolayer over time (29) and the trans-epithelial resistance (TER) in polarized WT and CFTR-defective cholangiocytes. As shown in figure 6A, the passage of FITC-dextrans across the paracellular compartment was significantly increased in Cftr-KO compared to WT monolayers, whereas TER measurements were significantly decreased in Cftr-KO monolayers (figure 6B), indicating that paracellular permeability and barrier function is altered in Cftr-KO cells. Cytoskeletal and paracellular changes are due to Src- and NF-κB-dependent inflammatory responses To test the effects of Src inhibition on the cytoskeletal and paracellular changes, we treated CF cholangiocytes with the Src inhibitor PP2 (0.1,1 and 10μM, not shown). As shown in figure 7A, treatment with PP2 significantly decreased the junctional defect (i.e afadin and ZO-1 cytoplasmic mislocalization) and rescued the single fiber organization of actin. Consistent with the rescue of cell junction defects by Src inhibition, treatment with PP2 significantly influenced the paracellular permeability of Cftr-KO cells with a dose-dependent effect. As shown in figure 7, as PP2 progressively inhibited Src (Y416) phosphorylation (figure 7C), paracellular permeability of Cftr-KO cells decreased (figure 7D). Inhibition of NF-κB using Bay 11-7082 had a similar rescue effect on the actin distribution and the paracellular permeability (supplementary figure 1A,B), suggesting that the apical junction changes are mediated by the increased Src-dependent NF-κB activation in CF cells. Furthermore, to exclude a potential effect of NF-κB on Src activation in Cftr-KO cholangiocytes, we measure Src phosphorylation at the tyrosine 419 of the catalytic domain by western blot in CF cells treated with the NF-κB inhibitor Bay 11-7082 (5 μM). As shown in supplementary figure 1C, phosphorylation level of Src was unchanged by Bay 11-7082 treatment, suggesting that Src activation is upstream of NF-κB. In vivo inhibition of SFK activity reduces biliary damage and inflammation in Cftr-KO mice treated with DSS To validate in vivo the role of Src inhibition and its possible use as a therapeutic target in CFLD, we treated Cftr-KO mice with DSS, in the presence or the absence of the specific SFK inhibitor PP2. It has been previously established (13, 14) that the increased intestinal permeability caused by DSS induced colitis favors the translocation of gut-derived endotoxins to the liver, causing damage and inflammation in the liver of Cftr-KO mice. At the end of the treatment, liver tissues were harvested and stained with K19, to quantify the biliary and progenitor cell compartment expansion as a marker of biliary damage and with CD45, to quantify the amount of infiltrating leukocytes in the portal space as a marker of inflammation. DSS-treated Cftr-KO mice developed the expected biliary damage with expansion of a small cholangiocyte ductular reactive component (36) and portal inflammatory reaction and extensive infiltration of mononuclear cells whereas no toxicity was shown in the group treated with PP2 alone. As shown in figure 8, in Cftr-KO mice treated with DSS in combination with PP2 (1 mg/Kg by i.p. daily), bile duct proliferation (A) and inflammatory cell infiltration (B) were significantly reduced as determined by computer-assisted morphometric analysis of K19 and CD45 positive areas respectively, suggesting a protective effect of Src inhibition. In addition, treatment with PP2 significantly reduced the amount of macrophages in DSS-treated Cftr-KO mice as shown by staining for macrophage marker F4/80 (supplementary figure 5). DISCUSSION Epithelial cells represent the first line of defense against external pathogens thanks to the innate immune system and the presence of a number of pattern recognition receptors (such as Toll-like receptors). This system must be tightly controlled to avoid excessive inflammation and damage that may occur even in the presence of physiological amounts of bacterial components (37, 38). In this study, we demonstrate a novel mechanism by which the protein CFTR controls TLR4 activation and inflammation in secretory epithelia. Our findings provide a mechanistic explanation to the observation that CFTR dysfunction affects innate immune pathways, in the liver, as well as in other epithelia in Cystic Fibrosis (39). We have recently shown that in CFTR-defective cholangiocytes, TLR4 phosphorylation at tyrosine 674 is increased as compared to normal cells (14) and that TLR4 signaling is up-regulated. Tyrosine phosphorylation of TLRs cytoplasmic domain takes part in the recruitment process of cytoplasmic adaptor proteins during activation of TLR4 downstream signaling (15). In immune cells (i.e. macrophages), members of the Src protein tyrosine kinase family (SFK) phosphorylate the TIR domain of TLR4 and increase the recruitment of downstream signal-competent adapter proteins (e.g MyD88) and kinases (e.g. IRAK-4 and IRAK-1). As a counter-regulatory mechanism, TLR4 tyrosine phosphorylation is reduced to avoid uncontrolled inflammation during continuous exposure to LPS (LPS tolerance)(15, 17). In this study we sought to understand whether activation of Src signaling plays a role in the pathologic inflammatory responses seen in CF cholangiocytes and to clarify the mechanism linking CFTR-deficiency to Src activation and cell dysfunction. Our findings indicate that CFTR associates in an apical membrane complex with proteins acting as constitutive negative regulators of Src (i.e Cbp and Csk). In CF, decreased CFTR protein at the membrane prevents the formation of this multiprotein complex, and consequently Src is able to self-activate and to phosphorylate TLR4, thereby sustaining an increased inflammatory response to LPS and other TLR agonists (Fiorotto/Strazzabosco, unpublished data). Interestingly, pharmacologic inhibition of Src in vitro reduces the LPS-induced NF-κB activation and cytokine secretion in CF biliary cells. CFTR is known to interact directly or indirectly with several proteins that regulate its channel activity (i.e AKAP, PKA), integrate signaling pathways (i.e adenosine 2b receptors, β2-adrenergic receptor) or coordinate other transport activities (i.e ENaC, ROMK) (3-5). Here we demonstrate that CFTR independently from its channel activity participates in the regulation of epithelial innate immunity by controlling the activity of Src. In fact, pharmacological inhibition of CFTR channel function in WT cells has no significant effects on Src activation and TLR4 phosphorylation. Conversely, chemical inhibition of Src significantly decreases the activation of NF-κB, downstream to TLR4 signaling and the dependent secretion of cytokines. The finding that inhibition of Src in vivo improves the inflammatory liver phenotype in Cftr-KO mice exposed to gut-derived endotoxins, provides a proof of concept that Src has a pathogenetic role in CF and is a possible target for treatment. In physiological conditions, Src activity is tightly regulated by Csk, (C-terminal Src kinase), a kinase that holds Src in an inactive conformation by phosphorylating its Y529 residue. To inactivate Src, Csk must translocate from the cytoplasm to the plasma membrane and be anchored in close proximity to the kinase, by the membrane adaptor Csk Binding Protein (Cbp)/PAG which resides in the lipid rafts and is anchored to the cytoskeleton through PDZ domains of ezrin-radix-moesin binding protein 50 (EBP50) (18, 19). EBP50 is an adapter protein localized at the apical region of epithelial cells (22, 23). EBP-50 acts as a scaffold protein that binds CFTR and cooperates in the formation of multiprotein complexes (40). Interestingly, CFTR and EBP50 are mislocalized in cholangiocytes of ezrin defective mice, which also present a cholestatic phenotype (41, 42). Co-immunoprecipitation, confocal microscopy and proximity ligation assay experiments show that in normal cholangiocytes, Cbp is physically linked to Csk and CFTR, and EBP50 to Csk and Cbp. These interactions are lost in Cftr-KO cholangiocytes, suggesting that CFTR insertion at the membrane regulates the assembly of a multi-protein complex that controls Src activation. Confocal microscopy experiments confirmed that EBP50 has a clear distribution on the apical membrane in the normal epithelium but is mislocalized and diffusely distributed in Cftr-KO cholangiocytes. This is consistent with the observation that in bile ducts of CF patients, EBP50 immunoreactivity is not localized at the apical membrane, but distributed throughout the cytoplasm (43). In addition to EBP50, also Cbp and Csk are mislocalized in CF cells suggesting a defect of the apical compartmentalization of these proteins. We also found evidence that lack of CFTR affects the apical junction complex of the biliary epithelium. In fact, the subcortical and peri-junctional F-actin cytoskeleton fails to form properly and ZO-1 and afadin, two key members of the tight and adherens junctions respectively, loose their junctional restriction and appear diffusely distributed in the cytoplasm of CFTR-defective cells. Consistent with the presence of junctional defects, epithelial permeability is significantly increased while transepithelial resistance is decreased in monolayers of Cftr-KO cells. In the biliary epithelium, the integrity of the barrier function is essential to prevent the back-diffusion of toxic bile acids that would cause peribiliary inflammation and fibrogenesis and aggravate the cholestasis. Noteworthy, barrier defects have been associated with several human inflammatory diseases, including cholestatic diseases (44, 45). Normal cell polarity requires epithelial barrier function and apical junctions integrity (34, 35). These structures are anchored into cytoskeletal components such as actin and myosin filaments (46). Several studies have shown that inflammation and production of inflammatory mediators affect the epithelial barrier function. Our group has previously shown that treatment with pro-inflammatory cytokines (i.e IL6, TNFα, IFNγ and IL1) increases the paracellular permeability in normal cholangiocytes (29). Here we show that treatment with Src or NF-kB inhibitors restores the F-actin distribution and normalizes trans-epithelial permeability in Cftr-KO cells, suggesting that these structural defects may be secondary to the increased NF-kB-dependent inflammatory response caused by Src activation. Our findings suggest that targeting tyrosine kinase activation and inflammation could be of potential therapeutic value in CF. In fact, administration of an Src inhibitor to our previously used model of CF cholangitis (Cftr-KO mice with DSS-induced colitis and portal endotoxemia (14)), significantly reduces biliary damage and inflammation, as compared to Cftr-KO mice treated only with DSS. In conclusion, this study provides strong evidence that CFTR regulates TLR-mediated responses in secretory epithelia, by controlling the activation of Src tyrosine-kinase. When CFTR is defective, the negative regulation of Src is lost and the tyrosine kinase is free to target TLR4/NFkB and increase its response to endotoxins. Moreover, aberrant TLR4 activation decreases the epithelial barrier function by destabilizing actin microfilaments and cell-junctional complexes (see cartoon in supplementary figure 6). In the biliary epithelium, decreased barrier function would increase the back-diffusion of toxic bile acids, resulting in further peribiliary inflammation and fibrogenesis. Given the effects of PP2 administration in vivo, and the availability of tyrosine kinase inhibitors already approved for use in several other human conditions (47), targeting SFK activity in the liver may prove to be a useful strategy for the treatment of CFLD. Supplementary Material Supp info Acknowledgements We thank Dr. Ruslan Medzithov, Yale School of Medicine, for helpful discussions, Dr. Nadia Ameen, Yale School of Medicine, for the CFTR antibody (AME4991) and Dr. John Riordan, University of North Carolina – Chapel Hill and Cystic Fibrosis Foundation Therapeutics, for the CFTR antibody (A596). Financial support Research reported in this publication was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under Award Number RO1 DK096096 to MS, P30 DK034989-Silvio O. Conte Digestive Diseases Research Core Centers to MS and RF, R01 NS069753 to P.Z.A. and partially RO1DK101528 to C.S.; by Fondazione Fibrosi Cistica (Grant #18-2012) and by Telethon (Grant #GGP12133) to M.S. and partially supported by PSC Partners Seeking a Cure Foundation (to RF). A.K. is supported by the Jay and Deanie Stein CDA for Cancer Research, Mayo Clinic. List of abbreviations CFTR cystic fibrosis transmembrane conductance regulator CF cystic fibrosis TLR toll-like receptor NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells Src Rous sarcoma oncogene cellular homolog LPS lipopolysaccharide cAMP Cyclic adenosine monophosphate ATP Adenosine triphosphate PKA Protein kinase A CFLD cystic fibrosis-associated liver disease SFK Src family kinases Csk Carboxy-terminal Src kinase Cbp Csk-binding protein/PAG PAG phosphoprotein associated with GEMs PDZ post synaptic density protein (PSD95) EBP50 ERM binding protein 50 G-CSF granulocyte colony-stimulating factor CXCL1 Chemokine (C-X-C motif) ligand 1 LIX LPS-induced CXC chemokine CXCL2 chemokine (C-X-C motif) ligand 2 ZO-1 zona occludens-1 protein AJ adherens junction TJ tight junction TER trans-epithelial resistance DSS dextran sodium sulfate K19 cytokeratin19 TIR toll-IL-1 receptor MyD88 Myeloid differentiation factor 88 IRAK Interleukin 1 receptor-associated kinase 1 AKAP A-kinase anchor protein ENaC epithelial sodium channel ROMK renal outer medullary potassium channel Figure 1 CFTR expression at the membrane is essential to modulate Src activity and TLR4 phosphorylation A) WT, DeltaF508 and Cftr-KO cholangiocytes were cultured in polarized conditions on transwell inserts. Src activity and TLR4 phosphorylation were determined by western blot using antibodies respectively against the active form of Src (pY418) or against the TLR4 tyrosine phosphorylation site (Y674). Bar graphs represent the optical density ratio of p-Src (pY418) vs total Src and p-TLR4 (pY674) vs total TLR4. A statistical significant increase in Src activity and TLR4 phosphorylation is shown in DeltaF508 and Cftr-KO cholangiocytes. Data represent mean ± SD of n=3 independent experiments (*p<0.05 vs WT). Representative western blot images were cropped from the same gel. B) Polarized WT cholangiocytes were treated with CFTR-inh172 for 1, 6 or 24 hours to inhibit the channel function activity. Src activity and TLR4 phosphorylation were determined by western blot. Bar graphs represent the optical density ratio of p-Src (pY418) vs total Src and p-TLR4 (pY674) vs total TLR4. No statistical differences were shown between the different treatments. Data represent mean ± SD of n=4 different experiments. C) Lysates from WT and Cftr-KO cells were immunoprecipitated with antibodies against Csk or Cbp and immunoblotted with specific antibodies against CFTR, Cbp and EBP50. The Western blot analysis shows that CFTR, Cbp, Csk and EBP50 are physically interacting in WT cells but not in Cftr-KO cells. Nuclear fraction lysates were used as negative controls. CFTR input is displayed from a shorter exposure of the same blot. D) Expression of CFTR, Cbp, Csk and EBP50 was evaluated by Western blot in total cell lysates from WT and Cftr-KO cholangiocytes. Figure 2 Proteins involved in Src regulation co-localize at the apical membrane of WT but not Cftr-KO cholangiocytes Confocal images of polarized WT and Cftr-KO cells single stained for EBP50 (A) or co-stained for CFTR-Cbp-Csk (B) or TLR4-Csk (C). A) EBP50 is localized on the apical membrane in WT cells while appears diffused and mislocalized in Cftr-KO cells. B) Z-stack sections show that CFTR, Cbp and Csk co-localize on the apical membrane in WT cells but not in CF cells. C) TLR4 shows the same distribution in WT and CF cells but does not co-localize with Csk in CF cells. D) Confocal imaging of in-situ proximity ligation assay, using EBP50 and Cbp antibodies, indicates interaction of the two proteins (green dots) in WT cells but not in CFTR-KO cells. Phalloidin staining (red) was used to show the localization of the interaction in proximity of F-actin fibers in x-z and y-z images. In the negative control the assay was performed using EBP50 antibody only. Figure 3 Treatment with the SFK inhibitor PP2 decreases LPS-induced NF-κB activation (A) and cytokine secretion (B) in Cftr-KO cholangiocytes Polarized cholangiocytes were treated with PP2 (10 μM), LPS (100 ng/ml) or their combination. A) NF-κB activity was assessed by western blot using an antibody against p65 in nuclear fractions. Histone-3 was used to normalize for the protein content. Bar graphs represent optical density quantification of n=4 experiments. B) Apical and basolateral media were analyzed by Luminex and data were normalized for the cell protein content. Treatment with PP2 significantly inhibited LPS-stimulated cytokine secretion in Cftr-KO cells. The plots represent the means of 5 independent experiments (*p<0.05; **p<0.01). Figure 4 F-actin cytoskeleton defect in Cftr-KO cells Confocal images of confluent monolayers of WT and Cftr-KO cholangiocytes were acquired with the same imaging setting. Distibution of phalloidin (F-actin) staining reveals a defect in the proper formation of the cortical actin ring with accumulation of F-actin at the tri-cellular junctions in Cftr-KO cells compared with WT cells. E-cadherin was used to define the junctional and lateral cell areas. Figure 5 Cell junction defect in Cftr-KO cells Confocal images of confluent monolayers of WT and Cftr-KO cholangiocytes were acquired with the same imaging setting. Staining for ZO-1 (B), a protein of tight-junctions and afadin (C), a protein of adherens junctions, shows that in Cftr-KO cells both proteins have lost their junctional restriction and appear diffusely distributed in the cytoplasm. Quantification of the fluorescence intensity revealed an increased cytoplasmic/junctional ratio in Cftr-KO cells of both proteins. Fluorescence intensity was measured in 10 cells per field and in 3 fields per sample (*p<0.05). P120 (B) and β-catenin (C) were used to define the entire lateral areas of cell-cell contact. Figure 6 Impaired epithelial barrier function in Cftr-KO monolayers A) Polarized monolayers of WT and Cftr-KO cholangiocytes were assayed for paracellular permeability to 10KDa FITC-dextrans. Concentration of dextrans diffused from the apical to the basolateral compartment was determined by linear regression (4-hour assay period) using known standard concentrations. Results represent the mean ± SD of 3 independent experiments in triplicate. B) WT and Cftr-KO cholangiocytes were seeded in transwell inserts. After reaching confluence TER was recorded daily for 9 days. Results represent the mean ± SD of 5 independent experiments. Figure 7 Cytoskeletal and paracellular defects are rescued by Src inhibition in Cftr-KO cholangiocytes A) Confluent monolayers of Cftr-KO cholangiocytes were treated with SFK inhibitor PP2 (0.1-1μM) for 2 hours and stained for ZO-1, afadin and phalloidin (F-actin) and analyzed at the confocal microscope. Images show the re-localization and restriction of ZO-1 and afadin to the junctions and the rescue of the normal subcortical distribution of F-actin after treatment with PP2. B) Quantification of the fluorescence intensity revealed a decreased cytoplasmic/junctional ratio in Cftr-KO cells, after PP2 treatment, of both proteins. Fluorescence intensity was measured in 10 cells per field and in 3 fields per sample (**p<0.01). C) Polarized monolayers of WT and Cftr-KO cholangiocytes were exposed to different concentrations of PP2 that significantly decreased Src activity in Cftr-KO cells as shown by the quantification of p-Src (pY418) vs total Src by Western Blot. D) Bar graph shows the quantification of FITC-dextrans paracellular permeability at the end of the treatment. Results represent the mean ± SD of 4 independent experiments in triplicate; (*p<0.05; **p<0.01). Figure 8 PP2 treatment reduces biliary damage and inflammation in Cftr-KO mice treated with DSS Cftr-KO mice were untreated (n=6) or either treated with DSS (n=5) or with DSS and PP2 (1 mg/Kg by i.p. daily)(n=5). At the end of the treatment, liver tissues were harvested and stained with the cholangiocyte-specific marker K19 (A) or the leukocyte specific marker CD45 (B). Computer-assisted morphometric analysis of K19 and CD45 positive areas shows that PP2 treatment significantly reduced the bile duct proliferation and inflammatory cell infiltration in Cftr-KO mice treated with DSS (*p<0.05 vs DSS only). Bar scale=50 um. Potential conflict of interests: Nothing to disclose Author names in bold designate shared co-first authorship. REFERENCES 1 Colombo C Battezzati PM Liver involvement in cystic fibrosis: primary organ damage or innocent bystander? J Hepatol 2004 41 1041 1044 15582140 2 Riordan JR Rommens JM Kerem B Alon N Rozmahel R Grzelczak Z Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA Science 1989 245 1066 1073 2475911 3 Li C Naren AP CFTR chloride channel in the apical compartments: spatiotemporal coupling to its interacting partners Integr Biol (Camb) 2010 2 161 177 20473396 4 Guggino WB The cystic fibrosis transmembrane regulator forms macromolecular complexes with PDZ domain scaffold proteins Proc Am Thorac Soc 2004 1 28 32 16113408 5 Guggino WB Stanton BA New insights into cystic fibrosis: molecular switches that regulate CFTR Nat Rev Mol Cell Biol 2006 7 426 436 16723978 6 Bruscia EM Zhang PX Satoh A Caputo C Medzhitov R Shenoy A Abnormal trafficking and degradation of TLR4 underlie the elevated inflammatory response in cystic fibrosis Journal of immunology 2011 186 6990 6998 7 Sturges NC Wikstrom ME Winfield KR Gard SE Brennan S Sly PD Monocytes from children with clinically stable cystic fibrosis show enhanced expression of Toll-like receptor 4 Pediatr Pulmonol 2010 45 883 889 20717938 8 John G Yildirim AO Rubin BK Gruenert DC Henke MO TLR-4-mediated innate immunity is reduced in cystic fibrosis airway cells Am J Respir Cell Mol Biol 2010 42 424 431 19502387 9 Cohn JA Strong TV Picciotto MR Nairn AC Collins FS Fitz JG Localization of the cystic fibrosis transmembrane conductance regulator in human bile duct epithelial cells Gastroenterology 1993 105 1857 1864 7504645 10 Colombo C Battezzati PM Strazzabosco M Podda M Liver and biliary problems in cystic fibrosis Semin Liver Dis 1998 18 227 235 9773423 11 Feranchak AP Sokol RJ Cholangiocyte biology and cystic fibrosis liver disease Semin Liver Dis 2001 21 471 488 11745036 12 Strazzabosco M Fabris L Spirli C Pathophysiology of cholangiopathies J Clin Gastroenterol 2005 39 S90 S102 15758666 13 Blanco PG Zaman MM Junaidi O Sheth S Yantiss RK Nasser IA Induction of colitis in cftr−/− mice results in bile duct injury Am J Physiol Gastrointest Liver Physiol 2004 287 G491 496 15064232 14 Fiorotto R Scirpo R Trauner M Fabris L Hoque R Spirli C Loss of CFTR affects biliary epithelium innate immunity and causes TLR4-NF-kappaB-mediated inflammatory response in mice Gastroenterology 2011 141 1498 1508 21712022 15 Chattopadhyay S Sen GC Tyrosine phosphorylation in Toll-like receptor signaling Cytokine Growth Factor Rev 2014 25 533 541 25022196 16 Okutani D Lodyga M Han B Liu M Src protein tyrosine kinase family and acute inflammatory responses Am J Physiol Lung Cell Mol Physiol 2006 291 L129 141 16581827 17 Medvedev AE Piao W Shoenfelt J Rhee SH Chen H Basu S Role of TLR4 tyrosine phosphorylation in signal transduction and endotoxin tolerance J Biol Chem 2007 282 16042 16053 17392283 18 Chong YP Mulhern TD Cheng HC C-terminal Src kinase (CSK) and CSK-homologous kinase (CHK)--endogenous negative regulators of Src-family protein kinases Growth Factors 2005 23 233 244 16243715 19 Ingley E Src family kinases: regulation of their activities, levels and identification of new pathways Biochim Biophys Acta 2008 1784 56 65 17905674 20 Hrdinka M Horejsi V PAG--a multipurpose transmembrane adaptor protein Oncogene 2014 33 4881 4892 24213579 21 Kawabuchi M Satomi Y Takao T Shimonishi Y Nada S Nagai K Transmembrane phosphoprotein Cbp regulates the activities of Src-family tyrosine kinases Nature 2000 404 999 1003 10801129 22 Bezprozvanny I Maximov A PDZ domains: More than just a glue Proc Natl Acad Sci U S A 2001 98 787 789 11158544 23 Fouassier L Duan CY Feranchak AP Yun CH Sutherland E Simon F Ezrin-radixin-moesin-binding phosphoprotein 50 is expressed at the apical membrane of rat liver epithelia Hepatology 2001 33 166 176 11124833 24 Moyer BD Denton J Karlson KH Reynolds D Wang S Mickle JE A PDZ-interacting domain in CFTR is an apical membrane polarization signal J Clin Invest 1999 104 1353 1361 10562297 25 Snouwaert JN Brigman KK Latour AM Malouf NN Boucher RC Smithies O An animal model for cystic fibrosis made by gene targeting Science 1992 257 1083 1088 1380723 26 Zeiher BG Eichwald E Zabner J Smith JJ Puga AP McCray PB Jr. A mouse model for the delta F508 allele of cystic fibrosis J Clin Invest 1995 96 2051 2064 7560099 27 Fiorotto R Spirli C Fabris L Cadamuro M Okolicsanyi L Strazzabosco M Ursodeoxycholic acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion Gastroenterology 2007 133 1603 1613 17983806 28 Spirli C Fiorotto R Song L Santos-Sacchi J Okolicsanyi L Masier S Glibenclamide stimulates fluid secretion in rodent cholangiocytes through a cystic fibrosis transmembrane conductance regulator-independent mechanism Gastroenterology 2005 129 220 233 16012949 29 Spirli C Nathanson MH Fiorotto R Duner E Denson LA Sanz JM Proinflammatory cytokines inhibit secretion in rat bile duct epithelium Gastroenterology 2001 121 156 169 11438505 30 Kumar S Tana A Shankar A Cystic fibrosis--what are the prospects for a cure? Eur J Intern Med 2014 25 803 807 25447947 31 Ma T Thiagarajah JR Yang H Sonawane ND Folli C Galietta LJ Thiazolidinone CFTR inhibitor identified by high-throughput screening blocks cholera toxin-induced intestinal fluid secretion J Clin Invest 2002 110 1651 1658 12464670 32 McNeer NA Anandalingam K Fields RJ Caputo C Kopic S Gupta A Nanoparticles that deliver triplex-forming peptide nucleic acid molecules correct F508del CFTR in airway epithelium Nat Commun 2015 6 6952 25914116 33 Scirpo R Fiorotto R Villani A Amenduni M Spirli C Strazzabosco M Stimulation of nuclear receptor peroxisome proliferator-activated receptor-gamma limits NF-kappaB-dependent inflammation in mouse cystic fibrosis biliary epithelium Hepatology 2015 62 1551 1562 26199136 34 Ivanov AI Parkos CA Nusrat A Cytoskeletal regulation of epithelial barrier function during inflammation Am J Pathol 2010 177 512 524 20581053 35 Meng W Takeichi M Adherens junction: molecular architecture and regulation Cold Spring Harb Perspect Biol 2009 1 a002899 20457565 36 Meng F Francis H Glaser S Han Y DeMorrow S Stokes A Role of stem cell factor and granulocyte colony-stimulating factor in remodeling during liver regeneration Hepatology 2012 55 209 221 21932404 37 Sansonetti PJ Medzhitov R Learning tolerance while fighting ignorance Cell 2009 138 416 420 19665961 38 Takeuchi O Akira S Toll-like receptors; their physiological role and signal transduction system Int Immunopharmacol 2001 1 625 635 11357875 39 Cohen TS Prince A Cystic fibrosis: a mucosal immunodeficiency syndrome Nat Med 2012 18 509 519 22481418 40 Fehon RG McClatchey AI Bretscher A Organizing the cell cortex: the role of ERM proteins Nat Rev Mol Cell Biol 2010 11 276 287 20308985 41 Fouassier L Fiorotto R Ezrin finds its groove in cholangiocytes Hepatology 2014 42 Hatano R Akiyama K Tamura A Hosogi S Marunaka Y Caplan MJ Knockdown of ezrin causes intrahepatic cholestasis by the dysregulation of bile fluidity in the bile duct epithelium Hepatology 2014 43 Fouassier L Rosenberg P Mergey M Saubamea B Claperon A Kinnman N Ezrin-radixin-moesin-binding phosphoprotein (EBP50), an estrogen-inducible scaffold protein, contributes to biliary epithelial cell proliferation Am J Pathol 2009 174 869 880 19234136 44 Sambrotta M Strautnieks S Papouli E Rushton P Clark BE Parry DA Mutations in TJP2 cause progressive cholestatic liver disease Nat Genet 2014 46 326 328 24614073 45 Lee SH Intestinal permeability regulation by tight junction: implication on inflammatory bowel diseases Intest Res 2015 13 11 18 25691839 46 Baum B Georgiou M Dynamics of adherens junctions in epithelial establishment, maintenance, and remodeling J Cell Biol 2011 192 907 917 21422226 47 Aleshin A Finn RS SRC: a century of science brought to the clinic Neoplasia 2010 12 599 607 20689754

PMC005xxxxxx/PMC5116009.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101170508 30450 J Thromb Haemost J. Thromb. Haemost. Journal of thrombosis and haemostasis : JTH 1538-7933 1538-7836 27546592 5116009 10.1111/jth.13477 NIHMS812350 Article Inorganic polyphosphate promotes cyclin D1 synthesis through activation of mTOR/Wnt/β-catenin signaling in endothelial cells Hassanian Seyed Mahdi 1 Ardeshirylajimi Abdolreza Dinarvand Peyman Rezaie Alireza R. Edward A. Doisy Department of Biochemistry and Molecular Biology, Saint Louis University School of Medicine, Saint Louis, MO Corresponding Author. Alireza R. Rezaie, Ph.D., Edward A. Doisy Department of Biochemistry and Molecular Biology, Saint Louis University School of Medicine, 1100 S. Grand Blvd., St. Louis, MO 63104, Tel: (314) 977-9240; Fax: (314) 977-9205; rezaiear@slu.edu 1 Current address: Department of Medical Biochemistry, School of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran 11 10 2016 5 10 2016 11 2016 01 11 2017 14 11 22612273 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Background Inorganic polyphosphate (polyP) elicits intracellular signaling responses in endothelial cells through activation of mTOR complexes 1 and 2. Glycogen synthase kinase 3 (GSK-3) is known to be a negative regulator of mTOR and Wnt/β-catenin signaling pathways. Objective The objective of this study was to investigate the effect of polyP on the expression, degradation and subcellular localization of the Wnt/β-catenin target gene, cyclin D1, in endothelial cells. Methods Regulation of cyclin D1 expression, phosphorylation and subcellular localization by polyP or platelet releasates was monitored in the absence and presence of pharmacological inhibitors and/or siRNA for specific molecules of the upstream mTOR/Wnt/β-catenin signaling network by established methods. Results Both purified polyP and boiled-platelet releasates induced the phosphorylation-dependent inactivation of GSK-3, thereby increasing the expression and nuclear localization, but inhibiting the degradation of cyclin D1. Inhibitors of mTORC1 (PI3K, AKT, PLC, PKC), rapamycin and siRNA for raptor (mTORC1-specific component) and β-catenin, all inhibited polyP-mediated regulation of cyclin D1 expression, phosphorylation and subcellular localization in endothelial cells. The signaling effect of polyP was effectively inhibited by the recombinant extracellular domain of the receptor for advanced glycation end products (RAGE) and/or by the RAGE siRNA. Specific pharmacological inhibitors and siRNA knockdown of ERK1/2 and NF-κB pathways indicated that polyP-mediated cyclin D1 expression and nuclear localization are IKKα- and ERK1/2-dependent, whereas its inhibitory effect on phosphorylation-dependent degradation of cyclin D1 is IKKβ-dependent. Conclusion We conclude that a RAGE-dependent polyP-mediated crosstalk between mTOR and GSK-3/Wnt/β-catenin signaling network can modulate important physiological processes in endothelial cells. Polyphosphate GSK-3 Wnt/β-catenin cyclin D1 mTOR endothelial cells Introduction Inorganic polyphosphates (polyP) are linear polymers of 3 to over 1,000 inorganic phosphate residues, linked together by high-energy phosphoanhydride bonds (1). High levels of medium size polyP polymers (~60–100 phosphate units) are stored in platelets which are known to be released to circulation upon their activation by thrombin or other physiological stimuli during activation of the blood coagulation cascade (2,3). Longer polyP polymers (>1000 phosphate units) may be synthesized by microorganisms under different environmental conditions (1,3). Recent results have established an important role for polyP in regulating numerous physiological processes including coagulation (3–5), tumor metastasis (6,7), proliferation (8), apoptosis (9,10) and inflammation (4,10–12). PolyP is known to activate mammalian target of rapamycin (mTOR) in breast cancer cells (13). We recently demonstrated polyP, upon interaction with the receptor for advanced glycation end products (RAGE) and P2Y1 purinergic receptor, mediates phosphorylation-dependent inactivation of the upstream regulatory tuberous sclerosis complex 1/2 (TSC1/2), thereby activating both mTOR complexes 1 (mTORC1) and 2 (mTORC2) in HUVECs (11,12). It has been established that the key negative regulator of Wnt/β-catenin signaling, glycogen synthase kinase 3β (GSK-3β), through activation of TSC1/2 can inactivate mTOR signaling, suggesting a crosstalk between the two signaling networks modulates metabolic and proliferative signaling responses in HUVECs (14,15). Upon expression and binding of Wnt proteins to the Frizzled family of receptors, GSK-3β is inactivated by phosphorylation, thus leading to activation of both mTOR and Wnt/β-catenin signaling pathways (15,16). GSK-3β is a kinase involved in degradation of β-catenin, thus as a consequence of GSK-3β inactivation, intact β-catenin can translocate into the nucleus, where it binds to specific transcription factors (i.e., T cell factor, TCF) to induce expression of Wnt/β-catenin target genes (16, 17). Cyclin D1 is a Wnt/β-catenin target gene, mainly involved in regulation of the G1 phase of the cell cycle (18). Cyclin D1 accumulates in the nuclei of cells during the G1 phase and exits into the cytoplasm during the S phase (19). It has been demonstrated that GSK-3β through phosphorylation of the nuclear cyclin D1 regulates cytoplasmic distribution and subsequent degradation of the cell cycle protein (20). In this study we investigated the role of polyP-70 in modulating Wnt/β-catenin signaling by analysis of polyP-mediated phosphorylation of GSK-3 and regulation of cyclin D1 synthesis in the absence and presence of siRNA for β-catenin and pharmacological inhibitors of specific signaling pathways. Results suggest that polyP-70, through phosphorylation-dependent inactivation of GSK-3, activates Wnt/β-catenin signaling, thereby regulating expression and subcellular localization of cyclin D1 in HUVECs. Possible physiological significance of these results was demonstrated by findings that platelet releasates exerted similar signaling effects in HUVECs by polyP-dependent mechanisms. Materials and Methods Reagents Wnt/β-catenin inhibitors, iCRT3 and PNU-74654, were from Santa Cruz Biotechnology (Santa Cruz, CA). MG-132, was obtained from Sigma-Aldrich Chemical Co., Inc. (St. Louis, MO). Vectashield mounting medium was from Vector Laboratories Inc. (Burlingame, CA). siRNA for β-catenin was from Dharmacon (Lafayette, CO). Rabbit antibodies to cyclin D1, GSK3α, GSK3β, phospho GSK3α, phospho GSK3β, phospho cyclin D1, eukaryotic initiation factor 4E (eIF4E) and Alexa-conjugated secondary antibodies were from Cell Signaling Technology Inc. (Beverly, MA). Anti-eIF4E (phospho S209) antibody was from Abcam Inc. (Cambridge, MA). Extracellular domain of receptor for advanced glycation end products (sRAGE) was prepared as described (11). All other reagents were provided as described (11,12). Scrambled siRNA was used as negative control for all siRNA experiments as described (11,12). PolyP-70 was a generous gift from Dr. James Morrissey (University of Illinois, Urbana). Platelets releasates were prepared by activation of human platelets with TRAP as described (21). Cell culture Primary human umbilical vein endothelial cells (HUVECs) and immortalized HUVECs (EA.hy926) were grown as described (11,12). Western-blot analysis Cells were treated with stimuli under different conditions (with or without transfection with siRNA and treatment with pharmacological inhibitors) and lysed with Pierce IP lysis buffer as described (12,22). Immunofluorescence staining To examine subcellular localization of cyclin D1, cells (2.5×105) were seeded on glass coverslips and fixed with 4% paraformaldehyde for 10min at room temperature. Next, cells were first incubated in blocking solution (2% BSA, 0.1% tritonX-100 in PBS) and then stained with either anti-cyclin D1 or phospho anti-cyclin D1 (Thr-286) monoclonal antibodies, followed by Alexa-conjugated secondary antibodies (Alexa Fluor 488 and 594 conjugates). Coverslips were mounted on glass slides with Vectashield mounting medium containing DAPI to counterstain cell nuclei. Coverslips were then sealed with a mixture of Vaseline and wax. Stained cells were visualized with a Nikon Optiphot-2 microscope fitted with appropriate fluorescence filters. Results PolyP-70 inactivates GSK-3 and promotes cyclin D1 expression PolyP-70 promoted GSK-3 phosphorylation in HUVECs by a time-dependent mechanism with optimal effect occurring after 15min of treatment (Fig. 1A). GSK-3 is encoded by two genes, GSK-3α and GSK-3β (23). GSK-3α and GSK3-β are inactivated by phosphorylation at Ser-21 and Ser-9, respectively, when Wnt proteins bind to Frizzled receptors to mediate activation of Wnt signaling (24,25). PolyP effectively phosphorylated both isoforms of GSK-3 (Fig. 1A). The Wnt signaling-dependent phosphorylation of GSK-3β inhibits its kinase activity, thereby preventing degradation of β-catenin by GSK-3β (15–17,23). The translocation of intact β-catenin to the nucleus, promotes transcription of Wnt/β-catenin target gene, cyclin D1 (18–20). In quiescent cells, in addition to degradation of β-catenin, GSK-3β is also involved in phosphorylation-dependent degradation of cyclin D1 (18). Results suggested that polyP-mediated inactivation of GSK-3β promotes overexpression of cyclin D1 protein (Fig. 1B), transcript (Fig. 1C) and its nuclear localization (Fig. 1D). PolyP-70 stimulated expression of cyclin D1 in both primary and transformed HUVECs. To provide further support for the hypothesis that polyP-70 activates Wnt/β-catenin signaling, expression of cyclin D1 in polyP-70-stimulated cells was studied with and without transfection of cells with β-catenin siRNA. Efficiency of siRNA knockdown of β-catenin is presented in Fig. 2A. siRNA for β-catenin effectively inhibited polyP-70-mediated up-regulation of cyclin D1 expression (Fig. 2B). Up-regulation of cyclin D1 in polyP-stimulated cells was also investigated in the presence of two different Wnt/β-catenin signaling inhibitors, iCRT3 and PNU-74654 which both bind β-catenin, disrupting the interaction of β-catenin with TCF transcription factor, thereby inhibiting expression of Wnt target genes (26). Consistent with β-catenin siRNA results, both inhibitors inhibited polyP-induced cyclin D1 expression (Fig. 2C,D). These results suggest polyP-70-induced cyclin D1 overexpression in is mediated through activation of Wnt/β-catenin signaling. PolyP-70 activates Wnt/β-catenin signaling by PI3K/AKT- and PLC/PKC/ERK-dependent mechanisms Next, we investigated the role of PI3K/AKT in polyP-mediated Wnt/β-catenin signaling by monitoring expression of cyclin D1 using pharmacological inhibitors of these signaling pathways. Pretreatment of cells with either wortmannin (PI3K inhibitor) or AKT VIII (AKT inhibitor) suppressed up-regulation of cyclin D1 in polyP-70-stimulated cells (Fig. 2E,F). PolyP mediates calcium release from intracellular stores through interaction with P2Y1 (11,12,27). It was found that calcium signaling is required for polyP-mediated Wnt/β-catenin signaling since the Ca2+ chelator, BAPTA-AM, inhibited the effect of polyP-70 on cyclin D1 overexpression (Fig. 2G). Inhibitors of PLC (U-73122) and PKC, bisindolylmaleimide I hydrochloride (BIS), also inhibited polyP-induced overexpression of cyclin D1 (Fig. 2H,I). These results suggest polyP-70 up-regulates cyclin D1 expression through activation of PI3K/AKT and PLC/PKC/Ca2+ signaling cascades. Next, polyP-70-mediated cyclin D1 overexpression was monitored in the absence and presence of siRNA for ERK1/2. Efficiency of gene knockdown was determined 48h post transfection (Fig. 2J). Results demonstrated ERK1/2 siRNA inhibits polyP-mediated up-regulation of cyclin D1 (Fig. 2K). Consistent with these results, pretreatment of cells with ERK1/2 inhibitor, PD-98059, also abrogated polyP-induced cyclin D1 expression (Fig. 2L). Our previous results indicated polyP exerts its modulatory effect through PI3K/AKT- and PLC/PKC/Ca2+-dependent activation of mTOR independent of ERK1/2 (12). Thus in contrast to mTOR activation, polyP-mediated up-regulation of Wnt/β-catenin signaling required ERK1/2 activation. We have demonstrated that none of the inhibitors used in this study has a toxic effect on EA.hy926 cells (12). PolyP-70 regulates phosphorylation and degradation of cyclin D1 GSK-3β phosphorylates cyclin D1 on Thr-286, thereby mediating its cytoplasmic degradation (20). Since polyP-70 inhibits GSK-3α/β, we investigated effect of polyP-70 on phosphorylation and degradation of cyclin D1 in the absence and presence of the proteasome inhibitor, MG-132. In the presence of MG-132, nuclear-localized phospho-cyclin D1 could be visualized by a specific anti-phospho-cyclin D1 antibody (Fig. 3A). However, co-treatment of cells with MG-132 and polyP-70 significantly decreased the phosphorylated form of cyclin D1 (Fig. 3A), suggesting polyP-70 inhibits phosphorylation-dependent degradation of cyclin D1. To further investigate this question, cells were treated with inhibitors of mTOR upstream signaling and phosphorylation of cyclin D1 was analyzed. Results suggest inhibition of PI3K/AKT and Ca2+-signaling abrogates the inhibitory effect of polyP-70 on cyclin D1 phosphorylation (Fig. 3B). In contrast to the suppressive effect of ERK1/2 inhibitor on cyclin D1, the inhibitory effect of polyP-70 on cyclin D1 phosphorylation was independent of ERK1/2 (Fig. 3B, bottom panel). In agreement with results obtained with the ERK1/2 inhibitor, siRNA for ERK1/2 had no effect on polyP-induced inhibition of cyclin D1 phosphorylation (Fig. 3C). Moreover, consistent with the inhibitory effect of PD-98059 (ERK1/2 inhibitor) on polyP-70-induced cyclin D1 overexpression, PD-98059 significantly suppressed nuclear localization of cyclin D1 in polyP-70-stimulated cells (Fig 3D). These results suggest that effects of polyP-70 on cyclin D1 expression and nuclear localization are both dependent on ERK1/2, whereas PolyP-mediated inhibition of cyclin D1 phosphorylation/degradation is independent of ERK1/2. PolyP-mediated cyclin D1 overexpression, localization and degradation are mTORC1- and NF-κB-dependent We recently demonstrated that polyP, through interaction with RAGE and P2Y1, activates both mTORC1 and mTORC2 in endothelial cells (12). GSK-3β through phosphorylation of β-catenin and TSC1/2 inhibits Wnt/β-catenin and mTOR signaling, respectively. Since polyP inactivated GSK-3β, we investigated the role of mTORC1 and mTORC2 in Wnt/βcatenin-dependent regulation of cyclin D1. Results indicated the mTORC1 inhibitor, rapamycin, abrogates polyP-induced up-regulation of cyclin D1 expression (Fig. 4A). Since rapamycin may also inhibit mTORC2 (28,29), this question was analyzed in cells transfected with siRNAs specific for components of either mTORC1 (raptor) or mTORC2 (rictor). Both raptor and rictor siRNAs effectively inhibited their expression (Fig. 4B), however, only siRNA for raptor, inhibited polyP-mediated overexpression of cyclin D1 (Fig. 4C), suggesting polyP-mediated Wnt/β-catenin and mTORC1 signaling are linked pathways. mTORC1 promotes protein synthesis through stimulation of phosphorylation of p70S6K and enhancing expression and phosphorylation-dependent release of eukaryotic initiation factor 4E (eIF-4E) from its inhibitory proteins (30). We previously demonstrated polyP effectively up-regulates phosphorylation of p70S6K through activation of mTORC1 in the absence of serum in endothelial cells (12). Analysis of polyP-70-mediated regulation of eIF-4E indicated polyP has no effect on expression of eIF-4E but promotes its phosphorylation (Fig. 4D). In light of findings that polyP activates NF-κB (10,11) and that there is interplay between mTORC1 and NF-κB pathways in polyP-stimulated cells (12), we investigated the role of NF-κB in Wnt/β-catenin-dependent regulation of cyclin D1 expression in polyP-70-stimulated cells. The IKKα-inhibitor, BAY11-7082, markedly decreased polyP-mediated cyclin D1 overexpression (Fig. 4E) whereas the IKKβ-inhibitor, BMS-345541, had no effect (Fig. 4F). In agreement with inhibitor results, while both IKKα and IKKβ siRNAs effectively inhibited their expression (Fig. 4G), only IKKα siRNA suppressed overexpression of cyclin D1 in the presence of polyP-70 (Fig. 4H). To determine whether polyP-70-mediated cyclin D1 nuclear localization and phosphorylation are NF-κB-dependent, cells were transfected with siRNAs for IKKα and IKKβ and localization and phosphorylation of cyclin D1 were analyzed. Results demonstrated that polyP-mediated nuclear localization of cyclin D1 is IKKα-dependent but the inhibitory effect of polyP on cyclin D1 phosphorylation is specifically dependent on IKKβ (Fig. 4 I,J). Next, involvements of RAGE and P2Y1 receptors in polyP-mediated cyclin D1 synthesis were investigated by both siRNA approaches and competitive binding studies using the extracellular domain of RAGE (sRAGE) (11). Results suggested while siRNA for either RAGE or P2Y1 significantly decreases the ability of polyP to induce cyclin D1, the combination of two siRNAs completely abrogates polyP-mediated cyclin D1 synthesis (Fig. 5A, lane 10), suggesting that similar to polyP-mediated mTOR activation, up-regulation of cyclin D1 synthesis is also mediated through polyP interacting with RAGE and P2Y1. Interestingly, RAGE siRNA by itself resulted in a dramatic decrease in cyclin D1 synthesis (Fig. 5A, lane 1) and in combination with P2Y1 siRNA, no cyclin D1 could be detected in transfected cells (Fig. 5A, lane 9), suggesting a key role for RAGE-signaling in cyclin D1 synthesis. In support of this hypothesis, sRAGE effectively inhibited cyclin D synthesis and polyP-70 did not up-regulate its expression in the presence of sRAGE (Fig. 5B), confirming siRNA results that RAGE is required for cell growth and that polyP induces cyclin D1 synthesis through stimulation of RAGE-signaling. Treatments of cells with polyP for up to 4h did not have any effect on the viability of endothelial cells as determined by the MTT assay as described (12) (Fig. 6A,B). Platelet releasates regulate GSK-3/Wnt/β-catenin signaling Activated platelets secrete polyP with polymer lengths of ~60–100 phosphate units (2,3). Expression, nuclear localization and phosphorylation of cyclin D1 by activated platelet- releasates were monitored employing the same assays described above. It was found that platelet releasates ratio of 0.04 to 0.1 (platelet releasates/total volume, estimated to be ~10 to 25µM, if the polyP concentration was expressed in terms of phosphate monomer) (12), phosphorylates GSK-3α/β by a dose-dependent manner (Fig. 7A). Since platelets were activated with TRAP, the possible stimulatory effect of a small amount of TRAP present in platelet releasates was analyzed. A TRAP concentration of 0.5µM, which exceeds the concentration of TRAP present in the highest platelet releasates ratio of 0.1 (~25µM) (12), did not have any effect on GSK-3 phosphorylation (Fig. 7A), excluding any contribution from trace amounts of TRAP present in platelet releasates. To establish the hypothesis that polyP derived from platelet releasates is responsible for GSK-3 phosphorylation, incubation of platelet releasates with specific polyP inhibitor EcPPXc (polyP-binding domain of Escherichia coli exopolyphosphatase) (31) or alkaline phosphatase (ALP) abrogated the signaling activity of both polyP-70 and platelet releasates (Fig. 7B). Furthermore, boiling platelet releasates for 30min prior to cell treatment did not impact the platelet releasates-mediated expression of cyclin D1, however, boiling releasates followed by treatment with either recombinant EcPPXc or ALP abrogated the signaling effect (Fig. 7B). It has been previously demonstrated that boiling platelet releasates for 30min denatures all proteins without negatively affecting the cofactor function of polyP (21). These results strongly suggest that platelet polyP is responsible for GSK-mediated cyclin D1 overexpression in endothelial cells. In agreement with results obtained with polyP-70, iCRT3, wortmannin, AKT VIII, U-73122, BIS, BAPTA-AM, PD-98059, BAY11-7082 (IKKα-inhibitor) and rapamycin all inhibited the platelet releasates-mediated cyclin D1 overexpression (Fig. 7C). As expected based on results presented above, BMS-345541 (IKKβ-inhibitor) had no inhibitory effect on the platelet releasates-mediated regulation of cyclin D1 expression (Fig. 7C). Similar to results obtained above with 25µM polyP-70, treatment of cells with a 0.1 ratio of boiled platelet releasates (~25µM) enhanced nuclear localization of cyclin D1 and that the effect was eliminated by PD-98059, rapamycin and BAY11-7082 but not by BMS-345541 (Fig. 7D,E). Moreover, consistent with the inhibitory effect of polyP-70 on cyclin D1 phosphorylation, incubation of cells with boiled platelet releasates (0.1 ratio) suppressed cyclin D1 phosphorylation in MG-132-treated cells by mTORC1- and IKKβ-dependent but ERK1/2- and IKKα-independent mechanisms. Taken together, these results suggest that both polyP-70 and platelet-derived polyP regulate GSK-3/Wnt/β-catenin signaling through the same mechanisms in endothelial cells. Discussion We have demonstrated here that polyP inhibits the kinase activity of GSK-3 by phosphorylation of both GSK-3α and GSK-3β isoforms in both primary and transformed HUVECs. The phosphorylation-dependent inhibition of GSK-3 by polyP results in stabilization of β-catenin and its subsequent translocation to the nucleus, thus inducing the expression of the β-catenin target gene, cyclin D1 (16–18). Consistent with this hypothesis, Wnt/β-catenin signaling inhibitors, iCRT3 and PNU-74654, both inhibited polyP-mediated up-regulation of cyclin D1 synthesis. Thus, polyP through inhibition of GSK-3 and stabilization of β-catenin/ cyclin D1 appears to play a key role in regulating metabolic and proliferative signaling responses in both primary and transformed HUVECs. The mechanism by which polyP modulates expression and subcellular localization of cyclin D1 is not known. Nevertheless, our recent results, showing that polyP is an effective stimulator of mTOR in endothelial cells (12), suggest that polyP, through activation of mTOR, may also modulate GSK/Wnt/β-catenin signaling. The polyP-induced mTOR signaling in endothelial cells has been found to be primarily mediated through polyP interacting with RAGE (11), a multi-ligand cell surface receptor involved in numerous pathophysiological processes including cell growth, proliferation, inflammation and cancer (32,33). We recently demonstrated that polyP can bind with high affinity to the RAGE ligands, HMGB1 and histones, and present the nuclear proteins to RAGE, thereby dramatically amplifying proinflammatory signaling responses (11). To determine whether polyP-mediated up-regulation of cyclin D1 expression is mediated through the same receptor signaling system, we used RAGE-specific siRNA to effectively down-regulate expression of RAGE in endothelial cells. Results suggested polyP induces cyclin D1 synthesis through the same mechanism since it was ineffective in up-regulating cyclin D1 expression in the presence of RAGE siRNA. In agreement with previous results that polyP interaction with P2Y1 further stimulates mTOR by inducing calcium signaling (11,12), the combination of RAGE and P2Y1 siRNAs completely inhibited cyclin D1 synthesis. Interestingly, RAGE siRNA by itself dramatically down-regulated cyclin D1 expression independent of polyP, suggesting RAGE signaling is required for cyclin D1 synthesis in endothelial cells during nutrient depletion. Stimulation of cells with polyP in the presence of RAGE siRNA did not produce a significant stimulatory effect on expression of cyclin D1. These results suggest that polyP through RAGE signaling regulates expression of cyclin D1. Further support for RAGE/mTOR-dependent regulation of cyclin D1 signaling was provided by the observation that pharmaceutical inhibitors of the upstream mTOR signaling network (PI3K/AKT and PLC/PKC/Ca2+) all abrogated regulatory effects of polyP on cyclin D1 expression, phosphorylation and subcellular localization. PolyP-mediated up-regulation of the cyclin D1 synthesis was mediated specifically through mTORC1 since siRNA for raptor but not rictor effectively inhibited this process. Results from several studies have indicated that RAGE siRNA arrests a number of human cancer cells in the G1 phase of the cell cycle, thereby inhibiting DNA synthesis, suggesting that RAGE signaling may be involved in promoting cancer growth and metastasis (34–37). Noting that tumor cells can secret significant amounts of HMGB1 and that HMGB1 promotes autophagy (37,38), it has been hypothesized that HMGB1-mediated autophagy through RAGE signaling plays an important role in allowing tumor cells to grow during nutrient depletion (38–41). In this context, we have shown that transformed HUVECs (EA.hy926 endothelial cells) secrete a significant amount of HMGB1 under serum-free growth conditions (42). Thus, it appears that these cells can acquire survival advantage during the serum free culture conditions through HMGB1/RAGE-mediated induction of autophagy, thereby up-regulating cyclin D1 synthesis. In support of this hypothesis, sRAGE, which binds with high affinity to HMGB1, effectively inhibited cyclin D1 synthesis in EA.hy926 cells. We have also demonstrated that polyP can bind to HMGB1 with high affinity and present the nuclear protein to the RAGE, thereby dramatically amplifying proinflammatory signaling responses (11). Several human tumor cell lines are known to synthesize and store a significant amount of polyP in their nuclei and other tissues (1,3,13,43,44). Results presented above raise the possibility that the tumor cell-derived polyP can make a positive contribution to the growth factor-independent proliferation and metastasis of cancer cells by a similar signaling mechanism. Thus, polyP establishes a crosstalk between RAGE, mTOR and Wnt/β-catenin signaling networks that can be critical for survival and regulation of key pathophysiological processes in vascular endothelial cells. The possible physiological significance of these results was demonstrated by findings that boiled human platelet releasates exert similar signaling effects in EA.hy926 endothelial cells by a polyP-dependent mechanism. It is worth noting that the aberrant regulation of both mTOR and Wnt/β-catenin signaling pathways are associated with a number of different tumor developments (16,30,45). A role for platelets in tumor metastasis has also been postulated (46). Thus, it is of interest to further investigate and to determine whether or not polyP, released by activated platelets, in addition to up-regulation of thrombin generation, coagulation and inflammation, can also participate in promoting tumor cell growth and metastasis under certain pathophysiological conditions. It should be emphasized that we monitored HMGB1 expression, RAGE signaling and platelet-releasates experiments only in transformed HUVECs, thus further studies will be required to validate these results in primary endothelial cells. We previously demonstrated that polyP through activation of mTOR also up-regulates the activation of NF-κB, suggesting that the two pathways are interlinked (12). It was interesting to note in the current study that a polyP-mediated crosstalk between NF-κB and Wnt/β-catenin signaling also modulated expression, phosphorylation and subcellular localization of cyclin D1. This is derived from observations that siRNAs for IKKα and IKKβ and their pharmacological inhibitors down-regulated polyP-mediated overexpression of cyclin D1 and/or abrogated the inhibitory effect of polyP on the phosphorylation-dependent degradation of the cell cycle protein. These two kinases appear to exert different modulatory effects on polyP-mediated up-regulation of cyclin D1. This is based on the observation that inhibition of IKKα, but not IKKβ, inhibited polyP-mediated overexpression and nuclear localization of cyclin D1. By contrast, inhibition of IKKβ, but not IKKα, abrogated inhibitory effect of polyP on the phosphorylation-dependent degradation of cyclin D1. These results provide some insight into the nature of an intricate and complex crosstalk between mTOR, GSK-3/Wnt/β-catenin and NF-κB and envision a key role for polyP in RAGE-dependent modulation of these signaling networks in endothelial cells. The mechanism through which polyP activates GSK-3/Wnt/β-catenin signaling is not known. It is however known that the pathway is activated when different Wnt ligands bind to Frizzled family of receptors, thereby inactivating GSK-3β by phosphorylation, thus leading to stabilization of β-catenin, its translocation to the nucleus and transcription of the Wnt/β-catenin target gene, cyclin D1 (15–18). Thus, a mechanism through which polyP can activate GSK-3/Wnt/β-catenin signaling is by up-regulation of the synthesis of Wnt proteins by endothelial cells. However, GSK-3 is also known to be phosphorylated/inactivated by AKT/PKB signaling (30,47), which is a pathway effectively activated by polyP (12). Moreover, we previously demonstrated that polyP, through activation of mTORC1, mediates phosphorylation and stimulation of p70S6 kinase (12), thereby increasing mRNA synthesis and translation of ribosomal proteins which promote protein synthesis during cell growth. In support of a role for polyP in promoting protein synthesis through mRNA translation, we discovered in the current study that polyP also promotes phosphorylation of the mRNA cap-binding protein eIF-4E, which has been shown to be involved in the initiation of translation of a subset of mRNAs (including cyclin D1) required for cell growth (48,49). Indeed, the overexpression of elF-4E in NIH 3T3 cells by transfection studies has been found to be associated with increased cyclin D1 mRNA expression and enhanced translation efficiency of cyclin D1 mRNA in serum-deprived cells (50). These results suggest that polyP can influence GSK-3/Wnt/β-catenin signaling through multiple steps, without requirement for Wnt proteins ligating their specific Frizzled receptors. Based on these results we propose the model presented in Fig. 8 depicting different mechanisms through which polyP may modulate the GSK-3/Wnt/β-catenin signaling. The interaction of polyP with RAGE and P2Y1 activates mTORC1 by PI3K/AKT and PLC/PKC dependent inhibition of TSC1/2 complex as we demonstrated previously (12). PolyP, through interaction with the same receptors, can activate GSK-3/Wnt/β-catenin signaling by at least four different mechanisms: 1) polyP-mediated AKT activation can lead to phosphorylation-dependent inhibition of GSK-3, thereby resulting in stabilization of β-catenin independent of Wnt-mediated Frizzled signaling, 2) polyP-mediated improvement of protein translation can result in augmentation of translation of mRNAs for cyclin D1 and/or Wnt proteins, 3) polyP through activation of IKKα/β can down-regulate GSK-3 activity, thereby stabilizing β-catenin and cyclin D1 independent of a direct Wnt/Frizzled signaling, and finally 4) polyP can directly interact with a Frizzled receptor, thereby activating the pathway by a mechanism similar to RAGE signaling. As indicated above, we have demonstrated that polyP can bind HMGB1 with high affinity to dramatically promote RAGE signaling (11). Thus, we hypothesize that low levels of HMGB1 synthesized by cells under stressed conditions can have a profound effect in modulating these signaling pathways in the presence of polyP (11). Further studies will be required to determine the exact mechanisms through which polyP establishes a crosstalk between mTOR and GSK-3/Wnt/β-catenin signaling network that culminates in elevated expression of cyclin D1 in endothelial cells. We thank Stephanie A. Smith from University of Illinois, Urbana for preparing the platelet releasates and Audrey Rezaie for proofreading the manuscript. Funding Sources This research was partly supported by grants awarded by the National Heart, Lung, and Blood Institute of the National Institutes of Health HL 101917 and HL 62565 to A.R.R. Figure 1 PolyP-70 induces GSK-3 phosphorylation and increases expression and nuclear localization of cyclin D1, in EA.hy926 endothelial cells. (A) Time course of polyP-mediated phosphorylation and inactivation of GSK-3α/β in EA.hy926 cells. (B) Cells were treated with polyP-70 (25 µM) at different time points followed by measuring the expression of cyclin D1. (C) Semiquantitative RT-PCR was performed as described before (22). Results were normalized to expression levels of GAPDH and presented as the fold difference relative to the control group at each time point. The results are shown as mean ± standard deviation of 3 different experiments as determined Student t test. *P<0.05; **P<0.01. (D) The same as B except that the effect of polyP-70 on nuclear localization of cyclin D1 was measured. Cells were stained with DAPI to visualize the nucleus (Blue) and anti-cyclin D1 antibody (Red) and then imaged by confocal microscopy. The scale for the microscopic figure is 20µm. Figure 2 PolyP-70-mediated expression of cyclin D1 in the absence and presence of siRNA or specific inhibitors for signaling molecules. (A) The efficiency of gene knockdown of β-catenin (>75%) was determined 48h post transfection by Western-blotting using a specific antibody. The (B) PolyP-mediated up-regulation of cyclin D1 was monitored after siRNA knockdown of β-catenin in EA.hy926 endothelial cells. (C–D) PolyP-mediated overexpression of cyclin D1 in the absence and presence of increasing concentrations of two different Wnt signaling inhibitors (iCRT3 and PNU-74654). (E) PolyP-mediated up-regulation of cyclin D1 in the absence and presence of increasing concentrations of PI3K inhibitor (wortmannin). (F) PolyP-mediated overexpression of cyclin D1 in the absence and presence of increasing concentrations of AKT inhibitor (AKT inhibitor VIII). (G) PolyP-mediated overexpression of cyclin D1 in the absence and presence of increasing concentrations of intracellular calcium chelator (BAPTA-AM). (H) PolyP-mediated up-regulation of cyclin D1 in the absence and presence of increasing concentrations of PLC inhibitor (U-73122). (I) PolyP-mediated overexpression of cyclin D1 in the absence and presence of increasing concentrations of PKC inhibitor (BIS). (J) EA.hy926 cells were transiently transfected with control siRNA or siRNA for ERK1/2 and the efficiency of gene knockdown (>75%) was determined 48h post transfection by Western-blotting using a specific antibody. (K) The same as J except that polyP-mediated overexpression of cyclin D1 was monitored after siRNA knockdown of ERK1/2. (L) PolyP-mediated overexpression of cyclin D1 in the absence and presence of increasing concentrations of ERK inhibitor (PD-98059). The results are shown as mean ± standard deviation of 3 different experiments. *P<0.05; **P<0.01. Figure 3 PolyP-70 inhibits phosphorylation and degradation of cyclin D1 in EA.hy926 endothelial cells. (A) Cells were treated with MG132 (25 µM) for 4h in the absence or presence of polyP-70 (25 µM) followed by measuring the phosphorylation of cyclin D1 using anti-phospho cyclin D1 antibody. (B) PolyP-mediated inhibition of cyclin D1 phosphorylation was monitored in the presence of specific inhibitors of mTOR upstream signaling pathway. In the presence of each inhibitor, cells were co-incubated with MG132 (25µM) and PolyP-70 (25µM) for 4h. (C) PolyP-mediated inhibition of cyclin D1 phosphorylation was monitored after siRNA knockdown of ERK1/2. (D) PolyP-mediated nuclear localization of cyclin D1 was measured in the absence and presence of ERK1/2 inhibitor (PD-98059). The scale for the microscopic figure is 20µm. Figure 4 PolyP-70-mediated regulation of expression, localization and degradation of cyclin D1 are dependent on mTORC1 and NF-κB signaling pathways. (A) PolyP-mediated overexpression of cyclin D1 was monitored in the presence of different concentrations of mTORC1 inhibitor (rapamycin). (B) EA.hy926 endothelial cells were transiently transfected with control siRNA or siRNA specific for either raptor or rictor and the efficiency of gene knockdown (>75%) was determined 48h post transfection. (C) The same as B except that polyP-mediated up-regulation of cyclin D1 was monitored after siRNA knockdown of either raptor or rictor. (D) The expression and phosphorylation of eIF-4E was monitored in EA.hy926 cells treated with different concentrations of polyP-70. (E) PolyP-mediated overexpression of cyclin D1 was monitored in the presence of different concentrations of IKKα inhibitor (BAY11-7082). (F) The same as D except that polyP-mediated up-regulation of cyclin D1 was monitored in the presence of different concentrations of IKKβ inhibitor (BMS-345541). (G) The same as B except that cells were transiently transfected with control siRNA or siRNA specific for either IKKα or IKKβ. (H) The same as C except that polyP-mediated up-regulation of cyclin D1 was monitored after siRNA knockdown of either IKKα or IKKβ. (I) PolyP-mediated nuclear localization of cyclin D1 was measured in EA.hy926 cells transiently transfected with control siRNA or siRNA specific for either IKKα or IKKβ. (J) The same as H except that polyP-mediated inhibition of cyclin D1 phosphorylation was monitored after siRNA knockdown of IKKα or IKKβ. The scale for the microscopic figure is 20µm. The results are shown as mean ± standard deviation of 3 different experiments. *P<0.05; **P<0.01. Figure 5 PolyP-mediated expression of cyclin D1 in EA.hy926 endothelial cells transfected with siRNAs for RAGE and/or P2Y1 or treated with recombinant soluble RAGE (sRAGE). (A) Cells were transiently transfected with control siRNA or siRNA specific for either RAGE or P2Y1 followed by stimulating cells with polyP-70 (25 µM) and monitoring the expression of cyclin D1. Lane 1, RAGE siRNA alone; lane 2, RAGE siRNA + polyP-70; lane 3, P2Y1 siRNA alone; lane 4, P2Y1 siRNA + polyP-70; lane 5, control siRNA alone; lane 6, control siRNA + polyp-70; lane 7, control (no siRNA, buffer only); lane 8, control + polyP-70; lane 9, RAGE and P2Y1 siRNAs alone combined; lane 10, RAGE and P2Y1 siRNAs combined + polyP-70. (B) Analysis of expression of cyclin D1 treated with sRAGE (2.5 µM). Lane 1, sRAGE alone; lane 2, sRAGE + polyP-70; lane 3, control (buffer only); lane 4, control + polyP-70. The efficiency of gene knockdown was >75% in all cases. The results are shown as mean ± standard deviation of 3 different experiments. *P<0.05 for lane 4 (P2Y1 siRNA + polyP), compared to control siRNA + polyP-70 in panel A. Figure 6 Viability of EA.hy926 cells treated with polyP-70. (A) The time course of cell survival was monitored by treating cells with polyP-70 (25 µM) followed by evaluating the viability by MTT assay as described (22). (B) The same as A except that the polyP-70 concentration-dependence of cell survival was monitored. The percentage of cell viability was calculated by the ratio of the OD540 of wells containing polyP-70 to the OD of the wells lacking polyP-70 × 100. The results are shown as mean ± standard deviation of 3 different experiments. *P<0.05. Figure 7 Platelet releasates regulate expression, localization and stability of cyclin D1 in EA.hy926 cells. (A) Dose response of platelet releasates (30 min) for inducing the phosphorylation of GSK-3α/β in EA.hy926 cells. PolyP-70 is shown as a control in the last lane. (B) Overexpression of cyclin D1 by boiled (30 min) platelet releasates (plt-rel, 0.1 ratio) is inhibited by co-incubation of platelet releasates with EcPPXc (250 µg/ml) and alkaline phosphatase (ALP 2 units/mL). PolyP-70 is shown as a control in the lanes 2–4. (C) Analysis of cyclin D1 overexpression by boiled (30 min) platelet releasates by the inhibitors of mTOR upstream signaling pathway. With all inhibitors, cells were treated with boiled platelet releasates co-incubated with each inhibitor for 8h. (D) Platelet releasates-mediated nuclear localization of cyclin D1 was measured in EA.hy926 cells co-incubated with ERK1/2 inhibitor (PD-98059), mTORC1 inhibitor (rapamycin), IKKα and IKKβ inhibitors (BAY11-7082 and BMS-345541 respectively). (E) The same as D except that platelet releasates-mediated inhibition of cyclin D1 phosphorylation was monitored. The scale for the microscopic figure is 20µm. The results are shown as mean ± standard deviation of 3 different experiments. *P<0.05; **P<0.01. Figure 8 Mechanism of polyP-mediated crosstalk between mTOR and GSK-3/Wnt/β-catenin signaling networks. PolyP, through interaction with RAGE and P2Y1 receptors, activates mTORC1 by PI3K/AKT and PLC/PKC/Ca2+ dependent inhibition of tumor suppressor TSC1/2 complex, thereby augmenting protein translation machinery by phosphorylation of p70S6K and eIF-4E. HMGB1 can dramatically promote these signaling reactions. PolyP by the same mechanism mediates the phosphorylation-dependent activation of IKK-α/β, thereby activating NF-kβ. PolyP can promote the expression of cyclin D1 by several different mechanisms: It can indirectly promote cyclin D1 synthesis by the phosphorylation-dependent inhibition of GSK-3 by either AKT and/or IKK-α/β, thereby preventing the ubiquitin- and proteasome-dependent degradation of β-catenin and cyclin D1 independent of Wnt-Frizzled receptor interaction (dashed blue lines). Alternatively, polyP can promote the translation of cyclin D1 mRNA and/or Wnt mRNAs, thereby leading to up-regulation Wnt proteins and their interaction with and activation of Frizzled receptors (dashed red arrows). Finally, polyP can directly interact with Frizzled (and LRP5/6) receptors to improve the expression of cyclin D1 (solid red arrow). HMGB1, high mobility group box 1; RAGE, receptor for advanced glycation end products; TSC1/2, tuberous sclerosis complex 1/2; GSK-3, glycogen synthase kinase 3; TCF, T cell factor; LRP5/6, Dvl, Dishevelled; LRP5/6, low-density lipoprotein receptor-related protein 5/6, mTORC1, mammalian target of rapamycin complex 1; eIF-4E, eukaryotic initiation factor 4E. Essentials Polyphosphate (polyP) activates mTOR but its role in Wnt/β-catenin signaling is not known. PolyP-mediated cyclin D1 expression (β-catenin target gene) was monitored in endothelial cells. PolyP and boiled platelet-releasates induced the expression of cyclin D1 by similar mechanisms. PolyP establishes crosstalk between mTOR and Wnt/β-catenin signaling in endothelial cells. Addendum S. M. Hassanian designed, performed experiments and wrote the manuscript. A. Ardeshirylajimi and P. Dinarvand performed experiments. A. R. Rezaie supervised the project and wrote the manuscript. Disclosure of Conflict of Interests The authors declare no competing financial interest. Supporting Information The source of reagents can be found in the online version of this article. References 1 Kornberg A Rao NN Ault-Riche D Inorganic polyphosphate: essential for growth and survival Annu Rev Biochem 2009 78 605 647 19344251 2 Ruiz FA Lea CR Oldfield E Docampo R Human platelet dense granules contain polyphosphate and are similar to acidocalcisomes of bacteria and unicellular eukaryotes J Biol Chem 2004 279 44250 44257 15308650 3 Morrissey JH Choi SH Smith SA Polyphosphate: an ancient molecule that links platelets, coagulation, and inflammation Blood 2012 119 5972 5979 22517894 4 Müller F Mutch NJ Schenk WA Smith SA Esterl L Spronk HM Schmidbauer S Gahl WA Morrissey JH Renné T Platelet polyphosphates are proinflammatory and procoagulant mediators in vivo Cell 2009 139 1143 1156 20005807 5 Smith SA Morrissey JH Polyphosphate as a general procoagulant agent J Thromb Haemost 2008 6 1750 1756 18665922 6 Galasso A Zollo M The Nm23-H1-h-Prune complex in cellular physiology: a 'tip of the iceberg' protein network perspective Mol Cell Biochemistry 2009 329 149 159 7 Tammenkoski M Koivula K Cusanelli E Zollo M Steegborn C Baykov AA Human metastasis regulator protein H-prune is a short-chain exopolyphosphatase Biochemistry 2008 47 9707 9713 18700747 8 Shiba T Nishimura D Kawazoe Y Onodera Y Tsutsumi K Nakamura R Ohshiro M Modulation of mitogenic activity of fibroblast growth factors by inorganic polyphosphate J Biol Chem 2003 278 26788 26792 12740373 9 Hernandez-Ruiz L Gonzalez-Garcia I Castro C Brieva JA Ruiz FA Inorganic polyphosphate and specific induction of apoptosis in human plasma cells Haematologica 2006 91 1180 1186 16956816 10 Bae JS Lee W Rezaie AR Polyphosphate elicits pro-inflammatory responses that are counteracted by activated protein C in both cellular and animal models J Thromb Haemost 2012 10 1145 1151 22372856 11 Dinarvand P Hassanian SM Qureshi SH Manithody C Eissenberg JC Yang L Rezaie AR Polyphosphate amplifies proinflammatory responses of nuclear proteins through interaction with receptor for advanced glycation end products and P2Y1 purinergic receptor Blood 2014 123 935 945 24255918 12 Hassanian SM Dinarvand P Smith SA Rezaie AR Inorganic polyphosphate elicits proinflammatory responses through activation of mTOR complexes 1 and 2 in vascular endothelial cells J Thromb Haemost 2015 5 860 871 13 Wang L Fraley CD Faridi J Kornberg A Roth RA Inorganic polyphosphate stimulates mammalian TOR, a kinase involved in the proliferation of mammary cancer cells Proc Natl Acad Sci (USA) 2003 100 11249 11254 12970465 14 Inoki K Ouyang H Zhu T Lindvall C Wang Y Zhang X Yang Q TSC2 integrates Wnt and energy signals via a coordinated phosphorylation by AMPK and GSK3 to regulate cell growth Cell 2006 126 955 968 16959574 15 Vigneron F Dos Santos P Lemoine S Bonnet M Tariosse L Couffinhal T Duplaà C Jaspard-Vinassa B GSK-3β at the crossroads in the signalling of heart preconditioning: implication of mTOR and Wnt pathways Cardiovasc Res 2011 1 49 56 16 Clevers H Wnt/beta-catenin signaling in development and disease Cell 2006 3 469 480 17 Behrens J von Kries JP Kuhl M Bruhn L Wedlich D Grosschedl R Birchmeier W Functional interaction of beta-catenin with the transcription factor LEF-1 Nature 1996 382 638 642 8757136 18 Takahashi-Yanaga F Sasaguri T GSK-3beta regulates cyclin D1 expression: a new target for chemotherapy Cell Signal 2007 4 581 589 19 Baldin VJ Lukas MJ Marcote MP Draetta G Cyclin D1 is a nuclear protein required for cell cycle progression in G1 Genes & Dev 1993 7 812 821 8491378 20 Diehl JA Cheng M Roussel MF Sherr CJ Glycogen synthase kinase-3β regulates cyclin D1 proteolysis and subcellular localization Genes & Dev 1998 15 3499 3511 21 Choi SH Smith SA Morrissey JH Polyphosphate is a cofactor for activation of factor XI by thrombin Blood 2011 118 6963 6970 21976677 22 Hassanian SM Dinarvand P Rezaie AR Adenosine regulates the proinflammatory signaling function of thrombin in endothelial Cells J Cell Physiol 2014 229 1292 1300 24477600 23 Soutar MP Kim WY Williamson R Peggie M Hastie CJ McLauchlan H Snider WD Gordon-Weeks PR Sutherland C Evidence that glycogen synthase kinase-3 isoforms have distinct substrate preference in the brain Neurochem 2010 114 974 983 24 Cross DA Alessi DR Cohen P Andjelkovich M Hemmings BA Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B Nature 1995 378 785 789 8524413 25 Li B Ryder J Su Y Zhou Y Liu F Ni B FRAT1 peptide decreases Abeta production in swAPP(751) cells FEBS Lett 2003 553 347 350 14572648 26 Hahne G Grossmann TN Direct targeting of β-catenin: Inhibition of protein-protein interactions for the inactivation of Wnt signaling Bioorg Med Chem 2013 21 4020 4026 23566764 27 Holmström KM Marina N Baev AY Wood NW Gourine AV Abramov AY Signalling properties of inorganic polyphosphate in the mammalian brain Nat Commun 2013 4 1362 23322050 28 Choo AY Yoon SO Kim SG Roux PP Blenis J Rapamycin differentially inhibits S6Ks and 4E-BP1 to mediate cell-type-specific repression of mRNA translation Proc Natl Acad Sci (USA) 2008 105 17414 17419 18955708 29 Sarbassov DD Ali SM Sengupta S Sheen JH Hsu PP Bagley AF Markhard AL Sabatini DM Prolonged rapamycin treatment inhibits mTORC2 assembly and Akt/PKB Mol Cell 2006 22 159 168 16603397 30 Laplante M Sabatini DM mTOR signaling at a glance J Cell Sci 2009 122 3589 3594 19812304 31 Werner TP Amrhein N Freimoser FM Specific localization of inorganic polyphosphate (polyP) in fungal cell walls by selective extraction and immunohistochemistry Fungal Genet Biol 2007 44 845 852 17320430 32 Fritz G RAGE: a single receptor fits multiple ligands Trends Biochem Sci 2011 36 625 632 22019011 33 Lotze MT Tracey KJ High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal Nat Rev Immunol 2005 5 331 342 15803152 34 Lata K Mukherjee TK Knockdown of receptor for advanced glycation end products attenuate 17α-ethinyl-estradiol dependent proliferation and survival of MCF-7 breast cancer cells Biochim Biophys Acta 2014 1840 1083 1091 24252278 35 Radia AM Yaser AM Ma X Zhang J Yang C Dong Q Rong P Ye B Liu S Wang W Specific siRNA targeting receptor for advanced glycation end products (RAGE) decreases proliferation in human breast cancer cell lines Int J Mol Sci 2013 14 7959 7978 23579957 36 Malik P Chaudhry N Mittal R Mukherjee TK Role of receptor for advanced glycation end products in the complication and progression of various types of cancers Biochim Biophys Acta 2015 1850 1898 1904 26028296 37 Kuhla A Ludwig SC Kuhla B Münch G Vollmar B Advanced glycation end products are mitogenic signals and trigger cell cycle reentry of neurons in Alzheimer's disease brain Neurobiol Aging 2015 36 753 761 25448604 38 Kang R Livesey KM Zeh HJ 3rd Lotze MT Tang D Metabolic regulation by HMGB1-mediated autophagy and mitophagy Autophagy 2011 7 1256 1258 21691146 39 Verma N Manna SK Advanced glycation end products (AGE) potently induce autophagy through activation of RAF kinase and nuclear factor kB (NF-kB) J Biol Chem 2016 291 1481 1491 26586913 40 Su Z Wang T Zhu H Zhang P Han R Liu Y Ni P Shen H Xu W Xu H HMGB1 modulates Lewis cell autophagy and promotes cell survival via RAGE-HMGB1-Erk1/2 positive feedback during nutrient depletion Immunobiology 2015 220 539 544 25578401 41 Xie J Méndez JD Méndez-Valenzuela V Aguilar-Hernández MM Cellular signalling of the receptor for advanced glycation end products (RAGE) Cell Signal 2013 25 2185 2197 23838007 42 Bae JS Rezaie AR Activated protein C inhibits high mobility group box 1 signaling in endothelial cells Blood 2011 118 3952 3959 21849480 43 Jimenez-Nuñez MD Moreno-Sanchez D Hernandez-Ruiz L Benítez-Rondán A Ramos-Amaya A Rodríguez-Bayona B Medina F Brieva JA Ruiz FA Myeloma cells contain high levels of inorganic polyphosphate which is associated with nucleolar transcription Haematologica 2012 97 1264 1271 22315501 44 Dedkova EN Blatter LA Role of β-hydroxybutyrate, its polymer poly-β-hydroxybutyrate and inorganic polyphosphate in mammalian health and disease Front Physiol 2014 5 260 25101001 45 Reya T Clevers H Wnt signalling in stem cells and cancer Nature 2005 434 843 850 15829953 46 Gay LJ Felding-Habermann B Contribution of platelets to tumour metastasis Nat Rev Cancer 2011 11 123 134 21258396 47 Qiao M Sheng S Pardee AB Metastasis and AKT activation Cell Cycle 2008 7 2991 2996 18818526 48 Rosenwald IB Kaspar R Rousseau D Gehrke L Leboulch P Chen JJ Schmidt EV Sonenberg N London IM Eukaryotic translation initiation factor 4E regulates expression of cyclin D1 at transcriptional and post-transcriptional levels J Biol Chem 1995 270 21176 21180 7673150 49 Rosenwald IB Lazaris-Karatzas A Sonenberg N Schmidt EV Elevated levels of cyclin D1 protein in response to increased expression of eukaryotic initiation factor 4E Mol Cell Biol 1993 13 7358 7363 8246956 50 Lazaris-Karatzas A Montine KS Sonenberg N Malignant transformation by a eukaryotic initiation factor subunit that binds to mRNA 5' cap Nature 1990 345 544 547 2348862

PMC005xxxxxx/PMC5116248.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101240003 32254 Semin Fetal Neonatal Med Semin Fetal Neonatal Med Seminars in fetal & neonatal medicine 1744-165X 1878-0946 27343151 5116248 10.1016/j.siny.2016.06.001 NIHMS797974 Article Necrotizing enterocolitis and preterm infant gut bacteria Warner Barbara B. a Tarr Phillip I. b a Fetal Care Center, Division of Newborn Medicine, Washington University School of Medicine, St Louis, MO, USA b Division of Gastroenterology, Hepatology, Nutrition; Pathobiology Research Unit; Department of Pediatrics, Washington University School of Medicine, St Louis, MO, USA * Corresponding author Address: Fetal Care Center, Division of Newborn Medicine, Washington, University School of Medicine, Campus Box 8116, 660 S, Euclid, St Louis, MO 63110, USA, Tel: +01 314 454 2531. warner_b@kids.wustl.edu (B.B. Warner) 29 6 2016 22 6 2016 12 2016 01 12 2017 21 6 394399 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Necrotizing enterocolitis remains an intractable consequence of preterm birth. Gut microbial communities, especially bacterial communities, have long been suspected to play a role in the development of necrotizing enterocolitis. Direct-from-stool nucleic acid sequencing technology now offers insights into the make-up of these communities. Data are now converging on the roles of Gram-negative bacteria as causative agents, despite the dynamic nature of bacterial populations, the varying technologies and sampling strategies, and the overall small sample sizes in these case–control studies. Bacteria that confer protection from necrotizing enterocolitis have not been identified across studies. The beneficial effect of probiotics is not apparent in infants with birth weights <1000 g (these infants are at highest risk of, and have the highest case fatality rate from, necrotizing enterocolitis). Further work should be directed to the modulating gut microbes, or the products they produce, to prevent this devastating complication of preterm birth. Necrotizing enterocolitis Gammaproteobacteria 1. Introduction The newborn gut microbiome is an area of intense and growing interest in perinatology. There is emerging appreciation of the roles played by gut microbes in intestinal health, and, indeed, in lifelong health. Most relevant to preterm infants in neonatal intensive care units (NICUs), several very important disorders are likely to originate from either abnormal proportions of microbial content (dysbiosis), or when a vulnerable host encounters a specific pathogen. In this review, we focus on early-in-life bacterial population assembly in the preterm infant gut, recent data on the biology and ecology of the bacterial community, and the role these microbes play in the development of necrotizing enterocolitis (NEC). Experimental data are mentioned as examples that corroborate or extend observations from the human host. 2. The in-population of the human infant gut by microbes Classic teaching holds that the human gut (i.e., the meconium) contains no bacteria, or at least no viable bacteria, at birth. However, recent data prompt reconsideration of this dogma. Mshvildadze et al. [1] identified bacterial sequences in freshly produced meconium, as have Heida et al. [2], and Ardissone et al. [3]. Stout et al. [4] have identified bacterial bodies on electron microscopy in the basal plate of placentas delivered via cesarean section. Aagaard et al. [5] reported bacterial 16S and metagenomic sequences in placentas, finding similarities between these sequences and those of bacteria resident in the mouth. Additional studies have identified bacterial sequences in amniotic fluid from term pregnancies delivered by elective cesarean section [6]. Notably, these papers rarely present evidence of viable bacteria in specimens putatively colonized based on nucleic acid sequencing (Table 1). The possibility that the fetal gut is colonized by bacteria before maternal membranes rupture is intriguing. However, low grade bacteremia occurs independent of pregnancy in healthy adults after brushing or flossing teeth, or defecation [11]. The finding of bacterial sequences in or on a newborn infant immediately after rupture of membranes may reflect colonization of the newborn while passing through the birth canal or being delivered through the skin. These sequences may also reflect nucleic acid remnants of viable bacteria that circulated in the mother's blood, but which have no replicative potential or relevance to the assembly of the earliest-in-life members of the bacterial community of the infant gut. An additional argument against prenatal colonization of the gut with bacteria acquired in utero is the observation that germ-free animals, which generally require immersion in iodine solution during the derivation process, are generated from mothers (usually mice) who are not free of germs. Iodine immersion would not sterilize the colonized gut, so if the gut is an intergenerational habitat for viable bacteria, we would expect that it would be impossible to derive germ-free animals. Recent publications illuminate rules of assembly for the human infant gut microbiome. Among healthy term and near-term infants, early gut colonization patterns are driven largely by delivery route and feeding patterns [12,13], with emerging data suggesting a role for the gut virome [10]. Infants born very preterm (<32 weeks), however, have a distinct set of exposures compared to those born near term. Parenteral antibiotic use is nearly universal during the first several days of life in preterm infants, feeding tubes are placed early, and enteral nutrition is commenced cautiously. In the days and weeks after birth, preterm infants reside almost exclusively in NICUs. These environments are designed to limit microbial transmission, and contact with bacteria is controlled to the extent possible. Visitors are restricted and often only parents and professional staff are permitted to touch the infant. Hand hygiene is stressed, line care is protocolized, and nutrition is either human breast milk (mother of infant or pasteurized donor pool), or sterilized liquid formula. There is no exposure to pets, or physical contact with other relatives. This microbiologically constrained biosphere offers a rich opportunity to study the transition of the neonatal gut from sterility or near sterility at birth to an organ that houses, for the rest of the life of the host, the greatest density of microbes in the human body. Not unexpectedly, bacterial communities assemble differently in the preterm gut than in the term infant gut. Whereas the in-population of the infant gut with bacteria is a fascinating ecologic event worthy of study in its own right, accumulating experimental data suggest that the earliest-in-life gut bacteria affect the future well-being of their hosts [14]. Hence, the study of these communities in infants is justifiable in order to determine whether the animal data are relevant for humans. However, pertinent to infants born very prematurely, there are additional compelling reasons to study the gut microbiome because of the high frequency with which these infants experience complications of premature birth that are plausibly associated with this biomass. The two most dire consequences in which gut bacteria could play major roles in outcome are NEC and late-onset neonatal bloodstream infections (P.I. Tarr and B.B. Warner, Chapter 4, this issue). The “normal” preterm infant gut microbiome has been characterized among infants born very preterm who were at risk of developing NEC, but who did not experience this event, i.e., controls. Until recently, these analyses used culture-based technology, or polymerase chain reaction amplification of DNA extracted from stool and testing for mobility in a gel. Most recently, advances in sequencing technology, expansion of ribosomal RNA gene databases, and metagenomic capabilities (DNA sequencing not confined to 16S rRNA gene regions) have made feasible the direct-from-stool amplification of extracted bacterial DNA. These approaches provide a less circuitous, and deeper and more economical, portrayal of bacterial populations in polymicrobial substances. In the targeted approach, conserved regions of the 16S ribosomal RNA gene of bacteria are primed and amplified, a technique employed in the NIH-sponsored Human Microbiome Project [15]. This targeted approach enables deep “censusing” of bacterial populations, as all such mass readouts are confined to the regions of the bacterial chromosome that identify the organism from which they are derived. We are aware of six publications from NICUs in eight different centers in which bacterial community assembly in “normal” preterm infants has been interrogated in depth using direct-from-stool sequencing. For the purposes of this review, our criteria for including such studies are those that included at least 100 stools from at least 25 subjects who did not develop NEC, and that the enumeration technology employed 16S rRNA gene or metagenomic sequencing (Table 2). Even though only one of the papers in Table 2 exclusively focused on defining the pattern of progression in children without NEC [20], data supplied in the others [16,18,21] were sufficient to confirm the findings of La Rosa et al. [20]. In that comprehensive study of preterm infants, 16S rRNA gene sequencing demonstrated a remarkably choreographed pattern: namely, the early-in-life gut bacterial content is predominated by Bacilli (despite their name, Bacilli are Gram-positive cocci such as staphylococci, streptococci, and enterococci). Bacilli are soon overtaken by Gram-negative facultative organisms (a diversity of genera and species within the Gammaproteobacteria class). This surge in Gammaproteobacteria is counterbalanced by a gradually increasing abundance of Clostridia (many genera and species) and Negativicutes (predominantly Veillonella). Overall, four bacterial classes (Bacilli, Gammaproteobacteria, Clostridia, and Negativicutes) account for >90% of the taxa present. Compared to the gut content of older children and adults, these preterm infant gut bacterial populations have much higher content of Gammaproteobacteria (one to two orders of magnitude difference), and approximately half the density of obligate anaerobes. There is a convergence on a consensus community structure by the equivalence of 33–36 weeks postmenstrual age (the sum of gestational age at birth, and day of life on which the sample was obtained). The content of this consensus community at this postmenstrual age (but not earlier) is independent of gestational age at birth. In particular, anaerobic bacteria gain abundance more rapidly in the gastrointestinal tracts of infants born least prematurely. This choreographed progression is punctuated unpredictably and substantially by short-lived changes in composition, before the communities self-revert to the choreographed progression. Such abrupt changes have been noted in older children and adults [12,13,15,22]. Unexpectedly, the factors believed to be influential in microbial community assembly (at least in children born after full-term gestation), namely mode of delivery (vaginal vs cesarean section), antibiotic administration in the aggregate, and feeds (breast milk), were either not determinative of bacterial content, or had only minimal or temporary influence on this progression. These non-associations between diverse exposures, each of which could logically be considered to influence bacterial community structure in the gut, prompts us to interpret that in the preterm human infant, the major driver of bacterial population assembly is intrinsic host biology or succession ecology rather than exogenous factors. However, we offer two caveats. First, as described above, the microbial exposures of very preterm infants differ considerably from those of infants born at term, in whom mode of delivery and breast-milk feeding appear to influence gut bacterial population assembly. Second, we wish to note that in a subsequent study of the St Louis Children's Hospital cohort [19], specific antibiotics (meropenem, cefotaxime, and ticarcillin–clavulanate), which were used in few subjects in La Rosa et al. [20], were associated with substantial directional changes in microbial content. It is also noteworthy that when metagenomic sequencing technology was applied [19], bacterial community population characteristics as previously defined by 16S rRNA gene sequencing were recapitulated [20]. 3. The gut microbiome and NEC Multiple lines of circumstantial evidence suggest that NEC is influenced by bacteria in the very preterm infant gut. Most notably, NEC does not occur in utero, and, in fact, rarely occurs before approximately day of life 10, after bacterial populations start to assemble in the newborn gut. Also, NEC is statistically and independently associated with increased antibiotic use, especially prolongation of antibiotics during the first week of life [23–25]. Moreover, H2 blockers, which could affect gut microbial populations by reducing the gastric acid line of defence against bacterial colonization, are associated with increased NEC risk [26]. Multiple studies suggest that probiotics prevent NEC (reviewed in [27]), but this benefit accrues chiefly to infants who weigh <1000 g at birth, and who are at lesser risk of experiencing NEC, and of dying from NEC, than those whose birthweights are <1000 g. Indeed, a recent large and well-conducted multi-center randomized control trial of Bifidobacterium breve BBG-001 failed to demonstrate any protective effect against NEC in a population in which most of the children weighed ≤1000 g at birth [28]. Whereas this failure might represent the use of a single probiotic instead of a combination of probiotics, and though the choice of the probiotic intervention could be subject to debate, it seems unlikely that we will soon identify viable microbes that can exert a profoundly protective effect against NEC, especially among infants whose birth weights are ≤1000 g. A review of the many taxa that have been associated with NEC is beyond the scope of this article. However, the diversity of incriminated species, the overall small numbers of subjects in these studies, and small effect sizes reported (often only in subgroup analysis), cast doubt on the existence of a specific mono-microbial driver of NEC [29]. Nonetheless, as reviewed above, direct-from-stool sequencing of DNA now offers new opportunities to compare cases with NEC to controls, to determine whether microbial populations are associated with this outcome. In the past decade, multiple groups have attempted to apply direct-from-stool sequencing to identify bacteria that might cause NEC. Some such attempts are summarized in Table 3, focusing on studies that utilized 16S rRNA amplification methods rather than culture or gel electrophoresis-based methods. These studies suggest that diverse bacterial taxa are associated with either risk of, or protection from, developing necrotizing enterocolitis among preterm infants. One interpretation is that there are center-specific differences in microbial drivers of NEC, as described for variability in gut microbial populations before the onset of bloodstream infections [35]. An alternative explanation is that the population biology of bacteria in the gut is exceptionally dynamic in the interval during which NEC occurs, which obligates the assembly of exceptionally large cohorts to study this disorder, and the need to interrogate an abundance of specimens prior to the event. Indeed, only one of the studies in Table 3 reported the analysis of >100 pre-NEC specimens. The dynamism of bacterial populations poses immense challenges. In the first 60 days of life, as described above, there is a week-by-week aggregate progression from Bacilli to Gammaproteobacteria predominance, while Clostridia slowly rise in abundance. In reality, the Clostridia class described by La Rosa et al. contains Clostridia and Negativicutes, because Negativicutes (Gram-negative obligate anaerobic bacteria) have recently been assigned their own class [20]. NEC generally does not present until after the second week of life, and risk extends to approximately day of life 60, with infants born most prematurely developing NEC later in this period of vulnerability [29]. To illustrate this challenge, stools from a case occurring on day of life 25 would ideally be compared to stools from a control group of infants produced on day of life 25, these controls having been born after the same gestational duration. However, control specimens for a case of NEC that occurs on day of life 45 would greatly differ in content from controls chosen on day of life 25, even if controlling for gestational age at birth. That is to say, the norm changes throughout the interval of risk, during which NEC can occur at any time. When one also takes into account the additional abrupt changes in populations, it is clear that a substantial number of subjects and specimens must be assembled to characterize the microbial population in children at risk in a case–control study. Notably, the larger studies tend to lean towards a predominance of Gram-negative bacteria as being drivers of NEC. Consensus protective organisms have generally not emerged from these large studies. A final complicating note is that specimens obtained immediately before NEC is apparent may reflect changes of NEC that are already under way before infants become visibly affected. It therefore seems prudent to “censor” sequence data from the hours preceding the onset of clinical NEC if trying to identify signatures well in advance of NEC that could be associated with this disorder. Interestingly, in one study in which specimens were analyzed late (tissue at resection) [30] or early (meconium) [2], anaerobic bacteria were associated with NEC. In the largest study (in terms of numbers of cases and numbers of pre-NEC stools analyzed) reported to date, an overrepresentation of Gammaproteobacteria was associated with NEC, whereas anaerobic bacteria, especially Negativicutes and secondarily Clostridia, were associated with control status (i.e., protection). Gammaproteobacteria risk has been suggested in several smaller studies [16,18,34]. In contrast, however, several publications employing direct-from-stool sequencing have not identified overabundant Gram-negative bacteria as a prelude to NEC [2,33]. Indirect data support that Gram-negative bacteria are causal in NEC pathogenesis. In animal models, toll-like receptor 4, the ligand for lipopolysaccharide, is believed to play a central role in mucosal injury [36], and antibiotics active against Gram-negative bacteria confer protection [37]. Moreover, anaerobic bacteria, in response to microbiota-accessible carbohydrates, generate anti-inflammatory short-chain fatty acids, notably acetate, propionate, and butyrate [38]. Several literature reviews [39,40] have evaluated studies in which infants were administered oral aminoglycosides in attempts to prevent NEC. Aminoglycosides would be active against Gammaproteobacteria in the gut, but not suppress anaerobic bacterial populations. In the aggregate, these studies [41–44] support the use of oral aminoglycosides to prevent NEC. However, because of concerns about selecting for aminoglycoside resistance [45] and of absorption of the oral aminoglycosides from the gut (albeit confined to very early in life before the incidence of NEC increases [46]), enteral antibiotics to prevent NEC are not widely used. It is interesting to note that the oral aminoglycosides were often discontinued in these studies before the time of life at which the most premature infants develop NEC. This timing raises the possibility that the beneficial effects of antibiotics in these studies might have been understated. Bacterial diversity – defined as the number of different taxa present, weighed according to their proportionality – is considered to reflect a healthy luminal microbial community in inflammatory bowel disease and C. difficile infections [47,48]. Even before these associations between lack of diversity and gut inflammation were reported, Claud and Walker proposed the hypothesis that diminished diversity of the premature infant gut could result in NEC [49]. Subsequent studies failed to find an association between lack of bacterial diversity and development of NEC [16–18,31,33,34,50], though again, as for NEC microbial associations, the numbers were limited. However, in a recent study [21], an association between subsequent development of NEC and comparatively lower gut bacterial diversity was noted. The difference appeared to be related to delayed or suppressed maturation of microbial diversity in infants who subsequently developed NEC, compared to those who did not. In other words, diversity slowly increased over the first 60 days of life in the controls but not the cases. However, this association is not straightforward: gut bacterial communities are exceptionally non-complex in preterm infants. Therefore, a change in the proportionality of one taxon is necessarily counterbalanced by a change in one or more of the few other taxa present. With only four dominant taxonomic “degrees of freedom,” it is difficult to attribute NEC to lack of gut bacterial diversity per se, versus an increase or a decrease in one or another taxon. In other words, it cannot be stated that lack of diversity is the driver of risk for NEC, versus an overrepresentation of Gammaproteobacteria, which directly ordains the lack of diversity in these sample sets. The role of bacterial diversity in protecting from NEC remains an intriguing hypothesis, however. 4. Conclusion NEC remains a catastrophic disorder. It is concerning that we have not had meaningful and durable improvements in incidence or outcomes of NEC in the nearly four decades since widespread recognition of this entity permeated neonatology. The finding of a microbial signature prior to development of NEC, and/or a protective signature in the form of obligate anaerobic bacteria, now offers new opportunities to prevent this devastating consequence of preterm birth. Funding sources: This work was supported by NIH Grants UH3AI083265, P30DK052574 (for the Biobank Core) and the Children's Discovery Institute of Washington University and the St Louis Children's Hospital. We are grateful to Ms Maida Redzic for assistance with manuscript preparation. Table 1 Data in support of prenatal colonization of the new-born gut with microbes. Study Subject of study Comments Jimenez et al. [7] Cord blood cultures of term neonates born by elective cesarean section Enterococci, streptococci, staphylococci, and propionibacterium recovered from cord blood Mshvildadze et al. [1] Meconium by 16S rRNA gene sequencing Viable bacteria not sought DiGiulio et al. [8] Amniotic fluid cultures and 16S rRNA gene sequencing in preterm deliveries 16S rRNA gene sequences identified in, and Mycoplasma hominis, Ureaplasma sp., Streptococcus agalactiae, Lactobacillus sp., Prevotella sp., Fusobacterium nucleatum, coagulase-negative Staphylococcus sp., Bacillus sp. (not anthrax), Peptostreptococcus sp., and Gardnerella vaginalis recovered from the amniotic fluid Rautava et al. [9] Bacterial DNA detected in amniotic fluid at time of elective cesarean section by 16S rRNA gene sequencing Viable bacteria not sought Stout et al. [4] Electron microscopy of placenta Bacteria identified in basal plate, no attempt to culture Aagaard et al. [5] 16S rRNA gene sequences and metagenomic sequences Bacterial sequences identified and reflected periodontal microbes, no attempt to culture Lim et al. [10] First in life stool (days 1–4) subjected to 16S rRNA gene sequencing and virome analysis Few bacterial species, many bacteriophages, based on sequence analysis Table 2 Studies of gut bacterial assembly in preterm infants without necrotizing enterocolitis (NEC). Study and location Sequencing technology No. Of subjects without NEC No. of specimens Conclusions about pattern of bacterial community assembly Zhou et al. [16], Brigham and Women's Hospital, Boston, MA, USA 16S rRNA gene sequencing 26 111 Increasing proportion of Clostridia over time, balanced by slowly diminishing proportion of Gram-negative genera, with little effect of antibiotics on this trend Ward et al. [17], Cincinnati, OH, and Birmingham, AL, USA Metagenomic sequencing 89 185 Clostridia class increases over time (specifically veillonella and C. freundii), with consistently high Proteobacteria (specifically E. coli) Shaw et al. [18], St Mary's Hospital, Queen Charlotte's and Chelsea Hospital, London,UKa 16S rRNA gene sequencing 44 369 Bifidobacteria and klebsiella increased in proportionality, and Gram-positive bacteria decreased in proportionality, over time Gibson et al. [19], St Louis Children's Hospital, St Louis, MO, USA Metagenomic sequencing 84 401 Some of these subjects and specimens were also analyzed in La Rosa et al. [21]. Notably, metagenomic sequencing recapitulated the 16S sequence analysis of this cohort in these two companion publications. La Rosa et al. [20], St Louis Children's Hospital, St Louis, MO, USA 16S rRNA Gene sequencing 58 922 Bacterial classes proceed from Bacilli to Gammaproteobacteria to Clostridia in these infants, but these populations are prone to changes in content. When infants near 33–36 weeks postconceptional age (i.e., an interval that is equivalent to the 3rd to the 12th week of age, in view of the wide range of gestational ages in this cohort), the populations converge on a consensus community, with ∼40% of the bacteria being obligate anaerobes (especially Clostridia and Negativicutes), and an equal percentage being Gammaproteobacteria. There was little or no effect of use of postnatal antibiotics, mode of delivery, or breast milk, and the community composition at this point. Warner et al. [21], St Louis Children's Hospital, St Louis, MO; Children's Hospital at Oklahoma University, Oklahoma City, OK; Kosair Children's Hospital, Louisville, KY, USA 16S rRNA gene sequencing 120 2720 Includes the 58 subjects without NEC and their 922 stools in La Rosa et al. [21]. Patterns in NICUs in Oklahoma City and in Louisville recapitulate those in St Louis cohort NICU, neonatal intensive care unit. a Based on data from Supplemental Table 1 in [18]. Table 3 Summary of studies in which DNA sequences obtained directly from stools have been used to associate bacterial risk and development of necrotizing enterocolitis (NEC), listed in ascending order according to number of pre-NEC stools sequenced. Study Sequencing technology No. of subjects without NEC No. of specimens from subjects without NEC No. of subjects with NEC No. of pre-NEC specimens from subjects who subsequently developed NEC Comments Brower-Sinning et al. [30], Pittsburgh, PA, USA 16S rRNA gene sequencing 10 10 9 9 Tissue analysis only, no pre-NEC samples; Proteobacteria, Clostridia associated with risk, as was diminished bacterial diversity Mai et al. [31], three University of Florida-affiliated NICUs, FL, USA 16S rRNA gene sequencing 9 18 9 18 Case stools demonstrated an increase in Proteobacteria, and a decrease in Firmicutes in the second of the paired samples (i.e., week before NEC) McMurtry et al. [32], Louisiana State University Health Sciences Center, Touro Infirmary, East Jefferson General Hospital and Children's Hospital of New Orleans, LA, USA 16S rRNA gene sequencing 74 74 21 21 Bacterial diversity and relative abundance of Clostridia was significantly lower in NEC specimens compared to controls Raveh-Sadka et al. [33] Pittsburgh, PA, USA Metagenomic sequencing 5 34 5 21 No clear association between bacterial content as identified by metagenomics and outcome; no microbiologic evidence of time–space clustering Heida et al. [2], Groningen, The Netherlands 16S rRNA gene sequencing of meconium and subsequent stools 22 57 11 30 Clostridium perfringens and Bacteroides dorei associated with risk, and staphylococi associated with protection. Torrazza et al. [34], Gainseville, FL, USA 16S rRNA gene sequencing 35 77 18 40 Novel sequence matching closest to Klebsiella pneumoniae during week 1 associated with subsequent development of NEC Ward et al. [17], Cincinnati, OH, and Birmingham, AL, USA Metagenomic sequencing 89 185 27 60 Specific sequence types of E. coli associated with NEC Sim et al. [18], St Mary's Hospital, Queen Charlotte's and Chelsea Hospital, London, UK 16S rRNA gene sequencing 44 369 22 88 Klebsiella, clostridium associated with risk; no microbiologic evidence of time–space clustering Zhou et al. [16], Brigham and Women's Hospital, Boston, MA, USA 16S rRNA gene sequencing 26 111 10 88 Age-specific differences identified, with early- and late-onset NEC having an association with Clostridia and Gammaproteobacteria, respectively Warner et al. [23], St Louis Children's Hospital, St Louis, MO; Kosair Children's Hospital, Louisville, KY; Children's Hospital at Oklahoma University, Oklahoma City, OK, USA 16S rRNA gene sequencing 120 2720 46 866 Gammaproteobacteria associated with risk, and Negativicutes associated with protection; lack of diversity is associated with risk NICU, neonatal intensive care unit. Practice points The causes of NEC are unknown. Judicious use of antibiotics and promotion of human milk use might lower the risk of NEC, but these interventions are unlikely to categorically reduce disease incidence, and are justifiable for multiple additional reasons. Research directions How can we anticipate the microbial community changes that lead to NEC? How can we modulate the gut microbial community to reduce bacteria-associated processesthat might lead to NEC? Conflict of interest statement: None declared. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Mshvildadze M Neu J Shuster J Theriaque D Li N Mai V Intestinal microbial ecology in premature infants assessed with non-culture-based techniques J Pediatr 2010 156 20 5 19783002 2 Heida FH van Zoonen AG Hulscher JB A necrotizing enterocolitis-associated gut microbiota is present in the meconium: results of a prospective study Clin Infect Dis 2016 62 863 70 26787171 3 Ardissone AN de la Cruz DM Davis-Richardson AG Meconium microbiome analysis identifies bacteria correlated with premature birth PloS One 2014 9 e90784 24614698 4 Stout MJ Conlon B Landeau M Identification of intracellular bacteria in the basal plate of the human placenta in term and preterm gestations Am J Obstet Gynecol 2013 208 226 e1 7 23333552 5 Aagaard K Ma J Antony KM Ganu R Petrosino J Versalovic J The placenta harbors a unique microbiome Sci Transl Med 2014 6 237ra65 6 Collado MC Rautava S Aakko J Isolauri E Salminen S Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid Sci Rep 2016 6 23129 27001291 7 Jimenez E Fernandez L Marin ML Isolation of commensal bacteria from umbilical cord blood of healthy neonates born by cesarean section Curr Microbiol 2005 51 270 4 16187156 8 DiGiulio DB Romero R Amogan HP Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation PloS One 2008 3 e3056 18725970 9 Rautava S Collado MC Salminen S Isolauri E Probiotics modulate host–microbe interaction in the placenta and fetal gut: a randomized, double-blind, placebo-controlled trial Neonatology 2012 102 178 84 22776980 10 Lim ES Zhou Y Zhao G Early life dynamics of the human gut virome and bacterial microbiome in infants Nature Med 2015 21 1228 34 26366711 11 Nikkari S McLaughlin IJ Bi W Dodge DE Relman DA Does blood of healthy subjects contain bacterial ribosomal DNA? J Clin Microbiol 2001 39 1956 9 11326021 12 Backhed F Roswall J Peng Y Dynamics and stabilization of the human gut microbiome during the first year of life Cell Host Microbe 2015 17 690 703 25974306 13 Palmer C Bik EM DiGiulio DB Relman DA Brown PO Development of the human infant intestinal microbiota PLoS Biol 2007 5 e177 17594176 14 Arrieta MC Stiemsma LT Amenyogbe N Brown EM Finlay B The intestinal microbiome in early life: health and disease Front Immunol 2014 5 427 25250028 15 Human Microbiome Project C Structure, function and diversity of the healthy human microbiome Nature 2012 486 7402 207 14 22699609 16 Zhou Y Shan G Sodergren E Weinstock G Walker WA Gregory KE Longitudinal analysis of the premature infant intestinal microbiome prior to necrotizing enterocolitis: a case–control study PloS One 2015 10 e0118632 25741698 17 Ward DV Scholz M Zolfo M Metagenomic sequencing with strain-level resolution implicates uropathogenic E. coli in necrotizing enterocolitis and mortality in preterm infants Cell Rep 2016 14 2912 24 26997279 18 Sim K Shaw AG Randell P Dysbiosis anticipating necrotizing enterocolitis in very premature infants Clin Infect Dis 2015 60 389 97 25344536 19 Gibson MK Wang B Ahmadi S Developmental dynamics of the preterm infant gut microbiota and antibiotic resistome Nature Microbiol in press 20 La Rosa PS Warner BB Zhou Y Patterned progression of bacterial populations in the premature infant gut Proc Natl Acad Sci USA 2014 111 12522 7 25114261 21 Warner BB Deych E Zhou Y Gut bacteria dysbiosis and necrotising enterocolitis in very low birthweight infants: a prospective case–control study Lancet 2016 387 10031 1928 36 26969089 22 Costello EK Lauber CL Hamady M Fierer N Gordon JI Knight R Bacterial community variation in human body habitats across space and time Science 2009 326 5960 1694 7 19892944 23 Cotten CM Taylor S Stoll B Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants Pediatrics 2009 123 58 66 19117861 24 Kuppala VS Meinzen-Derr J Morrow AL Schibler KR Prolonged initial empirical antibiotic treatment is associated with adverse outcomes in premature infants J Pediatr 2011 159 720 5 21784435 25 Alexander VN Northrup V Bizzarro MJ Antibiotic exposure in the newborn intensive care unit and the risk of necrotizing enterocolitis J Pediatr 2011 159 392 7 21489560 26 Guillet R Stoll BJ Cotten CM Association of H2-blocker therapy and higher incidence of necrotizing enterocolitis in very low birth weight infants Pediatrics 2006 117 e137 42 16390920 27 AlFaleh K Anabrees J Probiotics for prevention of necrotizing enterocolitis in preterm infants Cochrane Database Syst Rev 2014 4 CD005496 24723255 28 Costeloe K Hardy P Juszczak E Wilks M Millar MR Probiotics in Preterm Infants Study Collaborative Group Bifidobacterium breve BBG-001 in very preterm infants: a randomised controlled phase 3 trial Lancet 2016 387 10019 649 60 26628328 29 Neu J Walker WA Necrotizing enterocolitis N Engl J Med 2011 364 255 64 21247316 30 Brower-Sinning R Zhong D Good M Mucosa-associated bacterial diversity in necrotizing enterocolitis PloS One 2014 9 e105046 25203729 31 Mai V Young CM Ukhanova M Fecal microbiota in premature infants prior to necrotizing enterocolitis PloS One 2011 6 e20647 21674011 32 McMurtry VE Gupta RW Tran L Bacterial diversity and Clostridia abundance decrease with increasing severity of necrotizing enterocolitis Microbiome 2015 3 11 25810906 33 Raveh-Sadka T Thomas BC Singh A Gut bacteria are rarely shared by co-hospitalized premature infants, regardless of necrotizing enterocolitis development eLife 2015 4 34 Torrazza RM Ukhanova M Wang X Intestinal microbial ecology and environmental factors affecting necrotizing enterocolitis PloS One 2013 8 e83304 24386174 35 Taft DH Ambalavanan N Schibler KR Center variation in intestinal microbiota prior to late-onset sepsis in preterm infants PloS One 2015 10 e0130604 26110908 36 Hackam DJ Afrazi A Good M Sodhi CP Innate immune signaling in the pathogenesis of necrotizing enterocolitis Clin Dev Immunol 2013 2013 475415 23762089 37 Jensen ML Thymann T Cilieborg MS Antibiotics modulate intestinal immunity and prevent necrotizing enterocolitis in preterm neonatal piglets Am J Physiol Gastrointest Liver Physiol 2014 306 G59 71 24157972 38 Smith PM Howitt MR Panikov N The microbial metabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis Science 2013 341 6145 569 73 23828891 39 Bury RG Tudehope D Enteral antibiotics for preventing necrotizing enterocolitis in low birthweight or preterm infants Cochrane Database Syst Rev 2001 1 CD000405 11279690 40 Bell EF Preventing necrotizing enterocolitis: what works and how safe? Pediatrics 2005 115 173 4 15629996 41 Rowley MP Dahlenburg GW Gentamicin in prophylaxis of neonatal necrotising enterocolitis Lancet 1978 2 8088 532 42 Grylack LJ Scanlon JW Oral gentamicin therapy in the prevention of neonatal necrotizing enterocolitis. A controlled double-blind trial Am J Dis Child 1978 132 1192 4 362900 43 Egan EA Mantilla G Nelson RM Eitzman DV A prospective controlled trial of oral kanamycin in the prevention of neonatal necrotizing enterocolitis J Pediatr 1976 89 467 70 784926 44 Boyle R Nelson JS Stonestreet BS Peter G Oh W Alterations in stool flora resulting from oral kanamycin prophylaxis of necrotizing enterocolitis J Pediatr 1978 93 857 61 361939 45 Grylack L Neugebauer D Scanlon JW Effects of oral antibiotics on stool flora and overall sensitivity patterns in an intensive care nursery Pediatr Res 1982 16 509 11 7050868 46 Grylack L Boehnert J Scanlon J Serum concentrations of gentamicin following oral administration to preterm newborns Dev Pharmacol Therapeut 1982 5 47 52 47 Wills ES Jonkers DM Savelkoul PH Masclee AA Pierik MJ Penders J Fecal microbial composition of ulcerative colitis and Crohn's disease patients in remission and subsequent exacerbation PloS One 2014 9 e90981 24608638 48 Chang JY Antonopoulos DA Kalra A Decreased diversity of the fecal microbiome in recurrent Clostridium difficile-associated diarrhea J Infect Dis 2008 197 435 8 18199029 49 Claud EC Walker WA Hypothesis: inappropriate colonization of the premature intestine can cause neonatal necrotizing enterocolitis FASEB J 2001 15 1398 1403 11387237 50 Morrow AL Lagomarcino AJ Schibler KR Early microbial and metabolomic signatures predict later onset of necrotizing enterocolitis in preterm infants Microbiome 2013 1 13 24450576

PMC005xxxxxx/PMC5116260.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985117R 4816 J Immunol J. Immunol. Journal of immunology (Baltimore, Md. : 1950) 0022-1767 1550-6606 27799309 5116260 10.4049/jimmunol.1600576 NIHMS821782 Article Lung Cancer Subtypes Generate Unique Immune Responses Busch Stephanie E. * Hanke Mark L. * Kargl Julia *‡ Metz Heather E. * MacPherson David † Houghton A. McGarry *†§ * Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA † Human Biology Division, Fred Hutchinson Cancer Research Center, Seattle, WA ‡ Institute of Experimental and Clinical Pharmacology, Medical University of Graz, Graz, Austria § Division of Pulmonary and Critical Care, University of Washington, Seattle, WA. Corresponding Author: A. McGarry Houghton, M.D., Clinical Research Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue N. (D4-100), Seattle, WA 98109, Phone: 206.667.3175, Fax: 206.667.5255, houghton@fhcrc.org 21 10 2016 31 10 2016 1 12 2016 01 12 2017 197 11 44934503 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Lung cancer – the leading cause of cancer-related deaths worldwide – is a heterogeneous disease comprised of multiple histologic subtypes that harbor disparate mutational profiles. Immune-based therapies have shown initial promise in the treatment of lung cancer patients but are currently limited by low overall response rates. We sought to determine whether the host immune response to lung cancer is predicated, at least in part, by histologic and genetic differences, as such correlations would have important clinical ramifications. Using mouse models of lung cancer, we show that small cell lung cancer (SCLC) and lung adenocarcinoma (ADCA) exhibit unique immune cell composition of the tumor microenvironment. The total amount of leukocyte content was markedly reduced in SCLC compared to lung ADCA, which was validated in human lung cancer specimens. We further identified key differences in immune cell content using three models of lung ADCA driven by mutations in Kras, Tp53, and Egfr. Although Egfr-mutant cancers displayed robust myeloid cell recruitment, they failed to mount a CD8+ immune response. In contrast, Kras-mutant tumors displayed significant expansion of multiple immune cell types, including CD8+ cells, regulatory T cells, IL17A-producing lymphocytes, and myeloid cells. A human tissue microarray annotated for KRAS and EGFR mutations validated the finding of reduced CD8+ content in human lung adenocarcinoma. Taken together, these findings establish a strong foundational knowledge of the immune cell contexture of lung ADCA and SCLC and suggest that molecular and histological traits shape the host immune response to cancer. INTRODUCTION Despite decades of research, small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) remain among the world's deadliest diseases (1). SCLC, in which RB1 and TP53 mutations are common (2), accounts for 10-20% of lung cancer diagnoses (3). Over half of NSCLC cases are classified as lung adenocarcinomas (ADCA), in which KRAS, EGFR, and TP53 mutations are the predominant genetic drivers (4, 5). Although patients with EGFR mutations initially respond to targeted therapies, drug resistance typically develops within the first year (6). SCLC and KRAS-mutant ADCA have proven less tractable, as scant progress has been made towards the development of targeted therapeutics for these patients (7). Novel immunotherapeutic strategies have offered new hope for the management of NSCLC (8-10) and SCLC (11), but the clinical success of immunomodulatory agents will depend on a strong foundational knowledge of the cells that comprise the lung tumor microenvironment (TME) in these molecularly and histologically distinct diseases. Inflammation is a key attribute of neoplasia (12). The host immune response to cancer is an intricate web of both pro- and anti-tumorigenic signals (13). CD8+ T cells, also known as cytotoxic T lymphocytes (CTL), are the body's main immunological barrier to cancer, as they are capable of recognizing and killing tumor cells. However, CTL activity is curbed by tumor cells expressing immune checkpoint ligands (e.g. PD-L1), as well as an influx of immune-suppressive cells (i.e. regulatory T cells, macrophages, monocytes, and neutrophils) into the TME (14, 15). Newly developed immune checkpoint inhibitors (ICI; e.g. ipilimumab and nivolumab) seek to reverse this suppression and unleash an anti-tumor response (16). Although some lung cancer patients have experienced remarkable tumor regression upon commencing ICI therapy, overall response rates have peaked around 20% (9, 10). These disparate outcomes may in part be explained by findings that tumor immune cell composition and function vary between anatomical sites and histological origins (17). In lung cancer, for example, squamous cell carcinoma (SCC) patients have exhibited longer progression-free survival with ipilimumab treatment than have ADCA patients (8). However, even within histological subgroups, response rates vary widely, raising the question of what other tumor attributes might predict clinical outcome. The oncogenic functions of mutant RAS and EGFR in cancer include the production of pro-inflammatory cytokines, such as IL8, that help to shape the TME (18-21). TP53 has similarly demonstrated non-cell-autonomous behaviors during tumorigenesis (22, 23). The discrete impact of molecular signatures, such as KRAS, EGFR, TP53, and RB1 mutations, on the immune cell composition of lung cancer nevertheless remains largely undefined. To address this question, we profiled the TME of three genetically engineered mouse (GEM) models of NSCLC – KrasLSL-G12D, KrasLSL-G12D;Trp53Fl/Fl, and EgfrL858R – as well as the Rb1Fl/Fl;Trp53Fl/Fl model of SCLC. Here we show that the molecular and histological subtypes of lung cancer predict immune cell composition and may, therefore, demand specific immunotherapeutic regimens. MATERIALS AND METHODS Mice All animal experiments utilized aged-matched mice on approved IACUC protocols at the Fred Hutchinson Cancer Research Center. TetO-EgfrL858R mice (24) were obtained from the Mouse Models of Human Cancer Consortium on C57BL/6 background. Ccsp-rtTA mice (25) on FVB background were provided by Jeff Whitsett (University of Cincinnati). KrasLSL-G12D (i.e. Kras) (26), Trp53Fl/Fl (p53) (27), RORγtGFP (RORγt) (28), and Tcrd−/− (Tgd) (29) mice were obtained from Jackson Labs on C57BL/6 background. TetO-EgfrL858R;Ccsp-rtTA (Egfr), KrasLSL-G12D;Trp53Fl/Fl (Kp53), KrasLSL-G12D;RORγtGFP (K.RORγt), and KrasLSL-G12D;Tcrd−/− (K.Tgd) mice were generated by simple cross-breeding. Rb1Fl/Fl mice (30) were crossed with Trp53Fl/Fl to generate Rb1Fl/Fl;Trp53Fl/Fl (Rbp53) mice on a mixed C57BL/6x129 background. Egfr and single-transgene control mice (Ccsp-rtTA or TetO-EgfrL858R) were fed 200 mg/kg doxycycline-impregnated food (Harlan, Indianapolis, IN, USA). Kras, Kp53, p53, and wild-type (wt) C57BL/6 animals received an intratracheal dose of 2.5×107 pfu Adenoviral Cre Recombinase (AdCre; University of Iowa Viral Vector Core, Iowa City, IA, USA), as described (31). Each cohort was studied over a time course of 6, 10, and 14-weeks post initiation of doxycycline or infection with AdCre. Additional cohorts of K.RORγt and K.Tgd animals were similarly subjected to AdCre infection (2.5×107 pfu) and examined 14-weeks post initiation or when moribund. RbFl/Fl;Trp53Fl/Fl mice received 1×108 pfu AdCre; given the long latency period, Rbp53 animals were studied 9 months post-induction. CTLA4 antibody, clone 9D9 (MedImmune) or isotype control (mIgG2b) was administered to an additional cohort of Kras mice twice weekly via intraperitoneal injection for a total of four weeks – starting at 8 weeks post AdCre – at a dose of 10mg/kg. Tissue collection and histology Lung tissue specimens were collected and processed as described (32). Briefly, the left lung was ligated and snap-frozen for later analysis. The right lung was inflated with 10% neutral buffered formalin (NBF) at 25 cm H20 pressure before fixing in NBF overnight. 5-μm paraffin-embedded sections were stained for hematoxylin and eosin (H&E) or immunostained for CD45 (BD Bioscience, San Diego, CA, USA), FoxP3 (eBioscience, San Diego, CA, USA), or CD3 (Serotec, Raleigh, NC, USA) using 3,3”-diaminobenzidine development and hematoxylin counter-staining. Global adjustments to white balance, brightness and/or contrast were made to some photomicrographs using Photoshop (Adobe Systems, San Jose, CA, USA). Slides were imaged with an Eclipse 80i microscope (Nikon Instruments Inc., Melville, NY, USA), excluding the whole lobe images presented in Figure 1A, which were collected at 20X magnification with an Aperio digital pathology slide scanner (Leica Biosystems, Buffalo Grove, IL, USA). Total lung and tumor area (μm2) were measured from H&E stained slides using NIS-Elements Advanced Research software (Nikon). Results are expressed as % lung occupied by tumor ((area tumor ÷ area lung) × 100). Each lung was also scored for tumor grade, as described (33). FoxP3 and CD3 stained lung lobes (n = 5 mice/genotype) were scored for the presence or absence of cells within three locations: tumor-associated, tumor-infiltrating, or within a lymphoid aggregate (LA). Lung tissue single cell preparation Single-cell suspensions were generated from saline-perfused mouse lungs using mechanical disruption followed by 1 hr digestion at 37°C in RPMI-1640 containing 10% FCS and penicillin/streptomycin along with 80 U/ml DNase, 300 U/ml Collagenase Type 1 (both Worthington Biochemical Corporation, Lakewood, NJ, USA), and 60 U/ml hyaluronidase (Sigma, St Louis, MO, USA). Digested lungs were sheared through a 19g needle, strained through 70-μm nylon mesh, centrifuged, lysed (RBC), washed, strained through 40-μm mesh, centrifuged, and resuspended in DPBS + 2% FCS. Cell viability was determined using trypan blue staining and a TC20™ Automated Cell Counter (BioRad, Hercules, CA, USA). We obtained two SCLC and five ADCA surgical specimens, each with non-adjacent normal lung tissue, using an approved IRB file in association with FHCRC, University of Washington Medical Center, and Northwest BioTrust. Single-cell suspensions were generated using the above digestion protocol. Flow cytometry Single-cell suspensions were incubated with 1.0 μg/106 cells Mouse TruStain FcX™ or 1.0 μl/106 cells Human TruStain FcX™ (Biolegend, San Diego, CA, USA) prior to immunostaining. Twenty-seven fluorochrome-labeled antibodies were used among four multicolor panels for mouse specimens (all flow antibodies detailed in Supplemental Table I). Immunostaining was performed for 30 min on ice, protected from light. Dead cells were excluded with Fixable Viability Dye eFluor® 780 (FVD; eBioscience), per manufacturer's protocol. Stained cells were washed, fixed with IC Fixation Buffer (eBioscience), and stored at 4°C until analysis. Intracellular cytokine production was assessed using PMA (25 ng/ml), ionomycin (1 μg/ml; both from Sigma), and monensin (1.5 μl/ml; BD Bioscience) stimulation for 5 hr at 37°C, 5% CO2. An unstimulated sample was incubated without PMA and ionomycin. After stimulation, cells were washed, stained with FVD, fixed, and permeabilized with the Transcription Factor Buffer Set (BD) prior to immunostaining. Samples were analyzed on a LSR II flow cytometer with FACSDiva™ software (BD), recording ≥ 1×105 events per sample. Data were compensated and analyzed with FlowJo software (TreeStar, Ashland, OR, USA). Gates were defined by fluorescence-minus-one (FMO) samples and verified with appropriate isotype controls. The unstimulated control was used to define cytokine gates. Total cell content was calculated by multiplying the overall number of live cells recovered from each animal (i.e. the trypan-blue-negative hemacytometer count) by the percentage of live cells for each gated parameter. Cytokine-producing T cell subsets were calculated by multiplying the percent parent gate to the previously determined parent population count. Median fluorescence intensity (Med.F.I.) of NKG2D-BV421 and PDL1-BV421 parameters was calculated in FlowJo, with Med.F.I. of the relevant FMO control being subtracted from all experimental values for normalization. CFSE assay Splenocytes from wild-type mice were labeled with 50 μM CFSE (Molecular Probes, Eugene, OR, USA), per manufacturer's instructions. 1×106 CFSE-labeled splenocytes were transferred to 6-well tissue culture plates coated with anti-CD3/anti-CD28 antibodies (Biolegend). Cells were incubated with 200 μg homogenate – generated from 10-week Egfr, Kras, or wt lungs – at 37°C, 5% CO2 for 4 days. Lymphocyte proliferation was determined by harvesting the cells, staining with CD8a-BV421 (Biolegend) and FVD, and measuring ≥ 1×103 live CD8+ cells on a LSR II flow cytometer. Gene Expression Analysis Total RNA was isolated from frozen mouse lungs using TRIzol reagent (Life Technologies, Carlsbad, CA, USA) and subsequently purified with the RNeasy Mini Kit (Qiagen, Hilden, Germany). cDNA was generated from 2 μg total RNA using SuperScript II Reverse Transcriptase and Oligo(dT) (Life Technologies). The expression of indicated target genes was analyzed using a StepOnePlus Real Time PCR machine and TaqMan primer/probe sets (Applied Biosystems, Foster City, CA, USA), with all reactions run in triplicate. ΔCt values were calculated using Gapdh as the endogenous housekeeping gene. Tissue Microarray A lung ADCA cohort on tissue microarray (TMA), consisting of 135 cases, was obtained from the University of Pittsburgh Cancer Institute. Patient identifiers were removed and the study therefore considered “not human subjects” research (i.e. IRB exempt). Each case was previously annotated as either EGFR mutant (n = 31), KRAS mutant (n = 69), or wild-type for both EGFR and KRAS (n = 35). Formalin-fixed, paraffin-embedded sections were stained with an anti-human CD8 antibody (Abcam #ab4055, Cambridge, MA). Immunohistochemical (IHC)-stained TMA slides were scanned in brightfield at a 20X objective using NanoZoomer Digital Pathology System (Hamamatsu, Hamamatsu City, Japan). Twenty TMA cases (n = 5 EGFR and 16 KRAS) were excluded as lost, non-informative (e.g. poorly stained or non-tumorous), or exhibiting high inter-punch variability (i.e. SEM ≥ 50% of mean). The number of CD8+ cells per TMA core was recorded blind to genotype and normalized to core area. Individual core counts from 2 or more replicates were available for most cases, and CD8+ cell counts/mm2 were averaged across replicates. Cut-off values of low versus high CD8+ cell content were defined by the midpoint. Comparison of TMA cohorts was conducted using the Fisher's exact test with one-tailed P value. Statistical analysis Significant differences between experimental groups were determined in Prism 6 (GraphPad Software, La Jolla, CA, USA) using unpaired t-tests or, for comparing ≥ 3 groups, one-way ANOVA with indicated post-hoc test for correction of multiple comparisons. Incidence of CD3+ and/or FoxP3+ cells per lobe was compared between genotypes using the chi-square test. Unless indicated otherwise, data are presented as mean ± SEM. P < 0.05 was considered statistically significant. RESULTS Egfr- and Kras-driven ADCA induce a strong inflammatory response Lung tumor development and associated inflammation were assessed in TetO-EgfrL858R;Ccsp-rtTA (Egfr), KrasLSL-G12D (Kras) and KrasLSL-G12D;Trp53Fl/Fl (Kp53) mice. To allow for the dynamic assessment of tumor-associated immune responses, lung tumor-bearing animals and appropriate littermate controls were studied over a time course of 6, 10, and 14 weeks post initiation. Consistent with previous studies (24, 26, 34), Egfr, Kras, and Kp53 mice developed neoplastic lesions reminiscent of human disease, from benign hyperplasia and adenomas to malignant ADCA (Fig. 1A-C). Hyperplasia was observed at all time points, although it was less prevalent in Egfr mice than in Kras-mutant mice (data not shown). The introduction of a secondary mutation in Trp53 increased ADCA formation and amplified tumor growth. Accordingly, tumor burden in 10-week Kras mice was significantly less than observed in age-matched Kp53 mice (Fig. 1D; P = 0.0231). Analysis of Kp53 mice at the 14-week time point was precluded by early mortality, but Kras mice remained viable and exhibited 38% lung tumor burden. Body mass measurements, used to non-invasively monitor lung tumor-associated morbidity, correlated with tumor burden for all genotypes (Fig. 1E). Previous investigations of pulmonary inflammation have been largely based on the assessment of bronchoalveolar lavage (BAL) fluid. Although a well-accepted methodology, BAL studies confine analysis to the airway compartment and limit the number of immune cell types that can be identified. Therefore, to more thoroughly investigate the immune cell composition of the TME, we performed flow cytometric analyses on single-cell suspensions generated from whole lung tissues. Using the gating strategy shown in Figure 2A, we identified 13 unique leukocyte populations defined by 16 antibody markers (see Methods and Supplemental Table I). Kras, Kp53, and Egfr mutant mice all display a robust immune response, as evidence by 3 to 5-fold elevations in total CD45+ cell content as compared to normal lung (Fig. 2B-G). Macrophages were by far the most prevalent immune cell type in the lungs of the three murine ADCA models. Ten weeks after starting doxycycline, Egfr mice exhibited 11-fold increases in macrophage content compared to controls (Fig. 2B). Similarly, by 6 weeks post-Cre induction in Kp53 and 10 weeks in Kras animals, total macrophage cell counts had increased 14- and 17-fold (Fig. 2D and F). Of note, macrophage content increased with time only in Kras and Kp53 animals (Supplemental Table II). Since myeloid derived suppressor cells (MDSC) are comprised of monocytic (M-MDSC) and granulocytic (G-MDSC) subsets, we simply defined these cells as monocytes (CD11b+Ly6C+) and neutrophils (CD11b+Ly6G+). We observed small but significant differences in neutrophil content in 10- and 14-week Egfr (Fig. 2B and C) as well as 14-week Kras tumor-bearing lungs (Fig. 2E). By 6- and 10-weeks, tumor-associated neutrophil content increased 2- and 5-fold, respectively, in Kp53 animals compared to control (Fig. 2F and G). Indeed, neutrophil counts in Kp53 lungs were statistically increased compared to all other genotypes at the 10-week time point (Supplemental Table III). However, this neutrophil signature may merely reflect overall tumor burden, as statistical analysis of cohorts with approximately matched tumor area (i.e. 6-week Kp53, 10-week Kras, and 14-week Egfr) failed to identify significant differences in pulmonary PMN content. Impaired natural killer cell function in Kras-driven ADCA Natural killer (NK) cells are lymphocytes of the innate immune system that play an important role in the host defense against inhaled pathogens (35). NK cell counts in non-tumor bearing lungs were comparable to those of macrophages and granulocytes (Fig. 2). Unlike the myeloid cell expansion observed with ADCA development, however, NK populations remained largely unaltered in tumor-bearing lungs. When significant increases were identified (Fig. 2B, D-G), the fold changes were small, and NK cell counts decreased over time in Kras animals (Supplemental Table II). NK cells are required for effective tumor immunosurveillance (36), but cancer cells have developed multiple strategies to escape NK cell-mediated cytotoxicity, including downregulation of the natural killer group 2, member D receptor (NKG2D, also known as CD314) (37, 38). Surface expression of NKG2D on NK cells was significantly decreased in both Kras and Kp53 tumor-bearing animals at all time points, but exhibited no change in Egfr mice (Fig. 2H). Thus, these murine models of Kras-driven lung ADCA recapitulate an immune escape mechanism previously described in human lung cancer (39). Oncogenic drivers dictate lymphocyte recruitment into the ADCA microenvironment The majority of immunotherapeutic approaches are premised on the ability of the adaptive immune system to infiltrate tumors and identify tumor-specific antigens. Therefore, we comprehensively surveyed the lymphocyte subpopulations present within the TME. B cell populations remained unaltered in tumor-bearing Egfr mice (Fig. 2B and C) but increased at least 2-fold in Kras and Kp53 mice at all time points (Fig. 2D-G). CD3+ populations were significantly increased in multiple groups (Fig. 2B, D-H), but T cell counts were demonstrably greater in Kras and Kp53 compared to Egfr mice (Supplemental Table III), even after controlling for tumor burden. CD4+ helper T cell expansion was observed in both Kras-driven tumor models but was limited in Egfr mice (Fig. 3A-F). This pattern was reflected in regulatory T cell (Treg) content, which was significantly lower in tumor-bearing lungs from Egfr mice compared to Kras mice (Supplemental Table III). Interestingly, Treg content increased over time in both Kras and Kp53 animals (Supplemental Table II), the only cell type other than macrophages and neutrophils to exhibit such dynamic behavior. Expression of IFNγ by CD4+ T cells (TH1 cells) and IL17A by CD3+ T cells were also significantly upregulated in Kras and Kp53 animals compared to control at early and late time points (Fig. 3C-F), but exhibited little to no increase in Egfr animals (Fig. 3A and B). Direct comparison of tumor-bearing lungs from all genotypes found higher TH1 and CD3+IL17A+ cell counts in both Kras-driven models compared to the Egfr mice (Supplemental Table III). Despite repeated attempts, we were unable to identify IL4-positive TH2 cells in our animals (Supplemental Fig. 1); it remains unclear whether this deficiency reflects a true biological absence or, more likely, a technical barrier. To assess the spatial relationship between lymphocytes and tumor cells, we performed IHC staining for CD3 and FoxP3 on Egfr, Kras, and Kp53 tumor bearing mice. For each model of lung ADCA, immune cell location was assigned to three categories: 1) tumor-associated (i.e. TA, or peripheral), 2) tumor-infiltrating (TI), or 3) within a neighboring lymphoid aggregate (LA). Examples of each are provided in Figure 3G-J. Staining for CD3 illustrated key differences by genotype, as Egfr mutant mice displayed markedly less tumor-associated CD3+ T cell content than Kras and Kp53 specimens (Fig. 3K, left). Tumor-infiltrating CD3+ cells were identified in all ADCA models, although they were more prevalent in Kp53 mice (Fig. 3K, center). Similarly, CD3+ cells were present in all LAs, but LAs were significantly less common in Egfr mice, whereas they were uniformly present in the other genotypes (Fig. 3K, right). Approximately 15% of lobes from Egfr mice displayed the presence of tumor-associated Tregs, compared to about 40% for Kras mice (Fig. 3L, left). The primary location of Tregs in all genotypes was within LA structures (Fig. 3L, right), with essentially no tumor infiltration observed (Fig. 3L, center). Taken together, the IHC studies confirm the flow cytometry data showing that Egfr mice contain fewer Tregs than the other genotypes and, moreover, demonstrate a paucity of tumor-infiltrating lymphocytes, especially when compared to Kp53 tumors. CD8+ lymphocyte content and function differ by lung ADCA subtype CD8+ T cells are capable of detecting and discriminately eliminating tumor cells (40). Notably, CD8+ cell content differed significantly between models driven by mutant Kras versus Egfr. Expansion of CD8+ cells was observed at all time points in Kras and Kp53 mice but did not occur in the Egfr-mutant cohort (Fig. 3A-F). Even after controlling for tumor burden, Kras-mutant mice displayed greater CD8+ cell content than found in Egfr mice (Supplemental Table III). Therefore, we carried out a number of experiments in attempt to determine the mechanistic basis for this finding and to translate this observation to human disease. Initially, we performed quantitative real-time PCR (qPCR) for key CC and CXC chemokines known to impact immune cell recruitment. Notably, we identified an increase in Cxcl-9 and -10 in Kras mutant lungs compared to Egfr (Figure 3M), which may at least partially explain the differences in lymphocyte content seen between the two lung ADCA models. In order to translate these findings to human disease, we performed immunohistochemical staining for CD8 on a lung ADCA TMA annotated for KRAS (n = 53 cases) and EGFR (n = 26) mutational status. The frequency of KRAS and EGFR mutations present in the cases displaying high versus low CD8 content (the top and bottom 50% of cases, respectively) were assessed, and EGFR mutations were found to be significantly overrepresented in the CD8 low cohort (Fig. 4A). Specifically, 65.4% of EGFR-mutant cases were scored as CD8-low versus 41.5% of KRAS-mutant cases. Thus, similar to the findings in genetically engineered mouse models presented above, EGFR mutant lung adenocarcinomas exhibit reduced CD8+ lymphocyte infiltration in human lung cancers as compared to KRAS mutant lung ADCA. Because CD8+ responses can be blunted by immune checkpoint ligands, we measured PDL1 expression on both macrophages and tumor cells (EpCAM+) by flow cytometry. Interestingly, although PDL1 expression was decreased in tumor-bearing versus control lungs for both Egfr and Kras mice, there was no difference in PDL1 expression between oncogenic subtypes (Figure 4B). Since other tumor microenvironmental factors can perturb lymphocyte function, we assessed whether the TME of Egfr mice was more suppressive to lymphocyte proliferation than that found in Kras mice. Both Egfr and Kras tumor homogenates reduced CD8+ T cell proliferation using CFSE-labeled lymphocytes, but the Egfr homogenates were not found to be more suppressive than Kras (Fig. 4C). As the intriguing lack of CD8+ cell expansion in Egfr mutant tumors suggests a failure of CD8+ cell activation in Egfr mice, we performed a detailed assessment of T cell effector and memory status at the 10 and 14-week time points using the markers CD62L, CD44, and PD1 (Figure 4D). No evidence of CD8+ T cell activation was observed, as evidenced by lack of an increase in CD8+PD1+ cells in tumor bearing Egfr mice (Figure 4E, G). Additionally, the proportion of central memory (CD62L+CD44+) and effector/effector memory (CD62L−CD44+) CD8+ T cells was unchanged, with the majority of these cells still falling into the naïve (CD62L+CD44−) category in Egfr mice. In contrast, and consistent with the small but significant increase in helper T cells shown in Figure 3A-B, CD4+ T cells demonstrated a significant increase in effector/effector memory populations in both 10 and 14-week Egfr mice (Figure 4F, H). Given the robust increase in CD8+ T cell content observed in tumor-bearing lungs from Kras mice compared to control, we elected to assess the status and functionality of the lymphocyte populations in Kras mice using an CTLA4 mIgG2b (clone 9D9) antibody (MedImmune). Although administration of anti-CTLA4 increased the proportion of both CD8+ (Figure 4J) and CD4+ (Figure 4K) effector/effector memory cells, this cellular phenotype failed to translate into an altered tumor burden in Kras mice (Figure 4I). Thus, despite an activated CD8+ T cell response in Kras mutant mice, tumor progression continued unabated. Role of IL17A-producing γδ T cells in Kras-driven lung ADCA Given the robust expansion of IL17A+ T cells in Kras mutant tumor-bearing mice and the known pro-tumor role of TH17 cells in lung ADCA (41), we elected to examine the cellular sources of IL17A in the lungs of our Kras mutant mouse models. Surprisingly, the predominant source of IL17A was found to be γδ T cells rather than CD4+ TH17 cells (Fig. 5A-B). An attempt to interrogate the role of IL17A in lung tumorigenesis by crossing Kras mutant mice to mice lacking the transcription factor for IL17A (i.e. RORγt) (42) was stymied by the frequent occurrence of lymphoid neoplasms in these animals. The spontaneously arising lymphomas exhibited both thymic (Fig. 5C) and splenic (Fig. 5D) involvement as well as diffuse infiltration of the liver (Fig. 5E) and lungs (Fig. 5F). Previous investigations of a related mouse model of RORγt deficiency (43) similarly observed a high incidence of lymphoma but did not indicate the occurrence of the pulmonary metastases that, unfortunately, preclude the use of this model in lung tumorigenesis studies. Further efforts to interrogate the specific role of γδ T cells in Kras mutant lung ADCA revealed that deletion of γδ T cells impacted neither tumor burden (Fig. 5G) nor the immune cell composition of the tumor microenvironment (Fig. 5H). A paucity of tumor-infiltrating leukocytes in murine and human SCLC The immune cell composition present within the SCLC TME has not been previously investigated to any extent. Therefore, we profiled the immune content of SCLC, using cohorts of Rbp53 mice infected with AdCre. As described previously (44), lung tumors of mainly neuroendocrine histology arose within 40 to 50 weeks of AdCre administration (Fig. 6A). Flow cytometric analysis of SCLC tumor-bearing lungs (gated as shown in Fig. 2) identified a small but noteworthy inflammatory presence in Rbp53 mice compared to control (Fig. 6B). The total number of CD45+ leukocytes was increased 2-fold in SCLC, and tumor-associated CD45+ cells were also visible by IHC staining (Fig. 6C). Unlike Kras- and Egfr-mutant animals (Fig. 1A and B), SCLC tumors presented as large, discrete foci, and little hyperplasia was observed. CD45+ cells were consequently clustered at the periphery of the SCLC lesions (Fig. 6D), without the inflammatory “field effect” frequently observed in the ADCA models. Few tumor-infiltrating leukocytes were detected (Fig. 6E). The major immune component of SCLC was found to be CD3+ T lymphocytes (Fig. 6B). This population included a 7-fold increase in the number of γδ T cells and a strong trend toward increased CD4+ helper T cells (P = 0.0698). In marked contrast to the ADCA models, expansion of innate immune cells in SCLC tumor-bearing lungs was minimal, with only a 2-fold increase observed in macrophages and a non-significant increase in neutrophils (P = 0.0886). To further investigate this phenomenon, we compared the ratio of CD3+ T cells to myeloid cells (macrophages, neutrophils, monocytes, and eosinophils) and found a pronounced lymphocyte-dominant signature in SCLC versus all ADCA models (Fig. 6F). Egfr mice presented with the smallest CD3:myeloid ratio and were also significantly different from Kras mice. As part of an ongoing study of the immune composition of human lung cancer, we obtained two surgical specimens with confirmed small cell pathology. Since resecting SCLC is rarely clinically indicated, these two specimens represented a unique opportunity to measure the immune cell composition present within the SCLC TME. Therefore, we performed flow cytometry analyses on single cell suspensions generated from these two cases. Similar to the findings in the GEM models, tumor-associated inflammation was discernibly lower in SCLC compared to five ADCA specimens, as CD45+ cells comprised a mere 16.2% of live cells in the SCLC resections, compared to 81.9% in ADCA (Fig. 6G). DISCUSSION Lung cancer is a heterogeneous disease that can be divided into distinct subtypes based on both molecular and cellular characteristics (45). Herein we tested the hypothesis that these subtypes dictate the inflammatory response to cancer by immune-profiling the lung TME in a mouse model of SCLC and in three molecularly distinct models of NSCLC (summarized in Fig. 6H). We found that Egfr and Kras mutations give rise to distinct immune responses characterized by differential expansion of B cells, CD8+ T cells, Tregs, and IL17A-producing T cell populations. Although loss of Trp53 promoted malignancy, it had minimal effect on immune cell composition within the Kras TME. We further demonstrate that SCLC possesses an overall reduced inflammatory presence compared to NSCLC, and one in which lymphocytes predominate over myeloid lineage cells. Mutational profile and histological origin therefore actively shape the immune contexture of lung cancer, a finding that may have important clinical ramifications. The strong macrophage field responses that occur in mutant Kras- and Egfr-driven mouse ADCA are seldom observed in human lung cancer and represent a potential limitation of these GEM models. In Kras mice, this phenomenon of alveolar macrophages flooding the airspaces has been likened to desquamative interstitial pneumonitis (DIP), a rare interstitial lung disease with similar pathology (46). Moreover, the conditional mouse models studied herein utilize varied induction methodologies, i.e. adenoviral Cre recombinase and doxycycline-regulated transgene expression. Although we cannot exclude potential confounding effects of viral infection or doxycycline consumption on the tumor immune response, we have attempted to correct for these variables by using adenovirus- or doxycycline-exposed wild-type animals for the relevant control cohorts. Few gene-specific investigations of the mouse lung TME have been conducted, and no comprehensive effort has been made to compare and contrast different molecular and histological models of lung cancer. Our results do, however, validate findings from several earlier reports on lung tumor-associated inflammation. We were unsurprised to identify macrophages as the dominant immune cell presence in mouse ADCA given that strong tumor-associated macrophage responses have been previously identified in mutant Egfr (21) and Kras (19, 41, 47) mouse lung tumor models. Likewise, as we describe herein, neutrophils have been shown by others to be a modest but important component of Kras- but not Egfr-driven mouse lung ADCA (19, 21, 41). Since the majority of prior data in this regard relied on BAL cell counts, we employed flow cytometry to better define the quality of the immune response. Using this methodology, we found that recruitment of lymphoid lineage cells varies greatly between ADCA models, as EgfrL858R mice exhibited a paucity of B cells, CD8+ T cells, Tregs, and IL17A-producing T cells when compared to the Kras and Kras;Trp53 lung TME. The most clinically relevant finding in this study is the lack of a CD8+ lymphocyte response in both Egfr mutant mice and EGFR mutant human lung ADCA specimens when compared to KRAS mutant counterparts. Markers of effector/memory status failed to reveal any evidence of CD8+ T cell activation or differentiation in Egfr mice. This suggests that EGFR mutation may not elicit an antigen-driven immune response. Although the same could be said for mutant KRAS, we were able to demonstrate an increase in activated and effector-memory CD8+ cells in the Kras mouse model. Furthermore, tumor-infiltrating lymphocyte (TIL) populations that specifically target mutant KRASG12D have recently been identified in colon cancer (48). Despite the presence of activated CD8+ cells, tumor growth continued in Kras mice even with the addition of an anti-CTLA4 therapeutic antibody. Our interpretation of this data is that increases in activated CD8+ cells within the TME in Kras mutant mice will not impact tumor growth unless they are tumor reactive. Specifically in this case, tumor-derived chemokines, such as Cxcl-10 are likely to increase the number of tumor-associated lymphocytes. Whereas anti-CTLA4 antibody therapy drove an increase in effector T cells, these cells would not be expected to reduce tumor burden if they did not recognize a tumor-associated antigen. It is also possible that anti-CTLA4 monotherapy will prove ineffective but that combined immunotherapeutic regimens (e.g. anti-CTLA4 + anti-PDL1) would prove effective. With respect to human lung ADCA, the association of EGFR mutant cancers with never-smoker status would suggest that these tumors are genetically simplified and, unlike smoking-associated KRAS mutant cancers, may not possess sufficient mutational burden to harbor neo-antigens (49, 50). The mouse models described herein were not exposed to cigarette smoke or other carcinogens, eliminating this proposed explanation for the differential CD8+ responses we observed. At this time, however, we cannot exclude the possibility that KRAS mutant human and mouse lung ADCA have realized the same phenotype of high CD8+ T cell infiltration through different mechanisms of action. Tregs and IL17A+ T cells have emerged as important cell populations in multiple mouse models of cancer (41, 51, 52), and both cell types exhibited notable patterns of expression or localization in the murine ADCA models. Although TGFβ and IL6 generate gradients leading independently to Treg or TH17 differentiation, we observed concurrent increases in both populations in Kras and Kp53 mutant tumor-bearing lungs. Notably, we found the major source of IL17A in Kras mutant ADCA to be γδ T cells and not TH17 cells. Since IL17A deficiency has been previously shown to reduce lung tumor growth (41) and IL17A-producing γδ T cells are known to promote breast (53) and pancreatic neoplasia (52), these findings suggested to us that expansion of a pulmonary IL17A-producing γδ T cell subset might overshadow the tumor-surveillance role traditionally ascribed to γδ T cells (54). However, Kras mutant, γδ T cell-deficient mice displayed equivalent lung tumor burden and strikingly similar immune profiles to their γδ T cell-competent counterparts. Therefore, although IL17A is an important signaling component in the immune landscape of lung ADCA, IL17A+ γδ T cells appear to contribute little to the process of lung tumorigenesis. Tregs, in contrast, appear to play a particularly important role in the Kras mutant lung TME, as they were the only non-myeloid lineage population to expand over the course of tumor development. Moreover, Tregs display a unique anatomic location in lung ADCA. They are rarely associated with the tumor itself but are instead frequently found within lymphoid aggregate structures that are believed to function as a local site of antigen presentation and have been correlated with good clinical outcomes in NSCLC (55). The presence of Tregs in these structures has recently been shown to be detrimental to the generation of an effective immune response in murine lung ADCA (56), highlighting the importance of Treg targeting strategies for the clinical management of lung cancer patients. The immune cell composition of SCLC has not been well studied. Our findings in Rbp53 mice point to a less robust but more lymphocyte-predominant host immune response to murine SCLC than to ADCA. Moreover, when we analyzed the immune cell content of two human SCLC cases, we identified a strikingly similar immune profile of sparse CD45+ cell content. Although we acknowledge the inherent limitations of n = 2 studies, patients diagnosed with SCLC seldom undergo lung resection (57), making access to such specimens exceedingly rare. Solid tumor malignancies demonstrating the best responses to current ICI therapies have been those with high mutational burdens and/or a history of cigarette smoke exposure (e.g. melanoma, head and neck SCC, and urinary bladder cancer), both traits common to SCLC (2). In light of these correlations, it is tempting to speculate that SCLC patients would exhibit good responses to ICI therapy. However, initial reports suggest that success rates to anti-PD1 therapy in SCLC are at best only comparable to NSCLC (58, 59). Our preliminary findings with respect to the immune cell composition in SCLC suggest that the presence of redundant immune suppressive factors would not be a likely source of treatment failure, which is almost certainly an important concept in NSCLC. These findings point to potentially unique features of the SCLC TME (e.g. matrix protein composition) that will require additional study. A robust CD45+ immune response was observed in both the ADCA mouse models and human lung ADCA patients. Leukocytes account for nearly 75% of total cellular content in human ADCA, an even greater proportion than we identified in mice (~55%). With the exception of the aforementioned exaggerated macrophage responses, the robust and diverse immune landscape observed in GEM models of ADCA approximates that seen in human lung cancers (60). Driving mutations, such as in Egfr and Kras, substantially impact the TME through the release of bioactive molecules, which is very well reflected in these GEM models. One potential shortcoming of these models is the genetic simplicity of the tumors, which rely on a single driving mutation. In contrast, human NSCLC harbors an average of ~150 distinct mutations per case (61). Efforts are underway to construct mouse models of cancer that harbor a greater abundance of single nucleotide variations and, thus, potential neoantigens. However, EGFR mutant cancers in non-smokers are typically genetically simplified (5), such that the Egfr mutant mice described herein likely constitute an excellent representation of both the genetic component of the cancer cell and the immune composition of the TME. The emergence of immune checkpoint inhibitors has been a tremendous advance, but unfortunately the majority of lung cancer patients in clinical trials have failed to respond to ICI therapy (8-11). In addition to the PD-1/PD-L1 based drugs currently in use, novel ICI agents are likely to emerge in the near future. Our findings argue that the cellular and molecular characteristics of lung cancer may provide an important framework for patient-targeted immunotherapy. Furthermore, preclinical testing of future immunotherapy agents should be performed in genetically and histologically diverse model systems to enable the assessment of tumor subtype-specific efficacy. Supplementary Material 1 ACKNOWLEDGMENTS The authors thank the FHCRC Experimental Histopathology and Flow Cytometry shared resource facilities, the UW Histology and Imaging Core, and the Viral Vector Core Facility at the University of Iowa Carver College of Medicine. The authors would also like to thank MedImmune for supplying the anti-CTLA4 therapeutic antibody. Financial Support: This work was supported by NIH/NHLBI grant R01 HL108979 to A.M.H., European Commission FP7-PEOPLE-2012-IOF 331255 to J.K., and by the Fred Hutchinson Cancer Research Center. Abbreviations ADCA adenocarcinoma AdCre adenoviral Cre recombinase BAL bronchoalveolar lavage fluid GEM genetically engineered mouse ICI immune checkpoint inhibitor NSCLC non-small cell lung cancer PMN polymorphonuclear cell SCLC small cell lung cancer TMA tissue microarray TME tumor microenvironment Treg regulatory T cell Figure 1 Egfr, Kras and Kp53 mice develop lung tumors and associated inflammation. (A, B) All models chronologically develop atypical alveolar hyperplasia, adenoma, and adenocarcinoma 6, 10, and 14 weeks post tumor induction. Normal lung from a non-tumor bearing wild-type mouse is depicted in the lower right corner. H&E sections, scale bars = 2 mm (A) and 500 μm (B), except lower wild-type panel = 1 mm. (C) Spectrum of disease in murine ADCA models. Data are presented as percent of mice exhibiting ≥ 1 indicated lesion at each time point post-induction (n ≥ 5 mice per group). All genotypes exhibited hyperplasia at all time points examined (not shown). Analysis of 14-week Kras LSL/+;Trp53Fl/Fl mice was precluded by early mortality. (D) Percent tumor area was calculated at the indicated time points in a minimum of at least 3 representative lungs from Egfr, Kras, and Kp53 mice. (E) Body mass of tumor-bearing female mice (n ≥ 5) compared to non-tumor bearing littermate controls (n ≥ 3) at each time point post-induction. Figure 2 Flow cytometric analysis of the inflammatory response to lung tumorigenesis. (A) Representative dot plots demonstrate the strategy used to characterize the mouse lung TME. Single cell gates (not shown) were initially applied to remove doublet cells. All subsequent gating utilized a viability marker followed by gating on the CD45+ population. Ly6G identified neutrophils (PMN), while remaining myeloid cells were classified as macrophage (Mac; SiglecF+CD11c+), eosinophil (Eos; SiglecF+CD11c−) or monocyte (Mono; SiglecFloCD11bhiLy6C+) from the Ly6G− population. A size gate was applied for lymphocyte analysis followed by staining to identify B cells (CD3−CD19+), T cells (CD3+CD19−), and NK cells (CD3−CD19−NK1.1+). T cells were further classified into γδ T cells (γδTCR+), CD4 cells (CD4+CD8−), and CD8 cells (CD8+CD4−). T cell subtypes were identified as TH1 (CD4+IFNγ+), Treg (CD4+CD25+FoxP3+), and IL17A-producing T cells (CD3+IL17A+). Major lung immune cell populations in (B) Egfr mice 10 weeks post tumor induction (n ≥ 6), (C) Egfr at 14 weeks (n ≥ 4), (D) Kras at 10 weeks (n = 14), (E) Kras at 14 weeks (n = 11), (F) Kp53 at 6 weeks (n = 8), and (G) Kp53 at 10 weeks (n ≥ 8), compared to non-tumor bearing control mice (white bars, n ≥ 3). Early and late time points for each genotype are depicted with gray and black bars, respectively. Data for each cell type are displayed as the total number of live cells present within the mouse lung. (H) NKG2D median fluorescence intensity (i.e. Med.F.I.) on NK cells was examined in Egfr, Kras, and Kp53 mutant lungs compared to normal lung controls at indicated time points (n ≥ 3 per group). Asterisks indicate P < 0.05. Figure 3 Oncogenic drivers dictate lymphocyte recruitment into the ADCA microenvironment. CD3+ T lymphocyte subpopulations in (A) Egfr mice 10 weeks post tumor induction (n ≥ 6), (B) Egfr at 14 weeks (n = 6), (C) Kras at 10 weeks (n ≥ 10), (D) Kras at 14 weeks (n = 11), (E) Kp53 at 6 weeks (n = 8), and (F) Kp53 at 10 weeks (n ≥ 6), compared to non-tumor bearing control mice (white bars, n ≥ 4). Data for each cell type are displayed as the total number of live cells present within the mouse lung. (G-J) CD3 immunostaining revealed that tumor-associated T cells (G, arrowheads) were commonly located at the edges of neoplastic lesions. Infiltration of CD3+ T cells into the tumor mass is indicated with arrows (H); note that large clusters of TA CD3+ cells were also present in the periphery of this lesion. CD3+ lymphoid aggregates formed within the vicinity of a tumor (I, dashed line, Tu) and were typically associated with airways and/or blood vessels. Although FoxP3+ Tregs were seldom observed infiltrating tumor masses (J), they constituted a significant portion of cells present in lymphoid aggregates. Scale bar = 250 μm. (K-L) Quantification of immune cell localization in matched tumor burden 14-week Egfr (n = 6), 10-week Kras (n = 5), and 6-week Kp53 (n = 5) mice. Results are expressed as the percent of lung lobes that contained at least one occurrence of (K, left) tumor-associated CD3+ cells, (center) tumor-infiltrating CD3+ cells, (right) CD3+ cells in lymphoid aggregates, or (L, left) tumor-associated FoxP3+ cells, (center) tumor-infiltrating FoxP3+ cells, or (right) FoxP3+ cells in lymphoid aggregates. (M) Expression of cytokine and chemokine genes in tumor-bearing lungs from 14-week Egfr and 10-week Kras mice (n = 4 per genotype). Data are presented as mean 1/ΔCt with 95% CI. Asterisks indicate P < 0.05. Figure 4 CD8+ cell content and function correlates with lung ADCA subtype. (A) The number of CD8+ cells/mm2 was tabulated for each core section present on a TMA of lung ADCA cases annotated for EGFR and KRAS mutational status. Cases were ranked from lowest to highest CD8 content prior to unblinding for genotype. Shown are representative images of EGFR- (left) and KRAS-mutant (right) ADCA. Scale bar = 100 μm. 65.4% (17 of 26) of EGFR-mutant cases were scored as CD8-low versus 41.5% (22 of 53) of KRAS-mutant cases (Fisher's exact test, P = 0.0392). (B) PDL1 Med.F.I. was assessed on pulmonary EpCAM+ epithelial cells and macrophages from 10-week Egfr (n = 5) and 14-week Kras mice (n = 4). Expression compared to normal lung (n ≥ 4) is shown in bottom panels. (C) Splenocytes from non-tumor bearing wild-type mice were labeled with CFSE and incubated with protein homogenate generated from wild-type normal lung (NL) or tumor-bearing lung from 10 week Kras or Egfr mice, or with media alone. The cells were subsequently stained with anti-CD8 and a viability marker and analyzed for CFSE intensity; representative plots are shown. For each genotype (n ≥ 3) the percentage of proliferating CD8+ T cells was determined after normalization to the media control. Statistical differences were assessed by one-way ANOVA with Tukey's post-test. (D) Flow cytometric analysis of T cell function in Egfr mice compared to wild-type control, gated from single, live, CD45+CD3+ parent population. Lymphocytes were gated as CD62L+CD44− (i.e. Naïve), CD62L+CD44+ (TCM, central memory), and CD62L−CD44+ (TEM/Eff, effector memory/effector). PD1 expression was assessed on CD8+ T cells only. CD8+ and CD4+ T cell populations were examined at 10 (E, F) and 14 weeks (G, H) post-induction of mutant Egfr (n = 6 tumor-bearing lungs and n ≥ 3 controls per group), respectively. (I) Percent lung tumor area of Kras mice treated with anti-CTLA4 (n = 7) or isotype (n = 6) with representative H&E sections. Scale bar = 500 μm. (J, K) Flow cytometric analysis of CD8+ and CD4+ T cell populations in anti-CTLA4 and isotype-treated Kras mice (n = 5 per group). Asterisks indicate P < 0.05. Figure 5 IL17A cytokine production and impact in Kras mutant lung ADCA. (A) Representative dot plot demonstrating the relative production of IL17A by γδ T and non-γδ T cells; gated from single, live, CD45+CD3+ parent population. (B) Quantification of IL17A cytokine's cellular source in 14 week Kras tumor-bearing lungs (n = 5). (C-F) Spontaneously arising lymphomas occurred at high frequency in K.RORγt animals, commonly impacting (C) thymus, (D) spleen, (E) liver, and (F) lung tissues. H&E sections, scale bars = 250 μm, except lung panel (F) = 500 μm. (G) Representative H&E images from 14-week Kras and Kras.Tgd mice. Scale bar = 500 μm. (H) Flow cytometric analysis of lungs from tumor-bearing Kras and Kras.Tgd mice (n = 6 each) at 14 weeks post-induction. Data for each cell type are displayed as the total number of live cells present within the mouse lung. Asterisks indicate P < 0.05. Figure 6 Tumor-associated inflammation in SCLC. (A) Rbp53 mice develop SCLC tumors within 1 year of AdCre exposure. H&E section, scale bar = 1 mm. (B) Flow cytometric analysis of lungs from tumor-bearing Rbp53 mice (n = 4) and non-Cre exposed control mice (n = 3), approximately 10 months after tumor induction. IHC staining reveals that CD45+ immune cells are generally located in the SCLC tumor periphery (arrowheads, C), with some clustering of cells into organized lymphoid structures (arrows, D). Some large CD45+ cells, likely macrophages, were observed in alveolar spaces (arrowheads, D) proximal to the tumor. Little to no leukocyte tumor infiltration was observed (E). Scale bars = 1 mm (C) and 100 μm (D, E). (F) The ratio of CD3+ T cells to sum myeloid population in SCLC mice was significantly increased compared to three ADCA models, as assessed by one-way ANOVA with Tukey's post-test. (G) The percentage of CD45+ live cells present in two resected specimens of human SCLC was greatly reduced compared to five human ADCA specimens. (H) Leukocyte population summary of the flow cytometric analyses, shown as percent of live cells, for a representative normal mouse lung, 10-week Egfr, Kras, and Kp53 ADCA and mouse SCLC are displayed. Abbreviations: NL, non-adjacent normal lung; Tu, tumor. Asterisks indicate P < 0.05. REFERENCES 1 Howlader N Krapcho M Garshell J Miller D Altekruse SF Kosary CL Yu M Ruhl J Tatalovich Z Mariotto A Lewis DR Chen HS Feuer EJ Cronin KA SEER Cancer Statistics Review 1975-2011 National Cancer Institute Bethesda, MD. 2 Peifer M Fernandez-Cuesta L Sos ML George J Seidel D Kasper LH Plenker D Leenders F Sun R Zander T Menon R Koker M Dahmen I Muller C Di Cerbo V Schildhaus HU Altmuller J Baessmann I Becker C de Wilde B Vandesompele J Bohm D Ansen S Gabler F Wilkening I Heynck S Heuckmann JM Lu X Carter SL Cibulskis K Banerji S Getz G Park KS Rauh D Grutter C Fischer M Pasqualucci L Wright G Wainer Z Russell P Petersen I Chen Y Stoelben E Ludwig C Schnabel P Hoffmann H Muley T Brockmann M Engel-Riedel W Muscarella LA Fazio VM Groen H Timens W Sietsma H Thunnissen E Smit E Heideman DA Snijders PJ Cappuzzo F Ligorio C Damiani S Field J Solberg S Brustugun OT Lund-Iversen M Sanger J Clement JH Soltermann A Moch H Weder W Solomon B Soria JC Validire P Besse B Brambilla E Brambilla C Lantuejoul S Lorimier P Schneider PM Hallek M Pao W Meyerson M Sage J Shendure J Schneider R Buttner R Wolf J Nurnberg P Perner S Heukamp LC Brindle PK Haas S Thomas RK Integrative genome analyses identify key somatic driver mutations of small-cell lung cancer. Nat. Genet 2012 44 1104 1110 22941188 3 Govindan R Page N Morgensztern D Read W Tierney R Vlahiotis A Spitznagel EL Piccirillo J Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database. J. Clin. Oncol 2006 24 4539 4544 17008692 4 Ding L Getz G Wheeler DA Mardis ER McLellan MD Cibulskis K Sougnez C Greulich H Muzny DM Morgan MB Fulton L Fulton RS Zhang Q Wendl MC Lawrence MS Larson DE Chen K Dooling DJ Sabo A Hawes AC Shen H Jhangiani SN Lewis LR Hall O Zhu Y Mathew T Ren Y Yao J Scherer SE Clerc K Metcalf GA Ng B Milosavljevic A Gonzalez-Garay ML Osborne JR Meyer R Shi X Tang Y Koboldt DC Lin L Abbott R Miner TL Pohl C Fewell G Haipek C Schmidt H Dunford-Shore BH Kraja A Crosby SD Sawyer CS Vickery T Sander S Robinson J Winckler W Baldwin J Chirieac LR Dutt A Fennell T Hanna M Johnson BE Onofrio RC Thomas RK Tonon G Weir BA Zhao X Ziaugra L Zody MC Giordano T Orringer MB Roth JA Spitz MR Wistuba II Ozenberger B Good PJ Chang AC Beer DG Watson MA Ladanyi M Broderick S Yoshizawa A Travis WD Pao W Province MA Weinstock GM Varmus HE Gabriel SB Lander ES Gibbs RA Meyerson M Wilson RK Somatic mutations affect key pathways in lung adenocarcinoma. Nature 2008 455 1069 1075 18948947 5 The Cancer Genome Atlas Research Network Comprehensive molecular profiling of lung adenocarcinoma. Nature 2014 511 543 550 25079552 6 Pao W Chmielecki J Rational, biologically based treatment of EGFR-mutant non-small-cell lung cancer. Nat. Rev. Cancer 2010 10 760 774 20966921 7 Morgensztern D Campo MJ Dahlberg SE Doebele RC Garon E Gerber DE Goldberg SB Hammerman PS Heist RS Hensing T Horn L Ramalingam SS Rudin CM Salgia R Sequist LV Shaw AT Simon GR Somaiah N Spigel DR Wrangle J Johnson D Herbst RS Bunn P Govindan R Molecularly targeted therapies in non-small-cell lung cancer annual update 2014. J. Thorac. Oncol 2015 10 S1 63 25535693 8 Lynch TJ Bondarenko I Luft A Serwatowski P Barlesi F Chacko R Sebastian M Neal J Lu H Cuillerot JM Reck M Ipilimumab in combination with paclitaxel and carboplatin as first-line treatment in stage IIIB/IV non-small-cell lung cancer: results from a randomized, double-blind, multicenter phase II study. J. Clin. Oncol 2012 30 2046 2054 22547592 9 Topalian SL Hodi FS Brahmer JR Gettinger SN Smith DC McDermott DF Powderly JD Carvajal RD Sosman JA Atkins MB Leming PD Spigel DR Antonia SJ Horn L Drake CG Pardoll DM Chen L Sharfman WH Anders RA Taube JM McMiller TL Xu H Korman AJ Jure-Kunkel M Agrawal S McDonald D Kollia GD Gupta A Wigginton JM Sznol M Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N. Engl. J. Med 2012 366 2443 2454 22658127 10 Borghaei H Paz-Ares L Horn L Spigel DR Steins M Ready NE Chow LQ Vokes EE Felip E Holgado E Barlesi F Kohlhaufl M Arrieta O Burgio MA Fayette J Lena H Poddubskaya E Gerber DE Gettinger SN Rudin CM Rizvi N Crino L Blumenschein GR Jr. Antonia SJ Dorange C Harbison CT Graf Finckenstein F Brahmer JR Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N. Engl. J. Med 2015 373 1627 1639 26412456 11 Reck M Bondarenko I Luft A Serwatowski P Barlesi F Chacko R Sebastian M Lu H Cuillerot JM Lynch TJ Ipilimumab in combination with paclitaxel and carboplatin as first-line therapy in extensive-disease-small-cell lung cancer: results from a randomized, double-blind, multicenter phase 2 trial. Ann. Oncol 2013 24 75 83 22858559 12 Coussens LM Werb Z Inflammation and cancer. Nature 2002 420 860 867 12490959 13 Hanahan D Coussens LM Accessories to the crime: functions of cells recruited to the tumor microenvironment. Cancer Cell 2012 21 309 322 22439926 14 Pardoll DM The blockade of immune checkpoints in cancer immunotherapy. Nat. Rev. Cancer 2012 12 252 264 22437870 15 Gabrilovich DI Nagaraj S Myeloid-derived suppressor cells as regulators of the immune system. Nat. Rev. Immunol 2009 9 162 174 19197294 16 Helissey C Champiat S Soria JC Immune checkpoint inhibitors in advanced nonsmall cell lung cancer. Curr. Opin. Oncol 2015 27 108 117 25602683 17 Coussens LM Zitvogel L Palucka AK Neutralizing tumor-promoting chronic inflammation: a magic bullet? Science 2013 339 286 291 23329041 18 Sparmann A Bar-Sagi D Ras-induced interleukin-8 expression plays a critical role in tumor growth and angiogenesis. Cancer Cell 2004 6 447 458 15542429 19 Ji H Houghton AM Mariani TJ Perera S Kim CB Padera R Tonon G McNamara K Marconcini LA Hezel A El-Bardeesy N Bronson RT Sugarbaker D Maser RS Shapiro SD Wong KK K-ras activation generates an inflammatory response in lung tumors. Oncogene 2006 25 2105 2112 16288213 20 Wislez M Fujimoto N Izzo JG Hanna AE Cody DD Langley RR Tang H Burdick MD Sato M Minna JD Mao L Wistuba I Strieter RM Kurie JM High expression of ligands for chemokine receptor CXCR2 in alveolar epithelial neoplasia induced by oncogenic kras. Cancer Res 2006 66 4198 4207 16618742 21 Akbay EA Koyama S Carretero J Altabef A Tchaicha JH Christensen CL Mikse OR Cherniack AD Beauchamp EM Pugh TJ Wilkerson MD Fecci PE Butaney M Reibel JB Soucheray M Cohoon TJ Janne PA Meyerson M Hayes DN Shapiro GI Shimamura T Sholl LM Rodig SJ Freeman GJ Hammerman PS Dranoff G Wong KK Activation of the PD-1 pathway contributes to immune escape in EGFR-driven lung tumors. Cancer Discov 2013 3 1355 1363 24078774 22 Lujambio A Akkari L Simon J Grace D Tschaharganeh DF Bolden JE Zhao Z Thapar V Joyce JA Krizhanovsky V Lowe SW Non-cell-autonomous tumor suppression by p53. Cell 2013 153 449 460 23562644 23 Menendez D Shatz M Resnick MA Interactions between the tumor suppressor p53 and immune responses. Curr. Opin. Oncol 2013 25 85 92 23150340 24 Politi K Zakowski MF Fan PD Schonfeld EA Pao W Varmus HE Lung adenocarcinomas induced in mice by mutant EGF receptors found in human lung cancers respond to a tyrosine kinase inhibitor or to down-regulation of the receptors. Genes Dev 2006 20 1496 1510 16705038 25 Tichelaar JW Lu W Whitsett JA Conditional expression of fibroblast growth factor-7 in the developing and mature lung. J. Biol. Chem 2000 275 11858 11864 10766812 26 Jackson EL Willis N Mercer K Bronson RT Crowley D Montoya R Jacks T Tuveson DA Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001 15 3243 3248 11751630 27 Jonkers J Meuwissen R van der Gulden H Peterse H van der Valk M Berns A Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer. Nat. Genet 2001 29 418 425 11694875 28 Eberl G Marmon S Sunshine MJ Rennert PD Choi Y Littman DR An essential function for the nuclear receptor RORgamma(t) in the generation of fetal lymphoid tissue inducer cells. Nat. Immunol 2004 5 64 73 14691482 29 Itohara S Mombaerts P Lafaille J Iacomini J Nelson A Clarke AR Hooper ML Farr A Tonegawa S T cell receptor delta gene mutant mice: independent generation of alpha beta T cells and programmed rearrangements of gamma delta TCR genes. Cell 1993 72 337 348 8381716 30 Marino S Vooijs M van Der Gulden H Jonkers J Berns A Induction of medulloblastomas in p53-null mutant mice by somatic inactivation of Rb in the external granular layer cells of the cerebellum. Genes Dev 2000 14 994 1004 10783170 31 DuPage M Dooley AL Jacks T Conditional mouse lung cancer models using adenoviral or lentiviral delivery of Cre recombinase. Nat. Protoc 2009 4 1064 1072 19561589 32 Houghton AM Quintero PA Perkins DL Kobayashi DK Kelley DG Marconcini LA Mecham RP Senior RM Shapiro SD Elastin fragments drive disease progression in a murine model of emphysema. J. Clin. Invest 2006 116 753 759 16470245 33 Nikitin AY Alcaraz A Anver MR Bronson RT Cardiff RD Dixon D Fraire AE Gabrielson EW Gunning WT Haines DC Kaufman MH Linnoila RI Maronpot RR Rabson AS Reddick RL Rehm S Rozengurt N Schuller HM Shmidt EN Travis WD Ward JM Jacks T Classification of proliferative pulmonary lesions of the mouse: recommendations of the mouse models of human cancers consortium. Cancer Res 2004 64 2307 2316 15059877 34 Jackson EL Olive KP Tuveson DA Bronson R Crowley D Brown M Jacks T The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 2005 65 10280 10288 16288016 35 Culley FJ Natural killer cells in infection and inflammation of the lung. Immunology 2009 128 151 163 19740372 36 Waldhauer I Steinle A NK cells and cancer immunosurveillance. Oncogene 2008 27 5932 5943 18836474 37 Groh V Wu J Yee C Spies T Tumour-derived soluble MIC ligands impair expression of NKG2D and T-cell activation. Nature 2002 419 734 738 12384702 38 Wiemann K Mittrucker HW Feger U Welte SA Yokoyama WM Spies T Rammensee HG Steinle A Systemic NKG2D down-regulation impairs NK and CD8 T cell responses in vivo. J. Immunol 2005 175 720 729 16002667 39 Esendagli G Bruderek K Goldmann T Busche A Branscheid D Vollmer E Brandau S Malignant and non-malignant lung tissue areas are differentially populated by natural killer cells and regulatory T cells in non-small cell lung cancer. Lung Cancer 2008 59 32 40 17825949 40 Mahmoud SM Paish EC Powe DG Macmillan RD Grainge MJ Lee AH Ellis IO Green AR Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer. J. Clin. Oncol 2011 29 1949 1955 21483002 41 Chang SH Mirabolfathinejad SG Katta H Cumpian AM Gong L Caetano MS Moghaddam SJ Dong C T helper 17 cells play a critical pathogenic role in lung cancer. Proc. Natl. Acad. Sci. U. S. A 2014 111 5664 5669 24706787 42 Ivanov II McKenzie BS Zhou L Tadokoro CE Lepelley A Lafaille JJ Cua DJ Littman DR The orphan nuclear receptor RORgammat directs the differentiation program of proinflammatory IL-17+ T helper cells. Cell 2006 126 1121 1133 16990136 43 Ueda E Kurebayashi S Sakaue M Backlund M Koller B Jetten AM High incidence of T-cell lymphomas in mice deficient in the retinoid-related orphan receptor RORgamma. Cancer Res 2002 62 901 909 11830550 44 Meuwissen R Linn SC Linnoila RI Zevenhoven J Mooi WJ Berns A Induction of small cell lung cancer by somatic inactivation of both Trp53 and Rb1 in a conditional mouse model. Cancer Cell 2003 4 181 189 14522252 45 Chen Z Fillmore CM Hammerman PS Kim CF Wong KK Non-small-cell lung cancers: a heterogeneous set of diseases. Nat. Rev. Cancer 2014 14 535 546 25056707 46 Guo JY Karsli-Uzunbas G Mathew R Aisner SC Kamphorst JJ Strohecker AM Chen G Price S Lu W Teng X Snyder E Santanam U Dipaola RS Jacks T Rabinowitz JD White E Autophagy suppresses progression of K-ras-induced lung tumors to oncocytomas and maintains lipid homeostasis. Genes Dev 2013 27 1447 1461 23824538 47 Xu C Fillmore CM Koyama S Wu H Zhao Y Chen Z Herter-Sprie GS Akbay EA Tchaicha JH Altabef A Reibel JB Walton Z Ji H Watanabe H Janne PA Castrillon DH Rustgi AK Bass AJ Freeman GJ Padera RF Dranoff G Hammerman PS Kim CF Wong KK Loss of Lkb1 and Pten leads to lung squamous cell carcinoma with elevated PD-L1 expression. Cancer Cell 2014 25 590 604 24794706 48 Tran E Ahmadzadeh M Lu YC Gros A Turcotte S Robbins PF Gartner JJ Zheng Z Li YF Ray S Wunderlich JR Somerville RP Rosenberg SA Immunogenicity of somatic mutations in human gastrointestinal cancers. Science 2015 350 1387 1390 26516200 49 Subramanian J Govindan R Lung cancer in never smokers: a review. J. Clin. Oncol 2007 25 561 570 17290066 50 Govindan R Ding L Griffith M Subramanian J Dees ND Kanchi KL Maher CA Fulton R Fulton L Wallis J Chen K Walker J McDonald S Bose R Ornitz D Xiong D You M Dooling DJ Watson M Mardis ER Wilson RK Genomic landscape of non-small cell lung cancer in smokers and never-smokers. Cell 2012 150 1121 1134 22980976 51 Ganesan AP Johansson M Ruffell B Yagui-Beltran A Lau J Jablons DM Coussens LM Tumor-infiltrating regulatory T cells inhibit endogenous cytotoxic T cell responses to lung adenocarcinoma. J. Immunol 2013 191 2009 2017 23851682 52 McAllister F Bailey JM Alsina J Nirschl CJ Sharma R Fan H Rattigan Y Roeser JC Lankapalli RH Zhang H Jaffee EM Drake CG Housseau F Maitra A Kolls JK Sears CL Pardoll DM Leach SD Oncogenic Kras activates a hematopoietic-to-epithelial IL-17 signaling axis in preinvasive pancreatic neoplasia. Cancer Cell 2014 25 621 637 24823639 53 Coffelt SB Kersten K Doornebal CW Weiden J Vrijland K Hau CS Verstegen NJ Ciampricotti M Hawinkels LJ Jonkers J de Visser KE IL-17-producing gammadelta T cells and neutrophils conspire to promote breast cancer metastasis. Nature 2015 522 345 348 25822788 54 Hannani D Ma Y Yamazaki T Dechanet-Merville J Kroemer G Zitvogel L Harnessing gammadelta T cells in anticancer immunotherapy. Trends Immunol 2012 33 199 206 22364810 55 Dieu-Nosjean MC Antoine M Danel C Heudes D Wislez M Poulot V Rabbe N Laurans L Tartour E de Chaisemartin L Lebecque S Fridman WH Cadranel J Long-term survival for patients with non-small-cell lung cancer with intratumoral lymphoid structures. J. Clin. Oncol 2008 26 4410 4417 18802153 56 Joshi NS Akama-Garren EH Lu Y Lee DY Chang GP Li A DuPage M Tammela T Kerper NR Farago AF Robbins R Crowley DM Bronson RT Jacks T Regulatory T cells in tumor-associated tertiary lymphoid structures suppress anti-tumor T cell responses. Immunity 2015 43 579 590 26341400 57 van Meerbeeck JP Fennell DA De Ruysscher DK Small-cell lung cancer. Lancet 2011 378 1741 1755 21565397 58 Ott PA Fernandez MEE Hiret S Kim D-W Moss RA Winser T Yuan S Cheng JD Piperdi B Mehnert JM Pembrolizumab (MK-3475) in patients with extensive-stage small cell lung cancer: Preliminary safety and efficacy results from KEYNOTE-028 [abstract]. J. Clin. Oncol 2015 33 supp abstract nr 7502 59 Garon EB Rizvi NA Hui R Leighl N Balmanoukian AS Eder JP Patnaik A Aggarwal C Gubens M Horn L Carcereny E Ahn MJ Felip E Lee JS Hellmann MD Hamid O Goldman JW Soria JC Dolled-Filhart M Rutledge RZ Zhang J Lunceford JK Rangwala R Lubiniecki GM Roach C Emancipator K Gandhi L K.-. Investigators Pembrolizumab for the treatment of non-small-cell lung cancer. N. Engl. J. Med 2015 372 2018 2028 25891174 60 Suzuki K Kachala SS Kadota K Shen R Mo Q Beer DG Rusch VW Travis WD Adusumilli PS Prognostic immune markers in non-small cell lung cancer. Clin. Cancer Res 2011 17 5247 5256 21659461 61 Vogelstein B Papadopoulos N Velculescu VE Zhou S Diaz LA Jr. Kinzler KW Cancer genome landscapes. Science 2013 339 1546 1558 23539594

PMC005xxxxxx/PMC5116263.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9001516 21213 Trends Endocrinol Metab Trends Endocrinol. Metab. Trends in endocrinology and metabolism: TEM 1043-2760 1879-3061 27623245 5116263 10.1016/j.tem.2016.08.003 NIHMS812039 Article Linking the microbiota, chronic disease and the immune system Hand Timothy W. *1 Vujkovic-Cvijin Ivan 2 Ridaura Vanessa K. 2 Belkaid Yasmine 23 1 R.K. Mellon Institute for Pediatric Research, Children’s Hospital of Pittsburgh of UPMC, University of Pittsburgh, Pittsburgh, PA, 15224 2 Mucosal Immunology Section, Laboratory of Parasitic Diseases, NIAID/NIH, Bethesda, Maryland 20892, USA 3 National Institute of Allergy and Infectious diseases (NIAID) Microbiome Program, National Institutes of Health (NIH), Bethesda, Maryland 20892, USA * Correspondence addressed to: Timothy Hand (timothy.hand@chp.edu) or Yasmine Belkaid (ybelkaid@niaid.nih.gov) 23 8 2016 10 9 2016 12 2016 01 12 2017 27 12 831843 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Chronic inflammatory diseases are the most important causes of mortality in the world today and are on the rise. We now know that immune-driven inflammation is critical in the etiology of these diseases, though the environmental triggers and cellular mechanisms that lead to their development are still mysterious. Many chronic inflammatory diseases are associated with significant shifts in the microbiota towards inflammatory configurations, which can affect the host both by inducing local and systemic inflammation and by alterations in microbiota-derived metabolites. This review discusses recent findings suggesting that shifts in the microbiota may contribute to chronic disease via effects on the immune system. Microbiota dysbiosis metabolic syndrome inflammasome IMMUNITY, THE MICROBIOTA AND CHRONIC INFLAMMATORY DISEASE In the past decade our knowledge of chronic inflammatory disease (CID) has been transformed by a new understanding of the central role of the immune system in driving pathology. The immune system functions as a ‘rheostat for the entire body; seeking out physiological disturbances, (infectious or otherwise), and rectifying them via both inflammatory immune responses, but also critical repair/regulatory processes such as the clearance of dead cells and the restoration of the physical barrier [1, 2]. Thus, many of the most chronic and debilitating diseases afflicting humankind can be seen as a shift in the immune response away from repair/regulation, and towards immune-driven inflammatory responses. Some of these diseases are also associated with a significant shift in the composition of the microbiota [3–7]. The microbiota is a significant source of both nutritional metabolites and inflammatory innate immune signals [8]. Therefore, compositional shifts in the microbiota are of potential importance as modifiers and triggers of disease, in the context of genetic predisposition and environmental factors. CID can encompass both autoimmune disease, such as systemic lupus erythematosus or rheumatoid arthritis, that are obviously associated with defects in the immune system and responses against self antigens in addition to auto-inflammatory disorders such as metabolic syndrome and cardiovascular disease, that have more recently been associated with immune-driven inflammation and are not dependent upon autoantigen recognition. In this review. we will focus more on links between the microbiota, immunity and the non-autoimmune inflammatory diseases; though a link between the microbiota and many autoimmune diseases is undeniable and reviewed elsewhere [9, 10]. The incidence of many CID is on the rise worldwide, without a clearly defined explanation. It has been hypothesized that this rise in chronic inflammatory and metabolic diseases is due, at least in part, to a significant narrowing of bacterial diversity associated with the evolutionarily recent phenomena of large scale urban living [11]. Antibiotic usage, dietary modifications and infectious diseases all play important roles in causing alterations in the host’s associated microbiota. In this review we will discuss recent evidence linking CID with shifts in both local and systemic immune responses and disease promoting configurations of the microbiota. IMMUNE/MICROBIOTA INTERACTION Humans exist as metaorganisms consisting of host cells in addition to a vast consortium of microbial organisms that live on all of our barrier tissues. The highest density of organisms lives in the intestine, and these organisms provide a tremendous benefit to the host via the enzymatic processing of complex dietary constituents, such as fiber, into metabolites digestible by the host, in addition to many other enzymatic functions. Often these enzymatic pathways can involve multiple genes present in separate bacterial clades [12] and it is perhaps more appropriate to measure those genes that are broadly conserved and critical for function, known as the core metagenome, rather than individual strains of microbes whose maintenance is much more volatile [13]. As a result, a narrowing of organismal diversity may also lead to a loss of components of the metagenome, inducing possible deficits in functionality. Since there is great variation in the composition of the microbiota at lower phylogenetic levels, and significant gaps in our knowledge of the specific function of many strains, ‘dysbiosis’ of the microbiota is sometimes difficult to define and is contextual to the individual and their current state of health. Indeed it has been shown that the glycemic response to any particular nutritional input is unique to the individual and dependent upon the microbiota [14]. This study shows that although we have begun to understand how the host, their diet and their associated microbiota work together to influence health, the details appear to be quite complex and individualized. A microbiota that is ‘dysbiotic’ in one individual may be perfectly healthy in the context of different behavior and genetic predisposition. Without a more fundamental understanding of the core metagenome and how it affects health ubiquitously and comprehensively, describing ‘dysbiosis’ is a difficult task. Therefore, while it is undeniable that the microbiota can shift into states that promote inflammation, (thereby meeting the definition of dysbiosis) whether these states are universal is not clear, and thus we will forgo use of the term dysbiosis, in favor of describing those configurations of the microbiota associated to disease. The microbiome impacts multiple organ systems, including host immunity. The immune system is responsible for maintaining tissue homeostasis in the intestines as it must remain vigilant against invasive microbes while limiting overt inflammatory responses against the vast array of benign organisms that make up the microbiota [15]. To maintain this perilous relationship with the microbiota, the host has developed a series of immune mechanisms that severely limit and regulate the interaction between intestinal epithelia and the microorganisms that reside within the lumen [16]. This series of ‘checks and balances’ is absolutely critical to host health, because a breakdown of the intestine’s barrier function and immune homeostasis can contribute to inflammatory bowel disease (IBD) in addition to many other chronic diseases [6, 7, 17–19]. The host has also evolved multiple products that foster a healthy and diverse microbiota, including the secretion of IgA and bile acids and the fucosylation of the intestinal epithelium [20–22]. The exact mechanisms of these host factors in shaping the microbiota are not clear but are complex in that they both limit growth of some bacteria while benefiting others and are therefore part of a larger effort to maintain health in the microbiome [23]. There are multiple ways in which the immune system surveys the microbiota to detect alterations. One of the primary ways is through tonic sensation via innate immune pattern recognition receptors. Indeed, steady-state innate immune signaling via specific components of Bacteroides spp. have been shown to assist the development of a healthy immune system and antagonize the development of autoinflammatory disorders, such as asthma [24–26]. However, we now understand that the immune system also measures ‘keystone’ metabolites to determine the presence of a functional microbiota. This idea has been best demonstrated by several studies showing that Short-Chain Fatty Acids (SCFAs), which are metabolites derived from the microbiota-dependent breakdown of dietary fiber and are a key energy source for enterocytes, support immune homeostasis in the gastrointestinal tract. Specifically, SCFAs promote the induction and maintenance of regulatory Foxp3+ T cells (Tregs) in the colon and can limit experimentally induced colitis [27–29]. Critically, SCFA can also act systemically, leading to the diminishment of innate cell responses and pathology in a model of asthma [30, 31]. Similarly, a vitamin A metabolite, retinoic acid, whose production has been shown to be partly controlled by the microbiota, also supports immune function in a multitude of ways [32]. Microbiota-derived metabolites have also been shown to modulate the immune response through activation of the inflammasome and the production of IL-18 and anti-microbial peptides from enteric cells [33]. In addition to sensing of microbe-associated molecular patterns (MAMPs) and keystone metabolites, immune cells themselves rely on different metabolites as energy sources for their growth and survival. The most pertinent example of this idea is the preference of adipose tissue and fat metabolism for regulatory T cells (Tregs) and M2 anti-inflammatory macrophages that, as will be discussed in detail, may be critical to the etiology of metabolic syndrome [34]. As the microbiota affects the availability and absorption of nutrients via both their own enzymatic capabilities, and the modification of host products such as bile acids; this is an important and often overlooked effect of the microbiota on the development of disease [35]. THE MICROBIOTA AND CHRONIC DISEASES OF THE MUCOSA The intestine is home to the largest and most metabolically active microbial community in the human body. Therefore, it is perhaps unsurprising that the predominant inflammatory diseases of the gut, Crohn’s Disease (CD) and Ulcerative Colitis (UC) are intertwined with shifts in the microbiota. In both diseases, inflammatory immune responses in the intestine against the microbiota induce shifts towards more aggressive members of the population. The most salient example of this effect is the outgrowth of gram negative Proteobacteria, in particular the family Enterobacteriaceae in a subset of patients with IBD [3]. CD specifically has been associated with the outgrowth of opportunistic ‘Adherent-Invasive’ E. coli that can complicate the disease [36]. Additionally, IBD is also associated with a loss of benign gram positive Clostridia, such as Faecalbacterium prausnitzii, which are associated with health and the provision of metabolites such as SCFA from food [37]. Multiple animal models now show that a transmissible microbiota can predispose to IBD susceptibility [38, 39]. However, given the inconsistent efficacy of fecal transplant and antibiotic treatment for IBD it would be a significant overstatement to suggest that the microbiota is sufficient to induce the disease [19, 40–42]. One view on the etiology of these diseases is that barrier and immune dysfunction leads to an outgrowth of those organisms capable of withstanding the intestinal inflammatory response [43]. Indeed, blooms of E. coli are enabled by their ability to use inflammatory nitrogenous compounds as metabolites [44]. Unfortunately, the modules associated with survival in inflamed environments tend to be found in organisms with invasive capabilities and the outgrowth of these organisms lead to a further exacerbation of immune activation and inflammation [45–47]. For example, Clostridium difficile infection is a common complication of patients with IBD [48]. It stands to reason that the most inflammatory organisms are more likely to induce an IgA antibody response in the intestine [49]. Thus the fact that the IgA bound fraction of intestinal bacteria from IBD patients is sufficient to predispose to disease in a mouse model is excellent evidence for the hypothesis that a shift to an inflammatory configuration contributes to development of the disease [50]. However, interestingly none of the organisms discussed above was enriched in the IgA bound fraction of bacteria from IBD patients, and there was no ‘core’ IgA-bound microbiome, perhaps indicating that a loss of diversity in the ‘core’ microbiome in general is more important than the outgrowth of any particular organism. The model for chronic inflammation at barrier sites, wherein genetic predisposition, immune-driven inflammation and an altered microbiota combine to drive disease, is not unique to the intestine. One clear example is Cystic Fibrosis where defective salt transport leads to thickened mucus, reduced function in lung phagocytes and disrupted mucociliary transport, which allows for the overgrowth of Proteobacteria on all of the patient’s mucosal surfaces, most notably their lungs, leading to chronic inflammation and damage to the alveoli [51]. In addition, Atopic Dermatitis is associated with immune responses against Staphylococcus, explaining perhaps why antibacterial wraps are effective in combating severe cases [52]. THE MICROBIOTA, OBESITY AND METABOLIC SYNDROME Obesity and associated pathologies, including Type II Diabetes (T2D) and dyslipidemia, are a worldwide epidemic and affect an increasing number of people each year. The mechanism of T2D is now believed to be a block in insulin signaling driven by inflammatory cytokines [53]. In lean individuals, the fat is dominated by M2 macrophages, eosinophils, group 2 innate lymphoid cells and Tregs, all of which contribute in their own way to adipose tissue homeostasis [34]. In obese individuals, the adipose tissue is profoundly altered as it becomes invaded by inflammatory M1 macrophages that secrete large amounts of TNFα, IL-1β and IL-6, all of which contribute to an inhibition of PI3K/Akt signaling downstream of the insulin receptor [54]. Blockade of these inflammatory cytokines has shown modest efficacy in restoring insulin sensitivity in these patients, providing further support for this hypothesis [53]. Thus T2D is a metabolic disease whose symptoms are driven by alterations to the immune response, and identifying the underlying stressor would be of tremendous benefit to the development of novel therapeutics. It is undeniable that in most cases, the primary cause of these diseases is increased access to high calorie foods, in particular foods high in fat and simple sugars. Gordon and colleagues have now established that obesity is also associated with a configuration of the microbiota that has an increased fraction of Firmicutes, reduced Bacteroidetes, and is capable of providing additional calories to the host for a given caloric intake [5] (Figure 1). Interestingly, the obese microbiota is both a symptom and a contributor to disease, as transfer of this microbiota to germ-free mice leads to significant increases in adiposity in recipients [55]. Building from these studies, it has been demonstrated that the microbiota of obese patients actually produces significantly less SCFA than that of lean identical twins [56]. Whether these SCFAs are contributing to regulation of inflammation via dampening of the immune response through Tregs and innate immune cells, or via regulating other aspects of metabolism or satiety is not entirely clear, but will be an important area for future research. The microbiota can also block adiposity. Indeed, an additional finding of these studies is that gnotobiotic mice transplanted with the microbiota of obese patients can be invaded by a healthy “lean microbiota”, if the mice are placed on a standard diet low in fat [56]. As well, the presence of mucinophilic bacteria, Akkermansia muciniphila, has been associated with a reduction of inflammation and protection from T2D, in mouse models [57, 58] (Figure 1). Therefore, while the microbiota of patients with obesity resists weight loss, it is amenable to change and as such may represent a potential target for treatment. Beyond nutrient provision, the microbiota can also directly contribute to inflammatory cytokine expression at the core of T2D, via activation of the immune system. In mouse models, this has been clearly shown by the phenotype of TLR5 knockout mice. TLR5 knockout mice show a strikingly increased adiposity, insulin resistance and inflammatory cytokine production, while raised on standard mouse chow [59]. TLR5 is the innate immune receptor for bacterial flagellin, and mice deficient in this receptor have significantly increased populations of Enterobacteriaceae that can transmit many of the phenotypes in wild-type mice via fecal transfer [60] (Figure 1). Interestingly, the outgrowth of Enterobacteriaceae has also been associated with obesity and T2D in humans, though this finding is confounded by the effects of the diabetes medication, metformin [4, 61]. Much like CD, obesity, though not metabolic syndrome specifically, is correlated with reduced microbial diversity [62, 63]. Therefore, a shift towards a more invasive and inflammatory microbiota that induces changes to the immune milieu of the adipose compartment from M2 to M1 macrophage may contribute to the development of metabolic syndrome [64]. Indeed, a number of studies have linked obesity with a ‘leaky gut’, as measured by an increase in serum LPS, and increased intestinal adherence of Enterobacteriaceae [18, 65, 66]. In animal models, provision of a high fat diet and associated affects on the microbiota are sufficient to induce a significant increase in IFNγ expressing Th1 T cells in the small intestine which may traffic to associated adipose depots and directly affect inflammation [67]. The exact mechanism of Proteobacteria outgrowth in obesity and T2D remains unknown and will be an important area for future research. The effect of the microbiota on metabolic disease has also been demonstrated by experiments on animal models of liver disease and cirrhosis. The liver, via the hepatic portal vein, is directly downstream of the small intestine and acts as a barrier to the further trafficking of microbiota-derived products [68]. Mice deficient in components of the inflammasome harbor an altered microbiota with inflammatory potential [69]. When these animals are placed on specific diets associated with non-alcoholic steatohepatitis (NASH) their disease is pronounced and transmissible via the microbiota to co-housed animals [17]. Increased liver damage is dependent upon TLR signaling and TNFα production, and is associated with heightened presence of TLR ligands in the hepatic portal vein. Similar phenomena involving increased translocation of bacterial products may also contribute to other forms of cirrhosis due to alcohol abuse or as a complication of Cystic Fibrosis [70–72]. The microbiota and the immune system may also contribute to dyslipidemia and the cellular composition of the adipose tissue in a metabolically important way. The fat depots exists in at least two distinct states: (i) ‘white’ which functions to store energy, and (ii) brown/beige fat, which is critical for thermogenesis. Brown fat is present at birth, whereas beige fat develops later in life [73]. Instead of storing energy, beige adipocytes metabolize lipids rapidly to create heat and therefore is typically inversely correlated with adiposity throughout the body [74]. Interestingly, the presence of beige fat is supported directly via the local production of IL-4, IL-13, IL-25 and IL-33 from innate lymphoid cells, eosinophils, macrophages and stroma [75–77]. Since inflammation causes such a switch from from M2 to M1 immune cells downstream effects on adipocytes may explain with it is so detrimental to metabolic health. The microbiota does seem to also play a role in the decision to switch on a transcriptional program associated with brown fat in white adipose tissue, in murine models of hypothermia, where fat ‘beiges’ for heat production. Interestingly, the transfer of the microbiota from hypothermic mice is sufficient to induce beige fat in the transplant recipient [78, 79]. However, in this context, it is unclear whether the mechanism of hypothermic beiging is associated with immune cytokine production and adipose tissue inflammatory state. THE MICROBIOTA AND CARDIOVASCULAR DISEASE Cardiovascular disease (CVD) and in particular atherosclerosis is the most common cause of death in high-income countries. The root cause of atherosclerosis is blockages in coronary arteries, which is caused by plaques formed of fat deposits and the development of fat laden macrophages called foam cells. Development of atherosclerosis has long been associated with diets high in animal fats, cholesterol and red meat. Recently, Hazen and colleagues have identified that a chemical derivative of red meat, trimethylamine-N-oxide (TMAO) is a significantly more reliable indicator of risk for CVD than cholesterol, and may drive disease by effects on platelets [80, 81]. TMAO is the product of the bacterial breakdown of meat-derived compounds such as L-carnitine and choline, and the microbiota is absolutely necessary for its production, as germ-free mice are free of TMAO, even when fed a diet high in choline [82]. Plasma levels of TMAO can be correlated with specific taxa of the microbiota, but it will only be through a better understanding of the metagenome of patients suffering CVD that we will be able to understand if there is a particular configuration of the microbiome that contributes to CVD [83]. It is possible that the process of converting compounds to TMAO has some benefit for the microbes that carry it out, and that these organisms would be selected for by a diet high in meat. Indeed, the microbiome of vegans has a significantly reduced ability to convert L-carnitine to TMAO [83]. Chronic HIV infection provides another example of a human disease in which CVD risk is elevated, with a possible contributory role for the gut microbiota. HIV-infected subjects exhibit multiple metabolic symptoms, but critically, a significant increase in the incidence of CVD [84]. HIV infection is associated with an increased abundance of invasive, pro-inflammatory Proteobacteria in the gut microbiota including Enterobacteriaceae, along with a depletion of members of the SCFA-producing Clostridia clade; a microbial profile that superficially resembles that of IBD [85–87]. Possibly related to this shift in the microbiota, HIV-infected subjects also exhibit impaired mucosal barrier function, translocation of microbial products into systemic circulation, and increased markers of innate immune activation associated with risk for the development of CVD [88]. Indeed, the degree to which the HIV-associated gut microbiota profile is observed correlates with markers of circulating LPS as well as inflammatory biomarkers of CVD risk [85, 87, 89–91]. CVD may also be associated with shifts in the oral microbiota. Human periodontitis is characterized by an altered and expanded oral bacterial community [92]. In periodontitis patients, markers of systemic exposure to periodontal bacteria are elevated and correlated with increased risk of CVD across several cohorts [93, 94]. Indeed, mouse models reveal that atherogenesis is accelerated by oral colonization with the periodontitis-associated bacterium, Porphyromonas gingivalis and DNA from oral bacteria has been found in atherosclerotic plaques in mice and humans [95–97]. MODERN LIFE AND CHANGING HOST/MICROBIOTA RELATIONSHIPS As discussed, a multitude of evidence now suggests that shifts in the microbiota towards inflammatory and low diversity states can contribute to chronic disease. One of the primary ways that modern life has changed our relationship with our resident bacteria is the advent of antibiotics. The invention of antibiotics sparked a revolution that fundamentally changed morbidity and mortality associated with bacterial infections. However, it can be argued that antibiotics have been overused as palliatives instead of therapeutics and in agriculture, where they promote rapid weight gain in livestock [98]. Recently, concerns have been raised that the increased use of antibiotics may be contributing to the rise of metabolic syndrome [11]. Essentially, this hypothesis posits that the same effect that induces increased size in cows and pigs is acting on children via constitutive exposure to trace antibiotics present in meat and milk, and early-life administration to combat infection. In support of this idea, mice fed low dose antibiotics early in life grow 10% larger, and the difference is largely due to increases in adipose tissue [99, 100]. These differences in growth could be transferred with the microbiota and seemed to be associated to shifts in the mucosal immune response [100]. Importantly, after cessation of antibiotic treatment, the microbiota, which had deviated significantly from controls, returned to normal, implying that the negative impact of alteration in microbiota can be subsequently sustained by the immune system [100]. Although not formally shown, one might suspect that these immune effects are penetrating beyond the gastrointestinal compartment and acting upon adipose tissue homeostasis. It is not well understood why the microbiota that is resistant to β lactam antibiotics [100], would be more inflammatory and prone to the induction of metabolic issues. One intriguing possibility is that the most inflammatory members of the microbiota tend to be ‘successful colonizers’ with broad host ranges requiring genetic adaptability and therefore may be more accepting to horizontal gene transfer such as antibiotic resistance. Indeed, it has been shown that Enterobacteriaceae of different genera exchange plasmids at a very high rate in vivo under inflammatory conditions [101]. Another fundamental way that those in high-income countries may have developed altered relationship to the microbiota is through diet and exercise. Our ancestors likely ate diets high in fibrous plant material, and had limited access to the saturated fat and simple sugars that represent today the majority of calories in the ‘Western’ diet. As has already been discussed in this review, the Western diet can lead to metabolic syndrome and obesity and these effects are in part due to shifts in the microbiota that control nutrient uptake and inflammation in the adipose tissue. In addition, the western lifestyle, which tends to less physical activity contributes to shifts in the microbiota as well [102]. Western diets might be also selecting away from those organisms that are more closely adapted to the mammalian gut and dependent upon complex carbohydrates [103]. For example, the breakdown of complex carbohydrates to SCFA requires a cascade of enzymes that are not always present in a single organism and can require the cooperation of an ecological community [12]. The Western diet, taken to extremes, presents a situation where there is no selection for cooperative breakdown of complex molecules and only competition for simple sugars and energy rich fats. One could hypothesize that this situation would benefit the more aggressive and inflammatory members of the microbiota, in particular, Proteobacteria, who can divide in as little as twenty minutes and rely heavily on mono and disaccharides as carbon sources [104], and select against organisms that prefer complex carbohydrates as a carbon source. For example, a diet high in milk fat has been shown to favor the outgrowth of Deltaproteobacteria and predispose to significantly exacerbated disease in the IL-10 knockout mouse model of IBD [20]. Thus, the western diet may not only shift the microbiota towards increased provision of calories, but also may benefit those organisms that are least symbiotic and prone to inflammatory outgrowth. Worldwide urban life amongst millions of other human beings is a modern construction that we have not evolved to deal with. One of the primary ways that life has changed because of urbanization is infection. As opposed to our ancestors, who were beset by chronic parasitic infections and intermittent outbreaks of zoonoses [105, 106], our current population sizes and global travel patterns supports multiple endemic pathogens. Fortunately, in the last century the combined efficacy of sanitation, public health, antibiotics and vaccines have largely mitigated the lethal consequences of these pathogens but it is undeniable that the frequency and biological effect of infection has been profoundly changed. A primary example of this effect is the worldwide reduction in enteric helminth infection. Intestinal helminths used to be a ubiquitous of human life, but in high-income countries are now quite rare. ‘De-worming’ has obvious benefits, as infections with high burdens of worms can lead to malnutrition and anemia [107]. However, our immune system evolved in the context of exposure to these organisms and as a part of the ‘Hygiene Hypothesis’ their absence is believed to contribute to the rise of CID [108, 109]. Multiple findings in animal models have shown that co-infection with helminthes can affect the immune system both systemically and locally at the mucosal surface [110]. For example, helminth infection can support the production of Tregs at mucosal surfaces, which may prevent the development of food allergy and temper the immune response to viruses via shifting from type I to type II immunity, limiting immunopathology and aiding the healing process [111, 112]. Helminth infection can also promote immunoregulation via induction of increased SCFA production from the microbiota [113]. Animal models of IBD and chronic diarrhea have shown that transient helminth infection may represent an effective treatment [114, 115]. Trials in human patients with UC have been promising, but the identification of the products that drive regulatory immune functions is critical, because infection with live worms present obvious limitations [116]. Additionally, identifying direct host affects from those that act through the microbiota could also assist in developing the most effective therapeutics. Although improved sanitation has reduced the incidence of enteric infections, they remain a significant burden in high-income countries. For instance, the average child in North America will experience ten diarrheal episodes by their 5th birthday and children in low-income countries are exposed to an even higher number of enteric infections [117, 118] Enteric infection poses an issue for the mucosal immune system, because the microbiota and inciting organisms are not always easily discriminated and it has been hypothesized that infections could be triggering events for CID [119]. Indeed, in many cases bacterial pathogenicity is contextual to the complex relationship between the bacteria, the host and the microbiota. Multiple laboratories have now shown that infection causes a significant shift in the microbiota and in general, these shifts mimic those seen in IBD, characterized by increases in Proteobacteria, specifically Enterobacteriaceae, and a depletion of Firmicutes [45–47, 120]. Given that many enteric infections belong to the same phylogenetic groups as prominent members of the microbiota, a critical question is how the immune system discriminates between opportunistic members of the microbiota and the initiating infectious organism. This is made all the more difficult by the fact that many infections break the epithelial barrier leading to translocation of the microbiota into the host [47, 121, 122]. Recent work indicates that the immune system does not discriminate between the microbiota and the infectious organism and activates microbiota-specific T cells in a comparable manner to those specific to the pathogen [123]. How these microbiota-specific T cells are maintained long-term and possibly contribute to chronic disease is an active area of new research. Infection with enteric pathogens such as Yersinia pseudotuberculosis can also cause damage and induce ‘immunological scarring’, which can chronically affect homeostatic intestinal immune responses by deviating lymphatic traffic away from the lymph nodes and into the fat [124]. Fascinatingly, oral Yersinia infection is also associated to chronic pro-inflammatory shifts in the microbiota, possibly indicating a link with immunological scarring [125]. As well, it is interesting to consider whether this kind of atypical traffic of myeloid cells activated at barrier surfaces into adipose depots is somehow contributing to metabolic disease and IBD. Indeed, CD is characterized by the ‘wrapping’ of the gut with creeping fat, the presence of bacteria within the adipose tissue and the production of inflammatory adipokines [126, 127]. Thus taken together, these three effects of infection; shifts in the microbiota towards more inflammatory organisms, microbiota-specific memory, and immunological scarring, provides a framework within which current or cleared infections can act as the inciting event of a CID [119]. An example in which one or more of these microbiota-dependent factors may be acting, is Environmental Enteropathy (EE) in children. EE is a disease that is prevalent in sub-Saharan Africa and parts of the Indian subcontinent wherein malnutrition and poor sanitation allow for chronic infection leading to a malabsorption syndrome [128]. Recent studies have identified that Enterobacteriaceae and subsequent anti-microbiota immune responses are critical players in the etiology this disease [129, 130]. Fascinatingly, it took two years after restoration of a healthy diet and abatement of chronic enteric infection, for patients with EE to experience full restoration of intestinal absorption, implying that either immune memory against the microbiota and immunological scarring could be maintaining the phenotype [131]. CONCLUDING REMARKS AND FUTURE PERSPECTIVES It is now clear that the microbiota acts as a genetically distinct organ that is critical for the enzymatic digestion of food, promotion of immune homeostasis and prevention of enteric infection. Just like the host-derived organs of the body, the microbiota can be damaged, or, as commonly described, become dysbiotic, leading to a negative impact on dependent host systems. Alterations to the microbiota can certainly contribute to pathology and in cases where these changes include the outgrowth of opportunistic and innately inflammatory bacteria, complicate many diseases (see ‘Outstanding Questions’ and Figure 2). Many studies on the microbiota describe associations to a disease state, and lack clear causative mechanisms. Thus, in order to achieve experimental control, many microbiota studies are carried out in inbred mouse models, which lack both genetic diversity and life history that would fundamentally shape the interaction between the immune system and the microbiota. Additionally, experiments done in germ-free and antibiotic treated mice can be complicated by defects in immune development. This is important because a patient’s history of infection, pathology and treatment may also be critical to understanding their current disease state. Going forward, it is only through a more holistic understanding of the individualized genetic (host and microbial) and environmental triggers of complex diseases that we will be able to cure them. The authors would like to apologize that due to length requirements, not all work in this burgeoning field could be discussed and properly cited. This work was supported by NIAID K22 AI108719 (T.W.H) and the NIH intramural program (V.K.R, I.V-C and Y.B.). The authors would like to thank K. Gopalakrishna and J. Tometich for critical reading of the manuscript. Figure 1 The microbiota affects metabolic syndrome via the immune system a) Obesity is associated with an increase in Firmicutes and a decrease in Bacteroidetes. Firmicutes provide an increased amount of calories to the host by increased harvest of energy from the diet. b) Outgrowth of Proteobacteria is associated with metabolic syndrome and has been shown to increase the frequency of IFNγ-producing T cells in the host, which in turn is associated with increased concentrations of serum LPS. Serum LPS and IFNγ may then drive the development of pro-inflammatory M1 macrophages in the adipose tissue. M1 macrophages express significantly higher amounts of TNFα and IL-1β than the resident M2 macrophages of the gut, and both cytokines contribute to insulin resistance. Beige adipose cells further contribute to health by metabolizing lipids to heat instead of storing them and are supported directly and indirectly by the cytokines IL-4, IL-13, IL-25 and IL-33. c) Mucophilic bacteria Akkermansia combat many of the effects of Proteobacteria outgrowth, including IFNγ production, and can alleviate symptoms associated with metabolic syndrome in animal models. Figure 2 Potential linkages between the environment, microbiota, immune system and chronic inflammatory disease At homeostasis, the microbiota assists the host in converting the diet into metabolites that foster a healthy host/microbiota relationship. These metabolites bolster the barrier between the host and the microbiota, preventing systemic immunity and inflammation. Disruption of the microbiota due to environmental factors (such as diet, infection, antibiotics etc.) or host factors can lead to a narrowing of microbial diversity and a shift in the metabolites derived from the microbiota. Shifts in this relationship can also lead to increased invasion of host tissue by bacteria and bacterial products. Together, a shift in metabolites and increased translocation of microbial products is believed to contribute to immune activation at the core of chronic inflammatory disease (IBD = Inflammatory Bowel Disease; CVD = Cardiovascular Disease; T2D = Type II Diabetes; NASH/NAFLD = Non-alcoholic Steatohepatitis/Non-alcoholic Fatty Liver Disease). OUTSTANDING QUESTIONS Can we define a core metagenome that is associated with health and conversely deficient metagenomes that are dysbiotic in all hosts? Alternatively, is a healthy metagenome also contextual to host genetics and environment and shifting temporally? Can we define those aspects of modern living that are contributing to the increase in chronic inflammatory disease via induction of shifts in the microbiota? Is infection a trigger for chronic inflammatory disease via the modification of the microbiota/host relationship? If the microbiota is a critical contributor to disease, can we modify the microbiota with pre and probiotics to alleviate symptoms exacerbated by inflammatory organisms? If the immune response is critical, can we modulate the immune response downstream of the microbiota to modulate disease phenotypes? Can we catalog the microbiota-derived metabolites that affect the immune response and understand how the balance of these metabolites may allow for assessment of the health of the host/microbiome relationship. Shifts in the microbiota, such as those induced by early life antibiotic use, seem to have a systemic immune memory. Can these ‘memory’ factors be defined and more importantly, can we intervene, to prevent long-term health impacts TRENDS BOX Chronic inflammatory disease is associated with shifts in the microbiota; often towards reduced diversity and increased inflammatory character The immune system can identify shifts in the microbiota both by direct interaction with bacteria in addition to ‘sensing’ of microbiota-derived metabolites Immune activation associated with shifts in the microbiota can have far-reaching systemic effects and contribute to disease The incidence in chronic disease is on the rise and has been linked to a number of environmental factors, including antibiotic use, diet and infection. There is evidence that these factors act on the host via shifting the microbiota This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Chovatiya R Medzhitov R 2014 Stress, inflammation, and defense of homeostasis Mol Cell 54 281 288 24766892 2 Wynn TA Vannella KM 2016 Macrophages in Tissue Repair, Regeneration, and Fibrosis Immunity 44 450 462 26982353 3 Frank DN 2007 Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases Proc Natl Acad Sci U S A 104 13780 13785 17699621 4 Qin J 2012 A metagenome-wide association study of gut microbiota in type 2 diabetes Nature 490 55 60 23023125 5 Turnbaugh PJ 2006 An obesity-associated gut microbiome with increased capacity for energy harvest Nature 444 1027 1031 17183312 6 Berni Canani R 2015 The role of the commensal microbiota in the regulation of tolerance to dietary allergens Curr Opin Allergy Clin Immunol 15 243 249 25827065 7 Sandler NG Douek DC 2012 Microbial translocation in HIV infection: causes, consequences and treatment opportunities Nat Rev Microbiol 10 655 666 22886237 8 Belkaid Y Hand TW 2014 Role of the microbiota in immunity and inflammation Cell 157 121 141 24679531 9 Longman RS 2013 Microbiota: host interactions in mucosal homeostasis and systemic autoimmunity Cold Spring Harb Symp Quant Biol 78 193 201 24913313 10 Yurkovetskiy LA 2015 Microbiota and autoimmunity: exploring new avenues Cell Host Microbe 17 548 552 25974297 11 Blaser MJ Falkow S 2009 What are the consequences of the disappearing human microbiota? Nat Rev Microbiol 7 887 894 19898491 12 Flint HJ 2008 Polysaccharide utilization by gut bacteria: potential for new insights from genomic analysis Nat Rev Microbiol 6 121 131 18180751 13 Qin J 2010 A human gut microbial gene catalogue established by metagenomic sequencing Nature 464 59 65 20203603 14 Zeevi D 2015 Personalized Nutrition by Prediction of Glycemic Responses Cell 163 1079 1094 26590418 15 Belkaid Y 2013 Effector and memory T cell responses to commensal bacteria Trends Immunol 34 299 306 23643444 16 Hooper LV 2012 Interactions between the microbiota and the immune system Science 336 1268 1273 22674334 17 Henao-Mejia J 2012 Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity Nature 482 179 185 22297845 18 Amar J 2011 Intestinal mucosal adherence and translocation of commensal bacteria at the early onset of type 2 diabetes: molecular mechanisms and probiotic treatment EMBO Mol Med 3 559 572 21735552 19 Kaser A 2010 Inflammatory bowel disease Annu Rev Immunol 28 573 621 20192811 20 Devkota S 2012 Dietary-fat-induced taurocholic acid promotes pathobiont expansion and colitis in Il10−/− mice Nature 487 104 108 22722865 21 Kawamoto S 2014 Foxp3(+) T cells regulate immunoglobulin a selection and facilitate diversification of bacterial species responsible for immune homeostasis Immunity 41 152 165 25017466 22 Pickard JM 2014 Rapid fucosylation of intestinal epithelium sustains host-commensal symbiosis in sickness Nature 514 638 641 25274297 23 Wahlstrom A 2016 Intestinal Crosstalk between Bile Acids and Microbiota and Its Impact on Host Metabolism Cell Metab 24 41 50 27320064 24 Mazmanian SK 2005 An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system Cell 122 107 118 16009137 25 An D 2014 Sphingolipids from a symbiotic microbe regulate homeostasis of host intestinal natural killer T cells Cell 156 123 133 24439373 26 Olszak T 2012 Microbial exposure during early life has persistent effects on natural killer T cell function Science 336 489 493 22442383 27 Arpaia N 2013 Metabolites produced by commensal bacteria promote peripheral regulatory T-cell generation Nature 504 451 455 24226773 28 Furusawa Y 2013 Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells Nature 504 446 450 24226770 29 Smith PM 2013 The microbial metabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis Science 341 569 573 23828891 30 Macia L 2015 Metabolite-sensing receptors GPR43 and GPR109A facilitate dietary fibre-induced gut homeostasis through regulation of the inflammasome Nat Commun 6 6734 25828455 31 Maslowski KM 2009 Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43 Nature 461 1282 1286 19865172 32 Hall JA 2011 The role of retinoic acid in tolerance and immunity Immunity 35 13 22 21777796 33 Levy M 2015 Microbiota-Modulated Metabolites Shape the Intestinal Microenvironment by Regulating NLRP6 Inflammasome Signaling Cell 163 1428 1443 26638072 34 Mathis D 2013 Immunological goings-on in visceral adipose tissue Cell Metab 17 851 859 23747244 35 Buffie CG 2015 Precision microbiome reconstitution restores bile acid mediated resistance to Clostridium difficile Nature 517 205 208 25337874 36 Rolhion N Darfeuille-Michaud A 2007 Adherent-invasive Escherichia coli in inflammatory bowel disease Inflamm Bowel Dis 13 1277 1283 17476674 37 Sokol H 2009 Low counts of Faecalibacterium prausnitzii in colitis microbiota Inflamm Bowel Dis 15 1183 1189 19235886 38 Garrett WS 2010 Enterobacteriaceae act in concert with the gut microbiota to induce spontaneous and maternally transmitted colitis Cell Host Microbe 8 292 300 20833380 39 Elinav E 2011 NLRP6 inflammasome regulates colonic microbial ecology and risk for colitis Cell 145 745 757 21565393 40 Moayyedi P 2015 Fecal Microbiota Transplantation Induces Remission in Patients With Active Ulcerative Colitis in a Randomized Controlled Trial Gastroenterology 149 102 109 e106 25857665 41 Rossen NG 2015 Findings From a Randomized Controlled Trial of Fecal Transplantation for Patients With Ulcerative Colitis Gastroenterology 149 110 118 e114 25836986 42 Sokol H 2014 Probiotics and antibiotics in IBD Dig Dis 32 Suppl 1 10 17 25531348 43 Winter SE Baumler AJ 2014 Dysbiosis in the inflamed intestine: chance favors the prepared microbe Gut Microbes 5 71 73 24637596 44 Winter SE 2013 Host-derived nitrate boosts growth of E. coli in the inflamed gut Science 339 708 711 23393266 45 Craven M 2012 Inflammation drives dysbiosis and bacterial invasion in murine models of ileal Crohn’s disease PloS one 7 e41594 22848538 46 Heimesaat MM 2006 Gram-negative bacteria aggravate murine small intestinal Th1-type immunopathology following oral infection with Toxoplasma gondii J Immunol 177 8785 8795 17142781 47 Molloy MJ 2013 Intraluminal containment of commensal outgrowth in the gut during infection-induced dysbiosis Cell Host Microbe 14 318 328 24034617 48 Berg AM 2013 Clostridium difficile infection in the inflammatory bowel disease patient Inflamm Bowel Dis 19 194 204 22508484 49 Cullender TC 2013 Innate and adaptive immunity interact to quench microbiome flagellar motility in the gut Cell Host Microbe 14 571 581 24237702 50 Palm NW 2014 Immunoglobulin A coating identifies colitogenic bacteria in inflammatory bowel disease Cell 158 1000 1010 25171403 51 Huang YJ LiPuma JJ 2016 The Microbiome in Cystic Fibrosis Clin Chest Med 37 59 67 26857768 52 Boguniewicz M Leung DY 2011 Atopic dermatitis: a disease of altered skin barrier and immune dysregulation Immunol Rev 242 233 246 21682749 53 Gregor MF Hotamisligil GS 2011 Inflammatory mechanisms in obesity Annu Rev Immunol 29 415 445 21219177 54 McNelis JC Olefsky JM 2014 Macrophages, immunity, and metabolic disease Immunity 41 36 48 25035952 55 Turnbaugh PJ 2008 Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome Cell Host Microbe 3 213 223 18407065 56 Ridaura VK 2013 Gut microbiota from twins discordant for obesity modulate metabolism in mice Science 341 1241214 24009397 57 Everard A 2013 Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity Proc Natl Acad Sci U S A 110 9066 9071 23671105 58 Schneeberger M 2015 Akkermansia muciniphila inversely correlates with the onset of inflammation, altered adipose tissue metabolism and metabolic disorders during obesity in mice Sci Rep 5 16643 26563823 59 Vijay-Kumar M 2010 Metabolic syndrome and altered gut microbiota in mice lacking Toll-like receptor 5 Science 328 228 231 20203013 60 Carvalho FA 2012 Transient inability to manage proteobacteria promotes chronic gut inflammation in TLR5-deficient mice Cell Host Microbe 12 139 152 22863420 61 Forslund K 2015 Disentangling type 2 diabetes and metformin treatment signatures in the human gut microbiota Nature 528 262 266 26633628 62 Cotillard A 2013 Dietary intervention impact on gut microbial gene richness Nature 500 585 588 23985875 63 Le Chatelier E 2013 Richness of human gut microbiome correlates with metabolic markers Nature 500 541 546 23985870 64 Burcelin R 2012 Immuno-microbiota cross and talk: the new paradigm of metabolic diseases Semin Immunol 24 67 74 22265028 65 Martinez-Medina M 2014 Western diet induces dysbiosis with increased E coli in CEABAC10 mice, alters host barrier function favouring AIEC colonisation Gut 63 116 124 23598352 66 Cani PD 2007 Metabolic endotoxemia initiates obesity and insulin resistance Diabetes 56 1761 1772 17456850 67 Garidou L 2015 The Gut Microbiota Regulates Intestinal CD4 T Cells Expressing RORgammat and Controls Metabolic Disease Cell Metab 22 100 112 26154056 68 Balmer ML 2014 The liver may act as a firewall mediating mutualism between the host and its gut commensal microbiota Sci Transl Med 6 237ra266 69 Elinav E 2013 Analysis of microbiota alterations in inflammasome-deficient mice Methods Mol Biol 1040 185 194 23852605 70 Fiorotto R 2011 Loss of CFTR affects biliary epithelium innate immunity and causes TLR4-NF-kappaB-mediated inflammatory response in mice Gastroenterology 141 1498 1508 1508e1491 1495 21712022 71 Brenner DA 2015 Role of Gut Microbiota in Liver Disease J Clin Gastroenterol 49 Suppl 1 S25 27 26447960 72 Wang L 2016 Intestinal REG3 Lectins Protect against Alcoholic Steatohepatitis by Reducing Mucosa-Associated Microbiota and Preventing Bacterial Translocation Cell Host Microbe 19 227 239 26867181 73 Wu J 2012 Beige adipocytes are a distinct type of thermogenic fat cell in mouse and human Cell 150 366 376 22796012 74 Cohen P Spiegelman BM 2015 Brown and Beige Fat: Molecular Parts of a Thermogenic Machine Diabetes 64 2346 2351 26050670 75 Brestoff JR 2015 Group 2 innate lymphoid cells promote beiging of white adipose tissue and limit obesity Nature 519 242 246 25533952 76 Lee MW 2015 Activated type 2 innate lymphoid cells regulate beige fat biogenesis Cell 160 74 87 25543153 77 Qiu Y 2014 Eosinophils and type 2 cytokine signaling in macrophages orchestrate development of functional beige fat Cell 157 1292 1308 24906148 78 Suarez-Zamorano N 2015 Microbiota depletion promotes browning of white adipose tissue and reduces obesity Nat Med 21 1497 1501 26569380 79 Chevalier C 2015 Gut Microbiota Orchestrates Energy Homeostasis during Cold Cell 163 1360 1374 26638070 80 Zhu W 2016 Gut Microbial Metabolite TMAO Enhances Platelet Hyperreactivity and Thrombosis Risk Cell 165 111 124 26972052 81 Tang WH 2014 Prognostic value of elevated levels of intestinal microbe-generated metabolite trimethylamine-N-oxide in patients with heart failure: refining the gut hypothesis J Am Coll Cardiol 64 1908 1914 25444145 82 Wang Z 2011 Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease Nature 472 57 63 21475195 83 Koeth RA 2013 Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis Nat Med 19 576 585 23563705 84 Stein JH Hsue PY 2012 Inflammation, immune activation, and CVD risk in individuals with HIV infection JAMA 308 405 406 22820794 85 Dillon SM 2014 An altered intestinal mucosal microbiome in HIV-1 infection is associated with mucosal and systemic immune activation and endotoxemia Mucosal Immunol 7 983 994 24399150 86 Lozupone CA 2013 Alterations in the gut microbiota associated with HIV-1 infection Cell Host Microbe 14 329 339 24034618 87 Vujkovic-Cvijin I 2013 Dysbiosis of the gut microbiota is associated with HIV disease progression and tryptophan catabolism Sci Transl Med 5 193ra191 88 Brenchley JM 2006 Microbial translocation is a cause of systemic immune activation in chronic HIV infection Nat Med 12 1365 1371 17115046 89 Dinh DM 2015 Intestinal microbiota, microbial translocation, and systemic inflammation in chronic HIV infection J Infect Dis 211 19 27 25057045 90 Mutlu EA 2014 A compositional look at the human gastrointestinal microbiome and immune activation parameters in HIV infected subjects PLoS Pathog 10 e1003829 24586144 91 Vazquez-Castellanos JF 2015 Altered metabolism of gut microbiota contributes to chronic immune activation in HIV-infected individuals Mucosal Immunol 8 760 772 25407519 92 Costalonga M Herzberg MC 2014 The oral microbiome and the immunobiology of periodontal disease and caries Immunol Lett 162 22 38 25447398 93 Gmur R 1986 Double-blind analysis of the relation between adult periodontitis and systemic host response to suspected periodontal pathogens Infect Immun 52 768 776 3710586 94 Mustapha IZ 2007 Markers of systemic bacterial exposure in periodontal disease and cardiovascular disease risk: a systematic review and meta-analysis J Periodontol 78 2289 2302 18052701 95 Gibson FC 3rd 2004 Innate immune recognition of invasive bacteria accelerates atherosclerosis in apolipoprotein E-deficient mice Circulation 109 2801 2806 15123526 96 Koren O 2011 Human oral, gut, and plaque microbiota in patients with atherosclerosis Proc Natl Acad Sci U S A 108 Suppl 1 4592 4598 20937873 97 Lalla E 2003 Oral infection with a periodontal pathogen accelerates early atherosclerosis in apolipoprotein E-null mice Arterioscler Thromb Vasc Biol 23 1405 1411 12816879 98 Silbergeld EK 2008 Industrial food animal production, antimicrobial resistance, and human health Annu Rev Public Health 29 151 169 18348709 99 Cho I 2012 Antibiotics in early life alter the murine colonic microbiome and adiposity Nature 488 621 626 22914093 100 Cox LM 2014 Altering the intestinal microbiota during a critical developmental window has lasting metabolic consequences Cell 158 705 721 25126780 101 Stecher B 2012 Gut inflammation can boost horizontal gene transfer between pathogenic and commensal Enterobacteriaceae Proc Natl Acad Sci U S A 109 1269 1274 22232693 102 Clarke SF 2014 Exercise and associated dietary extremes impact on gut microbial diversity Gut 63 1913 1920 25021423 103 Sonnenburg ED Sonnenburg JL 2014 Starving our microbial self: the deleterious consequences of a diet deficient in microbiota-accessible carbohydrates Cell Metab 20 779 786 25156449 104 Kamada N 2012 Regulated virulence controls the ability of a pathogen to compete with the gut microbiota Science 336 1325 1329 22582016 105 McNeill WH 1976 Plagues and peoples Anchor Press 106 Trueba G Dunthorn M 2012 Many neglected tropical diseases may have originated in the Paleolithic or before: new insights from genetics PLoS Negl Trop Dis 6 e1393 22479653 107 Hotez PJ 2014 The global burden of disease study 2010: interpretation and implications for the neglected tropical diseases PLoS Negl Trop Dis 8 e2865 25058013 108 Weinstock JV 2015 Do We Need Worms to Promote Immune Health? Clin Rev Allergy Immunol 49 227 231 25326880 109 Allen JE Maizels RM 2011 Diversity and dialogue in immunity to helminths Nat Rev Immunol 11 375 388 21610741 110 Maizels RM 2012 Immune modulation and modulators in Heligmosomoides polygyrus infection Exp Parasitol 132 76 89 21875581 111 Grainger JR 2010 Helminth secretions induce de novo T cell Foxp3 expression and regulatory function through the TGF-beta pathway J Exp Med 207 2331 2341 20876311 112 Osborne LC 2014 Coinfection. Virus-helminth coinfection reveals a microbiota-independent mechanism of immunomodulation Science 345 578 582 25082704 113 Zaiss MM 2015 The Intestinal Microbiota Contributes to the Ability of Helminths to Modulate Allergic Inflammation Immunity 43 998 1010 26522986 114 Leung J 2012 Heligmosomoides polygyrus abrogates antigen-specific gut injury in a murine model of inflammatory bowel disease Inflamm Bowel Dis 18 1447 1455 22223533 115 Broadhurst MJ 2012 Therapeutic helminth infection of macaques with idiopathic chronic diarrhea alters the inflammatory signature and mucosal microbiota of the colon PLoS Pathog 8 e1003000 23166490 116 Helmby H 2015 Human helminth therapy to treat inflammatory disorders -where do we stand? BMC Immunol 16 12 25884706 117 Vernacchio L 2006 Diarrhea in American infants and young children in the community setting: incidence, clinical presentation and microbiology Pediatr Infect Dis J 25 2 7 16395094 118 Kosek M 2003 The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000 Bull World Health Organ 81 197 204 12764516 119 Cadwell K 2010 Virus-plus-susceptibility gene interaction determines Crohn’s disease gene Atg16L1 phenotypes in intestine Cell 141 1135 1145 20602997 120 Lupp C 2007 Host-mediated inflammation disrupts the intestinal microbiota and promotes the overgrowth of Enterobacteriaceae Cell Host Microbe 2 119 129 18005726 121 Meinzer U 2012 Yersinia pseudotuberculosis effector YopJ subverts the Nod2/RICK/TAK1 pathway and activates caspase-1 to induce intestinal barrier dysfunction Cell Host Microbe 11 337 351 22520462 122 Hasegawa M 2014 Interleukin-22 regulates the complement system to promote resistance against pathobionts after pathogen-induced intestinal damage Immunity 41 620 632 25367575 123 Hand TW 2012 Acute gastrointestinal infection induces long-lived microbiota-specific T cell responses Science 337 1553 1556 22923434 124 Fonseca DM 2015 Microbiota-Dependent Sequelae of Acute Infection Compromise Tissue-Specific Immunity Cell 163 354 366 26451485 125 Kamdar K 2016 Genetic and Metabolic Signals during Acute Enteric Bacterial Infection Alter the Microbiota and Drive Progression to Chronic Inflammatory Disease Cell Host Microbe 19 21 31 26764594 126 Peyrin-Biroulet L 2012 Mesenteric fat as a source of C reactive protein and as a target for bacterial translocation in Crohn’s disease Gut 61 78 85 21940721 127 Zulian A 2013 Differences in visceral fat and fat bacterial colonization between ulcerative colitis and Crohn’s disease. An in vivo and in vitro study PLoS One 8 e78495 24205244 128 Korpe PS Petri WA Jr 2012 Environmental enteropathy: critical implications of a poorly understood condition Trends Mol Med 18 328 336 22633998 129 Brown EM 2015 Diet and specific microbial exposure trigger features of environmental enteropathy in a novel murine model Nat Commun 6 7806 26241678 130 Kau AL 2015 Functional characterization of IgA-targeted bacterial taxa from undernourished Malawian children that produce diet-dependent enteropathy Sci Transl Med 7 276ra224 131 Lindenbaum J 1972 Subclinical malabsorption in developing countries Am J Clin Nutr 25 1056 1061 4562265

PMC005xxxxxx/PMC5116296.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0406041 4668 J Comp Neurol J. Comp. Neurol. The Journal of comparative neurology 0021-9967 1096-9861 27199256 5116296 10.1002/cne.24041 NIHMS788089 Article Patterns of cell death in the perinatal mouse forebrain Mosley Morgan 1 Shah Charisma 1 Morse Kiriana A. 2 Miloro Stephen A. 2 Holmes Melissa M. 3 Ahern Todd H. 2* Forger Nancy G. 1* 1 Neuroscience Institute, Georgia State University, Atlanta, GA 30302 2 Department of Psychology, Center for Behavioral Neuroscience, Quinnipiac University, Hamden, CT 06518 3 Department of Psychology, University of Toronto, Mississauga, Ontario L5L 1C6 * Co-Corresponding authors: Nancy G. Forger, Neuroscience Institute, Georgia State University, Atlanta, GA 30302, Phone: (404) 413-5888, nforger@gsu.edu, Todd. H. Ahern, Department of Psychology, Center for Behavioral Neuroscience, Quinnipiac University, Hamden, CT 06518, todd.ahern@quinnipiac.edu 2 6 2016 13 6 2016 1 1 2017 01 1 2018 525 1 4764 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The importance of cell death in brain development has long been appreciated, but many basic questions remain, such as what initiates or terminates the cell death period. One obstacle has been the lack of quantitative data defining exactly when cell death occurs. We recently created a “cell death atlas,” using the detection of activated caspase-3 (AC3) to quantify apoptosis in the postnatal mouse ventral forebrain and hypothalamus, and found that the highest rates of cell death were seen at the earliest postnatal ages in most regions. Here we have extended these analyses to prenatal ages and additional brain regions. We quantified cell death in 16 forebrain regions across nine perinatal ages from embryonic day (E) 17 to postnatal day (P) 11 and find that cell death peaks just after birth in most regions. We find greater cell death in several regions in offspring delivered vaginally on the day of parturition compared to those of the same post-conception age but still in utero at the time of collection. We also find massive cell death in the oriens layer of the hippocampus on P1, and in regions surrounding the anterior crossing of the corpus callosum on E18, as well as the persistence of large numbers of cells in those regions in adult mice lacking the pro-death Bax gene. Together, these findings suggest that birth may be an important trigger of neuronal cell death, and identify transient cell groups that may undergo wholesale elimination perinatally. Graphical Abstract Activated caspase-3 cell death brain development hypothalamus hippocampus nucleus accumbens bed nucleus of the stria terminalis oriens cingulate RRID:AB_231409 RRID:AB_2109645 RRID:AB_2314667 RRID:AB_2298772 Introduction Two waves of cell death shape normal mammalian brain development. The first is confined to the ventricular and subventricular proliferative zones, and eliminates many neural precursor cells within hours of their birth (Blaschke et al., 1996; 1998; Thomaidou et al., 1997; Kuan et al., 2000). The second wave, known as “post-mitotic cell death,” occurs throughout the brain and eliminates roughly 50% of the neurons that have migrated, differentiated, and begun to make axonal connections (Oppenheim et al., 1985; Buss et al., 2006). Post-mitotic cell death sculpts developing circuits, contributes to sex differences in neuronal cell number (Forger, 2009), and may identify a window of vulnerability during which exposure to injuries, infections, or variations in environmental factors have greatest impact (McDonald et al., 1988; Ikonomidou et al., 1989, 1999; Yakovlev et al., 2001; Olney, 2002). Although the importance of cell death in neural development has been appreciated for decades, many surprisingly basic questions remain (see Yamaguchi and Miura, 2015 for a recent review). For the most part, we still do not know what initiates the cell death period, what terminates it, or what accounts for the large regional differences in the magnitude of cell death in the developing brain. The classical view, that developmental neuronal cell death is a cell suicide program triggered by a lack of trophic factors (Purves, 1988; Barde, 1989; Burek and Oppenheim, 1996), is best supported for neurons with axonal connections in the periphery (i.e., motoneurons and ganglion cells), but less well established for the brain. The death of interneurons in the cerebral cortex, for example, may be intrinsically determined (i.e., independent of trophic factors; Southwell et al., 2012), and non-neuronal cells such as microglia may kill otherwise viable neurons in some developing brain regions (e.g., Marin-Teva et al. 2004, Wakselman et al. 2008). One obstacle to resolving basic questions about neuronal cell death is that, with the exception of a few well-studied regions (e.g., cerebellum, substantia nigra pars compacta, and some cortical regions; Jackson-Lewis et al., 2000; Verney et al., 2000; Stankovski et al., 2007; Cheng et al., 2011), systematic data on the timing and magnitude of postmitotic cell death in the mammalian brain has been lacking. We recently addressed this by creating a “cell death atlas of the postnatal mouse brain” (Ahern et al., 2013), comprising brains of male and female C57BL/6 mice collected during the first two postnatal weeks. Postmitotic neuronal cell death is crucially controlled by members of the Bcl-2 family of proteins (Yuan and Yankner, 2000; Roth and D’Sa, 2001), and culminates in the activation of caspases, including caspase-3 (Porter and Jänicke, 1999; Hengartner, 2000). Using immunohistochemical detection of activated caspase-3 (AC3) to detect dying cells, we found the highest density of cell death in the ventral forebrain and hypothalamus at the earliest age(s) examined (postnatal day (P)1 to P3; Ahern et al., 2013). This raised the possibility that “peak” cell death actually occurs earlier – i.e., at P0 or prenatally. Here we examined this possibility by extending the cell death atlas to earlier ages. In addition, we present both pre- and postnatal cell death data from brain regions not included in the previous study. We find that AC3 cell number is very low in most forebrain regions at embryonic day (E)17 and increases over the next few days, with remarkably steep rises in some areas. In most regions examined, the highest rates of cell death are observed just after birth. We also identify several forebrain regions with intense accumulations of AC3-positive cells perinatally, and confirm an excess of neurons in these regions in adult mice lacking the pro-death gene Bax. Taken together, these findings clearly define the period of post-mitotic cell death in many regions of the mouse forebrain, suggest a possible role for parturition in triggering neuronal cell death, and point to the perinatal elimination of several transient cell groups. Materials and Methods Mouse Breeding and Collection of Brains, E17 – P1 Methods throughout followed as closely as possible those used in the generation of the postnatal cell death atlas (Ahern et al., 2013). Wildtype C57BL/6J mice purchased from The Jackson Laboratory (Bar Harbor, Maine) were housed in a 12:12 light:dark cycle (lights on at 0600 h) at 22°C with food and water available ad libitum. All procedures were in accordance with National Institutes of Health animal welfare guidelines and approved by the Georgia State University Institutional Animal Care and Use Committee. The forebrain regions studied here differentiate as discernable nuclei between E15 - E17 in the mouse (Niimi et al., 1962; Creps, 1974; Henderson et al., 1999). We therefore collected brains on E17, E18, E19/P0 and P1, including a one-day overlap with the postnatal atlas (Ahern et al., 2013) to allow us to calibrate our counts with those reported previously. To generate timed pregnancies, males and females were paired overnight. The males were removed the following morning at ~0900 hours and females checked for sperm plugs; this was designated day E0. Both male and female offspring from a total of 17 litters were collected. All fetuses/pups were collected at the same time of day (between 1400–1500 h), and pups from at least three different litters were used for each time point with the exception of E17 (two litters). Fetuses collected on E17 and E18 were removed by Cesarean (C) section after briefly exposing the dam to CO2 (< 1 minute). Fetuses are refractory to CO2 exposure and are relatively unaffected by such short exposure times (Leary, 2013; Pritchett et al., 2005). A large incision was made in the dam’s ventral midline, fetuses were extruded from the uterine horns, and all brains collected within 10 minutes. At the time of collection on the 19th day after conception, some offspring had been born vaginally earlier that day, whereas others were still in utero and were delivered by C-section. In the analyses below, we refer to the combined group as “E19/P0.” When examined separately, the C-sectioned group is referred to as “E19” and the vaginally-born group as “P0.” All pups collected on P1 had been vaginally delivered on day 19 (P0). Fetuses and neonates of both sexes were rapidly decapitated, and brains removed and fixed overnight in 5% acrolein (Alfa Aesar, Ward Hill, MA) in 0.1M phosphate buffer (PB). The tissue was then transferred to 30% sucrose in 0.1M PB and stored at 4°C. Two series of 40µm coronal sections were collected, placed in cryoprotectant (30% sucrose, 1% polyvinylpyrrolidone, 30% ethylene glycol in 0.1M PB), and stored at −20°C until use. Previously Generated Material Postnatal Cell Death Atlas We collected new data from the previously generated postnatal atlas material (Ahern et al., 2013), including data from five hippocampal regions and the core and shell of the nucleus accumbens. In addition, some data from the postnatal cell death atlas that were published previously (Ahern et al., 2013) are reprinted below along with the expanded age range. The previously published data are clearly indicated as such in the figures. Bax +/+ and Bax −/− brains Bax is a pro-death member of the Bcl-2 protein family that is required for the death of developing neurons; virtually all naturally-occurring neuronal cell death is eliminated in mice with a targeted disruption of the Bax gene (White et al., 1998; Ahern et al., 2013). The Bax +/+ and Bax −/− brains examined in this study were drawn from previously collected brains of Bax knockout mice and their wildtype siblings (Forger et al., 2004; Holmes et al., 2009). The Bax deletion was on a C57BL/6 background and mice were adult at the time of sacrifice. For thionin-stained sections, animals were intracardially perfused with formalin, and brains were frozen sectioned at 30µm. For immunocytochemistry, animals were perfused with 4% paraformaldehyde and frozen sectioned at 30µm. Immunohistochemistry Alternate sections from all perinatal animals were immunohistochemically stained for AC3, closely following the protocol of Ahern et al. (2013). Free-floating sections were extensively rinsed in 1X Tris-buffered saline (TBS; pH 7.6), and immersed in 0.05M sodium citrate for 30 minutes. Sections were again rinsed, transferred to 0.1M glycine in 1X TBS for 30 minutes, rinsed, then incubated in a concentrated blocking solution (1X TBS, 20% normal goat serum, 1% hydrogen peroxide, 0.3% Triton-X). Sections were incubated overnight in primary antibody solution (Cleaved Caspase-3, RRID:AB_231409, Cell Signaling, Danvers, MA; 1:20,000 in 1XTBS, 2% normal goat serum, 0.3% Triton), then washed in a dilute blocking solution (1X TBS, 1% normal goat serum, 0.02% Triton-X), incubated for one hour in secondary antibody solution (goat anti-rabbit (Vector Laboratories, Burlingame, CA; 1:250, 1X TBS, 2% normal goat serum, 0.3% Triton-X), followed by rinses in 1X TBS with 0.2% Triton-X, before incubating for one hour in an ABC solution (Vectastain Elite ABC Kit, Vector Laboratories). Reagent A and B of the ABC kit were added to 1X TBS at a dilution of 0.8% based on pilot optimization runs. Sections were incubated for 2–5 minutes in diaminobenzidine-nickel (DAB) solution (Vector Laboratories), followed by rinses in 1X TBS. Sections were mounted onto microscope slides and counterstained with thionin. We also used immunohistochemical detection of the microglial marker, ionized calcium-binding adaptor molecule 1 (Iba1), to examine the distribution of microglia in a subset of animals on E18. Activated microglia with an amoeboid morphology are responsible for the phagocytosis of cell corpses, and accumulate in regions of high cell death (reviewed in Pont-Lezica et al., 2011; Wake et al., 2012). The immunohistochemical procedure followed that described for AC3 above, except that the primary antibody used was anti-Iba1 (RRID:AB_2314667, Wako Chemicals, Osaka, Japan, 1:20,000 in 1X TBS, 2% normal goat serum, 0.3% Triton), and the sodium citrate incubation was increased to one hour. To determine the phenotype of cell groups persisting in Bax −/− mice, we examined Bax+/+ and Bax −/− brains that were processed for double-label immunohistochemistry for NeuN and GFAP, markers of mature neurons and astrocytes, respectively. These brains came from animals in our previous study (Holmes et al., 2009) and were labeled using sequential staining for NeuN [1:1000 mouse anti-NeuN monoclonal antibody (RRID:AB_2298772, Chemicon International, Temecula, CA)] using a DAB reaction (brown reaction product), and GFAP (1:1000 rabbit anti-GFAP polyclonal antibody; RRID:AB_2109645, Chemicon International), using Vector SG (blue reaction product; Vector Laboratories), as described (Holmes et al., 2009). Antibody specificity Details of primary antibodies are given in Table 1. The AC-3 antibody used here recognizes the active fragment of cleaved caspase-3.. It detects a couplet on Western blots of the expected size and does so only when apoptosis-initiating proteases are active (Cheong et al., 2003). It does not cross-react with full-length (inactive) caspase-3 in human or mouse cells (Hu et al., 2000; Sammeta and McClintock, 2010). Moreover, virtually all labeling with this antibody is abolished in the forebrain of newborn mice lacking the pro-death gene Bax (Ahern et al., 2013), supporting its specific detection of dying cells in this material. The specificity of the GFAP antibody was demonstrated in Western blots of rat retina lysate (Chang et al., 2007) where it labeled a band of the expected size (50 kDa); quantification of the labeled band in Western blots also correlated with quantification of immunocytochemical labeling of cells with astrocytic morphology in tissue sections. The Iba1 antibody used here recognizes a single band of 17 kDa on immunoblots, corresponding to the size of Iba1 protein, and does not recognize neurons or astrocytes (Imai et al., 1996; Ito et al., 1998). It has been used extensively as a microglial marker in perinatal mouse brain (e.g., Nilsson et al., 2008; Shrivastava et al., 2012). The mouse monoclonal against neuron-specific nuclear antigen (NeuN) was originally made against cell nuclei purified from mouse brain (Mullen et al., 1992). The antibody does not recognize oligodendrocytes or astrocytes in mouse brain but labels neurons in vivo and in vitro. Western blotting with this antibody shows three bands of 46–48-kDa (Mullen et al., 1992). Digital analysis of AC3-labeled regions of interest All slides were digitally scanned using a NanoZoomer Digital Pathology slide scanner (Hamamatsu Photonics, Bridgewater, NJ) and analyzed using the Aperio ImageScope program (version 12.1.0.5029, Leica Biosystems, Nussloch, Germany) as previously (Ahern et al.; 2013). Digitized slides were coded and analyzed by investigators blind to group membership. We counted AC3+ cells bilaterally in a total of 16 brain regions. Regions were chosen that had clearly identifiable borders at all ages examined. For nine of the regions, AC3+ cells had previously been quantified from P1–P11, and we expanded the age range to E17 here. This included four hypothalamic regions [anteroventral periventricular nucleus (AVPV), medial preoptic nucleus (MPON), paraventricular nucleus (PVN), and suprachiasmatic nucleus (SCN)], the lateral and principal nuclei of the bed nuclei of the stria terminalis (BNSTl, BNSTp), two regions in the amygdala [central amygdala (CeA), and medial amygdala (MeA)], and the lateral septum (LS). In addition, we quantified cell death at nine ages in seven brain regions not previously examined in the postnatal atlas: the core and shell of the nucleus accumbens (NAcc core and NAcc shell) and five hippocampal regions [cornus ammonis (CA)1, CA2/3, CA1 oriens layer, CA2/3 oriens layer, and the dentate gyrus (DG)]. Newly generated data are plotted with solid lines in the figures below, whereas reprinted data are indicated by dotted lines. Most regions were examined in their entirety (i.e., in each section in which they occurred throughout their rostrocaudal extent). For the much larger hippocampal regions, we sampled four sections from the dorsal hippocampus of each animal. In the younger animals (E17-P1) these were four consecutive (alternate) sections, starting with the anterior-most section in which the DG was clearly visible. For older animals with larger hippocampi (~P3–P11), we identified the first section in which the DG was clearly visible (anterior anchor), and the first section in which the lateral extent of the hippocampus begins to dip ventrally (posterior anchor, i.e., Figure 69 in Paxinos et al. 2007); these sections and two additional sections spaced evenly between them were traced and counted. This allowed us to sample approximately the same rostro-caudal extent of the hippocampus in each subject. The NAcc shell could not be distinguished from the NAcc core prior to E18; counts on E17 therefore represent the combined area. An observer blind to the sex or age of the subject traced the outline of each region of interest using Aperio ImageScope, and the cross-sectional area was recorded. Labeled AC3+ cells within the region were counted as in Ahern et al. (2013). The measure “total AC3 cells” is the sum of these counts times two (because alternate sections were processed for AC3 immunohistochemistry). The “density of AC3+ cells” was calculated by dividing the total number of AC3+ cells by the total measured volume of each region, and is expressed as AC3+ cells per mm3. Cell number in the hippocampal oriens of Bax+/+ and Bax−/− mice We quantified the number of cells with a neuronal morphology (large and multipolar) in thionin-stained sections, as well as the number of NeuN+ and GFAP+ cells in immunohistochemically stained sections of the CA1 oriens of adult Bax+/+ and Bax−/− mice (N = 3 of each). Two consecutive sections were counted, and the first section was the most anterior section in which the DG was clearly visible in each animal. Cingulate cortex/indusium griseum and subcallosal sling Two striking clusters of AC3+ cells were visible with the naked eye in AC3-stained sections of late prenatal mice. One was located just dorsal and lateral to the midline crossing of the corpus callosum, in an area approximately corresponding to cingulate cortex area 2/indusium griseum (Cg2/IG; Adamek et al., 1984); the other was just ventral to the corpus callosum and may correspond to the subcallosal sling (Silver et al., 1982). Due to the sheer density of AC3 staining in these two areas, we were unable to individually count labeled cells. Instead, we captured images of the areas and used the thresholding function of ImageJ (Version 1.47; National Institutes of Health) to compare AC3 immunostaining across perinatal ages. The sections analyzed spanned from just anterior to the most rostral crossing of the corpus callosum to the appearance of the forceps major. Each region was outlined in ImageJ and the number of pixels above threshold determined. Threshold was based on background staining of an adjacent brain area with no AC3+ cells and was determined separately for each animal. Total number of pixels above threshold was then divided by the volume sampled, and converted to pixels above threshold per mm3. Photomicrographs For figures including photomicrographs, Adobe Photoshop was used to crop images, adjust brightness, and balance color. Statistical Analysis We included an overlapping day (P1) between analyses performed on the previously generated material (postnatal atlas; Ahern et al., 2013) and the new material (E17-P1), so that we could “calibrate” between the two data sets. Across the 16 brain regions examined, AC3+ cell densities on P1 in the two sets of material were highly correlated, with counts from the previously generated material accounting for 95% of the variance between brain regions in the new material (R2 = 0.95; df = 14, p < 0.0001). The datasets were therefore combined and the values plotted for P1 below represent the P1 animals from both collections. We first performed two-way ANOVAs (sex-by-age) for each brain region. No significant sex differences were found for any of the newly generated data presented here, so counts from males and females were combined and 1-way ANOVAs performed for each brain region. (We note that in the previously published data, there was an effect of sex on AC3+ cell number in the BNSTp only, with more AC3+ cells in females from about P5–P7; Ahern et al., 2013). Planned comparisons using Fisher’s least significant difference were performed only following significant main effects. The number of animals per age group varied from 12 (E17) to 35 (P1; the combined P1 animals from the postnatal atlas and new material). However, a given brain region was not included in the analysis if tissue damage or staining artifacts prevented an accurate count of AC3+ cells. The final median number of animals per age (followed by the range in parentheses) across all brain regions was as follows: E17, 6 (4–12); E18, 9 (6–12); E19/P0, 20 (12–31); P1, 29 (21–35); P3, P5, and P7, 23 (20–24); P9 and P11, 21 (19–22). Two-tailed, independent t-tests were used to compare the number of cells (thioninstained, NeuN+ or GFAP+) in the hippocampal oriens layer of Bax+/+ and Bax−/− mice. Results AC3+ cell density All 16 regions exhibited a significant effect of age on AC3+ cell density between E17-P11 (Table 2). For 13 of the 16 regions, AC3+ cells were very sparse on E17 and increased over the next several days (Figures 1 and 2). Highest AC3+ cell densities were seen between birth and P5, followed by a decline to very low levels in all regions on P9 and P11 (Figure 2). Exceptions to this pattern included the DG, CA1, and PVN: for each of these regions AC3+ cell density was high at E17 and remained high over the next several days (Figure 2E, G); density then declined between E19/P0 and P1 for the PVN (P < 0.0001), and between P1 and P3 for the DG and CA1 (P < 0.0001). The difference in AC3+ cell density between the CA1 and CA2/3 during late prenatal life was striking; this can be seen not only in the quantitative analysis (Figure 2G), but also in photomicrographs (Figure 4). As seen previously (Ahern et al., 2013), there were large differences in the density of AC3+ cells across brain regions. In the CeA, for example, densities were below 600 AC3+ cells per mm3 at all ages examined (Figure 2D), whereas peak densities of >5,000 AC3+ cells per mm3 were seen in the CA1 and CA2/3 oriens (Figure 2H). Total number of AC3+ Cells AC3+ density measures allow for comparisons of the rate of cell death across brain regions of very different overall size, but can obscure changes in the total number of dying cells when brain regions change significantly in size, as they do during perinatal development. Figure 3 presents total AC3+ cells for the 16 brain regions analyzed above; all regions showed a significant main effect of age on total number of AC3+ cell number (Table 1). As was seen for AC3+ cell density, the highest total numbers of AC3+ cells are seen soon after birth. In fact, the pattern is more consistent for total counts than it was for density: even in the regions with high densities of AC3+ cells prenatally (DG, CA1, and PVN), the total number of AC3+ cells is low at E17, with increases over the next several days (Figure 3E, G), followed by a decline to very low levels by P11. Peak total numbers of AC3+ cells were seen “early” (P0, P1) in the PVN, BNSTl, DG, CA1, CA2/3, CA1 oriens and CA2/3 oriens, and “late” (P5) in the BNSTp, AVPV, SCN, LS, MPON, CeA, NAcc core, and NAcc shell. Persistence of cells in the hippocampal oriens layer of Bax knockout mice As noted above, the CA1 and CA2/3 oriens layers of the hippocampus were remarkable for the exceptionally high density of AC3+ cells observed on P1 (~6,000 AC3+ cells per mm3), which was twice that of the next highest brain region (Figures 2H and 3H). Both oriens regions also exhibited abrupt, three- to four-fold increases in AC3+ cell density in the 24h between P0 and P1 (P < 0.0001 for both CA1 and CA2/3 oriens; Figures 2H, 3H and Figure 4). The hippocampal oriens layer is cell sparse in adulthood, and populated mainly by the basal dendrites of CA pyramidal neurons and their afferents (Reznikov, 1991). The very high rate of cell death in the newborn oriens therefore suggested the elimination of a transient cell population. To investigate this further, we examined the oriens layers in adult Bax−/− mice, in which developmental neuronal cell death is nearly completely prevented (White et al., 1998; Ahern et al., 2013). We find a large population of multipolar, neuronal-like cells in the CA1 and CA2/3 oriens layers of thionin-stained sections of Bax−/− mice; quantification indicates 8 times more cells in the CA1 oriens of Bax−/− than of wildtype mice (Figure 5; P < 0.0005). This far exceeds the ~1.5- to 2-fold increases in cell number seen in other brain and spinal cord regions of Bax knockouts (Deckwerth et al., 1996; Sun et al., 2003; Forger et al., 2004; Jacob et al., 2005). Double-labeling for NeuN and GFAP established that the large majority of rescued cells in the Bax knockouts are neuronal: the number of NeuN+ cells in the CA1 oriens of Bax−/− animals was 5.6-fold greater than the number in wild-type controls (Figure 5; P < 0.05), whereas the number of GFAP+ cells did not differ significantly between genotypes (not shown). This supports the existence of a transient layer of neurons in the hippocampal oriens that is eliminated by an explosive period of cell death coinciding with birth. Cingulate cortex area 2/ Indusium griseum and subcallosal sling Next we analyzed two forebrain areas that were not a priori identified as regions of interest, but stood out as having the highest levels of AC3 labeling in any brain region of perinatal animals, visible with the naked eye in stained sections. These cell groups were located just dorsal and ventral to the midline crossing of the corpus callosum and are identified here as Cg2/IG and the subcallosal sling (Figure 6A,B). In both regions we found significant effects of age on AC3+ cell labeling (P < 0.0001 and P = 0.0002 for the Cg2/IG and subcallosal sling, respectively; Figure 7), with the highest density of AC3+ cells seen on E18. The cluster in the Cg2/IG was remarkable for the abrupt increase between E17–E18 and the equally abrupt decrease between E18-E19/P0 (P < 0.0001 in both cases; Figure 7). The subcallosal sling displayed a similar pattern, but with more modest differences in cell death density between E17–E18 and E18–E19/P0 (P = 0.006 and P = 0.04, respectively). Few AC3+ cells were seen in these regions at P3 or later ages (not quantified). The morphology of AC3+ cells in the Cg2/IG could be seen more clearly in high-powered views of sections that were not counterstained (Figure 5C), and processes could be seen projecting into, or perpendicularly through, the corpus callosum in this material (Figure 5C). In support of the interpretation that the intense AC3 staining in the Cg2/IG and subcallosal sling is indicative of cell death, we observed many pyknotic cells in this regions in thionin-stained sections of E18 animals not processed for immunohistochemistry (not shown), as well as a dense accumulations of microglia with an activated morphology (Figure 6C). Examination of adult Bax −/− mice suggests an accumulation of thionin-stained and NeuN+ cells in the Cg2/IG region that far exceeds differences in cell density between Bax −/− and wildtype animals in adjacent regions (Figure 6E–H). As discussed below, the AC3 cells in the perinatal Cg2/IG may represent the elimination of a transient neuronal cell group that pioneers the midline crossing of the corpus callosum or hippocampal commissure. AC3+ density at 19 days post-conception in offspring vaginally born or still in utero Most C57BL/6 dams give birth at 19 days post-conception (e.g., Murray et al., 2010). At our usual collection time between 1400 and 1500 h on E19, some litters had been born vaginally earlier that day (P0 group), whereas others were still in utero (and removed by C-section; E19 group). The pups are combined as the E19/P0 group in the analyses above. When analyzed separately, there was no significant difference for most cell groups. However, we found greater AC3+ cell density in vaginally delivered pups in the suprachiasmatic nucleus (P = 0.033), AVPV (P = 0.013) and the subcallosal sling (P = 0.003) (Figure 7). No brain regions showed the opposite pattern. Body weights between the groups did not differ (1.32 ± 0.02 g and 1.33 ± 0.02 g for E19 and P0 offspring, respectively; two-tailed t-test, P = 0.756). Discussion Despite the fact that the role of cell death in brain development is now textbook material, systematic analyses of cell death in the developing mammalian brain have been lacking. With few exceptions, most previous studies have quantified dying cells in one, or at most a few, brain regions at one or a few ages. Here we present cell death dynamics in 16 forebrain regions across 9 perinatal ages, providing the most comprehensive view to date of the patterning of post-mitotic neuronal cell death in the mouse brain. These data may help guide the search for factors that account for regional differences in the magnitude of cell death or that trigger the onset and termination of the cell death period. Limitations We opportunistically chose to quantify brain regions with clearly defined borders at all ages in the current study and, although we examined a relatively large number of areas, there are obviously many more that we did not. Missing from this analysis, for example, are any thalamic or cortical regions (with the exception of Cg2). Limited data on the timing of cell death in the mouse cortex have previously been published and analyses of multiple cortical areas using the current material are in progress (T. Ahern, unpublished). We believe the regions quantified here to be reasonably representative because the lack of AC3+ cells at the earliest ages examined (E17–E18), and peak rates of cell death during first few days of postnatal life, were clearly widespread phenomena throughout the forebrain based on qualitative assessments of the material. Another limitation of this work is that the identification of cell death relied almost exclusively on the detection of AC3. This could be a concern if some dying cells do not express AC3, or if AC3 immunoreactivity labels cells that are not undergoing apoptosis. Although caspase-independent cell death has been described for neuronal precursor cells or after neuronal injury (Zhan et al., 2001; Rideout and Stefanis, 2001; Cregan et al., 2002), it does not, as far as we know, occur during naturally-occurring death of post-mitotic neurons. The other possibility (non-apoptotic roles for AC3 in our labeled cells) is not likely to be a concern at the ages studied here: although AC3 is involved in dendritic pruning in Drosophila and synaptic plasticity in weanling rats (Li et al., 2010; Hyman and Yuan, 2012), the large majority of cells expressing caspase-3 in the neonatal rodent nervous system also show signs of pyknosis (Srinivasan et al., 1998; Zacharaki et al., 2010; Zuloaga et al., 2011). More important, almost all (>95%) activated caspase-3 (AC3) cells are eliminated throughout the perinatal forebrain in mice lacking the pro-death gene, Bax (Ahern et al., 2013). Double labeling studies are limited in number, but indicate that the majority of AC3+ cells in the perinatal rodent brain are neuronal (Zuloaga et al., 2011), in accord with our observation that many of the AC3+ cells in perinatal animals in the current study had a neuronal morphology. In addition, the “extra” cells in Bax−/− mice were NeuN+ in two brain regions that had high AC3+ cell densities perinatally. Taken together, AC3 immunoreactivity in the late prenatal and neonatal brain is a reliable marker for naturally occurring cell death and primarily labels dying neurons. Does birth orchestrate cell death? The observation that cell death peaks within a few days of birth in most forebrain regions raises the possibility that parturition orchestrates patterns of neuronal cell death. The final stage of mammalian gestation is characterized by major hormonal shifts and a state of sterile inflammation (Soares and Talamantes, 1984; Bernal, 2001; Thomson et al., 1999; Golightly et al., 2011); these are followed by the mechanical stimuli of a vaginal delivery, a transition to breathing air, and marked changes in energy metabolism, feeding, temperature regulation, and new challenges to sensory-motor systems (Turgeon and Meloche, 2009; Hillman et al., 2012). Peri-parturitional events trigger major adaptive changes in peripheral organs that are required for adaptation to ex utero life (e.g., Thilaganathan et al., 1994; Fowden et al., 1998; Jain and Eaton, 2006) and could also influence brain development, including the timing or magnitude of neuronal cell death. Alternatively, cell death may be pre-programmed and occur at a set time after conception regardless of the timing of birth. As far as we know, data are not currently available to discriminate between these two possibilities. Whether birth plays a role in controlling neuronal cell death could be addressed by examining mice in which birth is experimentally advanced or delayed by hormonal or genetic manipulations (Sugimoto et al., 1997; Langenbach et al., 1997; Gross et al., 1998; Jeff et al., 2000; Hashimoto et al., 2010). Similarly, by carefully equating time ex utero after a vaginal or cesarean delivery, one could determine whether mode of birth makes a difference. Both types of studies have the potential to identify novel factors controlling the timing and magnitude of cell death in the mammalian brain. Exceptions to the “rule” Exceptions to the pattern of peak rates of cell death soon after birth were seen in the DG, PVN, and CA1, which all had high densities of AC3+ cells prenatally.; even in these regions, however, total AC3+ cell number conformed to the pattern. Our counts of total AC3+ cells in the DG are in good agreement with those reported previously for perinatal mice (Kim et al., 2009). The DG forms quite late, with the earliest neurons born at E16 (Bayer, 1980; Altman and Bayer, 1990; Stanfield and Cowan, 1979). As would be expected, DG volume increases steeply during the next few days (a nearly 5-fold increase (487%) in volume in the three days between E17 and P1, based on our material; data not shown). Similarly, although many neurons in the CA1 of rodents are born by E16, neurogenesis continues in this region through the last days of gestation and first few postnatal days (Stanfield and Cowan, 1979; Soriano et al., 1989; Bayer, 1980; Bowers et al., 2010). Because regions of neurogenesis are often associated with high cell turnover (Blaschke et al., 1998; Morshead and van der Kooy, 1992; Jabès et al., 2010; Kim et al., 2011), the relatively late histogenesis of the DG and CA1 could explain the high density of AC3+ cells during late prenatal life. The PVN, on the other hand, forms quite early in the mouse (Karim and Sloper, 1980), so late histogenesis cannot explain high rates of cell death prenatally. The PVN is one of the most important autonomic control centers in the brain, and plays a central role in the stress response. Many studies have demonstrated “programming” effects of late prenatal stressors on the hypothalamic-pituitary-adrenal axis and, in particular, long-term changes in PVN function (e.g., McCormick et al., 1995; Hossain et al., 2008; Zohar and Weinstock, 2011). Late prenatal stress also increases apoptosis in the fetal PVN of rats (Fujioka et al., 1999; 2003; Tobe et al., 2005). It is possible that the cell turnover occurring just prior to birth in the PVN of mice contributes to the susceptibility of this brain region to prenatal programming. Unusually high rates of cell death represent the elimination of transient cell groups in the hippocampus, subcallosal sling, and cingulate cortex One unexpected finding of the current study was the large number of AC3+ cells in the oriens layers of the hippocampus (normally a cell-poor region in adulthood). The CA1 and CA2/3 oriens layers of the hippocampus had by far the highest densities of AC3+ cells in any of the brain regions quantified in this or the previous study (Ahern et al., 2013). Also of note were the abrupt changes in the number and density of AC3+ cells in the oriens regions: a 300–400% increase in AC3+ cells between P0 and P1, followed by a return to baseline several days later. Because a population of large neurons persisted in the oriens layers in Bax−/− mice, this suggests the relatively sudden elimination of a transient cell group in the developing hippocampus. The elimination of cell groups that may play a temporary role during development of the vertebrate nervous system has been described previously (e.g., Whitlock and Westerfield, 1998; Reyes et al., 2004; Jung et al., 2008). As far as we can tell, the rapid die-off of a cell group in the hippocampal oriens has not previously been described, although Stanfield and Cowan (1979) noted a decrease in cell density in the stratum oriens and stratum radiatum between E18 and P1 in mice, based on appearance in a Nissl stain. The identity or function of the eliminated cells is not known and is an interesting question for future study. A massive accumulation of AC3+ cells was also seen in the subcallosal sling and (especially) Cg2/IG at E18; in fact, the density of AC3+ cells in these regions was so great that we could not count them individually. Both the subcallosal sling and Cg2/IG have been implicated in the early formation of the corpus callosum (CC) or hippocampal commissure (HC). The subcallosal sling (previously referred to as the “glial sling,” but renamed when it was shown to contain neurons) is present before the first CC axons cross the midline (E15–E16 in the mouse) and may function to guide pioneer axons that grow along its surface (Silver et al., 1982; Shu and Richards, 2001; Shu et al., 2003). The appearance of pyknotic cells just before birth in the subcallosal sling of the mouse has been reported previously (Hankin et al., 1988); our analyses confirm the interpretation that the disappearance of this structure is due to cell death. Niquille et al. (2009) more recently proposed the elimination of a transient population of neurons thought to guide axons crossing the CC. Although the cells they describe were located primarily within the CC itself, we note that some of the AC3+ labeling they observed on P0 (Niquille et al., 2009; Supplemental figure 3) appears ventrolateral to the CC and may also correspond, in part, to what we have identified here as the dying off of the subcallosal sling. Based on work in the rat, cells in the Cg2 region were proposed to be the first cortical neurons to project their axons to the contralateral hemisphere (i.e., to pioneer the CC; Koester and O’Leary, 1994). Prenatally, however, the CC is closely apposed to the slightly more ventral HC, and work in mice suggests that the early projecting cingulate neurons instead enter the HC (Ozaki and Wahlten, 1998). Regardless of which interpretation is correct, it is interesting that both Koester and O’Leary (1994) and Ozaki and Wahlsten (1998) raised the possibility that the Cg2 neurons that pioneer the midline crossing are later eliminated by apoptosis, although neither group favored that explanation. Many AC3+ labeled cell processes in the current study coursed perpendicular to the CC, in accord with the suggestion of Ozaki and Wahlsten (1998) that they projected through the CC to the more ventrally located HC; other AC3+ fibers ran parallel to the CC, however. The fact that so many cells in the Cg2/IG were AC3 labeled, and that this area was also filled with pyknotic cells and amoeboid microglia, suggests that many are eliminated at ~E18, after the CC and HC have been established. Conclusions Much of the recent work on cell death in the developing brain pertains to injury-induced cell death. Many years after its discovery, however, there is clearly much that remains to be learned about the control of naturally-occurring cell death during brain development. The findings presented here, and previously (Ahern et al., 2013), demonstrate that the bulk of developmental cell death occurs within several days of birth. Whether birth (parturition) plays a causative role in timing cell death, although a seemingly basic question, is not known. The signals that trigger the elimination of cells in the Cg2/IG and SCS on E18 or hippocampal oriens cells on P1 also are not known, and are areas for future investigation. Supported by: NIMH R01-068482 (NGF), a Georgia State University Brains and Behavior Seed Grant (NGF), a Quinnipiac University College of Arts & Sciences Seed Grant (THA), and NSERC Discovery Grant 402633 (MMH). Figure 1 Photomicrographs of coronal sections immunohistochemically stained for activated caspase-3 (AC3) show many more dying cells on postnatal day (P) 1 than on embryonic day (E) 17. A, C, and E are from mice on E17; B, D, and F are from P1 mice. A and B are taken at the level of the nucleus accumbens; C and D are taken at the level of the lateral nucleus of the bed nucleus of the stria terminalis; E and F are higher magnification views of the regions indicated by the boxes in C, D. Scale bars = 100 µm in A–D and 25 µm in E,F. Abbreviations: ac, anterior commissure; CC, corpus callosum; LV, lateral ventricle; 3V, third ventricle. Figure 2 Mean (±SEM) density of AC3+ cells in 16 regions of the mouse forebrain from embryonic day 17 (E17) to postnatal day 11 (P11). Filled symbols (E17-P1) represent data from new material generated for this study and open symbols (P1–P11) are data from previously generated material (Ahern et al., 2013). Previously published data from Ahern et al. (2013) are shown by dotted lines (A–D and PVN in E) and reprinted here with the permission of John Wiley & Sons; new data collected from the material generated in Ahern et al., (2013) are plotted in solid lines (DG in E, and F–H). Abbreviations: AVPV, anteroventral periventricular nucleus; BNSTl and BNSTp, lateral and principal nuclei of the bed nuclei of the stria terminalis; CA1, cornus ammonis 1; CA2/3, cornus ammonis 2 and 3; CeA, central amygdala; DG, dentate gyrus; LS, lateral septum; MeA, medial amygdala; MPON, medial preoptic nucleus; NAcc, nucleus accumbens; PVN, paraventricular nucleus of the hypothalamus; SCN, suprachiasmatic nucleus. Figure 3 Mean (±SEM) total number of AC3+ cells in 16 regions of the mouse forebrain from embryonic day 17 (E17) to postnatal day 11 (P11). Filled symbols (E17-P1) represent data from new material generated for this study and open symbols (P1–P11) are data from previously generated material (Ahern et al., 2013). Previously published data from Ahern et al. (2013) are shown by dotted lines (A–D and PVN in E) and reprinted here with the permission of John Wiley & Sons; new data collected from the material generated in Ahern et al., (2013) are plotted in solid lines (DG in E, and F–H). Abbreviations: AVPV, anteroventral periventricular nucleus; BNSTl and BNSTp, lateral and principal nuclei of the bed nuclei of the stria terminalis; CA1, cornus ammonis 1; CA2/3, cornus ammonis 2 and 3; CeA, central amygdala; DG, dentate gyrus; LS, lateral septum; MeA, medial amygdala; MPON, medial preoptic nucleus; NAcc, nucleus accumbens; PVN, paraventricular nucleus of the hypothalamus; SCN, suprachiasmatic nucleus. Figure 4 Photomicrographs of coronal sections through the hippocampus from embryonic day (E) 17 through postnatal day (P) 5. Sections were immunohistochemically stained for activated caspase-3 (AC3) and counterstained with thionin. A: E17, B: E18, C: E19/P0, D: P1, E: P3, F: P5. A peak of cell death in the hippocampus is evident from P1–P3. Although not shown here, few to no AC3+ cells were present in the hippocampus of the older animals (P7, P9, and P11) examined in this study. Scale bars = 100 µm. Figure 5 The elimination of developmental cell death reveals a prominent cell layer in the hippocampal oriens layers. A, C. Photomicrographs of thionin-stained sections of the hippocampus of adult wild-type (Bax+/+) and Bax −/− mice. B, D. Higher magnification views of the boxed areas in A and C. Many large, neuron-like cells are found in the CA1 and CA2/3 oriens of the Bax knockouts. E, F. Photomicrographs of the hippocampal oriens of adult Bax+/+ (E) and Bax−/− (F) mice double-labeled for NeuN and GFAP. An accumulation of NeuN cells is seen in the Bax−/− mice. Scale bars = 200 µm in A, B, E, and F; 25 µm in C and D. Quantification of thionin stained (G) and NeuN+ (H) sections of the CA1 oriens layer indicates many more cells with a neuronal morphology in Bax −/− than Bax+/+ mice. Figure 6 Striking accumulations of AC3+ cells are found in the cingulate cortex 2 / indusium griseum (Cg2/IG) region on E18. A, B. Photomicrographs of sections immunohistochemically stained for activated caspase 3 (AC3) and counterstained with thionin on E17 (A) and E18 (B). Photos in both cases are taken just posterior to the anterior-most crossing of the corpus callosum. Note the sudden increase in AC3+ cells in the Cg2/IG on E18 (B). C. Higher magnification view of the Cg2/Ig on E18 in a section without counterstain. AC3+ labeled cell bodies are seen in the Cg2/Ig (upper left) and several processes of labeled cells can be see running perpendicular to the corpus callosum (arrows). D. Immunohistochemistry for Iba1 on E18, in a section rostral to B, shows an accumulation of amoeboid microglia just dorsal and ventral to the CC in the Cg2/IG and subcallosal sling, respectively (arrows). E, F, G. H. Thionin-stained (E, F) and NeuN/GFAP double-stained (G, H) sections of the Cg2/IG region in adult wild-type (Bax+/+; E, G) and Bax −/− (F, H) mice. Arrows point to an accumulation of cells in the Bax −/− animals that is greatly reduced in the wild-type mice. Arrowhead in (H) points to an accumulation of cells around the lateral ventricle as previously described in Bax−/− animals (Kim et al., 2007). Abbreviations: LV, lateral ventrical; CC, corpus callosum. Scale bars = 100 µm in A and B, 10 µm in C, 200 µm in D; scale bar in E is 50 µm for E and F and scale bar in G is 100 µm for G and H. Figure 7 Quantification of AC3+ labeling in the cingulate cortex area 2/indusium griseum (Cg2/IG) and subcallosal sling (SCS) from embryonic day (E) 17 to postnatal day (P) 1. Highest rates of cell death were seen in both regions on E18. *, significantly different from E17 (p < 0.0001 and p < 0.001 for Cg2/Ig and SCS, respectively). Figure 8 Vaginally delivered E19/P0 pups exhibited significantly higher cell death density than cesarean delivered pups on E19/P0 in the (A) suprachiasmatic nucleus (SCN), (B) anteroventral periventricular nucleus of the hypothalamus (AVPV), and (C) subcallosal sling (SCS). Table 1 Primary Antibodies Antibody Immunogen Manufacturer data Concentration Activated caspase-3 (AC3) Peptide sequence CRGTELDCGIETD, which is adjacent to Asp175 in human caspase-3 Cell Signaling Technology, Cat# 9661, Lot 36, RRID:AB_231409, Polyclonal antibody raised in rabbit 1:20,000 Glial fibrillary acidic protein (GFAP) Purified bovine GFAP Chemicon, Cat# AB5804, Lot 0512018283, RRID:AB_2109645, Polyclonal antibody raised in rabbit 1:1,000 Ionized calcium binding adaptor molecule 1 (Iba1) Synthetic peptide corresponding to C- terminal sequence PTGPPAKKAISELP of Iba1 Wako, Cat# 27030, Lot PDJ0493, RRID:AB_2314667, Polyclonal antibody raised in rabbit 1:20,000 Neuronal nuclei (NeuN) Purified cell nuclei from mouse brain Chemicon, Cat# MAB377, Lot 0601019159, RRID:AB_2298772, Monoclonal antibody raised in mouse 1:1,000 Table 2 Results of ANOVAs examining the effects of age on cell death density and total cell death for 16 regions of interest. All p-values were significant (p < 0.0001). AC3+ CELL DENSITY TOTAL AC3+ CELLS REGION F (df) F (df) AVPV 12.43 (8,165) 15.63 (8,165) BNSTL 53.86 (8,179) 33.12 (8,179) BNSTP 31.18 (8,175) 35.23 (8,175) CA1 48.04 (8,164) 25.21 (8,164) CA1 OR 29.01 (8,165) 27.44 (8,165) CA2/3 30.29 (8,164) 30.03 (8,164) CA2/3 OR 50.79 (8,165) 36.84 (8,165) CEA 17.56 (8,161) 17.72 (8,161) DG 26.15 (8,166) 23.85 (8,166) LS 55.33 (8,143) 24.52 (8,143) MEA 66.99 (8,168) 26.18 (8,168) MPON 34.99 (8,144) 24.62 (8,144) NACC CORE 35.81 (8,150) 32.85 (8,150) NACC SHELL 25.55 (8,150) 41.48 (8,150) PVN 21.93 (8,184) 10.39 (8,184) SCN 19.80 (8,187) 27.38 (8,187) The authors quantify cell death in 16 regions of the mouse forebrain during late prenatal and early postnatal life. They find peak cell death just after birth in most regions. They also describe massive cell death perinatally in regions surrounding the corpus callosum, and the persistence of neurons in those regions in adult Bax knockout mice. Conflict of Interest Statement. The authors have nothing to declare. Role of Authors. Study concept and design: MM, THA, NGF; Acquisition of data: MM, CS, KAM, SAM, THA, NGF; Analysis and interpretation of data: MM, THA, NGF; Drafting of the manuscript: NGF; Critical revision of the manuscript for important intellectual contribution: MM, THA, NGF; Obtained funding: THA, NGF; Administrative, technical, and material support: MM, CS, KAM, SAM; Study supervision: THA, NGF. REFERENCES Adamek GD Shipley MT Sanders MS The indusium griseum in the mouse: architecture, Timm's histochemistry and some afferent connections Brain Res Bull 1984 12 657 668 6206929 Ahern TH Krug S Carr AV Murray EK Fitzpatrick E Bengston L McCutcheon J De Vries GJ Forger NG Cell death atlas of the postnatal mouse ventral forebrain and hypothalamus: effects of age and sex J Comp Neurol 2013 521 2551 2569 23296992 Altman J Bayer SA Migration and distribution of two populations of hippocampal granule cell precursors during the perinatal and postnatal periods J Comp Neurol 1990 301 365 381 2262596 Barde YA Trophic factors and neuronal survival Neuron 1989 2 1525 1534 2697237 Bayer SA Development of the hippocampal region in the rat. I. Neurogenesis examined with 3H-thymidine autoradiography J Comp Neurol 1980 190 87 114 7381056 Bernal AL Overview of current research in parturition Exp Physiol 2001 86 213 222 11429638 Blaschke AJ Staley K Chun J Widespread programmed cell death in proliferative and postmitotic regions of the fetal cerebral cortex Development 1996 122 1165 1174 8620843 Blaschke AJ Weiner JA Chun J Programmed cell death is a universal feature of embryonic and postnatal neuroproliferative regions throughout the central nervous system J Comp Neurol 1998 396 39 50 9623886 Bowers JM Waddell J McCarthy MM A developmental sex difference in hippocampal neurogenesis is mediated by endogenous oestradiol Biol Sex Differ 2010 1 8 21208470 Burek MJ Oppenheim RW Programmed cell death in the developing nervous system Brain Pathol 1996 6 427 446 8944315 Buss RR Sun W Oppenheim RW Adaptive roles of programmed cell death during nervous system development Annu Rev Neurosci 2006 29 1 35 16776578 Chang ML Wu CH Jiang-Shieh YF Shieh JY Wen CY Reactive changes of retinal astrocytes and Müller glial cells in kainate-induced neuroexcitotoxicity J Anat 2007 210 54 65 17229283 Cheng XS Li MS Du J Jiang QY Wang L Yan SY Yu DM Deng JB Neuronal apoptosis in the developing cerebellum Anat Histol Embryol 2011 40 21 27 21231956 Cheong JW Chong SY Kim JY Eom JI Jeung HK Maeng HY Lee ST Min YH Induction of apoptosis by apicidin, a histone deacetylase inhibitor, via the activation of mitochondria-dependent caspase cascades in human Bcr-Abl-positive leukemia cells Clin Cancer Res 2003 9 5018 5027 14581377 Cregan SP Fortin A MacLaurin JG Callaghan SM Cecconi F Yu SW Dawson TM Dawson VL Park DS Kroemer G Slack RS Apoptosis-inducing factor is involved in the regulation of caspase-independent neuronal cell death J Cell Biol 2002 158 507 517 12147675 Creps ES Time of origin in preoptic and septal areas of the mouse: An autoradiographic study J Comp Neurol 1974 157 161 244 4416651 Deckwerth TL Elliott JL Knudson CM Johnson EM Jr Snider WD Korsmeyer SJ BAX is required for neuronal death after trophic factor deprivation and during development Neuron 1996 17 401 411 8816704 Forger NG Rosen GJ Waters EM Jacob D Simerly RB de Vries GJ Deletion of Bax eliminates sex differences in the mouse forebrain Proc Natl Acad Sci U S A 2004 101 13666 13671 15342910 Forger NG Control of cell number in the sexually dimorphic brain and spinal cord J Neuroendocrinol 2009 21 393 399 19207822 Fowden AL Li J Forhead AJ Glucocorticoids and the preparation for life after birth: are there long-term consequences of the life insurance? Proc Nutr Soc 1998 57 113 122 9571716 Fujioka T Fujioka A Endoh H Sakata Y Furukawa S Nakamura S Materno-fetal coordination of stress-induced Fos expression in the hypothalamic paraventricular nucleus during pregnancy Neuroscience 2003 118 409 415 12699777 Fujioka T Sakata Y Yamaguchi K Shibasaki T Kato H Nakamura S The effects of prenatal stress on the development of hypothalamic paraventricular neurons in fetal rats Neurosci 1999 92 1079 1088 Golightly E Jabbour HN Norman JE Endocrine immune interactions in human parturition Mol Cell Endocrinol 2011 335 52 59 20708653 Gross GA Imamura T Luedke C Vogt SK Olson LM Nelson DM Sadovsky Y Muglia LJ Opposing actions of prostaglandins and oxytocin determine the onset of murine labor Proc Natl Acad Sci U S A 1998 95 11875 11879 9751758 Hankin MH Schneider BF Silver J Death of the subcallosal sling is correlated with formation of the cavum septi pellucidi J. Comp. Neurol 1988 272 191 202 2456310 Hashimoto H Eto T Endo K Itai G Kamisako T Suemizu H Ito M Comparative study of doses of exogenous progesterone administration needed to delay parturition in Jcl:MCH(ICR) mice Exp Anim 2010 59 521 524 20660999 Henderson RG Brown AE Tobet SA Sex differences in cell migration in the preoptic area/anterior hypothalamus of mice J Neurobiol 1999 41 252 266 10512982 Hengartner MO The biochemistry of apoptosis Nature 2000 407 770 776 11048727 Hillman N Kallapur SG Jobe A Physiology of transition from intrauterine to extrauterine life Clin Perinatol 2012 39 769 783 23164177 Holmes MM McCutcheon J Forger NG Sex differences in NeuN- and androgen receptor-positive cells in the bed nucleus of the stria terminalis are due to Bax-dependent cell death Neurosci 2009 158 1251 1256 Hossain A Hajman K Charitidi K Erhardt S Zimmermann U Knipper M Canlon B Prenatal dexamethasone impairs behavior and the activation of the BDNF exon IV promoter in the paraventricular nucleus in adult offspring Endocrinol 2008 149 6356 6365 Hyman BT Yuan J Apoptotic and non-apoptotic roles of caspases in neuronal physiology and pathophysiology Nat Rev Neurosci 2012 13 395 406 22595785 Ikonomidou C Mosinger JL Salles KS Labruyere J Olney JW Sensitivity of the developing rat brain to hypobaric/ischemic damage parallels sensitivity to N-methyl-aspartate neurotoxicity J Neurosci 1989 9 2809 2818 2671294 Ikonomidou C Bosch F Miksa M Bittigau P Vöckler J Dikranian K Tenkova TI Stefovska V Turski L Olney JW Blockade of NMDA receptors and apoptotic neurodegeneration in the developing brain Science 1999 283 70 74 9872743 Imai Y Ibata I Ito D Ohsawa K Kohsaka S A novel gene iba1 in the major histocompatibility complex class III region encoding an EF hand protein expressed in a monocytic lineage Biochem Biophys Res Commun 1996 224 855 862 8713135 Ito D Imai Y Ohsawa K Nakajima K Fukuuchi Y Kohsaka S Microglia-specific localisation of a novel calcium binding protein, Iba1 Mol Brain Res 1998 57 1 9 9630473 Jackson-Lewis V Vila M Djaldetti R Guegan C Liberatore G Liu J O'Malley KL Burke RE Przedborski S Developmental cell death in dopaminergic neurons of the substantia nigra of mice J Comp Neurol 2000 424 476 488 10906714 Jacob DA Bengston CL Forger NG Effects of Bax gene deletion on muscle and motoneuron degeneration in a sexually dimorphic neuromuscular system J Neurosci 2005 25 5638 5644 15944391 Jain L Eaton DC Physiology of fetal lung fluid clearance and the effect of labor Semin Perinatol 2006 30 34 43 16549212 Jabès A Lavenex PB Amaral DG Lavenex P Quantitative analysis of postnatal neurogenesis and neuron number in the macaque monkey dentate gyrus Eur. J. Neurosci 2010 31 273 285 20074220 Jeff R Bibhash PC Naoko B Xuemei Z Jason MD Sudhansu DK Coordinated regulation of fetal and maternal prostaglandins directs successful birth and postnatal adaptation in the mouse Proc Natl Acad Sci U S A 2000 97 9759 9764 10944235 Jung AR Kim TW Rhyu IJ Kim H Lee YD Vinsant S Oppenheim RW Sun W Misplacement of Purkinje cells during postnatal development in Bax knock-out mice: a novel role for programmed cell death in the nervous system? J Neurosci 2008 28 2941 2948 18337425 Karim MA Sloper JC Histogenesis of the supraoptic and paraventricular neurosecretory cells of the mouse hypothalamus J Anat 1980 130 341 347 7400041 Kim WR Chun SK Kim TW Kim H Ono K Takebayashi H Ikenaka K Oppenheim RW Sun W Evidence for the spontaneous production but massive programmed cell death of new neurons in the subcallosal zone of the postnatal mouse brain Eur J Neurosci 2011 33 599 611 21219476 Kim WR Kim Y Eun B Park OH Kim H Kim K Park CH Vinsant S Oppenheim RW Sun W Impaired migration in the rostral migratory stream but spared olfactory function after the elimination of programmed cell death in Bax knock-out mice J Neurosci 2007 27 14392 14403 18160647 Koester SE O'Leary DD Axons of early generated neurons in cingulate cortex pioneer the corpus callosum J Neurosci 1994 14 6608 6620 7965064 Kuan CY Roth KA Flavell RA Rakic P Mechanisms of programmed cell death in the developing brain Trends Neurosci 2000 23 291 297 10856938 Langenbach R Morham SG Tiano HF Loftin CD Ghanayem BI Chulada PC Mahler JF Davis BJ Lee CA Disruption of the mouse cyclooxygenase 1 gene. Characteristics of the mutant and areas of future study Adv Exp Med Biol 1997 407 87 92 9321936 Leary SL American Veterinary Medical Association AVMA guidelines for the euthanasia of animals 2013 2013 Available from: https://www.avma.org/KB/Policies/Documents/euthanasia.pdf Li Z Jo J Jia J-M Lo S-C Whitcomb DJ Song J Cho K Sheng M Caspase-3 activation via mitochondria is required for long-term depression and AMPA receptor internalization Cell 2010 141 859 871 20510932 Marín-Teva JL Dusart I Colin C Gervais A van Rooijen N Mallat M Microglia promote the death of developing Purkinje cells Neuron 2004 41 535 547 14980203 McCormick CM Smythe JW Sharma S Meaney MJ Sex-specific effects of prenatal stress on hypothalamic-pituitary-adrenal responses to stress and brain glucocorticoid receptor density in adult rats Dev Brain Res 1995 84 55 61 7720217 McDonald JW Silverstein FS Johnston MV Neurotoxicity of N-methyl-D-aspartate is markedly enhanced in developing rat central nervous system Brain Res 1988 459 200 203 3048538 Morshead CM van der Kooy D Postmitotic death is the fate of constitutively proliferating cells in the subependymal layer of the adult mouse brain J Neurosci 1992 12 249 256 1729437 Mullen RJ Buck CR Smith AM NeuN, a neuronal specific nuclear protein in vertebrates Development 1992 116 201 211 1483388 Murray SA Morgan JL Kane C Sharma Y Heffner CS Lake J Donahue LR Mouse gestation length is genetically determined PLoS ONE 2010 5 e12418 20811634 Niimi K Harada I Kusaka Y Kishi S The ontogenetic development of the diencephalon of the mouse Tokushima J Exp Med 1962 8 203 238 Nilsson I Lindfors C Fetissov SO Hökfelt T Johansen JE Aberrant agouti-related protein system in the hypothalamus of the anx/anx mouse is associated with activation of microglia J Comp Neurol 2008 507 1128 1140 18098136 Niquille M Garel S Mann F Hornung JP Otsmane B Chevalley S Parras C Guillemot F Gaspar P Yanagawa Y Lebrand C Transient neuronal populations are required to guide callosal axons: a role for semaphorin 3C PLoS One 2009 7 e1000230 Olney JW New insights and new issues in developmental neurotoxicology Neurotoxicology 2002 23 659 668 12520755 Oppenheim RW Naturally occurring cell death during neural development Trends Neurosci 1985 8 487 493 Ozaki HS Wahlsten D Timing and origin of the first cortical axons to project through the corpus callosum and the subsequent emergence of callosal projection cells in mouse J Comp Neurol 1998 400 197 206 9766399 Paxinos G Halliday G Watson C Koutcherov Y Wang HQ Atlas of the Developing Mouse Brain at E17.5, P0, and P6 2007 San Diego Elsevier Pont-Lezica L Bechade C Belarif-Cantout Y Pascual O Bessis A Physiological roles of microglia during development J Neurochem 2011 119 901 908 21951310 Porter AG Jänicke RU Emerging roles of caspase-3 in apoptosis Cell Death Differ 1999 6 99 104 10200555 Pritchett K Corrow D Stockwell J Smith A Euthanasia of neonatal mice with carbon dioxide Comp Med 2005 55 275 281 16089177 Purves D Body and Brain: A Trophic Theory of Neural Connections 1988 Harvard University Press Reyes R Haendel M Grant D Melancon E Eisen JS Slow degeneration of zebrafish Rohon-Beard neurons during programmed cell death Dev Dyn 2004 229 30 41 14699575 Reznikov KY Cell proliferation and cytogenesis in the mouse hippocampus Adv Anat Embryol Cell Biol 1991 122 1 82 1927657 Rideout HJ Stefanis L Caspase inhibition: a potential therapeutic strategy in neurological diseases Histol Histopathol 2001 16 895 908 11510981 Roth KA D'Sa C Apoptosis and brain development Ment Retard Dev Disabil Res Rev 2001 7 261 266 11754520 Shu T Richards LJ Cortical axon guidance by the glial wedge during the development of the corpus callosum J Neurosci 2001 21 2749 2758 11306627 Shu T Li Y Keller A Richards LJ The glial sling is a migratory population of developing neurons Development 2003 130 2929 2937 12756176 Silver J Lorenz SE Wahlsten D Coughlin J Axonal guidance during development of the great cerebral commissures: descriptive and experimental studies, in vivo, on the role of preformed glial pathways J Comp Neurol 1982 210 10 29 7130467 Soares MJ Talamantes F Pre-parturitional changes in serum prolactin, placental lactogen, growth hormone, progesterone, and corticosterone in the C3H/HeN mouse J Dev Physiol 1984 6 423 429 6501813 Soriano E Cobas A Fairén A Neurogenesis of glutamic acid decarboxylase immunoreactive cells in the hippocampus of the mouse. I: Regio superior and regio inferior J Comp Neurol 1989 281 586 602 2708583 Southwell DG Paredes MF Galvao RP Jones DL Froemke RC Sebe JY Alfaro-Cervello C Tang Y Garcia-Verdugo JM Rubenstein JL Baraban SC Alvarez-Buylla A Intrinsically determined cell death of developing cortical interneurons Nature 2012 491 109 113 23041929 Shrivastava K Gonzalez P Acarin L The immune inhibitory complex CD200/CD200R is developmentally regulated in the mouse brain J Comp Neurol 2012 520 2657 2675 22323214 Srinivasan A Roth KA Sayers RO Shindler KS Wong AM Fritz LC Tomaselli KJ In situ immunodetection of activated caspase-3 in apoptotic neurons in the developing nervous system Cell Death Differ 1998 5 1004 1016 9894607 Stanfield BB Cowan WM The development of the hippocampus and dentate gyrus in normal and reeler mice J Comp Neurol 1979 185 423 459 86549 Stankovski L Alvarez C Ouimet T Vitalis T El-Hachimi KH Price D Deneris E Gaspar P Cases O Developmental cell death is enhanced in the cerebral cortex of mice lacking the brain vesicular monoamine transporter J Neurosci 2007 27 1315 1324 17287506 Sugimoto Y Yamasaki A Segi E Tsuboi K Aze Y Nishimura T Oida H Yoshida N Tanaka T Katsuyama M Hasumoto K Murata T Hirata M Ushikubi F Negishi M Ichikawa A Narumiya S Failure of parturition in mice lacking the prostaglandin F receptor Science 1997 277 681 683 9235889 Sun W Gould TW Vinsant S Prevette D Oppenheim RW Neuromuscular development after the prevention of naturally occurring neuronal death by Bax deletion J Neurosci 2003 23 7298 7310 12917363 Thilaganathan B Meher-Homji N Nicolaides KH Labor: an immunologically beneficial process for the neonate Am J Obstet Gynecol 1994 171 1271 1272 7977532 Thomaidou D Mione MC Cavanagh JF Parnavelas JG Apoptosis and its relation to the cell cycle in the developing cerebral cortex J Neurosci 1997 17 1075 1085 8994062 Thomson AJ Telfer JF Young A Campbell S Stewart CJ Cameron IT Greer IA Norman JE Leukocytes infiltrate the myometrium during human parturition: further evidence that labour is an inflammatory process Hum Reprod 1999 14 229 236 10374126 Tobe I Ishida Y Tanaka M Endoh H Fujioka T Nakamura S Effects of repeated maternal stress on FOS expression in the hypothalamic paraventricular nucleus of fetal rats Neurosci 2005 134 387 395 Turgeon B Meloche S Interpreting neonatal lethal phenotypes in mouse mutants: insights into gene function and human diseases Physiol Rev 2009 89 1 26 19126753 Verney C Takahashi T Bhide PG Nowakowski RS Caviness VS Jr Independent controls for neocortical neuron production and histogenetic cell death Dev Neurosci 2000 22 125 138 10657705 Wake H Moorhouse AJ Miyamoto A Nabekura J Microglia: actively surveying and shaping neuronal circuit structure and function Trends Neurosci 2012 36 209 217 23260014 Wakselman S Béchade C Roumier A Bernard D Triller A Bessis A Developmental neuronal death in hippocampus requires the microglial CD11b integrin and DAP12 immunoreceptor J Neurosci 2008 28 8138 8143 18685038 White FA Keller-Peck CR Knudson CM Korsmeyer SJ Snider WD Widespread elimination of naturally occurring neuronal death in Bax-deficient mice J Neurosci 1998 18 1428 1439 9454852 Whitlock KE Westerfield M A transient population of neurons pioneers the olfactory pathway in the zebrafish J Neurosci 1998 18 8919 8827 9786997 Yakovlev AG Ota K Wang G Movsesyan V Bao WL Yoshihara K Faden AI Differential expression of apoptotic protease-activating factor-1 and caspase-3 genes and susceptibility to apoptosis during brain development and after traumatic brain injury J Neurosci 2001 21 7439 7446 11567033 Yamaguchi Y Miura M Programmed cell death in neurodevelopment Dev Cell 2015 32 478 490 25710534 Yuan J Yankner BA Apoptosis in the nervous system Nature 2000 407 802 809 11048732 Zacharaki T Sophou S Giannakopoulou A Dinopoulos A Antonopoulos J Parnavelas JG Dori I Natural and lesion-induced apoptosis in the dorsal lateral geniculate nucleus during development Brain Res 2010 1344 62 76 20471376 Zhan RZ Wu C Fujihara H Taga K Qi S Naito M Shimoji K Both caspase-dependent and caspase-independent pathways may be involved in hippocampal CA1 neuronal death because of loss of cytochrome c from mitochondria in a rat forebrain ischemia model J Cereb Blood Flow Metab 2001 21 529 540 11333363 Zohar I Weinstock M Differential effect of prenatal stress on the expression of corticotrophin-releasing hormone and its receptors in the hypothalamus and amygdala in male and female rats J Neuroendocrinol 2011 23 320 328 21306450 Zuloaga DG Carbone DL Hiroi R Chong DL Handa RJ Dexamethasone induces apoptosis in the developing rat amygdala in an age-, region-, and sex-specific manner Neuroscience 2011 199 535 547 22008524

PMC005xxxxxx/PMC5116300.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8805635 27439 Immunol Allergy Clin North Am Immunol Allergy Clin North Am Immunology and allergy clinics of North America 0889-8561 1557-8607 27712766 5116300 10.1016/j.iac.2016.06.008 NIHMS795565 Article Eosinophils and Mast Cells in Aspirin-Exacerbated Respiratory Disease Steinke John W. 1 Payne Spencer C 3 Borish Larry 2 1 Asthma and Allergic Disease Center, Carter Center for Immunology Research, Department of Medicine, University of Virginia Health System Charlottesville, VA 2 Asthma and Allergic Disease Center, Carter Center for Immunology Research, Department of Microbiology, University of Virginia Health System Charlottesville, VA 3 Department of Otolaryngology – Head and Neck Surgery, University of Virginia Health System Charlottesville, VA 2 Corresponding Author: Larry Borish, Asthma and Allergic Disease Center, Box 801355, University of Virginia Health System, Charlottesville, VA 22908, Telephone #: (434)-243-6570; Fax #: (434)-924-5779; lb4m@virginia.edu 17 6 2016 13 9 2016 11 2016 01 11 2017 36 4 719734 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Aspirin-exacerbated respiratory disease (AERD) is explained in part by over-expression of pro-inflammatory mediators including 5-lipoxygenase and leukotriene C4 synthase (LTC4S) that results in constitutive over-production of cysteinyl leukotrienes (CysLTs). Mast cells and eosinophils are two cell types that have important roles in mediating many of the effects observed in this disease. Increased levels of both interleukin (IL-4) and interferon (IFN)-γ are present in the tissue of AERD subjects. Previous studies demonstrated that IL-4 is primarily responsible for the upregulation of LTC4S by mast cells. Our studies demonstrate that IFN-γ, but not IL-4 drives this process in eosinophils. We also extend to both IL-4 and IFN-γ the ability to upregulate CysLT receptors. Prostaglandin E2 (PGE2) acts to prevent CysLT secretion by inhibiting mast cell and eosinophil activation. PGE2 concentrations are reduced in AERD and studies confirm that this reflects diminished expression of cyclooxygenase (COX)-2, a process again that is driven by IL-4. Thus, IL-4 and IFN-γ acting on eosinophils and mast cells together play an important pathogenic role in generating the phenotype of AERD. This review will examine the overall role that eosinophils and mast cells contribute to the pathophysiology of AERD. eosinophil mast cell leukotriene cyclooxygenase prostaglandin aspirin-exacerbated respiratory disease arachidonic acid Introduction In 1968, the term Samter’s triad was coined and was defined by the presence of nasal polyps, aspirin sensitivity, and asthma 1, however this disease is now referred to as aspirin-exacerbated respiratory disease (AERD) as asthma is not always present despite reactions to aspirin. AERD comprises as many as 7% of adult-onset asthmatics and up to 12–14% of adult asthmatics with severe asthma 2, 3. This disorder is characterized by the unique intolerance to aspirin and other non-selective cyclooxygenase (COX) inhibitors 4–6. Other characteristics include hypereosinophilia, both in circulation and in the tissue, a tendency to develop de novo in adulthood 5, 7–9 and often an absence of identifiable atopy 5, 7. Sinusitis is present in this disorder, the degree of which is often severe and associated with complete or near complete opacification of the sinus cavity 9. Though not a requirement, when asthma is present it often progresses in severity and is associated with aggressive airway remodeling 10. During aspirin reactions, many mediators are released including cysteinyl leukotrienes (CysLT), tryptase, eosinophil cationic protein (ECP) and prostaglandin D2 (PGD2) suggesting both mast cell and eosinophil activation 11–13. Recently, aspirin was shown to directly activate both of these cell types ex vivo potentiating mediator release 14. A predominant physiological feature of AERD is the robust over-production and over-responsiveness to CysLTs (inflammatory) 12, 15 while at the same time there is under-production and under-responsiveness to the anti-inflammatory lipid mediator PGE2 16–18. These CysLTs have important pro-inflammatory and pro-fibrotic effects that contribute to the asthma severity and to the extensive hyperplastic sinusitis and nasal polyposis 9, 19, 20. And, conversely, the down-regulation of PGE2 pathways reduces the constraints that would normally act to attenuate these pro-inflammatory pathways 21. This review will focus on the role that eosinophils and mast cells play in contributing to these cardinal features of AERD. Eosinophil and mast cell numbers in AERD Chronic sinusitis is now recognized as a collection of disorders that result from inflammation of the sinuses and in many cases can be separated into different types based on the cellular infiltrate. One distinguishing feature in the nasal polyps that often form in association with chronic sinusitis is the presence or absence of eosinophils and amongst eosinophilic polyps a distinction can be made between AERD, allergic fungal sinusitis (AFS) and chronic hyperplastic eosinophilic sinusitis (CHES) 22, 23. Within the eosinophilic polyps, AERD has more than twice the number of eosinophils in the polyp tissue than AFS or CHES, implicating them as important cells in the disease process 23. Examination of bronchial biopsies from AERD subjects also revealed highly elevated eosinophil numbers in comparison to aspirin-tolerant asthmatics and non-asthmatics 24. These eosinophils were in an activated state as evidenced by the presence of secretory ECP 24. We have reported lower numbers of mast cells in nasal polyps from eosinophilic sinus disease as compared to healthy tissue by both toluidine blue and chloroacetate (chymase) staining, and there was no difference between aspirin tolerant and AERD groups 23. This contrasts somewhat with a previous report that found no difference in mast cell numbers in nasal polyps from AERD groups when compared to allergic or non-allergic subjects via tryptase staining 25. The differences in the results of these studies may reflect the use of different markers of mast cells (chymase vs. tryptase) or the stratification of the groups, the latter study not taking into account eosinophilic infiltration into the polyp tissue. Regardless, there do not appear to be more mast cells in nasal polyps in subjects with AERD. However, this result may be erroneous – given the high expression of mast cell-derived mediators – and perhaps reflects the inability to stain for activated, granule-depleted mast cells (so-called “phantom” mast cells). Similarly, when the lungs have been examined, as with NPs, fewer numbers of mast cells have been found in AERD subjects compared to non-asthmatic controls using immunohistochemistry to stain for tryptase positive cells 24. Another study examining the bronchial mucosa found increased numbers of tryptase-positive mast cells only in subjects with non-aspirin sensitive asthma: again, AERD and healthy controls paradoxically had similar numbers 26. Development of Eosinophils and Mast Cells Eosinophils develop from pluripotent hematopoietic stem cells in bone marrow that initially differentiate into eosinophil/basophil progenitors or colony forming units (Eo/B CFU) (Figure 1). Eo/B CFU are mononuclear cells that express CD34, CD35, and interleukin (IL)-5 receptors (CD125) that are capable of responding to appropriate cytokine signals allowing differentiation into mature basophils and eosinophils 27, 28. Eo/B CFU are increased in numbers in both the blood and bone marrow of allergic patients and further increases in their number are observed following allergen exposure 28. These progenitors are also observed in nasal polyp tissue 29. Several transcription factors including GATA-1, PU.1 and C/EBP are induced in response to appropriate cytokine signals and become involved in the development of the eosinophil lineage and eosinophil-associated genes 30–32. In vitro eosinophil differentiation experiments have demonstrated that GATA-1 is the primary transcription factor responsible for this eosinophil lineage specification 33. Three cytokines, IL-3, IL-5 and granulocyte macrophage colony-stimulating factor (GM-CSF) play the most important roles in the regulation of eosinophil development (Figure 1). The function of IL-3 is the broadest as it leads to the expansion of a variety of cell types including monocytes, megakaryocytes, erythrocytes, basophils, neutrophils and eosinophils 27. GM-CSF acts in a similar fashion, albeit with more mature precursor cells, inducing the formation of macrophages, neutrophils and eosinophils 34. IL-5 is the cytokine responsible for selective terminal differentiation of eosinophils 35 and stimulates the release of eosinophils from the bone marrow into peripheral circulation 36. GM-CSF, IL-3 and the chemokines CCL11, CCL24, and CCL26 (eotaxins) are also involved in eosinophil homeostasis and play an important role upon arrival of the eosinophil at a tissue location. In addition, an IFN-γ-induced transcription factor ICSBP can also drive the differentiation of eosinophils 37. As IFN-γ is routinely present in allergic inflammation and, in our studies, was particularly upregulated in AERD 38, this led us to speculate on the role of IFN-γ being able to contribute to eosinophilia. Using an in vitro model with CD34+ hematopoietic progenitors, we demonstrated the capacity of IFN-γ, acting in synergy with IL-5, to promote the survival and differentiation of mature bi-lobed, CCR3- and Siglec-8-expressing eosinophils 38 confirming prior studies 39 regarding the influence of this cytokine on eosinophil-mediated inflammation. As with eosinophils, mast cells are derived from pluripotent hematopoietic CD34+ cells from the bone marrow 40, however mast cells do not fully mature until they reach their final tissue destination with the exception of mast cell leukemia. IL-3 and IL-6 increase early CD34+ progenitor cell numbers and begin the differentiation process, however it is binding of stem cell factor (SCF) to its receptor c-Kit (CD117) that is the master growth and differentiation factor for human mast cells (Figure 1) 41. SCF is produced primarily by stromal cells 42 and it can either be released as a soluble growth factor or expressed on the cell surface of these cells. It is the expression and binding of SCF at tissue sites that cause the CD34+ precursor cells to arrest and terminally differentiate into mast cells. Cysteinyl leukotriene over-production and over-responsiveness in AERD Arguably, the best-characterized molecules associated with AERD are the CysLTs. A unique characteristic of the disease is the over-production of CysLTs in the resting state and a tremendous surge in CysLT production in response to aspirin and other non-selective cyclooxygenase inhibitors that target COX-1 43. Included in this list are the non-selective non-steroidal anti-inflammatory drugs (NSAIDs) as well as other inhibitors of COX-1 44, 45. The high CysLT levels in AERD reflect increased expression of the primary synthesis enzymes 5-lipoxygenase (5-LO) and more importantly leukotriene C4 synthase (LTC4S) (Figure 2). Increased expression of these enzymes is observed in the lungs, sinuses and nasal polyps of AERD subjects with eosinophils and resident mast cells being the primary cells expressing the enzymes 16, 19, 24, 46. Not only do AERD subjects produce more CysLTs, but they also demonstrate an increased sensitivity to CysLTs 47. Initially, two CysLT receptors were identified and were distinguished from each other by their differing potency for the CysLTs: CysLT1 receptors primarily respond to LTD4 whereas CysLT2 receptors respond equally to LTD4 and LTC4. Neither receptor responds well to LTE4. Acting through CysLT1 and inhibited by the CysLT2 receptor, CysLTs induce mast cell proliferation through activation of c-kit and extracellular signal-regulated kinase 48. In AERD sinus tissue, high levels of CysLT1 were found in comparison to healthy tissue and following aspirin desensitization, the CysLT1 levels returned to normal 49. CysLT1 receptors are also prominently expressed on airway smooth muscle 50 and these receptors mediate a portion of the CysLT-induced bronchospasm associated with aspirin challenges or desensitizations 51–54 as demonstrated by the ability of leukotriene receptor antagonists to attenuate much of the bronchospasm that occurs during these procedures. While these findings made it appear that CysLT1 was the only important leukotriene receptor in AERD, several observations suggested they were only partially involved in the pathogenesis of disease. The most abundant leukotriene found in the circulation and airway is LTE4 with C4 and D4 being rapidly converted to E4, thus limiting their duration of action in vivo. Inhalation of LTE4 by asthmatic and AERD subjects potentiates airway hyperresponsiveness to subsequent challenges with histamine, the effect of which could be blocked with indomethacin 47, 55, 56. In the bronchial mucosa of asthmatics, inhalation of LTE4, but not LTD4, causes recruitment of eosinophils, basophils and mast cells into the tissue 57, 58. Mice in which both the CysLT1 and CysLT2 receptors have been deleted show enhanced skin swelling in response to intracutaneous LTE4 in comparison to mice with these genes intact 59. These studies led to the exploration for and ultimate identification of additional CysLT receptors that selectively respond to LTE4 (GPR99 and P2Y12) 59–62. The functional role of these LTE4 receptors in AERD and the utility in targeting them as a therapeutic option are areas of active research, although diminished synthesis of CysLTs and, by extension, of LTE4, could explain the superior efficacy of 5-lipoxygenase inhibitors in treating this disorder 63. Underappreciated role for PGD2 in AERD PGD2 and its metabolites have been found in blood and urine following aspirin challenge 11, 13, 64 and the conventional thought is that it is primarily mast cell derived, however reports demonstrate that eosinophils are also a source 65, 66. Synthesis of PGD2 is regulated by hematopoietic prostaglandin D2 synthase (hPGDS) (Figure 2) and when secreted it binds two receptors CRTH2 (DP2) and DP1 that are expressed on numerous cell types. Activities of PGD2 binding include stimulation of cell migration (including eosinophils and innate lymphoid type 2 (ILC2) cells), bronchoconstriction, vasodilation (flushing), and cellular activation and differentiation 67–70. A role for PGD2 in the pathogenesis of AERD is supported by studies in chronic sinusitis. Expression of hPGDS has been observed in polyps, specifically in eosinophils 66, 71, 72. The degree of hPGDS expression correlated with eosinophil number and severity of disease. These studies did not examine AERD subjects, but given the high levels of eosinophilic infiltrate in AERD tissue, it is likely that hPGDS levels would have correlated as well. Recent work from our lab demonstrated that amongst CRS syndromes, in AERD the highest levels of hPGDS transcripts and proteins expression were observed and with eosinophils being the predominant cell type where expression was localized 66. As discussed in other chapters, aspirin desensitization followed by high-dose aspirin therapy is often used to treat AERD, however not all patients tolerate the desensitization protocol. It has recently been shown that those subjects who cannot be desensitized express higher basal levels of PGD2 (and thromboxane) in their serum and urine 64. During the desensitization procedure, these patients had a surge of PGD2 release but not thromboxane, whereas those who were successfully desensitized had decreased thromboxane and unchanged PGD2 levels 64. In this study, the correlation between PGD2 and eosinophil number in the polyp tissue was not performed. Baseline PGD2 levels may serve as a marker to identify those who will successfully undergo desensitization. PGE2 and PGE2 receptor dysregulation in AERD PGE2 displays both pro- and anti-inflammatory functions reflecting its ability to interact with 4 distinct receptors (EP1-4) each having various activating or inhibitory functions. However, it is the role of PGE2 acting through anti-inflammatory EP2 receptors to block eosinophil and mast cell degranulation that is central to the pathogenesis of AERD. AERD patients constitutively produce low levels of PGE2 16, 73 attenuating the anti-inflammatory constraints provided by this lipid in this basal state. The further reduction of tissue PGE2 concentrations by aspirin and other NSAIDs through COX-1 inhibition precipitates the activation of eosinophils and mast cells in AERD, as demonstrated by the ability of inhaled PGE2 to protect against these reactions 74, 75. This sensitivity of AERD patients to low tissue PGE2 concentrations is amplified by their reduced expression of the anti-inflammatory EP2 receptor 18. Serra-Pages and colleagues demonstrated that the ratio of EP2 to EP3 receptors on the surface of a mast cell influences the activation potential of these cells when the high affinity IgE receptor (FcεRI) is stimulated in a PGE2-containing milieu 76. Through examination of various mast cell lines, the authors found that those with high levels of EP2 could suppress FcεRI activation in the presence of PGE2, but when EP3 levels were high FcεRI activation of mast cells was enhanced. It is likely that the low EP2/EP3 ratio on mast cells and possibly eosinophils in AERD contributes to this disease as any PGE2 that is available would preferentially signal through the EP3 receptor and activate these cells, thus contributing to the pro-inflammatory cascade. Several studies have investigated the mechanism behind the reduced levels of PGE2 in AERD and, perhaps not surprisingly, have correlated this with a decrease in the relevant upstream metabolic enzymes. The production of PGE2 from arachidonic acid (Figure 2) involves the sequential synthesis of PGG2/PGH2 by the two cyclooxygenase enzymes (COX-1 and COX-2) followed by the synthesis of PGE2 by the microsomal PGE2 synthases (mPGES-1, mPGES-2) and cytosolic PGE2 synthase (cPGES). It is mPGES-1 that is most relevant to PGE2 production in inflammatory disorders such as AERD as it is the enzyme primarily functionally coupled to COX-2 77. COX-2 mRNA and protein expression are markedly diminished in AERD 16, 17, 78. Our studies have confirmed this diminished expression of COX-2 79. We found no significant change in COX-1 and a trend towards diminished mPGES-1 expression and that this is driven in part by IL-4. Diminished COX-2 expression and the reduced capacity to synthesize PGE2 contributes to the severity of inflammation observed in AERD and accentuates the sensitivity of these individuals to the further inhibition of PGE2 synthesis associated with aspirin and other NSAIDs. With this relative absence of COX-2, AERD subjects become dependent upon COX-1 for the PGE2 that is necessary to restrain mast cell and eosinophil activation. Thus, most AERD patients tolerate selective COX-2 inhibitors 80, supporting this concept regarding the unique importance of COX-1-derived PGE2. Cytokine expression in AERD Numerous studies have examined the cytokine milieu found in the tissue of eosinophilic sinusitis and AERD. Many of these studies have reported expression of a Th2-like immune profile (IL-4, IL-13 and IL-5), similar to other allergic diseases with eosinophilic infiltrate, though few have specifically focused on AERD 81–88. However, numerous observations suggest that in contrast to eosinophilic sinusitis patients who tolerate aspirin, AERD appears to have a mixed Th2- and Th1-like cytokine milieu characterized by prominent expression of IFN-γ. The first study to suggest this examined non-allergic sinusitis patients and demonstrated enhanced IFN-γ expression in this cohort: a group in which AERD is likely to be over-expressed 89. Since this initial report, high levels of IFN-γ in chronic sinusitis have been reported by other investigators, although AERD subjects were not specifically separated or recruited for the studies 90. One study specifically addressed IFN-γ expression in AERD and found enhanced levels of IFN-γ in circulating CD8+ cells compared to aspirin-tolerant controls 91. The concept that AERD reflects a mixed Th2- and Th1-like cytokine profile was confirmed in our recent study in which NP tissue derived from AERD subjects was contrasted with those obtained from aspirin tolerant and control subjects by their over-expression of IFN-γ mRNA transcripts and protein 38. To our surprise, we found that eosinophils themselves were the most important source for this cytokine, which is consistent with previous studies that demonstrated IFN-γ can be expressed by eosinophils in substantial quantities 92–94. In contrast to our findings, a recent report did not find elevated levels of IFN-γ protein in nasal polyps from AERD subjects, however this group did not perform flow cytometry, immunohistochemistry or quantitative PCR to verify their results 88. Cytokine dysregulation of LTC4S and CysLT receptors Under non-inflammatory conditions, mast cells express moderate levels of LTC4S that can be increased greatly following stimulation by IL-4 (but not by IL-5 or IL-13) 95. It has also been observed that IFN-γ also has the capability of upregulating LTC4S expression in umbilical cord-derived mast cell progenitors (Joshua Boyce, personal communication). While mast cells are capable of synthesizing CysLTs, previous studies 24 have demonstrated that eosinophils are the more important cell type over-expressing LTC4S in AERD. Examining a battery of cytokines (including IL-3, IL-4, IL-5, GM-CSF, IL-1, TNF-α and IFN-γ), we were unable to demonstrate an ability of any of these cytokines to modulate LTC4S expression on circulating eosinophils. This may be due to their terminal differentiation state and short life span. We have also failed to demonstrate the ability of IL-4 to increase LTC4S expression on eosinophils differentiated from progenitors in the presence of IL-3 and IL-5. However, when the progenitors were incubated with IFN-γ during the differentiation stage, a significant increase in LTC4S expression was observed 38. The increase in LTC4S expression translated into increased capacity of these newly differentiated eosinophils to secrete CysLTs upon activation by aspirin 14. Another mechanism of CysLT production involves the transcellular conversion of LTA4 by adherent platelets expressing LTC4S 96. The frequency of platelet-adherent eosinophils, neutrophils and monocytes are markedly elevated in the blood of AERD subjects. Any LTA4 that is released into the extracellular space can be captured by the platelets and converted to CysLTs. It has been estimated that adherent-platelets contribute up to 50% of the total LTC4S activity in blood and thus would represent a significant source of the CysLTs found in AERD 96. CysLT receptor expression is regulated by numerous cytokines including IL-4 and IFN-γ, but also IL-5 and IL-13. IL-4 increases expression of CysLT1 on mast cells 97, 98 and monocytes 99. In our studies, IL-4 also increased the expression of both CysLT1 and CysLT2 on T and B lymphocytes and eosinophils 100. We also demonstrated robust upregulation of CysLT1 and CysLT2 receptors in response to IFN-γ on T cells and eosinophils 100. In recent studies, examination of eosinophils derived from CD34+ progenitors have also demonstrated the ability of IFN-γ to upregulate CysLT1 and CysLT2 receptor expression 38. There have been no reports of cytokine modulation of other CysLT receptors. In summary, in the AERD cytokine environment, both mast cells and eosinophils are primed to produce and respond to the CysLTs that are produced during the disease process. Cytokine dysregulation of PGE2 Synthesis and EP2 Receptors Cytokine regulation of the PGE2 synthesis pathway in AERD has not been thoroughly investigated. However, in other model systems, the influence of IL-4 on COX-2 expression has been reported. Inhibition of COX-2 expression by IL-4 was noted using peripheral blood monocytes, alveolar macrophages and non-small cell lung cancer cells 101–103. We performed studies on nasal polyp-derived fibroblasts and mononuclear phagocytic cells. Monocytes were utilized both as representative inflammatory cells, but also because PGE2 is their dominant prostaglandin product. Similar to other findings, significant inhibition of COX-2 and also mPGES-1 (but not COX-1) mRNA and protein expression was observed following stimulation with IL-4 on both the monocytes and nasal polyp-derived fibroblasts 79. Inhibition of COX-2 and mPGES-1 synergize to result in dramatically less stimulated PGE2 secretion by monocytes 79. IL-13 has been reported to have a similar effect on airway epithelial cells 104. Thus, in addition to enhancing the CysLT pathways, IL-4 and IL-13 contribute to the AERD phenotype through inhibition of the PGE2 pathway. The role of IFN-γ modulation of the prostaglandin pathway is unclear as its action appears to be cell-type specific. IFN-γ can induce COX-2 mRNA in most inflammatory cells 105, 106 whereas it decreases COX-2 expression in placental 107 and intestinal epithelial cancer cells 108. Actions of IFN-γ on other parts of the PGE2 pathway have not been studied in detail. Activation of eosinophils and mast cells by aspirin It was unknown in AERD how aspirin triggered the release of pro-inflammatory mediators. While, as noted, inhibition of COX releases the protective constraints provided by PGE2, this alone does not explain the positive signaling driving cell activation. In a particularly robust murine model of AERD, aspirin sensitivity is induced by the knocking out of the mPGES-1 gene 109. In this model, the positive – activating – signal is provided by allergic inflammation, a mechanism not likely to be relevant in AERD, at least in those AERD patients who are not atopic. We speculated that aspirin and other NSAIDs had the inherent capacity to directly activate eosinophils and mast cells. When tested, both eosinophils and mast cells generated Ca+2 fluxes following stimulation with water soluble lysine aspirin (LysASA) 14. Similar results were observed with eosinophil activation measured by EDN release and eosinophil and mast cell secretion of PGD2 14, 66. To our surprise, when eosinophils from control, aspirin tolerant, and AERD subjects were compared, no differences were observed in levels of mediator release. Our explanation as to why hypersensitivity reactions due to mediator release caused by aspirin/NSAIDs are not observed in these control cohorts reflects alterations in their PGE2 sensitivity, specifically the decreased capacity to produce and respond to this anti-inflammatory mediator observed in AERD. We speculate that the higher expression of PGE2 as well as its anti-inflammatory EP2 receptor acts to prevent the acute reactions to aspirin/NSAIDs in controls and aspirin tolerant asthmatics 21, 74. An additional explanation for the absence of clinical symptoms in these control cohorts is that when activated with aspirin, their circulating eosinophils produced very low levels of CysLTs 14 in contrast to the robust levels found in AERD subjects following aspirin ingestion 43. As mentioned earlier, AERD sinonasal and lung tissue is characterized by high numbers of eosinophilic hematopoietic progenitor (CD34+IL-5Rα+) cells 29, 110. We therefore investigated whether eosinophils differentiated from progenitor cells in the presence of IFN-γ would recapitulate the sensitivity to aspirin displayed by tissue eosinophils in vivo in AERD. After maturation with IFN-γ, the mature eosinophils displayed increased gene expression for both LTC4S and hPGDS 14, 66. Consistent with the increase in LTC4S gene expression, CysLT secretion was dramatically increased upon LysASA activation 14. In addition to increased CysLT production, these IFN-γ matured eosinophils displayed enhanced PGD2 production when stimulated with LysASA 66. Summary: Towards a generalized model for the induction of the AERD phenotype While our understanding that aspirin causes reactions and that AERD is a debilitating disease have been known for many years, the exact mechanisms driving AERD and how to treat it are still largely unknown. Our work and that of others have demonstrated the importance of both eosinophils and mast cells as drivers of the disease leading to increased expression of pro-inflammatory mediators and reflecting the loss of protective PGE2. The ultimate result is constitutive over-production of and over-responsiveness to mediators by eosinophils and mast cells in the basal state in AERD, with the uncontrolled release of mediators when these cells are directly triggered by aspirin. The increased recognition of the cellular components and mechanisms of action in AERD provides an opportunity to develop alternative targeted therapeutic approaches aimed at dampening the severe impacts of this disease. Supported by NIH grants R01AI057438, R56AI120055, and U01AI100799 Abbreviations 5-LO 5-lipoxygenase AERD aspirin exacerbated respiratory disease AFS allergic fungal sinusitis CHES chronic hyperplastic eosinophilic sinusitis COX cyclooxygenase CysLT cysteinyl leukotriene ECP eosinophil cationic protein Eo/B CFU eosinophil/basophil progenitors or colony forming units GM-CSF granulocyte macrophage colony-stimulating factor IFN interferon IL interleukin LT leukotriene LTC4S leukotriene C4 synthase NP nasal polyposis NSAID non-steroidal anti-inflammatory drugs PG prostaglandin PGDS prostaglandin D2 synthase PGES prostaglandin E2 synthase SCF stem cell factor STAT signal transducer and activator of transcription Figure 1 Eosinophil and mast cell development from CD34 progenitor cell Through the actions of IL-3, IL-5 and GM-CSF, CD34 progenitor cells mature into eosinophils. Mast cells develop following stimulation with IL-3, IL-6 and SCF from CD34 progenitor cells. Figure 2 Pathway depicting metabolites of arachidonic acid important in AERD Following conversion to arachidonic acid by phospholipase A2, further processing occurs via either the prostaglandin pathway mediated by COX-1/COX-2 or the leukotriene pathway mediated by 5-lipoxygenase. Red lettering shows genes inhibited by IL-4 and pink lettering show genes stimulated by IL-4 and IFN-γ. Key Points AERD is a disease of overproduction and hyper-responsiveness to lipid mediators. Mast cells and eosinophils are key driver of AERD pathogenesis through production of pro-inflammatory mediators following aspirin stimulation. Due to their involvement, therapies that target mast cells and eosinophils may be useful in providing clinical benefit in AERD. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Samter M Beers RF Jr Intolerance to aspirin. Clinical studies and consideration of its pathogenesis Ann Intern Med 1968 68 975 83 5646829 2 Mattos JL Woodard CR Payne SC Trends in common rhinologic illnesses: analysis of U.S. healthcare surveys 1995–2007 Int Forum Allergy Rhinol 2011 1 3 12 22287302 3 Rajan JP Wineinger NE Stevenson DD White AA Prevalence of aspirin-exacerbated respiratory disease among asthmatic patients: A meta-analysis of the literature J Allergy Clin Immunol 2015 135 676 81 e1 25282015 4 Szczeklik A Stevenson DD Aspirin-induced asthma: Advances in pathogenesis and management J Allergy Clin Immunol 1999 104 5 13 10400832 5 Szczeklik A Nizankowska E Clinical features and diagnosis of aspirin induced asthma Thorax 2000 55 S42 S4 10992556 6 Berges-Gimeno MP Simon RA Stevenson DD The natural history and clinical characteristics of aspirin-exacerbated respiratory disease Ann Allergy Asthma Immunol 2002 89 474 8 12452205 7 Vally H Taylor ML Thompson PJ The prevalence of aspirin intolerant asthma (AIA) in Australian asthmatic patients Thorax 2002 57 569 74 12096197 8 Steinke JW Huyett P Payne SC Negri J Borish L Cytokines and lymphocytes in eosinophilic sinusitis: prominent role for interferon-γ submitted 2011 9 Mascia K Borish L Patrie J Hunt J Phillips CD Steinke JW Chronic hyperplastic eosiniphilic sinusistis as a predictor of aspirin-exacerbated respiratory disease Ann Allergy Asthma Immunol 2005 94 652 7 15984597 10 Mascia K Haselkorn T Deniz YM Miller DP Bleecker ER Borish L Aspirin sensitivity and severity of asthma: evidence for irreversible airway obstruction in patients with severe or difficult-to-treat asthma J Allergy Clin Immunol 2005 116 970 5 16275362 11 Fischer AR Rosenberg MA Lilly CM Callery JC Rubin P Cohn J Direct evidence for a role of the mast cell in the nasal response to aspirin n aspirin-sensitive asthma J Allergy Clin Immunol 1994 94 1046 56 7798537 12 Daffern PJ Muilenburg D Hugli TE Stevenson DD Association of urinary leukotriene E4 excretion during aspirin challenges with severity of resiratory responses J Allergy Clin Immunol 1999 104 559 64 10482828 13 Bochenek G Nagraba K Nizankowska E Szczeklik A A controlled study of 9alpha,11beta-PGF2 (a prostaglandin D2 metabolite) in plasma and urine of patients with bronchial asthma and healthy controls after aspirin challenge The Journal of allergy and clinical immunology 2003 111 743 9 12704352 14 Steinke JW Negri J Liu L Payne SC Borish L Aspirin activation of eosinophils and mast cells: implications in the pathogenesis of aspirin-exacerbated respiratory disease J Immunol 2014 193 41 7 24890720 15 Antczak A Montuschi P Kharitonov S Gorski P Barnes PJ Increased exhaled cysteinyl-leukotrienes and 8-isoprostane in aspirin-induced asthma American Journal of Respiratory and Critical Care Medicine 2002 166 301 6 12153961 16 Perez-Novo CA Watelet JB Claeys C van Cauwenberge P Bachert C Prostaglandin, leukotiene, and lipoxin balance in chronic rhinosinusitis with and without nasal polyposis J Allergy Clin Immunol 2005 115 1189 96 15940133 17 Picado C Fernandez-Morata JC Juan M Roca-Ferrer J Fuentes M Xaubet A Cyclooxygenase-2 mRNA is downexpressed in nasal polyps from aspirin-sensitive asthmatics Am J Respir Crit Care Med 1999 160 291 6 10390414 18 Ying S Meng Q Scadding G Parikh A Corrigan CJ Lee TH Aspirin-sensitive rhinosinusitis is associated with reduced E-prostanoid 2 receptor expression on nasal mucosal inflammatory cells J Allergy Clin Immunol 2006 117 312 8 16461132 19 Steinke JW Bradley D Arango P Crouse CD Frierson H Kountakis SE Cytseinyl leukotriene expression in chronic hyperplastic sinusitis-nasal polyposis: Importance to eosinophilia and asthma J Allergy Clin Immunol 2003 111 342 9 12589355 20 Beller TC Friend DS Maekawa A Lam BK Austen KF Kanaoka Y Cysteinyl leukotriene 1 receptor controls the severity of chronic pulmonary inflammation and fibrosis Proc Natl Acad Sci USA 2004 101 3047 52 14970333 21 Steinke JW Editorial: Yin-Yang of EP receptor expression Journal of leukocyte biology 2012 92 1129 31 23204260 22 Adamjee J Suh YJ Park HS Choi JH Penrose JF Lam BK Expression of 5-lipoxygenase and cyclooxygenase pathway enzymes in nasal polyps of patients with aspirin-intolerant asthma J Pathol 2006 209 392 9 16583357 23 Payne SC Early SB Huyett P Han JK Borish L Steinke JW Evidence for distinct histologic profile of nasal polyps with and without eosinophilia The Laryngoscope 2011 121 2262 7 21898422 24 Cowburn AS Sladek K Soja J Adamek L Nizankowska E Szczeklik A Overexpression of leukotriene C4 synthase in bronchial biopsies from patients with aspirin-intolerant asthma J Clin Invest 1998 101 834 46 9466979 25 Park HS Nahm DH Park K Suh KS Yim HE Immunohistochemical characterization of cellular infiltrate in nasal polyp from aspirin-sensitive asthmatic patients Ann Allergy Asthma Immunol 1998 81 219 24 9759797 26 Corrigan CJ Napoli RL Meng Q Fang C Wu H Tochiki K Reduced expression of the prostaglandin E2 receptor E-prostanoid 2 on bronchial mucosal leukocytes in patients with aspirin-sensitive asthma J Allergy Clin Immunol 2012 129 1636 46 22418066 27 Boyce JA Friend D Matsumoto R Austen KF Owen WF Differentiation in vitro of hybrid eosinophil/basophile granulocytes: autocrine function of an eosinophil intermediate J Exp Med 1995 182 49 57 7540656 28 Denburg JA Sehmi R Saito H Pil-Seob J Inman MD O’Bryne PM Systemic aspects of allergic disease: bone marrow responses J Allergy Clin Immunol 2000 106 S242 S6 11080738 29 Kim YK Uno M Hamilos DL Beck L Bochner B Schleimer RP Immunolocalization of CD34 in nasal polyp. Effect of topical corticosteroids Am J Respir Cell Mol Biol 1999 20 388 97 10030836 30 Nerlov C Graf T PU. 1 induces myeloid lineage commitment in multipotent hematopoietic progenitors Genes Dev 1998 12 2403 12 9694804 31 Nerlov C McNagny KM Doderlein G Kowenz-Leutz E Graf T Distinct C/EBP functions are required for eosinophil lineage commitment and maturation Genes Dev 1998 12 2413 23 9694805 32 McNagny K Graf T Making eosinophils through subtle shifts in transcription factor expression J Exp Med 2002 195 F43 F7 12045250 33 Hirasawa R Shimizu R Takahashi S Osawa M Takayanagi S Kato Y Essential and instructive roles of GATA factors in eosinophil development J Exp Med 2002 195 1379 86 12045236 34 Nishijima I Nakahata T Hirabayashi Y Inoue T Kurata H Miyajima A A human-GM-CSF receptor expressed in transgenic mice stimulates proliferation and differentiation of hemopoietic progenitors to all lineages in response to human GM-CSF Mol Cell Biol 1995 6 497 508 35 Sanderson CJ Interleukin-5, eosinophils, and disease Blood 1992 79 3101 9 1596561 36 Collins PD Marleau S Griffiths-Johnson DA Jose PJ Williams TJ Cooperation between interleukin-5 and the chemokine eotaxin to induce eosinophil accummulation in vivo J Exp Med 1995 182 1169 74 7561691 37 Milanovic M Terszowski G Struck D Liesenfeld O Carstanjen D IFN consensus sequence binding protein (Icsbp) is critical for eosinophil development J Immunol 2008 181 5045 53 18802108 38 Steinke JW Liu L Huyett P Negri J Payne SC Borish L Prominent Role of Interferon-γ in Aspirin-Exacerbated Respiratory Disease Journal of Allergy and Clinical Immunology 2013 132 856 65 e3 23806637 39 de Bruin AM Buitenhuis M van der Sluijs KF van Gisbergen KP Boon L Nolte MA Eosinophil differentiation in the bone marrow is inhibited by T cell-derived IFN-gamma Blood 2010 116 2559 69 20587787 40 Kirshenbaum AS Goff JP Semere T Foster B Scott LM Metcalfe DD Demonstration that human mast cells arise from a progenitor cell population that is CD34(+), c-kit(+), and expresses aminopeptidase N (CD13) Blood 1999 94 2333 42 10498605 41 Kirshenbaum AS Goff JP Kessler SW Mican JM Zsebo KM Metcalfe DD Effect of IL-3 and stem cell factor on the appearance of human basophils and mast cells from CD34+ pluripotent progenitor cells J Immunol 1992 148 772 7 1370517 42 Anderson DM Lyman SD Baird A Wignall JM Eisenman J Rauch C Molecular cloning of mast cell growth factor, a hematopoietin that is active in both membrane bound and soluble forms Cell 1990 63 235 43 1698558 43 Christie PE Tagari P Ford-Hutchinson AW Charlesson S Chee P Arm JP Urinary leukotriene E4 concentrations increase after aspirin challenge in aspirin-sensitive asthmatic subjects Am Rev Respir Dis 1991 143 1025 9 1850964 44 Cardet JC White AA Barrett NA Feldweg AM Wickner PG Savage J Alcohol-induced respiratory symptoms are common in patients with aspirin exacerbated respiratory disease The journal of allergy and clinical immunology In practice 2014 2 208 13 24607050 45 Payne SC Alcohol’s effect in AERD related to polyphenols J Allergy Clin Immunol Pract 2014 in press 46 Sampson AP Cowburn AS Sladek K Profound overexpression of leukotriene C4 synthase in bronchial biopsies from aspirin-intolerant asthmatic patients Int Archives Allergy Immunology 1997 113 355 7 47 Arm JP O’Hickey S Spur BW Lee TH Airway responsiveness to histamine and leukotriene E4 in subjects with aspirin-induced asthma Am Rev Respir Dis 1989 140 148 53 2546469 48 Jiang Y Borrelli LA Kanaoka Y Bacskai BJ Boyce JA CysLT2 receptors interact with CysLT1 receptors and down-modulate cysteinyl leukotriene dependent mitogenic responses of mast cells Blood 2007 110 3263 70 17693579 49 Sousa AR Parikh A Scadding G Corrigan CJ Lee TH Leukotriene-receptor expression on nasal mucosal inflammatory cells in asprin-sensitive rhinosinusitis N Engl J Med 2002 347 1493 9 12421891 50 Lynch KR O’Neill GP Liu Q Im DS Sawyer N Metters KM Characterization of the human cysteinyl leukotriene CysLT1 receptor Nature 1999 399 789 93 10391245 51 Christie PE Smith CM Lee TH The potent and selective sulfidopeptide leukotriene antagonists, SK&F 104353, inhibits aspirin-induced asthma Am Rev Resir Dis 1991 144 957 8 52 Dahlen B Treatment of aspirin-intolerant asthma with antileukotrienes Am J Respir Crit Care Med 2000 161 S137 S41 10673243 53 Walch L Norel X Back M Gascard JP Dahlen SE Brink C Pharmacological evidence for a novel cysteinyl-leukotriene receptor subtype in human pulmonary artery smooth muscle Br J Pharmacol 2002 137 1339 45 12466244 54 White A Bigby T Stevenson D Intranasal ketorolac challenge for the diagnosis of aspirin-exacerbated respiratory disease Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology 2006 97 190 5 55 Christie PE Hawksworth R Spur BW Lee TH Effect of indomethacin on leukotriene4-induced histamine hyperresponsiveness in asthmatic subjects Am Rev Respir Dis 1992 146 1506 10 1333740 56 Christie PE Schmitz-Schumann M Spur BW Lee TH Airway responsiveness to leukotriene C4 (LTC4), leukotriene E4 (LTE4) and histamine in aspirin-sensitive asthmatic subjects Eur Respir J 1993 6 1468 73 8112440 57 Laitinen LA Laitinen A Haahtela T Vilkka V Spur BW Lee TH Leukotriene E4 and granulocytic infiltration into asthmatic airways Lancet 1993 341 989 90 8096945 58 Gauvreau GM Parameswaran KN Watson RM O’Byrne PM Inhaled leukotriene E(4), but not leukotriene D(4), increased airway inflammatory cells in subjects with atopic asthma American Journal of Respiratory and Critical Care Medicine 2001 164 1495 500 11704602 59 Maekawa A Kanaoka Y Xing W Austen KF Functional recognition of a distinct receptor preferential for leukotriene E4 in mice lacking the cysteinyl leukotriene 1 and 2 receptors Proc Natl Acad Sci U S A 2008 105 16695 700 18931305 60 Nonaka Y Hiramoto T Fujita N Identification of endogenous surrogate ligands for human P2Y12 receptors by in silico and in vitro methods Biochem Biophys Res Commun 2005 337 281 8 16185654 61 Paruchuri S Tashimo H Feng C Maekawa A Xing W Jiang Y Leukotriene E4-induced pulmonary inflammation is mediated by the P2Y12 receptor J Exp Med 2009 206 2543 55 19822647 62 Kanaoka Y Maekawa A Austen KF Identification of GPR99 protein as a potential third cysteinyl leukotriene receptor with a preference for leukotriene E4 ligand The Journal of biological chemistry 2013 288 10967 72 23504326 63 Dahlen B Nizankowska E Szczeklik A Benefits from adding the 5-lipoxygenase inhibitor zileuton to conventional therapy in aspirin-intolerant asthmatics Am J Respir Critc Care Med 1998 157 1187 94 64 Cahill KN Bensko JC Boyce JA Laidlaw TM Prostaglandin D(2): a dominant mediator of aspirin-exacerbated respiratory disease J Allergy Clin Immunol 2015 135 245 52 25218285 65 Luna-Gomes T Magalhaes KG Mesquita-Santos FP Bakker-Abreu I Samico RF Molinaro R Eosinophils as a novel cell source of prostaglandin D2: autocrine role in allergic inflammation Journal of immunology 2011 187 6518 26 66 Feng X Negri J Baker MG Payne SC Borish L Steinke JW Eosinophil production of PGD2 in aspirin-exacerbated respiratory disease 2016 67 Hirai H Tanaka K Yoshie O Ogawa K Kenmotsu K Takamori Y Prostaglandin D2 selectively induces chemotaxis in T helper type 2 cells, eosinophils, and basophils via seven-transmembrane receptor CRTH2 J Exp Med 2001 193 255 61 11208866 68 Satoh T Moroi R Aritake K Urade Y Kanai Y Sumi K Prostaglandin D2 plays an essential role in chronic allergic inflammation of the skin via CRTH2 receptor J Immunol 2006 177 2621 9 16888024 69 Schratl P Royer JF Kostenis E Ulven T Sturm EM Waldhoer M The role of the prostaglandin D2 receptor, DP, in eosinophil trafficking J Immunol 2007 179 4792 9 17878378 70 Stinson SE Amrani Y Brightling CE D prostanoid receptor 2 (chemoattractant receptor-homologous molecule expressed on TH2 cells) protein expression in asthmatic patients and its effects on bronchial epithelial cells J Allergy Clin Immunol 2015 135 395 406 25312757 71 Okano M Fujiwara T Yamamoto M Sugata Y Matsumoto R Fukushima K Role of prostaglandin D2 and E2 terminal synthases in chronic rhinosinusitis Clin Exp Allergy 2006 36 1028 38 16911359 72 Hyo S Kawata R Kadoyama K Eguchi N Kubota T Takenaka H Expression of prostaglandin D2 synthase in activated eosinophils in nasal polyps Arch Otolaryngol Head Neck Surg 2007 133 693 700 17638783 73 Schmid M Gode U Schafer D Wigand ME Arachidonic acid metabolism in nasal tissue and peripheral blood cells in aspirin intolerant asthmatics Acta Otolaryngol 1999 119 277 80 10320091 74 Sestini P Armetti L Gambaro G Pieroni MG Refini RM Sala A Inhaled PgE2 prevents aspirin-induced bronchoconstriction and urinary LTE4 excretion in aspirin-sensitive asthma American J Respiratory Critical Care Medicine 1996 153 572 5 75 Feng C Beller EM Bagga S Boyce JA Human mast cells express multiple EP receptors for prostaglandin E2 that differentially modulate activation responses Blood 2006 107 3243 50 16357326 76 Serra-Pages M Olivera A Torres R Picado C de Mora F Rivera J E-prostanoid 2 receptors dampen mast cell degranulation via cAMP/PKA-mediated suppression of IgE-dependent signaling J Leukoc Biol 2012 77 Murakami M Nakashima K Kamei D Masuda S Ishikawa Y Ishii T Cellular prostaglandin E2 production by membrane-bound prostaglandin E synthase-2 via both cyclooxygenases-1 and -2 The Journal of biological chemistry 2003 278 37937 47 12835322 78 Gosepath J Brieger J Mann WJ New immunohistologic findings on the differential role of cyclooxygenase 1 and cyclooxygenase 2 in nasal polposis Am J Rhinol 2005 19 111 6 15921208 79 Steinke JW Culp JA Kropf E Borish L Modulation by aspirin of nuclear phospho-signal transducer and activator of transcription 6 expression: Possible role in therapeutic benefit associated with aspirin desensitization J Allergy Clin Immunol 2009 124 724 30 e4 19767084 80 Stevenson DD Szczeklik A Clinical and pathologic perspectives on aspirin sensitivity and asthma J Allergy Clin Immunol 2006 118 773 86 quiz 87–8 17030227 81 Bachert C Wagenmann M Hauser U Rudack C IL-5 synthesis is upregulated in human nasal polyp tissue J Allergy Clin Immunol 1997 99 837 42 9215253 82 Bachert C Gevaert P van Cauwenberge P Nasal polyposis- A new concept on the formation of polyps ACI International 1999 11 130 5 83 Minshall EM Cameron L Lavigne F Leung DYM Hamilos D Garcia-Zepada A Eotaxin mRNA and protein expression in chronic sinusitis and allergen-induced nasal responses in seasonal allergic rhinitis Amer J Resp Cell Mol Biol 1997 17 683 90 84 Hamilos DL Leung DYM Huston DP Kamil A Wood R Hamid Q GM-CSF, IL-5, and RANTES immunoreactivity and mRNA expression in chronic hyperplastic sinusitis with nasal polyposis Clin Exp Allergy 1998 28 1145 52 9761019 85 Kamil A Ghaffar O Lavigne F Taha R Renzi PM Hamid Q Comparison of inflammatory cell profile and Th2 cytokine expression in the ethmoid sinuses, maxillary sinuses, and turbinates of atopic subjects with chronic sinusitis Otolaryngol Head Neck Surg 1998 18 804 9 86 Riechelmann H Deutschle T Rozsasi A Keck T Polzehl D Burner H Nasal biomarker profiles in acute and chronic rhinosinusitis Clin Exp Allergy 2005 35 1186 91 16164446 87 Van Bruaene N Perez-Novo CA Basinski TM Van Zele T Holtappels G De Ruyck N T-cell regulation in chronic paranasal sinus disease J Allergy Clin Immunol 2008 121 1435 41 41 e1 3 18423831 88 Stevens WW Ocampo CJ Berdnikovs S Sakashita M Mahdavinia M Suh L Cytokines in Chronic Rhinosinusitis. Role in Eosinophilia and Aspirin-exacerbated Respiratory Disease Am J Respir Crit Care Med 2015 192 682 94 26067893 89 Hamilos DL Leung DYM Wood R Cunningham L Bean DK Yasruel Z Evidence for distinct cytokine expression in allergic versus nonallergic chronic sinusitis J Allergy Clin Immunol 1995 96 537 44 7560666 90 Van Zele T Claeys S Gevaert P Van Maele G Holtappels G van Cauwenberge P Differentiation of chronic sinus diseases by measurement of inflammatory mediators Allergy 2006 61 1280 9 17002703 91 Shome GP Tarbox J Shearer M Kennedy R Cytokine expression in peripheral blood lymphocytes before and after aspirin desensitization in aspirin-exacerbated respiratory disease Allergy Asthma Proc 2007 28 706 10 18201436 92 Kanda A Driss V Hornez N Abdallah M Roumier T Abboud G Eosinophil-derived IFN-γ induces airway hyperresponsiveness and lung inflammation in the absence of lymphocytes J Allergy Clin Immunol 2009 124 573 82 19539982 93 Akuthota P Xenakis JJ Weller PF Eosinophils: Offenders or General Bystanders in Allergic Airway Disease and Pulmonary Immunity? J Innate Immun 2011 3 113 9 21228563 94 Spencer LA Szela CT Perez SAC Kirchhoffer CL Neves JS Radke AL Human eosinophils constitutively express multiple Th1, Th2, and immunoregulatory cytokines that are secreted rapidly and differentially J Leukoc Biol 2009 85 117 23 18840671 95 Hsieh FH Lam BK Penrose JF Austen KF Boyce JA T helper cell type 2 cytokines coordinately regulate immunoglobulin E-dependent cysteinyl leukotriene production by human cord blood-derived mast cells: profound induction of leukotriene C4 synthase expression by interleukin 4 J Exp Med 2001 193 123 33 11136826 96 Laidlaw TM Kidder MS Bhattacharyya N Xing W Shen S Milne GL Cysteinyl leukotriene overproduction in aspirin-exacerbated respiratory disease is driven by platelet-adherent leukocytes Blood 2012 119 3790 8 22262771 97 Mellor EA Austen KF Boyce JA Cysteinyl leukotrienes and uridine diphosphate induce cytokine generation by human mast cells through an interleukin 4-regulated pathway that is inhibited by leukotriene receptor antagonists J Exp Med 2002 195 583 92 11877481 98 Mellor EA Frank N Soler D Hodge MR Lora JM Austen KF Expression of the type 2 receptor for cysteinyl leukotrienes (CysLT2R) by human mast cells: Functional distinction from CysLT1R Proc Natl Acad Sci U S A 2003 100 11589 93 13679572 99 Thivierge M Stankova J Rola-Pleszczynski M IL-13 and IL-4 up-regulate cysteinyl leukotriene 1 receptor expression in human monocytes and macrophages J Immunol 2001 167 2855 60 11509632 100 Early SB Barekzi E Negri J Hise K Borish L Steinke JW Concordant modulation of cysteinyl leukotriene receptor expression by IL-4 and IFN-g on peripheral immune cells Am J Respir Cell Mol Biol 2007 36 715 20 17272825 101 Yano T Hopkins HA Hempel SL Monick M Hunninghake GW Interleukin-4 inhibits lipopolysaccharide-induced expression of prostaglandin H synthase-2 in human alveolar macrophages J Cell Physiol 1995 165 77 82 7559810 102 Dworski R Sheller JR Differential sensitivites of human blood monocytes and alveolar macrophages to the inhibition of prostaglandin endoperoxide synthase-2 by interleukin-4 Prostaglandins 1997 53 237 51 9167211 103 Cui X Yang SC Sharma S Heuze-Vourc’h N Dubinett SM IL-4 regulates COX-2 and PGE2 production in human non-small cell lung cancer Biochem Biophys Res Commun 2006 343 995 1001 16574063 104 Trudeau J Hu H Chibana K Chu HW Wescott JY Wenzel SE Selective downregulation of prostaglandin E2-related pathways by the TH2 cytokine IL-13 J Allergy Clin Immunol 2006 117 1446 54 16751012 105 Matsuura H Sakaue M Subbaramaiah K Kamitani H Eling TE Dannenberg AJ Regulation of cyclooxygenase-2 by interferon gamma and transforming growth factor alpha in normal human epidermal keratinocytes and squamous carcinoma cells. Role of mitogen-activated protein kinases J Biol Chem 1999 274 29138 48 10506169 106 Kim J Yoon Y Jeoung D Kim YM Choe J Interferon-gamma stimulates human follicular dendritic cell-like cells to produce prostaglandins via the JAK-STAT pathway Mol Immunol 2015 66 189 96 25818476 107 Hanna N Bonifacio L Reddy P Hanna I Weinberger B Murphy S IFN-gamma-mediated inhibition of COX-2 expression in the placenta from term and preterm labor pregnancies Am J Reprod Immunol 2004 51 311 8 15212685 108 Klampfer L Huang J Kaler P Sasazuki T Shirasawa S Augenlicht L STAT1-independent inhibition of cyclooxygenase-2 expression by IFNgamma; a common pathway of IFNgamma-mediated gene repression but not gene activation Oncogene 2007 26 2071 81 17016440 109 Liu T Laidlaw TM Katz HR Boyce JA Prostaglandin E2 deficiency causes a phenotype of aspirin sensitivity that depends on platelets and cysteinyl leukotrienes Proceedings of the National Academy of Sciences of the United States of America 2013 110 16987 92 24085850 110 Denburg JA Haemopoietic mechanisms in nasal polyposis and asthma Thorax 2000 55 S24 S5 10992551

PMC005xxxxxx/PMC5116320.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8219232 5883 Neurol Clin Neurol Clin Neurologic clinics 0733-8619 1557-9875 27720002 5116320 10.1016/j.ncl.2016.06.009 NIHMS793069 Article Alzheimer Disease and its Growing Epidemic: Risk Factors, Biomarkers and the Urgent Need for Therapeutics Hickman RA 1* Faustin A 124* Wisniewski T 1234∞ 1 Department of Pathology, New York University School of Medicine, Alexandria ERSP, 450 East 29th Street, New York, NY 10016, USA 2 Department of Neurology, New York University School of Medicine, Alexandria ERSP, 450 East 29th Street, New York, NY 10016, USA 3 Department of Psychiatry, New York University School of Medicine, Alexandria ERSP, 450 East 29th Street, New York, NY 10016, USA 4 Center for Cognitive Neurology, New York University School of Medicine, Alexandria ERSP, 450 East 29th Street, New York, NY 10016, USA ∞ Corresponding author: Thomas Wisniewski, MD, Thomas.wisniewski@nyumc.org * Equally contributed 9 6 2016 4 8 2016 11 2016 01 11 2017 34 4 941953 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. SYNOPSIS Alzheimer disease represents one of the greatest medical challenges that face this century; the condition is becoming increasingly prevalent worldwide and as yet, no effective treatments have been developed for this terminal disease. Since the disease manifests at a late stage after a long period of clinically silent neurodegeneration, knowledge of the modifiable risk factors and the implementation of biomarkers is crucial in the primary prevention of the disease and pre-symptomatic detection of AD, respectively. This review discusses the growing epidemic of AD and antecedent risk factors in the disease process. Disease biomarkers are discussed and the implications that this may have for the treatment of this currently incurable disease. Alzheimer disease risk factors biomarkers epidemiology INTRODUCTION Alzheimer disease represents one of the greatest medical challenges that face this century; the condition is becoming increasingly prevalent worldwide and as yet, no effective treatments have been developed for this terminal disease. In the United States in 2015, over five million people suffered with AD, costing over 170 billion dollars. Since the disease manifests at a late stage after a long period of clinically silent neurodegeneration, knowledge of the modifiable risk factors and the implementation of biomarkers is crucial in the primary prevention of the disease and pre-symptomatic detection of AD, respectively. This review discusses the growing epidemic of AD and antecedent risk factors in the disease process. Disease biomarkers are discussed and the implications that this may have for the treatment of this currently incurable disease. EPIDEMIOLOGY Alzheimer disease (AD) is the most common dementia in the elderly and is a growing epidemic across the globe. Although the risks associated with developing AD are multifactorial, the greatest risk factor by far is aging1. The age-specific risk of AD dramatically increases as individuals get older; findings from the Framingham study in the early 1990s showed that the incidence doubles every five years up to the ages of 89 years2. Age-dependent increases have been seen in other studies3–5. Unsurprisingly therefore, with global reductions in fertility and extended life expectancies, the number of patients with AD is expected to increase as populations age6. In the United States, it is estimated that approximately 5.3 million people had AD in 2015; 5.1 million people being 65 years and older and approximately 200,000 people under the age of 65 years with early onset AD (EOAD) 7–9. It is estimated that the number of new cases of AD and other dementias will at least double by 2050 and substantially increase the socioeconomic burden worldwide7, 10. In 2010, it was estimated that dementia afflicted 35.6 million people worldwide, many of which will have AD, with the projection that this figure will double every twenty years11. The incidence of AD is generally lower in many less economically developed countries than in North America and Europe, however, sharp rises in prevalence have been predicted and seen in China, India and Latin America12, 13. The effect of this increasing dementia has obvious socioeconomic consequences for each country affected, through costs of hospital care and also of caregivers. In the USA, the total payments were estimated at $226 billion of which Medicare and Medicaid provided 68% 7, 14, whilst out-of-pocket expenses for patients and their families were expected to be $44 billion7. CLASSIFICATION AND STAGING Revised criteria and guidelines by the National Institute on Aging and the Alzheimer Association published in 2011 (NIA-AA) have recognized three stages of Alzheimer disease: preclinical Alzheimer disease, mild cognitive impairment (MCI) due to Alzheimer disease and dementia due to Alzheimer disease8, 15. These are described as follows: Preclinical AD: pre-symptomatic of AD with early AD-related brain changes as detected by neuroimaging or other biomarker studies; Mild cognitive impairment (MCI) due to AD: mild cognitive decline but still able to perform activities of daily living; Dementia due to AD: cognitive decline is more pronounced and interferes with activities of daily living7. With this classification in mind, it follows that the actual number of individuals with active disease are gross underestimates because they are based on approximations of diagnosed symptomatic patients and largely ignores the vast number of individuals who are preclinical, in whom the disease process is active but asymptomatic16. This long pre-clinical phase of AD is characterized by progressive neuronal loss, the formation of neurofibrillary tangles (NFT) and the deposition of amyloid plaques within the brain17–20. Although the exact pathogenesis of AD is debated, the prevailing hypothesis is that the neurodegeneration is the result of the amyloid cascade, in which aberrant digestion and processing of the amyloid precursor protein (APP) results in the accumulation of neurotoxic Aβ oligomeric proteins21–24. These proteins aggregate to form the insoluble amyloid plaques that are seen at microscopic examination of autopsy brains of AD patients. BIOMARKERS It is widely believed therefore that future therapeutics should be introduced during the preclinical and MCI stages of the disease course so as to preserve the existing functioning neural networks7, 25. In order to provide effective recognition of preclinical AD, there will likely need to be widespread implementation of disease biomarkers, such as in national screening programs targeting specific age groups and other high-risk categories. Such screening may prove popular as many patients are keen to know their disease status at an earlier time point26–28. Current disease biomarkers focus on indicators of cerebral amyloidosis or synaptic dysfunction. Markers of brain amyloidosis include reduced CSF Aβ42 concentrations and increased amyloid tracer uptake on positron emission tomography (PET) 29. These changes are followed after a period of time with markers of neuronal injury, notably increased CSF tau levels and brain atrophy on magnetic resonance imaging (MRI) 30, 31. PET tau imaging has great promise as a biomarker and may be able to provide estimates of pathological disease stage32. Validation is needed prior to introduction of these tests into the clinical setting, but would certainly be useful in providing a more complete overview of the current number of patients with AD and also in evaluating pre-clinical therapeutic response. RISK FACTORS The modifiable and non-modifiable risk factors of AD are important because they provide insight into the predispositions of the disease process prior to onset and also provides stratification of individuals who may be at increased risk. Besides aging, which, as discussed previously, is the most significant risk factor; other determinants of AD include genetic risk factors, and non-genetic, modifiable risk factors. NON-MODIFIABLE GENETIC RISK FACTORS Recent genome wide association studies (GWAS) have revealed many new genes that increase the risk of developing AD33. This review, however, considers the most commonly discussed genetic influences on the disease, notably mutations in ApoE, APP and presenilin mutations. ApoE Of all of the mutations identified in AD, genome wide association studies have demonstrated that it is the ε4 allele of the APOE gene that poses the greatest risk for AD34–38. ApoE is a 34 kDa astrocytic protein that is encoded on chromosome 19q13. The apoE gene has three alleles that result in the production of ε2, ε3 and ε4 isoforms. One of its principal functions within the CNS is the delivery of cholesterol to neurons via the ApoE receptor39. ApoE3 is the most common variant, present in approximately 60% of the population and is regarded as having no altered risk in AD7, 35, 36. However, the next most common allele is the ε4 followed by the ε2 allele. ApoE heterozygosity with ApoE4/E3 or ApoE4 homozygosity confers a significantly risk of developing AD, from three-fold to eight-twelve fold, respectively. In approximately 40% of cases of AD, ApoE4 is identified 40. Furthermore, patients with ApoE4 have poorer cognitive performance in childhood and tend to develop the disease significantly earlier than those with ApoE341. In a population based study, patients who had suffered a head injury and carried the ApoE4 allele had a ten times increased risk of developing AD, unlike those without the allele who were at two-fold increased risk42. The MIRAGE study demonstrated that patients who have suffered head injury are at markedly increased risk of developing AD43. Curiously, the ε2 isoform bestows a decreased risk of AD than the ε3 allele36. It comes as no surprise therefore that the ε2 allele is over-represented in centenarians44. Triggering Receptor on Myeloid Cells 2 Discovery of the triggering receptor on myeloid cells 2 (TREM2) allele as a rare genetic predisposition for AD has sparked interest because of its role in inflammation45. TREM2 is a receptor found on microglia that is important in phagocytosis and dampening the CNS immune response46. Mutation in TREM2 is rare, however, the most common receptor mutation (R47H) increases the risk of LOAD by approximately twofold. Furthermore, mutations in TREM2 are associated with more severe degrees of atrophy in AD than those without47. Mutations in TREM2 that increase the risk and severity of AD may result from derangements in neuroinflammation and amyloid clearance. APP and Presenilin Mutations Early onset familial AD (EOAD), which usually begins in patients younger than 65 years of age, represents less than 1% of cases of AD48. EOAD is often caused by autosomal dominant mutations such as mutations in amyloid precursor protein (APP), presenilin-1, and presenilin-2 genes48. Mutations in proteins that are involved in the synthesis of Aβ result in downstream overproduction of the pathological Aβ. APP is encoded on chromosome 21q21.3 and comprises 3 transcript variants, the most common of which protein within the CNS is 695 amino acids long49. Over 30 coding APP mutations have been identified that usually result in an autosomal dominant EOAD because of increased Aβ production, shifts in synthesis of pathologic Aβ1–42, or production of Aβ that may have increased susceptibility to aggregation50. Of interest, not all APP mutations result in AD preponderance; actually, one mutation was found to be protective50. Presenilin is one of the proteins that constitute the active site of γ-secretase and therefore mutations alter the efficacy of this enzyme increasing the amount of Aβ1–42 production. Presenilin mutations account for the majority of cases of familial AD48. Down Syndrome Down syndrome (DS) is the most common chromosomal abnormality with an incidence of 1 per 733 live births and is characterized by trisomy 2149. Since APP is encoded on chromosome 21q21.3, this results in three copies of the APP protein. This increased abundance of APP expression, production of Aβ is considered to be one of the mechanisms as to why many of these patients develop EOAD. Given that the lifespan of patients with Down’s syndrome is now 55–60 years of age, approximately 70% of patients will suffer from AD51. Cardiovascular Health A large body of evidence suggests that cardiovascular disease increases the risk of dementia. Studies that have investigated patients with clinical and subclinical cardiovascular disease have poorer cognitive function than those without52, 53. Cortical ischemic changes can increase the risk of dementia54. However, studying the role of cardiovascular disease and AD is complicated by several issues, notably that extensive cardiovascular disease and dementia may preclude from a clinical diagnosis of AD and may instead favor a diagnosis of multi-infarct dementia55. Studies have shown mixed results with regard to the influence of hypertension and this is in part due to differences in study design54, 56–58. Observational studies however have generally shown that increased hypertension are associated with cognitive decline and an increased likelihood of developing AD, possibly through blood vessel injury, protein extravasation, neuronal injury and subsequent Aβ accumulation59. Diabetes Mellitus Diabetes mellitus (DM) is associated with an increased risk of cognitive decline and AD later in life. Observational studies of type 2 DM (T2DM) have found that T2DM nearly doubles the risk of AD60–62. In the Religious Orders Study, 824 individuals who were older than 55 years of age, were evaluated for cognitive decline and AD and found that those with DM had a 65% increased risk of developing AD after a mean 5.5 year period63. The cognitive decline was found to be mainly in perceptual speed. Several meta-analyses have further confirmed an increased risk of AD in DM. The biological mechanism for this association may relate to competition of Aβ and insulin for insulin degrading enzyme, thereby reducing Aβ clearance. Alternatively, increased Aβ aggregation has been demonstrated through increased age-related glycation end-products that can occur in DM. Anti-diabetic therapies in patients with DM and cognitive impairment and also in patients with AD have shown improvement in cognition, which may be related to the anti-inflammatory properties of these drugs. Traumatic Brain Injury Traumatic brain injury (TBI) is a growing public health concern worldwide because the incidence is rising and it carries a significant healthcare and socioeconomic burden for society64–68. For patients who survive TBI, the average life expectancy is considerably shortened and many cases of TBI suffer chronic neurological and psychological morbidity that reduces quality of life 69–72. More data is now showing that there are ongoing chronic changes within the brain following TBI and that these ensuing processes may result in further damage with possible neurodegenerative sequelae73. The first documentation of a syndrome that directed attention towards a neurodegenerative phenomenon after head injury was ‘punch drunk syndrome’. This syndrome described degenerative changes after repeated episodes of sub-lethal head injuries in professional boxers 74. This condition is now termed chronic traumatic encephalopathy (CTE) and afflicts a diverse range of people including professional and amateur players of contact sports as well as veterans 75–78. CTE has pathological features that overlap with AD and TBI is recognized to shorten the time to onset of AD79. Furthermore, it is now considered that TBI is the most significant environmental risk factor for AD76, 80. Recent data has demonstrated that in both long-standing TBI and AD there is chronic inflammation within the brain parenchyma and this persistent inflammatory milieu within the brain parenchyma could be where the pathophysiology of TBI and AD converge81, 82. Following TBI the amyloid levels increase due to several factors. Firstly, APP expression is noticeably increased post-TBI83, 84. APP is particularly prominent at the axon terminals where there has been axonal transection and axonal transection is known to occur even in mild cases of TBI 85, 86. Secondly, β-secretase and γ-secretase, enzymes that both contribute to the digestion of APP and formation of Aβ are also upregulated87–89. These increases in both substrate and enzymes, results in increased deposition of amyloid at the axon bulbs and offers one explanation as to how the risk of AD is increased after TBI 90. The Influence of Neprilysin and TBI Removal of cerebral amyloid is likely to be multifactorial, involving partly passive diffusion of soluble amyloid, active transport mechanisms and cellular digestion91, 92. The degree of amyloid pathology post-TBI and in AD is particularly influenced by neprilysin. Neprilysin is a membrane zinc metalloprotease that is capable of digesting Aβ peptide and thus has the capability of reducing the amyloid load within the brain93. Neprilysin knockout mice demonstrate increased amyloid burden in a gene-dose dependent correlation94. Johnson and colleagues demonstrated that in post-mortem subjects, the degree of amyloid burden was most in patients who had more than 41 GT repeats in the promoter region of the neprilysin gene, which was considered to be related to defective amyloid clearance 95. Curiously, neprilysin expression increases post-TBI and this may be a mechanism by which Aβ and amyloid plaques are cleared months after injury, despite increased intra-axonal APP and presenilin-1 expression 96. With age-related reduction in neprilysin, the balance between formation of amyloid and its breakdown may shift towards accumulation of amyloid and this may be responsible for the preponderance to AD post-TBI97. There is also likely to be a contribution to amyloid breakdown from microglial activation. Neprilysin, metalloproteinase-9 and several other factors that are released by healthy microglia, digest Aβ98. There is a heightened neuro-inflammatory response following TBI that persists, and the activation of microglia most likely releases factors that will assist in the digestion of Aβ. However, with aging, the efficacy of microglial breakdown is likely to be lost, and may even accentuate the accumulation, thereby causing a gradual shift toward accumulation of amyloid in the dynamic Aβ turnover99. Previous Amyloid Exposure One of the most concerning developments over the past few years has been the accumulation of evidence that suggests infectivity of amyloid in a prion-like fashion. In a recent case series of iatrogenic CJD, a proportion of patients who received homogenized human pituitaries for growth hormone replacement were found to have significant cerebral amyloid angiopathy at autopsy, to an extent that was inconsistent with age100. Given that pituitaries may have amyloid deposits, there is the possibility that amyloid could seed through peripheral injection with proteopathic spread over subsequent decades100, 101. The proteopathic spread of amyloid in the brain has been demonstrated in numerous animal models and in human AD pathological staging102. The main fear that stem from these findings is that iatrogenic infection may occur from re-used surgical instruments, since amyloid is difficult to remove from metal devices103. Further research is needed in this area in order to gauge the significance of these findings on amyloid infectivity. Protective Factors In general, environmental influences that are anti-inflammatory appear to be beneficial at reducing the likelihood of developing AD. Low calorie diets that are sustained for a protracted period of time are associated with reduced free radical production and increased brain neurogenesis and BDNF concentrations, all of which are recognized to promote healthier brain aging44, 104. Data regarding diets that are rich in antioxidants and polyunsaturated fatty acids (PUFA) have proved inconclusive with some studies demonstrating a reduction in the risk of AD, whilst others showing no such association105–107. Other protective influences include cognitive stimulation and a high educational achievement, which improves cognitive reserve108, 109. Physical exercise does appear to reduce the risk of developing dementia and can show improvements in cognition in patients with dementia110–114. Perspectives While the incidence of AD appears to be increasing worldwide, the age-specific risk of developing AD in high income countries may be decreasing. Improvements in diet, exercise, education and management of chronic conditions, such as DM, appear to be improving the individual age-specific risk of AD within the USA115. However, in view of longer life expectancies and worldwide increases in the prevalence of other risk factors, such as obesity and DM, the incidence of AD is most likely set to increase considerably with significant socioeconomic impact. Future work on the development and validation of biomarkers and on the introduction of therapeutics into the preclinical phase of AD has the greatest promise in the effective treatment of this otherwise incurable disease. KEY POINTS Alzheimer disease is increasing in prevalence worldwide Many individuals have preclinical AD without symptoms Biomarkers are currently in development for detecting preclinical AD that may be amenable to novel therapies Addressing modifiable risk factors should also help to reduce the prevalence of AD in the future None of the authors have anything to disclose. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Esiri MM Chance SA Cognitive reserve, cortical plasticity and resistance to Alzheimer’s disease Alzheimer’s research & therapy 2012 4 2 7 2 Bachman DL Wolf PA Linn RT Incidence of dementia and probable Alzheimer’s disease in a general population: the Framingham Study Neurology 1993 43 3 Pt 1 515 519 8450993 3 Vardarajan BN Faber KM Bird TD Age-specific incidence rates for dementia and Alzheimer disease in NIA-LOAD/NCRAD and EFIGA families: National Institute on Aging Genetics Initiative for Late-Onset Alzheimer Disease/National Cell Repository for Alzheimer Disease (NIA-LOAD/NCRAD) and Estudio Familiar de Influencia Genetica en Alzheimer (EFIGA) JAMA neurology 2014 71 3 315 323 24425039 4 Edland SD Rocca WA Petersen RC Dementia and Alzheimer disease incidence rates do not vary by sex in Rochester, Minn Archives of neurology 2002 59 10 1589 1593 12374497 5 Ott A Breteler MM van Harskamp F Prevalence of Alzheimer’s disease and vascular dementia: association with education The Rotterdam study Bmj 1995 310 6985 970 973 7728032 6 Division UNDoEaSAP World Population Ageing 2013 2013 7 2015 Alzheimer’s disease facts and figures Alzheimer’s & dementia: the journal of the Alzheimer’s Association 2015 11 3 332 384 8 Hebert LE Weuve J Scherr PA Alzheimer disease in the United States (2010–2050) estimated using the 2010 census Neurology 2013 80 19 1778 1783 23390181 9 Association As Early-onset dementia: a national challenge, a future crisis Washington, DC Alzheimer’s Association 2006 10 Brookmeyer R Johnson E Ziegler-Graham K Forecasting the global burden of Alzheimer’s disease Alzheimer’s & dementia: the journal of the Alzheimer’s Association 2007 3 3 186 191 11 Prince M Bryce R Albanese E The global prevalence of dementia: a systematic review and metaanalysis Alzheimer’s & dementia: the journal of the Alzheimer’s Association 2013 9 1 63 75 e62 12 Kalaria RN Maestre GE Arizaga R Alzheimer’s disease and vascular dementia in developing countries: prevalence, management, and risk factors Lancet neurology 2008 7 9 812 826 18667359 13 Ferri CP Prince M Brayne C Global prevalence of dementia: a Delphi consensus study Lancet (London, England) 2005 366 9503 2112 2117 14 Hurd MD Martorell P Langa KM Monetary costs of dementia in the United States The New England journal of medicine 2013 369 5 489 490 15 Jack CR Jr Albert MS Knopman DS Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease Alzheimer’s & dementia: the journal of the Alzheimer’s Association 2011 7 3 257 262 16 McKhann G Drachman D Folstein M Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease Neurology 1984 34 7 939 944 6610841 17 Nelson PT Alafuzoff I Bigio EH Correlation of Alzheimer disease neuropathologic changes with cognitive status: a review of the literature J Neuropathol Exp Neurol 2012 71 5 362 381 22487856 18 Consensus recommendations for the postmortem diagnosis of Alzheimer’s disease The National Institute on Aging, and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease Neurobiol Aging 1997 18 4 Suppl S1 2 9330978 19 Brier MR McCarthy JE Benzinger TL Local and distributed PiB accumulation associated with development of preclinical Alzheimer’s disease Neurobiology of aging 2016 38 104 111 26827648 20 Saito S Yamamoto Y Ihara M Mild Cognitive Impairment: At the Crossroad of Neurodegeneration and Vascular Dysfunction Current Alzheimer research 2015 12 6 507 512 26027811 21 Hardy J Allsop D Amyloid deposition as the central event in the aetiology of Alzheimer’s disease Trends in pharmacological sciences 1991 12 10 383 388 1763432 22 Tanzi RE Bertram L Twenty years of the Alzheimer’s disease amyloid hypothesis: a genetic perspective Cell 2005 120 4 545 555 15734686 23 Wisniewski T Goni F Immunotherapeutic approaches for Alzheimer’s disease Neuron 2015 85 6 1162 1176 25789753 24 Di Paolo G Kim TW Linking lipids to Alzheimer’s disease: cholesterol and beyond Nature reviews Neuroscience 2011 12 5 284 296 21448224 25 Bloudek LM Spackman DE Blankenburg M Review and meta-analysis of biomarkers and diagnostic imaging in Alzheimer’s disease Journal of Alzheimer’s disease: JAD 2011 26 4 627 645 21694448 26 Wikler EM Blendon RJ Benson JM Would you want to know? Public attitudes on early diagnostic testing for Alzheimer’s disease Alzheimer’s research & therapy 2013 5 5 43 27 Neumann PJ Cohen JT Hammitt JK Willingness-to-pay for predictive tests with no immediate treatment implications: a survey of US residents Health economics 2012 21 3 238 251 22271512 28 Kopits IM Chen C Roberts JS Willingness to pay for genetic testing for Alzheimer’s disease: a measure of personal utility Genetic testing and molecular biomarkers 2011 15 12 871 875 21749214 29 Jack CR Jr Knopman DS Jagust WJ Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade Lancet neurology 2010 9 1 119 128 20083042 30 Toledo JB Xie SX Trojanowski JQ Longitudinal change in CSF Tau and Abeta biomarkers for up to 48 months in ADNI Acta neuropathologica 2013 126 5 659 670 23812320 31 Vemuri P Wiste HJ Weigand SD Serial MRI and CSF biomarkers in normal aging, MCI, and AD Neurology 2010 75 2 143 151 20625167 32 James OG Doraiswamy PM Borges-Neto S PET Imaging of Tau Pathology in Alzheimer’s Disease and Tauopathies Frontiers in neurology 2015 6 38 25806018 33 Karch CM Goate AM Alzheimer’s disease risk genes and mechanisms of disease pathogenesis Biological psychiatry 2015 77 1 43 51 24951455 34 Corder EH Saunders AM Strittmatter WJ Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families Science 1993 261 5123 921 923 8346443 35 Qiu C Kivipelto M von Strauss E Epidemiology of Alzheimer’s disease: occurrence, determinants, and strategies toward intervention Dialogues Clin Neurosci 2009 11 2 111 128 19585947 36 Raber J Huang Y Ashford JW ApoE genotype accounts for the vast majority of AD risk and AD pathology Neurobiology of aging 2004 25 5 641 650 15172743 37 Holtzman DM Herz J Bu G Apolipoprotein E and apolipoprotein E receptors: normal biology and roles in Alzheimer disease Cold Spring Harbor perspectives in medicine 2012 2 3 a006312 22393530 38 Spinney L Alzheimer’s disease: The forgetting gene Nature 2014 510 7503 26 28 24899289 39 Bu G Apolipoprotein E and its receptors in Alzheimer’s disease: pathways, pathogenesis and therapy Nature reviews Neuroscience 2009 10 5 333 344 19339974 40 Farrer LA Cupples LA Haines JL Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium Jama 1997 278 16 1349 1356 9343467 41 Deary IJ Whiteman MC Pattie A Apolipoprotein e gene variability and cognitive functions at age 79: a follow-up of the Scottish mental survey of 1932 Psychology and aging 2004 19 2 367 371 15222832 42 Mayeux R Ottman R Maestre G Synergistic effects of traumatic head injury and apolipoprotein-epsilon 4 in patients with Alzheimer’s disease Neurology 1995 45 3 Pt 1 555 557 7898715 43 Guo Z Cupples LA Kurz A Head injury and the risk of AD in the MIRAGE study Neurology 2000 54 6 1316 1323 10746604 44 Esiri MM Ageing and the brain The Journal of pathology 2007 211 2 181 187 17200950 45 Guerreiro R Wojtas A Bras J TREM2 variants in Alzheimer’s disease The New England journal of medicine 2013 368 2 117 127 23150934 46 Rohn TT The triggering receptor expressed on myeloid cells 2: “TREM-ming” the inflammatory component associated with Alzheimer’s disease Oxidative medicine and cellular longevity 2013 2013 860959 23533697 47 Rajagopalan P Hibar DP Thompson PM TREM2 and neurodegenerative disease The New England journal of medicine 2013 369 16 1565 1567 48 Blennow K de Leon MJ Zetterberg H Alzheimer’s disease Lancet (London, England) 2006 368 9533 387 403 49 Bayer TA Cappai R Masters CL It all sticks together--the APP-related family of proteins and Alzheimer’s disease Molecular psychiatry 1999 4 6 524 528 10578233 50 Jonsson T Atwal JK Steinberg S A mutation in APP protects against Alzheimer’s disease and age-related cognitive decline Nature 2012 488 7409 96 99 22801501 51 Hartley D Blumenthal T Carrillo M Down syndrome and Alzheimer’s disease: Common pathways, common goals Alzheimer’s & dementia: the journal of the Alzheimer’s Association 2015 11 6 700 709 52 Breteler MM Claus JJ Grobbee DE Cardiovascular disease and distribution of cognitive function in elderly people: the Rotterdam Study BMJ 1994 308 6944 1604 1608 8025427 53 Kuller LH Shemanski L Manolio T Relationship between ApoE, MRI findings, and cognitive function in the Cardiovascular Health Study Stroke 1998 29 2 388 398 9472879 54 Reitz C Brayne C Mayeux R Epidemiology of Alzheimer disease Nature reviews Neurology 2011 7 3 137 152 21304480 55 Stampfer MJ Cardiovascular disease and Alzheimer’s disease: common links Journal of internal medicine 2006 260 3 211 223 16918818 56 Kivipelto M Helkala EL Hanninen T Midlife vascular risk factors and late-life mild cognitive impairment: A population-based study Neurology 2001 56 12 1683 1689 11425934 57 Launer LJ Masaki K Petrovitch H The association between midlife blood pressure levels and late-life cognitive function. The Honolulu-Asia Aging Study Jama 1995 274 23 1846 1851 7500533 58 Swan GE Carmelli D Larue A Systolic blood pressure tracking over 25 to 30 years and cognitive performance in older adults Stroke 1998 29 11 2334 2340 9804644 59 Kalaria RN Vascular basis for brain degeneration: faltering controls and risk factors for dementia Nutrition reviews 2010 68 Suppl 2 S74 87 21091952 60 Leibson CL Rocca WA Hanson VA The risk of dementia among persons with diabetes mellitus: a population-based cohort study Annals of the New York Academy of Sciences 1997 826 422 427 9329716 61 Luchsinger JA Tang MX Stern Y Diabetes mellitus and risk of Alzheimer’s disease and dementia with stroke in a multiethnic cohort American journal of epidemiology 2001 154 7 635 641 11581097 62 Ott A Stolk RP van Harskamp F Diabetes mellitus and the risk of dementia: The Rotterdam Study Neurology 1999 53 9 1937 1942 10599761 63 Arvanitakis Z Wilson RS Bienias JL Diabetes mellitus and risk of Alzheimer disease and decline in cognitive function Archives of neurology 2004 61 5 661 666 15148141 64 Roozenbeek B Maas AI Menon DK Changing patterns in the epidemiology of traumatic brain injury Nature reviews Neurology 2013 9 4 231 236 65 Maas AI Stocchetti N Bullock R Moderate and severe traumatic brain injury in adults Lancet neurology 2008 7 8 728 741 18635021 66 Langlois JA Rutland-Brown W Wald MM The epidemiology and impact of traumatic brain injury: a brief overview The Journal of head trauma rehabilitation 2006 21 5 375 378 16983222 67 Coronado VG McGuire LC Sarmiento K Trends in Traumatic Brain Injury in the U.S. and the public health response: 1995–2009 Journal of safety research 2012 43 4 299 307 23127680 68 Bruns J Jr Hauser WA The epidemiology of traumatic brain injury: a review Epilepsia 2003 44 Suppl 10 2 10 69 Dutton RP Stansbury LG Leone S Trauma mortality in mature trauma systems: are we doing better? An analysis of trauma mortality patterns, 1997–2008 J Trauma 2010 69 3 620 626 20093983 70 Selassie AW McCarthy ML Ferguson PL Risk of posthospitalization mortality among persons with traumatic brain injury, South Carolina 1999–2001 The Journal of head trauma rehabilitation 2005 20 3 257 269 15908825 71 Ferguson PL Smith GM Wannamaker BB A population-based study of risk of epilepsy after hospitalization for traumatic brain injury Epilepsia 2010 51 5 891 898 19845734 72 Fazel S Wolf A Pillas D Suicide, Fatal Injuries, and Other Causes of Premature Mortality in Patients With Traumatic Brain Injury: A 41-Year Swedish Population Study JAMA Psychiatry 2014 73 Masel BE DeWitt DS Traumatic brain injury: a disease process, not an event J Neurotrauma 2010 27 8 1529 1540 20504161 74 Martland HS Punch drunk JAMA 1928 91 15 1103 1107 75 McKee AC Stern RA Nowinski CJ The spectrum of disease in chronic traumatic encephalopathy Brain: a journal of neurology 2013 136 Pt 1 43 64 23208308 76 Shively S Scher AI Perl DP Dementia resulting from traumatic brain injury: what is the pathology? Archives of neurology 2012 69 10 1245 1251 22776913 77 Goldstein LE Fisher AM Tagge CA Chronic traumatic encephalopathy in blast-exposed military veterans and a blast neurotrauma mouse model Science translational medicine 2012 4 134 134ra160 78 McKee AC Cantu RC Nowinski CJ Chronic traumatic encephalopathy in athletes: progressive tauopathy after repetitive head injury Journal of neuropathology and experimental neurology 2009 68 7 709 735 19535999 79 Plassman BL Havlik RJ Steffens DC Documented head injury in early adulthood and risk of Alzheimer’s disease and other dementias Neurology 2000 55 8 1158 1166 11071494 80 Fleminger S Oliver DL Lovestone S Head injury as a risk factor for Alzheimer’s disease: the evidence 10 years on; a partial replication Journal of neurology, neurosurgery, and psychiatry 2003 74 7 857 862 81 Johnson VE Stewart JE Begbie FD Inflammation and white matter degeneration persist for years after a single traumatic brain injury Brain: a journal of neurology 2013 136 Pt 1 28 42 23365092 82 Smith C Gentleman SM Leclercq PD The neuroinflammatory response in humans after traumatic brain injury Neuropathology and applied neurobiology 2013 39 6 654 666 23231074 83 Pierce JE Trojanowski JQ Graham DI Immunohistochemical characterization of alterations in the distribution of amyloid precursor proteins and beta-amyloid peptide after experimental brain injury in the rat The Journal of neuroscience: the official journal of the Society for Neuroscience 1996 16 3 1083 1090 8558237 84 Van den Heuvel C Blumbergs PC Finnie JW Upregulation of amyloid precursor protein messenger RNA in response to traumatic brain injury: an ovine head impact model Experimental neurology 1999 159 2 441 450 10506515 85 Gentleman SM Nash MJ Sweeting CJ Beta-amyloid precursor protein (beta APP) as a marker for axonal injury after head injury Neuroscience letters 1993 160 2 139 144 8247344 86 Oppenheimer DR Microscopic lesions in the brain following head injury Journal of neurology, neurosurgery, and psychiatry 1968 31 4 299 306 87 Cribbs DH Chen LS Cotman CW Injury induces presenilin-1 gene expression in mouse brain Neuroreport 1996 7 11 1773 1776 8905662 88 Nadler Y Alexandrovich A Grigoriadis N Increased expression of the gamma-secretase components presenilin-1 and nicastrin in activated astrocytes and microglia following traumatic brain injury Glia 2008 56 5 552 567 18240300 89 Blasko I Beer R Bigl M Experimental traumatic brain injury in rats stimulates the expression, production and activity of Alzheimer’s disease beta-secretase (BACE-1) Journal of neural transmission 2004 111 4 523 536 15057522 90 Clinton J Roberts GW Gentleman SM Differential pattern of beta-amyloid protein deposition within cortical sulci and gyri in Alzheimer’s disease Neuropathology and applied neurobiology 1993 19 3 277 281 8355814 91 Deane R Sagare A Zlokovic BV The role of the cell surface LRP and soluble LRP in blood-brain barrier Abeta clearance in Alzheimer’s disease Current pharmaceutical design 2008 14 16 1601 1605 18673201 92 Turner AJ Fisk L Nalivaeva NN Targeting amyloid-degrading enzymes as therapeutic strategies in neurodegeneration Annals of the New York Academy of Sciences 2004 1035 1 20 15681797 93 Iwata N Tsubuki S Takaki Y Identification of the major Abeta1–42-degrading catabolic pathway in brain parenchyma: suppression leads to biochemical and pathological deposition Nature medicine 2000 6 2 143 150 94 Iwata N Tsubuki S Takaki Y Metabolic regulation of brain Abeta by neprilysin Science 2001 292 5521 1550 1552 11375493 95 Johnson VE Stewart W Graham DI A neprilysin polymorphism and amyloid-beta plaques after traumatic brain injury Journal of neurotrauma 2009 26 8 1197 1202 19326964 96 Chen XH Johnson VE Uryu K A lack of amyloid beta plaques despite persistent accumulation of amyloid beta in axons of long-term survivors of traumatic brain injury Brain pathology 2009 19 2 214 223 18492093 97 Hellstrom-Lindahl E Ravid R Nordberg A Age-dependent decline of neprilysin in Alzheimer’s disease and normal brain: inverse correlation with A beta levels Neurobiology of aging 2008 29 2 210 221 17098332 98 Yan P Hu X Song H Matrix metalloproteinase-9 degrades amyloid-beta fibrils in vitro and compact plaques in situ The Journal of biological chemistry 2006 281 34 24566 24574 16787929 99 Hickman SE Allison EK El Khoury J Microglial dysfunction and defective beta-amyloid clearance pathways in aging Alzheimer’s disease mice The Journal of neuroscience: the official journal of the Society for Neuroscience 2008 28 33 8354 8360 18701698 100 Jaunmuktane Z Mead S Ellis M Evidence for human transmission of amyloid-beta pathology and cerebral amyloid angiopathy Nature 2015 525 7568 247 250 26354483 101 Irwin DJ Abrams JY Schonberger LB Evaluation of potential infectivity of Alzheimer and Parkinson disease proteins in recipients of cadaver-derived human growth hormone JAMA neurology 2013 70 4 462 468 23380910 102 Eisele YS Obermuller U Heilbronner G Peripherally applied Abeta-containing inoculates induce cerebral beta-amyloidosis Science 2010 330 6006 980 982 20966215 103 Prusiner SB Biology and genetics of prions causing neurodegeneration Annual review of genetics 2013 47 601 623 104 Mattson MP Maudsley S Martin B BDNF and 5-HT: a dynamic duo in age-related neuronal plasticity and neurodegenerative disorders Trends in neurosciences 2004 27 10 589 594 15374669 105 Kalmijn S Launer LJ Ott A Dietary fat intake and the risk of incident dementia in the Rotterdam Study Ann Neurol 1997 42 5 776 782 9392577 106 Roberts RO Cerhan JR Geda YE Polyunsaturated fatty acids and reduced odds of MCI: the Mayo Clinic Study of Aging Journal of Alzheimer’s disease: JAD 2010 21 3 853 865 107 Engelhart MJ Geerlings MI Ruitenberg A Diet and risk of dementia: Does fat matter?: The Rotterdam Study Neurology 2002 59 12 1915 1921 12499483 108 Carlson MC Helms MJ Steffens DC Midlife activity predicts risk of dementia in older male twin pairs Alzheimer’s & dementia: the journal of the Alzheimer’s Association 2008 4 5 324 331 109 Fratiglioni L Wang HX Brain reserve hypothesis in dementia Journal of Alzheimer’s disease: JAD 2007 12 1 11 22 17851191 110 Groot C Hooghiemstra AM Raijmakers PG The effect of physical activity on cognitive function in patients with dementia: A meta-analysis of randomized control trials Ageing research reviews 2016 25 13 23 26607411 111 Abbott RD White LR Ross GW Walking and dementia in physically capable elderly men Jama 2004 292 12 1447 1453 15383515 112 Fratiglioni L Paillard-Borg S Winblad B An active and socially integrated lifestyle in late life might protect against dementia Lancet neurology 2004 3 6 343 353 15157849 113 Rovio S Kareholt I Helkala EL Leisure-time physical activity at midlife and the risk of dementia and Alzheimer’s disease Lancet neurology 2005 4 11 705 711 16239176 114 Akbaraly TN Portet F Fustinoni S Leisure activities and the risk of dementia in the elderly: results from the Three-City Study Neurology 2009 73 11 854 861 19752452 115 Langa KM Is the risk of Alzheimer’s disease and dementia declining? Alzheimer’s research & therapy 2015 7 1 34

PMC005xxxxxx/PMC5116424.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 100883691 21720 Cell Microbiol Cell. Microbiol. Cellular microbiology 1462-5814 1462-5822 27206578 5116424 10.1111/cmi.12617 NIHMS794247 Article The Borrelia burgdorferi CheY3 Response Regulator is Essential for Chemotaxis and Completion of its Natural Infection Cycle Novak Elizabeth A. 1*^ Sekar Padmapriya 2^ Xu Hui 1 Moon Ki Hwan 1 Manne Akarsh 1 Wooten R. Mark 2 Motaleb Md. A. 1# 1 Department of Microbiology and Immunology, Brody School of Medicine, East Carolina University, Greenville, North Carolina, USA 2 Department of Medical Microbiology and Immunology, University of Toledo College of Medicine, Toledo, Ohio, USA # Corresponding Author: MD A. MOTALEB, Department of Microbiology and Immunology, East Carolina University Brody School of Medicine, 600 Moye Blvd, Greenville, NC 27834, USA, motalebm@ecu.edu * Current address: University of Pittsburgh School of Medicine, Pittsburgh, PA 15224. ^ E.A. Novak and P. Sekar contributed equally as 1st authors for this reported work 14 6 2016 11 7 2016 12 2016 01 12 2017 18 12 17821799 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. SUMMARY Borrelia burgdorferi possesses a sophisticated and complex chemotaxis system but how the organism utilizes this system in its natural enzootic life cycle is poorly understood. Of the three CheY chemotaxis response regulators in B. burgdorferi, we found that only deletion of cheY3 resulted in an altered motility and significantly reduced chemotaxis phenotype. Though ΔcheY3 maintained normal densities in unfed ticks, their numbers were significantly reduced in fed ticks compared to the parental or cheY3-complemented spirochetes. Importantly, mice fed upon by the ΔcheY3-infected ticks did not develop a persistent infection. Intravital confocal microscopy analyses discovered that the ΔcheY3 spirochetes were motile, but appeared unable to reverse direction and perform the characteristic backward-forward motility displayed by the parental strain. Subsequently, the ΔcheY3 became “trapped” in the skin matrix within days of inoculation, were cleared from the skin needle-inoculation site within 96 hours post-injection, and did not disseminate to distant tissues. Interestingly, although ΔcheY3 cells were cleared within 96 hours post-injection, this attenuated infection elicited significant levels of B. burgdorferi-specific IgM and IgG. Taken together, these data demonstrate that cheY3-mediated chemotaxis is crucial for motility, dissemination, and viability of the spirochete both within and between mice and ticks. INTRODUCTION Bacterial cells possess sophisticated signal transduction pathways that detect changes in specific parameters of their dynamic external environment, allowing them to respond appropriately to the fluctuating environment (Falke et al., 1997, Skerker et al., 2005). Chemotaxis, the cellular movement in response to chemical gradients (Wadhams et al., 2004), is one behavior that bacteria alter based on their external environment (Wolanin et al., 2002). This movement empowers bacteria to approach and remain in beneficial environments or escape from noxious ones by modulating their swimming behaviors. There are a vast array of signals, including nutrient concentrations, osmolarity, temperature, oxygen, and pH changes, which bacteria integrate together and translate into a specific response (Falke et al., 1997), either benign or pathogenic (Wadhams et al., 2004). Bacterial chemotaxis pathways are regulated by a complex two-component signal transduction system (Falke et al., 1997, Bourret et al., 2010, Hazelbauer, 2012). The system starts with the membrane-bound chemoreceptors—methyl-accepting chemotaxis proteins (MCPs)—that sense external stimuli. The MCPs are coupled (via the coupling protein, CheW) with CheA, a histidine kinase. The catalytic activity of CheA is mediated by a ligand (either attractant or repellant) bound to the MCPs. Once active, CheA utilizes ATP to autophosphorylate and then proceeds to phosphorylate the response regulator, CheY. Phosphorylated CheY (CheY-P) then binds to the flagellar switch proteins FliM and FliN, regulating the direction of rotation of the flagella (Welch et al., 1993, Toker et al., 1997, Djordjevic et al., 1998, Sarkar et al., 2010). Borrelia burgdorferi, the causative agent of Lyme disease (Burgdorfer et al., 1982, Lane et al., 1991), is the most common arthropod-borne disease in the United States and Europe (Mead, 2015). Chemotaxis and motility genes comprise approximately 5–6% of the genome of B. burgdorferi, and we have shown that motility is required for every stage of the infectious life cycle of B. burgdorferi (Sultan et al., 2013, Motaleb et al., 2015, Sultan et al., 2015). In nature, B. burgdorferi cycles between the Ixodes tick vector and a mammalian host (Burgdorfer et al., 1982, Levine et al., 1985, Lane et al., 1991, Tsao, 2009, Brisson et al., 2012). Completion of the enzootic cycle requires that B. burgdorferi traverse through dense and complex tissues within tick and vertebrate hosts; the spirochetes must migrate from the midgut to the salivary glands within the tick to allow transmission to the next host during tick feeding (Dunham-Ems et al., 2009, Radolf et al., 2012, Sultan et al., 2013), as well as navigate through the skin matrix of a vertebrate host after deposition via the tick-bite to reach a multitude of target tissues before migrating back to a feeding tick (Ribeiro et al., 1987, Dunham-Ems et al., 2009, Pal, 2010, Radolf et al., 2012, Sultan et al., 2013). The spirochete must complete these tasks all by simultaneously detecting its current environment, determining its next optimal direction, and evading the immune systems of both hosts. Indeed, a recent global signature-tagged mutagenesis study identified that mutations in chemotaxis and motility genes were most often associated with loss in infectivity of B. burgdorferi, signifying their importance for its enzootic lifecycle (Lin et al., 2012). While chemotaxis has been extensively studied in Escherichia coli and Salmonella enterica (Armitage, 1999, Bren et al., 2000, He et al., 2014), the chemotaxis system of B. burgdorferi is poorly understood and differs greatly from those prototypic systems. The Lyme disease spirochete is relatively long (10 to 20 μm) and thin (0.3 μm) with a distinctive flat-wave morphology, and motility is generated by rotation of the periplasmic flagella (Charon et al., 2012). Between 7 to 11 periplasmic flagella are attached near each cellular pole, causing the motility/chemotaxis behavior of B. burgdorferi and other spirochetes to be unique and complex. Tracking of B. burgdorferi swimming in vitro has described that spirochetes perform run, flex, and reverse swimming modes. Runs occur when the periplasmic flagellar motors at one pole rotate in the direction opposite that of the motors at the other pole (motors at one end rotate in clock-wise whereas motors at other end rotate counter clock-wise). The flex is a non-translational (i.e. no net motility) mode and is thought to be equivalent to the E. coli tumble. During the flex, the motors at both poles rotate in the same direction, i.e. both rotate in clock-wise (CW) or counter clock-wise (CCW). Spirochetal reversal occurs in translating (i.e. motile) cells when the motors at each end reverse their direction of rotation. For spirochetes to swim toward an attractant, the organisms must be able to coordinate the rotation of the motors at the two separate poles of the cell that are located at a considerable distance from one another (often greater than 10 μm). One of the questions related to spirochete chemotaxis is how the organisms are able to achieve this coordination (Li et al., 2002, Motaleb et al., 2005, Motaleb et al., 2011b, Charon et al., 2012, Wolgemuth, 2015). The genome of B. burgdorferi encodes multiple homologs of several chemotaxis genes (e.g. two cheA, three cheW, three cheY, two cheB, and two cheR genes), making it much more complex than E. coli or S. enterica (Fraser et al., 1997, Charon et al., 2012, Motaleb et al., 2015). In several species of pathogenic bacteria, the CheY response regulator has been shown to play an important role in chemotaxis, which is also needed for virulence (Butler et al., 2005, McGee et al., 2005, Antunez-Lamas et al., 2009, Lertsethtakarn et al., 2011). To date, only two studies have shown that chemotaxis—specifically involving the histidine kinase cheA2 and phosphatase enhancer cheD—are essential for the infectious life cycle of B. burgdorferi (Sze et al., 2012, Moon et al., 2016). Amino acid sequence analysis indicates that B. burgdorferi CheY1, CheY2, and CheY3 share 25–37% identity with each other. Moreover, these proteins share 32%, 38%, and 25% amino acid sequence identity with E. coli CheY, respectively (Motaleb et al., 2011b). Importantly, all of the functional residues of the E. coli CheY response regulator were found to be conserved in CheY1, CheY2, and CheY3, suggesting that they all are potential chemotaxis response regulators. Previous reports also indicate that cheY3, but not cheY1 or cheY2, is important for motility and chemotaxis in vitro (Motaleb et al., 2011b). Moreover, the function of cheY3 cannot be substituted by the other cheYs in B. burgdorferi. Importantly, those cheY mutants were constructed in a high-passage, avirulent strain that cannot be evaluated in the tick vector or vertebrate hosts (Motaleb et al., 2011b). We hypothesize that cheY3 is crucial for one or more stages of the enzootic cycle. The goal of this study is to utilize a cheY3-mutated B. burgdorferi generated in a virulent genetic background to delineate the importance of CheY3 for the different host environments encountered by these bacteria. Our findings are significant in describing the deficiencies in motility and chemotaxis abilities exhibited by the ΔcheY3 strain and delineating the essential nature of this gene for every stage of the tick-mouse infection cycle. Based on our data, we propose a model indicating the importance of chemotaxis (and motility) during the enzootic life cycle of B. burgdorferi. RESULTS Construction and confirmation of ΔcheY3 and cheY3+ strains B. burgdorferi genome sequence and transcriptional analyses indicated that the cheY3 gene is located on the large linear chromosome and is transcribed as a polycistronic mRNA using a σ70 promoter (Li et al., 2002). Previous reports also indicate that only cheY3 is essential for chemotaxis (Motaleb et al., 2011b). However, those studies were done in vitro under one condition, using a high-passage, non-infectious clone that cannot be evaluated in the tick-mouse infectious cycle (Motaleb et al., 2011b). To determine the importance of cheY3 in the pathogenic life cycle of B. burgdorferi, we constructed a ΔcheY3 in the low-passage B31-A3 strain using a promoterless kanamycin resistance cassette, Pl-Kan (Figure 1A) (Sultan et al., 2010). PCR data confirmed the deletion of the cheY3 gene (data not shown, see below). The cheY3 gene is located at the end of its operon, thus integration of Pl-Kan in place of the cheY3 gene is unlikely to exhibit a polar effect on the expression of downstream genes. However, we complemented the ΔcheY3 in cis by genomic integration to ensure the phenotype of the mutant was solely attributed to the inactivation of the cheY3 (Figure 1A). Mutation and complementation of cheY3 were confirmed by western blotting as demonstrated by the absence or presence of CheY3 production, respectively (Figure 1B). Since cheY3 is in a polycistronic operon, we verified expression of other gene products (cheX and cheA2) in that operon. As shown in Figure 1B, expression of CheA2 and CheX proteins were not affected in ΔcheY3 bacteria. Furthermore, linear and circular endogenous plasmids of the mutant and complemented clones were verified by PCR and found that all clones retained the plasmids seen in the parental wild-type (WT) cells (data not shown). In vitro phenotypes of ΔcheY3 The growth, motility, and chemotaxis phenotypes of ΔcheY3 were assessed in vitro using various approaches. Dark-field microscopic analysis indicated that WT and cheY3+ cells exhibited run-flex/pause-reverse swimming patterns, whereas ΔcheY3 cells constantly ran in a single direction without reversals or switching their swimming direction. Additionally, semi-solid plating analysis showed that ΔcheY3 formed significantly smaller swarming colonies compared to WT or cheY3+ (Figure 2A). Despite the altered motility and swarming deficiencies, ΔcheY3 had no growth defect in liquid culture relative to WT or the cheY3+ (data not shown). Additionally, the chemotactic response of the ΔcheY3 was measured by using a quantitative capillary-tube chemotaxis assay. As expected, the chemotactic ability of the mutant was significantly reduced when compared to the WT or complemented cells (Figure 2B). The complemented cells exhibited a chemotactic response that was higher than the WT cells. We currently do not understand why this response was enhanced, but we speculate that the cheY3+ cells assessed for this assay were taken from a growth phase that was more chemotactic than the WT cells even though we were careful to collect all clones from the same growth phase. Together, these data are the same as reported previously, which indicate that cheY3 is important for chemotaxis as well as for governing motility in B. burgdorferi in vitro (Motaleb et al., 2011b). ΔcheY3 spirochetes are unable to survive in mice by needle injection To evaluate the ability of the ΔcheY3 to establish infection in a mammalian host, groups of C3H/HeN mice were needle-inoculated with WT, ΔcheY3, or cheY3+, and multiple tissues (i.e. ear, bladder, and joint tissues from each mouse) were collected at 4 weeks post-injection to determine bacterial outgrowth (Table 1). WT and complemented cells were re-isolated from one or more tissues assessed from all mice in those groups, demonstrating that these cells established an infection. Alternatively, ΔcheY3 spirochetes were not re-isolated from any mouse tissues, even when infected with 1×106 bacteria (Table 1). These results indicate that the cheY3 chemotaxis response regulator is essential for B. burgdorferi infection in a mammalian host. ΔcheY3 spirochetes are unable to infect mice by Ixodes scapularis tick bite Tick-mouse-tick infection studies were performed to determine if ΔcheY3 spirochetes are able to colonize, replicate, and transmit from infected ticks to naïve C3H/HeN mice. Since ΔcheY3 were unable to infect mice via needle inoculation, infecting ticks via feeding on infected mice was not feasible. Alternatively, we performed tick-immersion studies in order to artificially infect naïve larvae, and then the ticks were allowed to feed on naïve mice to determine the ability of the spirochetes to transmit from infected ticks to naïve mice. Additionally, the burden of spirochetes per tick was evaluated to determine the replication and survival of the spirochetes within the fed larvae 7 days post-repletion. Immunofluorescence assay using a FITC-conjugated antibody specific for B. burgdorferi demonstrated that spirochetes were intact within fed ticks (data not shown). However, while artificial immersion achieved a 90% infection rate of ticks for all strains (data not shown), the burden of ΔcheY3 spirochetes per tick was significantly less when compared to the WT or the cheY3+ cells, as determined by tick plating (Figure 3A). This result indicates that the non-chemotactic ΔcheY3 has a decreased ability to survive or multiply within fed larvae. Four weeks after tick-feeding, mice were euthanized and animal tissues were cultivated to re-isolate B. burgdorferi in order to determine transmission from tick to mouse. No tissues from mice fed on by ΔcheY3-infected ticks showed bacterial outgrowth whereas 4 out of 5 mice fed upon by WT- or cheY3+-infected larvae showed regrowth (Figure 3B). These assays were also performed using encapsulated artificially immersed nymphs that produced similar results—reduced survivability of the mutants in fed but not in unfed ticks (Figures 4A, B). Moreover, we assessed the skin containing the tick-bite site, as well as ear, joint, and bladder tissues for bacterial outgrowth; PCR was also performed to detect spirochetes genomes. The artificially ΔcheY3-immersed nymphs were unable to establish an infection in any of the mice they fed upon, as the mutant bacteria were not reisolated by outgrowth in culture media or detected by PCR from those tissues, whereas five out of six mice were positive for WT bacteria and three out of six mice were positive for the cheY3+ spirochetes (Figure 4C). Together, these results indicate that chemotaxis is crucial for optimal survival of spirochetes in ticks and infection of mice by tick-bite. Intravital microscopy shows that ΔcheY3 motility behavior within skin is similar to its in vitro phenotype The finding that the ΔcheY3 was unable to infect mice via needle-injection or tick-bite suggested this mutant would demonstrate similar deficiencies in motility/chemotaxis in vivo as observed in vitro. To address this, intravital confocal fluorescence microscopy was used to directly observe GFP-expressing strains of WT and ΔcheY3 within the intact skin tissues of living mice. No notable differences were observed in the shapes of the WT GFP-expressing B. burgdorferi (WT-eGFP) and ΔcheY3-GFP-expressing B. burgdorferi (ΔcheY3-eGFP) spirochetes, suggesting that loss of this chemotaxis protein does not alter the morphology of B. burgdorferi within skin tissues. Assessment of time-lapse images taken at 6 hours post-injection showed that the majority of WT bacteria displayed a back-and-forth (B/F) motion (see Experimental Procedures section for a description of different motility events), with only a few showing directed runs (Figure 5B and Video 1) (Harman et al., 2012). Alternatively, while the majority of the ΔcheY3 mutants appeared non-motile, those that did display translational motility were unable to reverse their direction; even when they did stop moving, they were only able to continue movement in the same direction (Figure 5B and Video 2). Viewing the non-motile ΔcheY3 at a higher magnification showed that the majority of these mutants appeared to be trying to move forward, but could only make a minimal gain as they appeared stuck and were returned to the same spot, resulting in no net movement. This suggests that certain tissue components were hindering B. burgdorferi movement, and since the ΔcheY3 cannot reverse direction to find an unhindered route, they lose translational motility. Based on our described motility patterns (see Experimental Procedures), WT bacteria maintained the B/F motion as the major motility behavior over a period of 48 hours, whereas the majority of the ΔcheY3 bacteria were non-translational by 6 hours and almost all were non-motile at 48 hours (Figure 5C), suggesting the ΔcheY3 mutants are either stuck or become non-motile and/or cleared. Since a significant number of ΔcheY3 bacteria were non-translational by 6 hours, the velocity of bacteria that could translocate (translational, includes both run and B/F) and the ones that could not (non-translating bacteria defined above) were calculated separately. The average velocity of translating WT bacteria was 217.7 μm/min while translating ΔcheY3 bacteria was 183.4 μm/min (Figure 5D). Thus, even though ΔcheY3 were unable to reverse direction, they could still achieve velocities similar to that of the WT bacteria in skin. The velocity of bacteria at later times post-injection was not calculated, since the majority the ΔcheY3 bacteria (>90%) were non-translational after 6 hours post-inoculation. These observations suggest that the loss of cheY3 affects the motility pattern, but not the velocity potential of B. burgdorferi within murine skin. ΔcheY3 bacteria fail to disseminate from the skin injection site to other target tissues and are cleared within 4 days post-injection To determine the persistence of the mutant in mice, ΔcheY3-eGFP and WT-eGFP bacteria were injected intradermally into ear skin and confocal microscopy images were collected at the indicated times post-injection for manual enumeration of the spirochetes in each image. The burden of WT bacteria appeared to initially decrease between 6 hours and 24 hours post-injection, but subsequently increased dramatically between 48 hours and 96 hours post-injection (Figure 6A, right panel). Alternatively, the burden of ΔcheY3 steadily decreased after 6 hours post-injection, such that none of the mutants were visible by 96 hours post-injection, suggesting that the murine immune system cleared the ΔcheY3 from skin tissues within 96 hours post-injection. Although the same number of WT and ΔcheY3 spirochetes was injected, there appeared to be more ΔcheY3 than WT bacteria per viewing field at 6 hours post-injection (Figure 6A). We speculate that these differences are artificial due to more ΔcheY3 spirochetes becoming stuck at the injection site compared to WT spirochetes, since WT are capable of efficiently migrating through skin. These data indicate that loss of cheY3 leads to increased clearance of B. burgdorferi within murine skin tissues. Although the ΔcheY3 appeared to be cleared relatively quickly at the inoculation site, it was still feasible that their limited mobility could allow dissemination to distant tissues via the bloodstream or other routes. To address this, B6 mice were intradermally injected with WT-eGFP or ΔcheY3-eGFP spirochetes and a number of target tissues (e.g. distant back skin, ankles, and heart) were harvested at the indicated times post-injection for bacterial enumeration by qPCR (Figure 6B). ΔcheY3 were cleared from all ear tissues (i.e. the injection site) by 96 hours post-injection, whereas WT spirochetes persisted in ear tissues (Figure 6B; top left), corroborating our intravital microscopy results (Figure 6A). Assessment of all distant tissues (i.e. back skin, heart, and ankles) indicated that WT B. burgdorferi were detected in all tested tissues by day 14–28 post-injection. However, no ΔcheY3 were detected in any of these tissues at any times post-injection, implying that these mutants did not escape the injection site but were instead cleared from the local ear tissues by the murine immune system (Figure 6B). This is consistent with the tick-mouse infection studies (Figure 3C and 4C), where ΔcheY3 spirochetes could not be re-isolated or detected from mouse tissues harvested 4–28 days after infected-ticks fed to repletion. Thus, loss of cheY3 leads to early clearance of B. burgdorferi in mice and an inability to disseminate to distant tissues. Infection with ΔcheY3 is sufficient to elicit B. burgdorferi-specific antibodies Typically during any infection, an immunocompetent vertebrate host initiates a primary antibody response that consists mainly of IgM antibodies. As this initial response starts to wane, the host then generates a stronger secondary antibody response that is primarily comprised of pathogen-specific IgG antibodies. Interestingly, B. burgdorferi infection of both mice and humans elicits a continuously increasing IgM response during an active infection together with an IgG response that is not well-maintained after the infection is cleared (Kalish et al., 2001, Hastey et al., 2012, Elsner et al., 2015). When the antisera from our current experiments were assessed by ELISA, WT bacteria elicited a strong B. burgdorferi-specific IgM response that continued to increase even after 60 days post-injection (Figure 7A), similar to those studies listed previously. Alternatively, ΔcheY3 infection elicited IgM levels similar to WT ≤28 post-injection, but these levels subsequently decrease by day 62 (Figure 7A). B. burgdorferi-specific IgG levels generated in response to the ΔcheY3 were also similar to WT ≤28 post-injection (Figure 7B); however the WT IgG levels continue to increase between days 28 and 62 post-injection, whereas antibody levels against ΔcheY3 did not increase after day 28, but was maintained for the duration of this experiment. Interestingly, the IgG levels against a non-motile ΔmotB (lacks the MotB motor stator), which we previously reported were cleared within 48–72 hours post-injection and elicited a minimal B. burgdorferi-specific antibody response, were lower than both WT and ΔcheY3 at day 62 post-injection (Sultan et al., 2015). Thus, although the IgM response is abbreviated in the ΔcheY3-infected mice, the infection is still sufficient to generate a robust B. burgdorferi-specific IgG response that persists well after bacterial clearance. DISCUSSION The enzootic lifecycle of B. burgdorferi requires that it cycles between a tick vector and a vertebrate host (Burgdorfer et al., 1982, Welch et al., 1993, Toker et al., 1997). This necessitates that the spirochete sense and assess its external environment, and subsequently responds in an appropriate fashion by changing its cellular behavior and/or gene expression accordingly. The chemosensory system of B. burgdorferi would presumably aid in such sensing as well as traversing a path between the tick vector and the mammalian host. Previous results indicate that all of the chemotaxis genes in the flaA-cheA2-cheW3-cheX-cheY3 operon are important for chemotaxis in vitro (Li et al., 2002, Motaleb et al., 2005, Motaleb et al., 2011b, Zhang et al., 2012). Additionally, previous studies have demonstrated that CheA2 is capable of both autophosphorylating and transferring that phosphoryl group to CheY3 (Motaleb et al., 2005). Therefore, it is plausible that CheA2 and CheY3 comprise a two-component system in B. burgdorferi, and that CheA2 is the cognate kinase for CheY3. The in vivo studies with the ΔcheY3 demonstrated that burden of the ΔcheY3 in fed but not unfed ticks was significantly reduced compared to the burden of parental spirochetes (Figure 4A, B). The reason for this observed reduced burden in ticks is unknown. It is possible that since the ΔcheY3 is non-chemotactic and only able to swim in one direction it is easily identified and captured by the innate immune system of the tick or that ingested blood factors more efficiently cleared the mutant organisms (Ribeiro et al., 1990, Kern et al., 2011, Hajdusek et al., 2013). Based on our previous investigations with motility and cyclic-di-GMP mutants, we proposed that back-and-forth motility is crucial to protect the spirochetes in the fed ticks (Sultan et al., 2010, Pitzer et al., 2011, Sultan et al., 2013, Novak et al., 2014, Motaleb et al., 2015, Sultan et al., 2015). The same proposal can be applied here, as the ΔcheY3 spirochetes are unable to reverse their swimming patterns. Nevertheless, further studies are needed in order to determine the cause for the reduced burden of the ΔcheY3 in ticks. The inability of the chemotaxis-deficient ΔcheY3 to establish an infection in mice is not completely understood, but could be due to the inability of the mutant to disseminate to target organs and/or evade the host cellular immune responses, as these mutant cells are not able to relay its CheY3-associated signals to the flagellar motor, and thus would not respond appropriately to certain chemotactic signals within their microenvironment (Table 1 and Figure 3C, 4C). We noticed that the in vivo phenotype of the mutant was partially restored in the cheY3+ cells (Figure 4C) even though we complemented the mutant by genomic reconstitution, and the CheY3 protein expression was restored to WT level (Figure 1). The reason for the partial restoration could be due to the loss of some of the endogenous B. burgdorferi plasmids in part of the population of the cheY3+ cells (Sultan et al., 2013). Nonetheless, our current data suggest that the failure of ΔcheY3 to switch its swimming behavior is causing the mutant to become trapped in the skin tissues. Skin dermis is a meshwork composed of a variety of ECM molecules, including collagen, elastin, and an extrafibrillar matrix made up of glycosaminoglycans (GAGs), proteoglycans, and glycoproteins (Leong et al., 1998, Parveen et al., 2000, Jarvelainen et al., 2009). WT B. burgdorferi appear able to detect barriers in its motility pathway and subsequently reverse or change its direction before taking an adjacent path that might be comprised of less dense tissue. Alternatively, the ΔcheY3 is unable to reverse and adjust its direction, and thus will eventually become trapped after translocating into a dense area, indicating that back-and-forth movement is essential for maneuvering around complex structures in the skin tissue (Moriarty et al., 2008, Norman et al., 2008). This is supported by the significant reduction of ΔcheY3 in skin by 24 hours and complete clearance by 96 hours post-injection (Figure 6A), suggesting that these non-translating spirochetes were quickly recognized and efficiently cleared by innate immune components. There are currently no commercially available vaccines to protect against B. burgdorferi infection. Many of the individual purified proteins tested as vaccines do not confer complete protection (Hanson et al., 1998, Exner et al., 2000, Hagman et al., 2000, Nogueira et al., 2012, Floden et al., 2013). Passive transfer of serum from mice (or humans) injected with WT bacteria or with certain B. burgdorferi proteins seem to provide limited protection (Fikrig et al., 1994, Barthold et al., 1997, Hanson et al., 1998, Fikrig et al., 2000, Hagman et al., 2000, Floden et al., 2013, Small et al., 2014), suggesting that antibodies can provide protection if generated against critical antigens. Attenuated or killed vaccines have been successful in preventing many bacterial and viral infections (Mc, 1948, Mekalanos, 1994, Jin et al., 2015), but there are few studies screening non-infectious B. burgdorferi mutants for potential protective abilities. Immunization with killed or an aflagellar B. burgdorferi mutant conferred protection to naïve mice only if the mice were challenged early with WT bacteria (Johnson et al., 1986, Sadziene et al., 1996). But similar to passive immunization results, the protective ability of these attenuated (or killed) strains was decreased if the mice were challenged late (≥ 90 days post-immunization) (Johnson et al., 1986, Sadziene et al., 1996). One of the reasons surmised for this failure to achieve long-lasting immunity was that these mutants did not persist more than 24 hours post-injection, which is insufficient for generating an effective antibody response, particularly since it theoretically would not provide enough time for the mutants to upregulate critical virulence proteins that are expressed during vertebrate infection. Our observation that the ΔcheY3 survived ≤96 hours post-injection suggested that these mutants might persist long enough to upregulate more vertebrate host-specific virulence factors that could elicit antibodies capable of conferring protection from subsequent challenges. Our studies showed that infection with the ΔcheY3 generated similar levels of spirochete-specific IgM and IgG as WT B. burgdorferi through day 28 post-infection before reaching a significant baseline level that was maintained at least through day 62 post-infection. Notably, these levels were also significantly higher at all times tested than those generated against a non-motile ΔmotB strain that was shown to be cleared within 48–72 hours post-infection (Sultan et al., 2015). This implies that infection with the ΔcheY3 could potentially generate sufficient levels of spirochete-specific antibodies to confer protection against future challenges with fully virulent B. burgdorferi strains. If so, such studies might also allow the identification of individual antigens that can elicit protective antibodies, potentially leading to a vaccine based on recombinant proteins. Further studies are needed to demonstrate that infection with the mutant protects against WT infection. As mentioned, B. burgdorferi contains three homologs of the cheY gene, which are located in separate operons (Motaleb et al., 2011b). However, lack of cheY3 was not compensated for by the other cheY homologs, and evidence suggests that cheY1 and cheY2 are not important for motility and chemotaxis in vitro (Motaleb et al., 2011b). So, what then is the function of these additional cheY genes? Many bacterial genomes contain multiple copies of a chemotaxis cheY gene and at least one of them is shown to be dedicated to control motility, similar to what we report in this communication with cheY3. The function of most of the additional genes is unknown (in any bacterium). There are a few instances in which multiple chemotaxis-like signaling systems appear to have very different roles in the same species and those additional cheY genes may control other cellular non-chemotactic processes (Kato et al., 1999, Szurmant et al., 2004, Whitchurch et al., 2004, Berleman et al., 2005). Consequently, it is plausible that B. burgdorferi CheY2 or CheY1 controls some non-chemotactic cellular processes such as serving as virulence determinants rather than acting as a classical chemotaxis response regulator, like CheY3 (H. Xu et al—manuscript under review; our unpublished observations). In summary, we have shown that the CheY3 response regulator is an essential component of the chemosensory system in B. burgdorferi, and that CheY3 is important for successful completion of the enzootic lifecycle and for continued perpetuation of Lyme disease. Based on our data, we propose that the chemosensory system of B. burgdorferi, and by extension motility, is critical during mammalian infection, dissemination, and possibly transmission from tick to the vertebrate host (Figure 8) (Motaleb et al., 2015). The motility/chemotaxis system is likely to be “active” or “on” during dissemination and persistent infection of the mammalian host, as well as during the tick’s blood-meal to allow navigation from the mouse into the tick and subsequent colonization of the mid-gut. However, after the nutrients from the blood-meal are depleted during the molt, B. burgdorferi must shut down flagellar rotation and chemotaxis to conserve energy, as this (motility/chemotaxis) system is unnecessary during molting (Figure 8) (Motaleb et al., 2015). Motility and chemotaxis must then be turned back “on” during nymphal feeding for subsequent transmission of the spirochetes. Although the latest time point post-injection analyzed by intravital microscopy in this study was 96 hours, we have examined WT-eGFP bacteria constantly swimming in the murine host for >2 years post-injection (R. M. Wooten and M. A. Motaleb, unpublished data). Accordingly, we propose that continuous chemotaxis/motility activities are necessary for persistent mammalian infection and are still required even after B. burgdorferi reaches its target colonization tissues in the mammal host. As outlined in Figure 8, CheY proteins are phosphorylated by CheA-P. CheY3-P then binds to the flagellar switch proteins to alter swimming behavior until it is dephosphorylated by the CheX phosphatase (Motaleb et al., 2005, Pazy et al., 2010). Based on our data, only CheY3 appears to act as a classical response regulator, as it contributes to motility and chemotaxis. Further studies are needed to not only elucidate the environmental signals that trigger the “active/on” or “inactive/off” status of the chemotaxis/motility system, but also the exact molecular components that transduce these signals within the spirochete. EXPERIMENTAL PROCEDURES Ethics statement East Carolina University and University of Toledo are both accredited by the International Association for the Assessment and Accreditation of Laboratory Animal Care. All animal procedures received Institutional Animal Care and Use Committee approvals and were in accordance with federal guidelines for the care and use of laboratory animals. Mouse strains All intravital microscopy experiments were performed utilizing C57BL/6 (B6; National Cancer Institute); all other experiments were completed using C3H/HeN mice (Charles River Laboratories, Raleigh, NC). These mouse strains contain equivalent susceptibilities to infection with B. burgdorferi and have been shown to contain similar bacterial numbers in most tissues during persistent infection, even though they can yield different levels of disease severity (Ma et al., 1998, Brown et al., 1999b, Weis et al., 1999). Bacterial strains and growth conditions Low-passage, virulent B. burgdorferi strain B31-A3 was utilized as the WT clone throughout this study (Elias et al., 2002). This clone was used to infect naïve C3H/HeN mice and then was reisolated from the mouse tissues, subcloned, and used as the parental clone for all of our subsequent studies. The genome of this strain is known to contain 12 linear and 9 circular plasmids, for a total of 21 plasmids, in addition to a 960-kbp linear chromosome (Fraser et al., 1997, Casjens et al., 2000). This strain lacks circular plasmid 9 (cp9), but still maintains its infectivity in tick-mouse infection studies (Elias et al., 2002, Jewett et al., 2009). Amplified genomic DNA from B31-A3 was used as the foundation to construct the ΔcheY3 (explained below). B. burgdorferi cells were grown in liquid Barbour-Stoenner-Kelly (BSK-II) medium, and cells were plated using plating BSK (P-BSK), which was prepared using 0.5% agarose (Motaleb et al., 2007, Stewart et al., 2008b, Sultan et al., 2013). Cells were grown at 35°C in a 2.5% CO2 incubator, as previously described (Elias et al., 2002, Smith et al., 2003, Motaleb et al., 2007). When required, culture and plating medium were supplemented with appropriate antibiotics at the following concentrations: 200 μg/ml kanamycin and 100 μg/ml streptomycin. The endogenous plasmid contents from all B. burgdorferi strains were confirmed before commencing any in vivo study involving ticks or mice. Construction of the ΔcheY3, complement, and green-fluorescent protein (GFP) strains The B. burgdorferi cheY3 gene was identified from the genomic sequence of B. burgdorferi and was annotated as bb0672 (441 bp) (Fraser et al., 1997). Construction of the cheY3-deletion plasmid, electroporation, and plating conditions were described previously (Sultan et al., 2010, Pitzer et al., 2011). Briefly, three pieces of DNA fragments were amplified separately by PCR: cheY3 gene plus adjacent DNA from the 5′-end (1 kb), the promoterless kanamycin resistance cassette, and the 3′-flanking DNA of cheY3 (1 kb). PCR primer sequences are not shown here but can be obtained upon request. These three pieces of fragments were gel purified and then ligated by overlapping PCR, as described in detail (Motaleb et al., 2011a). The PCR product was then ligated into the pGEM-T Easy vector (Promega, Inc.), yielding pCheY3-Pl-Kan-Easy. DNA containing cheY3-Pl-Kan was linearized by NotI restriction digestion to remove the ampicillin marker of the vector and electroporated into competent B31-A3 cells to obtain mutants (Motaleb et al., 2000, Sultan et al., 2013). Transformants were screened by PCR for proper recombination of the cheY3 inactivation cassette. The ΔcheY3 GFP-expressing strain was constructed in the same manner as the ΔcheY3, except the linearized DNA was electroporated into a B31-A3-GFP expressing strain, as described previously (Sultan et al., 2015). In order to complement the cheY3 mutation, the cheY3 and cheX gene were amplified from genomic DNA along with adjacent flanking DNA using the primers (5′-3′): CheY3.FORWARD (TTGGATGCTGCTTCTTCGG) and CheY3.REVERSE (GCTGCTTGCATTGTTAG). The resulting PCR product was ligated into the pGEM-T Easy vector, yielding RCY3-Easy. The PflgB-aadA cassette, which confers resistance to streptomycin (Frank et al., 2003), was similarly amplified with engineered HindIII sites by PCR using the following primers (5′-3′): Flg-F (AAGCTTCCCGAGTTCAAGGAAGAT) and Strep-R (AAGCTTATTATTTGCCGACTACCTTGG). The cassette was then inserted into the unique HindIII site after the cheY3 gene. DNA containing cheX-cheY3-aadA was linearized by restriction digestion and was then electroporated into competent ΔcheY3 cells in order to obtain complemented cells. Transformants were selected with streptomycin. Western blot analysis was used to confirm the inactivation and restoration of CheY3 in the mutant and complemented cells, respectively, and that no disruption of other gene products in the operon had occurred as a result of the genetic manipulation (see below). Linear and circular plasmid contents of all B. burgdorferi transformants were confirmed by PCR using primers described previously (Elias et al., 2002, Sultan et al., 2010, Pitzer et al., 2011). SDS-PAGE and immunoblot analyses Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting with an enhanced chemiluminescent detection method (GE Health Inc.) were performed as reported previously (Motaleb et al., 2000). A Bio-Rad protein assay kit determined the concentration of protein in cell lysates. Unless otherwise noted, 5 μg of lysate protein was subjected to SDS-PAGE and immunoblotting using specific antibodies. Antibodies kindly provided by other investigators included the following: anti-DnaK by J. Benach (State University of New York [SUNY], Stony Brook, NY), and polyclonal E. coli anti-CheA by R. Silversmith (University of North Carolina, Chapel Hill, NC). CheY3 and CheX antibodies are described elsewhere (Motaleb et al., 2005, Motaleb et al., 2011b). Mouse serum collected at indicated times post-injection was utilized as the ‘primary antibody’ in western blot analysis to determine the B. burgdorferi-specific antibody response. Dark-field microscopy and swarm plate assays Exponentially growing B. burgdorferi clones (1–3×107 cells/ml) were imaged using a Zeiss Imager M1 dark-field microscope connected to a digital camera to determine morphology and motility. Swarming ability of individual bacterial colonies of each strain was determined by plating no more than 50 cells into a petri dish (95 mm × 15 mm) containing semi-solid P-BSK (0.35% agarose) diluted 1:10 in Dulbecco’s phosphate-buffered saline (Motaleb et al., 2000, Motaleb et al., 2007, Motaleb et al., 2011a). Plates were incubated for three weeks at which time colony diameters were measured. At least 11 colony diameters were measured for each strain in each assay. Capillary tube chemotaxis assay The chemotactic ability of B. burgdorferi was quantitatively determined as previously described (Motaleb et al., 2011b) with slight modifications. Briefly, WT, ΔcheY3, or cheY3 complement (cheY3+) B. burgdorferi were grown to approximately 1×108 cells/ml. Spirochetes were counted using a Petroff-Hausser chamber, and the cells needed for the assay (1×107 cells/ml) were collected by centrifuging for 15 minutes at 1,800 × g at room temperature. The cells were washed with PBS and then gently resuspended in motility buffer (136.9 mM NaCl, 8.10 mM Na2HPO4, 2.7 mM KCl, 1.47 mM KH2PO4, 2% recrystallized BSA (Sigma-Aldrich Co.), 0.1 mM EDTA, pH 7.4) to obtain a final solution of 1×107 cells/ml, 0.5% methylcellulose (400 mesh; Sigma Aldrich Inc.). This suspension (0.2 ml) was added into each 1.5 ml eppendorf tube chemotaxis chamber with a perforated lid. Capillary tubes were filled with control solution (motility buffer) or attractant solution (100 mM N-Acetylglucosamine in motility buffer, 0.22μm filter-sterilized) and sealed on one end with clay. Excess fluids on the outside of the capillary tubes were wiped off, and the capillary tubes were inserted into the appropriate eppendorf tube chemotaxis chambers (one capillary tube per chamber; 3–5 replicates per strain). Tubes were incubated at 35°C for 2 hours, at which point the tubes were carefully wiped off to remove excess liquid on the outside, and the contents was expelled into a clean eppendorf tube. The contents of each chemotaxis tube were plated individually using P-BSK to determine colony-forming units (CFU). The plates were incubated at 35°C for 2–5 weeks (until colonies appeared). Each strain was repeated three times with each repeat containing 3 replicates, and the results are expressed as the average attractant/buffer ratio. An increase in the number of spirochetes equal to or greater than twice that of the buffer control was considered significant. Mouse infection studies using injected B. burgdorferi Six-to-seven weeks old C3H/HeN mice were used for infection studies, as previously described (Elias et al., 2002, Stewart et al., 2008a, Sultan et al., 2013). In order to determine the infectious ability of the spirochetes, mice (n = 5 or 2 per strain) were injected subcutaneously with WT (5×103), ΔcheY3 (5×103 or 1×106), or cheY3+ cells (5×104). The number of spirochetes was determined using a Petroff-Hausser chamber and verified by colony forming units (CFUs) by plating. At four weeks post-injection, mice were euthanized and ear skin, bladder, and tibiotarsal joints were harvested and placed in BSK-II broth for up to 35 days to allow bacterial outgrowth from the animal tissues (Elias et al., 2002, Grimm et al., 2003, Stewart et al., 2008a), which is the direct determination of the ability of spirochetes to infect mice by tick bite and disseminate throughout the body. The presence of spirochetes in the growth medium was determined by dark-field microscopy (Elias et al., 2002, Grimm et al., 2003, Grimm et al., 2004, Sultan et al., 2013). Tick-mouse studies Transmission of spirochetes from infected ticks to naïve mice was assessed using tick-mouse infection assays (Policastro et al., 2003, Battisti et al., 2008, Stewart et al., 2008a). Naïve Ixodes scapularis larvae were purchased from Oklahoma State University. Naïve larvae were artificially inoculated by immersion in equal-density, exponential-phase (5×107 cells/ml) cultures of B. burgdorferi clones, as previously described (Policastro et al., 2003, Battisti et al., 2008, Stewart et al., 2008a). Ticks were subsequently fed to repletion on mice (3 mice per strain; ~200 larvae/mouse). After 5–7 days, fed ticks were collected once they dropped off mice. At 7 days post-repletion, a subset of ticks was individually dissected and the isolated midguts were analyzed by immunofluorescence (IFA; see below) for the presence of spirochetes (Stewart et al., 2008a). A second subset of ticks was surface-sterilized with 3% H2O2 and 70% ethanol, individually crushed in BSK-II medium, and plated in P-BSK to determine the number of viable spirochetes per tick (CFUs). Results are expressed as the spirochete burden per tick, where each dot is representative of a single tick (n = 12). A third subset of ticks was crushed individually and genomic DNA was extracted using the DNeasy blood and tissue kit, according to the manufacturer’s instructions (Qiagen Inc.). The DNA from each tick was then utilized for PCR to determine spirochete-positive ticks. Spirochete burdens in each spirochete-positive tick were determined by using quantitative real-time PCR (qPCR) using primers specific for the B. burgdorferi enolase gene, as described previously (Yang et al., 2004, Zhang et al., 2009, Pitzer et al., 2011). Copies of the B. burgdorferi enolase gene per tick were extrapolated from a standard curve generated using a known amount of plasmid DNA containing the enolase gene as the template. To determine spirochete transmission and infection of the C3H/HeN mice, fed animals were euthanized four weeks post-repletion followed by bacterial outgrowth analysis. The presence of spirochetes in the growth medium was determined by dark-field microscopy (Sultan et al., 2010, Sultan et al., 2011). Transmission of spirochetes to mice by encapsulated nymphs Naïve nymphs were artificially infected by immersion, as described above. Nymphs were allowed to feed on mice using a capsulated system, as previously described (Mulay et al., 2009, Patton et al., 2011, Sultan et al., 2013). Mice were anesthetized and 15–20 nymphs were confined to capsules affixed to the shaved back of a naïve C3H/HeN mouse (n = 3 per strain per assay). The ticks were allowed to feed to repletion and then collected from the capsules. At 7 days post-repletion, spirochete burdens were determined from individually crushed ticks by qPCR, as described above. The results are expressed as mean ± SEM from at least 4 spirochete-positive tick data per clone per assay. At 68 hours or 4 days post-repletion, mice were euthanized and the tick-bite sites were extensively washed. A section of skin comprising the tick-feeding site was excised, rinsed in 70% isopropanol, and cut into equal portions. Part of the tick-bite site skin, ear, bladder, and joint tissues were cultured separately in BSK-II medium for up to 35 days to determine bacterial outgrowth, and the other remaining tissues were processed for PCR to detect B. burgdorferi DNA using enolase gene-specific primers (Pitzer et al., 2011). Immunofluorescence Assay IFAs were set up as previously described (Sultan et al., 2013). Briefly, ticks were individually dissected in phosphate-buffered saline (PBS)-5 mM MgCl2 on Teflon-coated microscopic slides. Dissected tick contents were then 10-fold serially diluted to avoid quenching by hemin in the blood (Policastro et al., 2003, Sultan et al., 2010). The slides were air dried and blocked with 0.75% bovine serum albumin (BSA) in PBS-5 mM MgCl2 for 30 min. The slides were then washed twice with PBS-5 mM MgCl2, and spirochetes were detected using goat anti-B. burgdorferi antisera labeled with fluorescein isothiocyanate (1:100 dilution; Kirkegaard & Perry Laboratories, Inc.). Images were captured using a Zeiss Axio Imager M1 microscope connected to a digital camera. Intravital microscopy Two days prior to an experiment, the outer ear surface of the mice was depilated (Nair), rinsed immediately with H2O, and allowed to rest. Low-passage cultures of WT-eGFP and ΔcheY3-eGFP were counted to contain the desired B. burgdorferi inoculum (106 bacteria) in 10μl BSK-II medium. B6 mice were anesthetized, and the desired numbers of GFP-expressing bacteria were injected intradermally in a 10μl bolus using a 31G insulin syringe via the dorsal ear surface of the mouse. The mice were allowed to rest for 6h and then re-anesthetized for imaging at the indicated times. Imaging was performed using an Olympus FV1000 laser confocal microscope system. For imaging, the mouse was placed on a 37°C heated imaging stage to mount the ear on a coverslip (with the template) with the dorsal surface down. For bacterial enumeration, as well as motility patterns and velocity determination of the bacteria, 2D images were collected using a 20× dry objective with a 2× optical zoom at 1 frame/1100 milliseconds for 66 seconds (60 images) at different times post-injection. At least two time-lapse images were collected for each section of the template (=10 time-lapse images per ear). For bacterial morphology studies, images were collected between 2–6 hours post-injection using a 60× water immersion objective with an additional optical zoom. Image Analysis for intravital microscopy For bacterial enumeration and motility assessment, the first of the 60 images in a dataset was selected and the bacteria in that image were counted and tracked manually. Motility patterns and velocity were calculated for bacteria present in the viewing field since the first frame. For motility patterns, the movement of any bacteria that remained within the viewing field for ≥ 5 seconds was manually assessed and categorized as one of three groups: (1) Run—bacteria continue to move in one direction without reversing or switching the swimming direction throughout the entire time of assessment, (2) Back/Forth (B/F)—bacteria reversed/switched their direction of motion at least once throughout the entire time of assessment, and (3) Non-translational—bacteria had no net movement, which includes both non-motile bacteria and those that could move slightly, but were unable to generate net movement (translocate). Time of assessment of the bacteria was defined as the total time the bacteria were visible in the viewing field, which ranged from 5 seconds minimum up to 60 seconds. For calculating the velocity of the bacteria, all the 60 images of the dataset were assessed using MetaMorph software (Molecular Devices), and every bacterium noted on the first image was tracked through 60 images and the average velocity of the bacterium was calculated. For tracking, one end of the spirochete was chosen and it was followed until that end is not visible for more than 1 frame. Observations for bacteria that stayed in the field of view for less than 5 frames were discarded. Quantitative measurement of B. burgdorferi in murine skin and other target tissues The bacterial numbers in murine tissues were quantified as previously described (Brown et al., 1999a, Morrison et al., 1999). Briefly, naïve anesthetized B6 mice were intradermally injected with 5×104 bacteria per animal via the dorsal ear surface. Multiple tissues including the entire ears, back skin, entire ankle joints, and heart were collected at various times post-injection. The tissues were appropriately processed to isolate DNA, as described (Brown et al., 1999a, Morrison et al., 1999, Sultan et al., 2015). qPCR was performed using a Light Cycler 96 (Roche Diagnostics) and Syber Green detection. Mouse DNA levels were determined by amplifying the nidogen gene and B. burgdorferi DNA was determined by amplifying the flaB gene. Copy numbers for mouse and B. burgdorferi genomes present in each sample were calculated by extrapolation to standard curves using LightCycler software (Roche Diagnostics). Normalizing B. burgdorferi genomes to 1000 mouse genomes represented the final B. burgdorferi numbers. The primers used to detect mouse nidogen were nido.F (5′-CCA GCC ACA GAA TAC CAT CC-3′) and nido.R (5′-GGA CAT ACT CTG CTG CCA TC-3′). The oligonucleotide primers used to detect B. burgdorferi flaB were flaB.F (5′-TTG CTG ATC AAG CTC AAT ATA ACC A -3′) and flaB.R (5′-TTG AGA CCC TGA AAG TGA TGC -3′). B. burgdorferi-specific antibody detection by ELISA from intravital microscopy studies Serum was collected at the indicated times by either retro-orbital bleeding or exsanguination, and immunoglobulin (Ig) content was assessed using previously described methods (Brown et al., 1999a). For detection of B. burgdorferi-specific antibodies, 96-well high-binding ELISA plates (Costar) were coated with B. burgdorferi sonicate or goat anti-mouse total immunoglobulin (IgG+IgM+IgA; Southern Biotech). For B. burgdorferi sonicates, the WT bacteria were grown at 33ºC and then shifted to 37ºC for the final overnight culture (Schwan et al., 2000). This temperature-shifted culture was then resuspended in PBS for sonication within a closed sterile tube using a bath sonicator (Sonifier® Cell Disruptor). The protein content of the sonicated sample was measured by Bicinchoninic acid assay (BCA; Pierce Thermo Scientific) and stored at −80ºC. The B. burgdorferi ELISA plates were made by coating appropriate wells with 5μg/ml of the sonicate (for sample and blank) or goat anti-mouse total Ig (to provide a purified Ig standard) in 0.1M carbonate-bicarbonate Buffer (pH 9.5) overnight. Serial dilutions of individual sera from uninfected or infected mice were added to the B. burgdorferi-coated plates overnight, washed to remove unbound antibodies, and bound murine Ig was detected using goat anti-mouse IgG conjugated to biotin (Southern Biotech). After incubation for 2–3 hours at room temperature, the samples were visualized by adding avidin-conjugated Horseradish Peroxidase (avidin-HRP; Vector Labs) for 30 minutes followed by adding the color solution (0.4mg/ml of O-phenylenediamine and 0.01% H2O2 in citrate buffer, pH 5) and inhibiting the reaction with 1N hydrochloride. The content was quantified by comparison to standard curves constructed using the appropriate purified mouse Ig isotype (Southern Biotech). Statistical analyses The significance difference between the mean values of the groups for each experiment was analyzed as follows. When comparing WT, ΔcheY3, and cheY3+ (either the cheY3+ or ΔmotB) data was checked with D’Agostino-Pearson omnibus normality test to determine if the values come from a Gaussian distribution. The data that passed normality was analyzed via a multiple-comparison analysis using a one-way analysis of variance (ANOVA), followed by a Tukey’s post hoc test; if the data did not pass normality, a multiple-comparison analysis was performed by using a Kruskal-Wallis test, followed by Dunn test. P < 0.05 is considered significant. For all intravital and detection of GFP-expressing bacteria in mice, normality was checked using Shapiro-Wilk normality test. If data was normal than the difference between WT and ΔcheY3 was assessed via an unpaired t-test, whereas non-normal data was assessed via a Mann Whitney test. Supplementary Material Supp Video S1 Video 1. WT B. burgdorferi motility in mouse skin 1×106 WT-eGFP B. burgdorferi were injected intradermally into mouse ear skin and time-lapse images were collected using an Olympus FV1000 laser confocal microscope system at 6 hours post-injection. 2D images were collected using a 20× dry objective with a 2X optical zoom at 1frame/1100 milliseconds for 66 seconds (60 images). The video is sped up two times the “real-time” speed. Supp Video S2 Video 2. ΔcheY3 B. burgdorferi motility in mouse skin 1×106 ΔcheY3-eGFP B. burgdorferi were injected intradermally into mouse ear skin and time-lapse images were collected using an Olympus FV1000 laser confocal microscope system at 6 hours post-injection. 2D images were collected using a 20× dry objective with a 2X optical zoom at 1frame/1100 milliseconds for 66 seconds (60 images). The video is sped up two times the “real-time” speed. We thank Dr. Ruth Silversmith for reagents and the Medical Entomology and Zoonoses Ecology team at Public Health, England for the tick’s illustrations. We also would like to acknowledge John Presloid for technical help in performing the murine infection and imaging studies. This research was supported by National Institute of Allergy and Infectious Diseases grants (1R21AI113014), National Institute of Arthritis and Musculoskeletal and Skin Diseases grant (1R01AR060834), and an American Heart Association Pre-doctoral Fellowship 14PRE20490177 (PS). Figure 1 Construction and complementation of the ΔcheY3 A. WT B. burgdorferi genome arrangement of flaA operon containing cheY3 (labeled as WT chromosome). The Pl-Kan (aph1) cassette was inserted after deleting the cheY3 gene by allelic exchange (ΔcheY3 chromosome). The mutant was complemented in cis by genomic reconstitution by inserting a WT copy of the cheY3 gene flanked by the PflgB-aadA cassette (cheY3+ chromosome). Arrows indicate the direction of transcription. DNAs/Plasmids are not drawn to scale. B. Immunoblot analysis of B. burgdorferi cells probed with the indicated antibodies. The CheY3 protein expression was inhibited in the mutant, but restored in the complemented cheY3+ cells as confirmed by using cell lysates from the indicated clones probed with anti-CheY3. CheY3 protein is approximately 14 kDa. The cheA2 and cheX genes are located in the same operon as the targeted cheY3, however, the expression of those gene products were not altered in the mutant or the complemented cells (see anti-CheA and anti-CheX blots). DnaK was used as a loading control. Figure 2 In vitro phenotypes of the ΔcheY3 A. ΔcheY3 colonies exhibit a significantly reduced swarming ability on semi-solid agar plates when compared to the parental WT or complemented cells (P = 0.0001). Values are mean ± SD from at least 11 individual colonies per strain. Statistical analysis was performed by using ANOVA followed by Tukey’s Multiple Comparisons test. A P<0.05 between strains is considered significant. B. The ΔcheY3 is deficient in chemotaxis as determined by the capillary tube chemotaxis assay using N-Acetylglucosamine as an attractant. A capillary tube filled with buffer without any attractant was used as a control. An increase in the number of spirochetes equal to or greater than twice that of the buffer control was considered “chemotactic”. The numbers on top of each vertical bar is the fold-increase over the buffer control. Results shown are mean ± SEM from three independent studies with three replicates per strain per assay. Figure 3 CheY3 is important for tick-mouse infectious cycle A. The burden of ΔcheY3 was significantly less in fed larval ticks compared to WT or the complemented cheY3+. Naive larval ticks were artificially infected by immersion using in vitro-grown spirochetes. Larvae were crushed individually on day 7 post-repletion, and the spirochetal density per larva was determined by plating on semi-solid growth media followed by counting viable spirochetes (CFU). The p-values were determined using Kruskal-Wallis ANOVA test followed by Dunn Multiple Comparisons test. Results shown are the spirochete burden per larva, where each dot is representative of a single tick (n = 12). The line denotes the mean of the entire group. A P<0.05 between strains is considered significant. B. CheY3 is important for establishing infection in mice by tick-bite. Naïve C3H/HeN mice were fed upon by artificially-infected larvae (~200 larvae/mouse, 3 mice per clone). Four weeks post-repletion, mice were euthanized to determine bacterial outgrowth from ear, joint, and bladder tissues. Figure 4 Mice are not infected by ΔcheY3-infected nymph bite The burden of the mutant was significantly less in fed, but not in unfed nymphs when compared to the parental cells (A, unfed; B, fed nymphs). Naïve nymphs were artificially infected and then allowed to feed on separate naïve mice (n=3 per assay). Seven days after feeding, nymphs (fed and unfed) were processed for PCR analysis to determine spirochete-positive ticks, and subsequently qPCR to determine the number of spirochete genomes using enolase gene-specific primers (five spirochete-positive nymphs per clone). Results shown are mean ± SEM. Statistical analysis was performed using ANOVA test followed by Tukey Multiple Comparisons test. A P<0.05 between strains is considered significant. C. Naïve mice fed by the ΔcheY3-infected nymphs were not able to establish persistent infection. Fifteen infected encapsulated nymphs per mouse were allowed to feed. Four days (or 68h, data not shown) after feeding, mice were euthanized to detect B. burgdorferi DNA from tick bite site skin, ear, joint, and bladder tissues by PCR (one half of each tissue or one joint tissue from each mouse). To validate the PCR data, the other half of each of those tissues was processed for bacterial outgrowth analysis (not shown). WT and cheY3+ spirochetes were detected only in the tick-bite site skin tissues, as expected, at this early time points when B. burgdorferi cells generally do not disseminate to the distant tissues. Figure 5 Comparison of motility patterns in murine ear tissue between ΔcheY3 and WT A. Morphology of WT and ΔcheY3 B. burgdorferi was observed in vivo in ear skin tissue of mouse using intravital microscopy technique. 1×106 WT-eGFP or ΔcheY3-eGFP B. burgdorferi was injected intradermally into ear skin and images were collected between 2-6 hours post-injection. Both WT (i–ii) and ΔcheY3 (iii–iv) bacteria demonstrate similar characteristic flat-wave morphologies in vivo. Scale bar is 5μm. B–C. 1×106 WT-eGFP or ΔcheY3-eGFP B. burgdorferi (Bb) were injected as above and time-lapse images were collected at different times post-injection. B. Representative images of ΔcheY3 (left) and WT (right) motility path tracked at 6h post-injection using MetaMorph. Colored lines (except green) are the tracks of the bacteria. C. For both strains, the % bacteria performing run, back-and-forth (B/F) and no translational motility was calculated. The majority of WT bacteria at all times performed B/F, whereas most of the ΔcheY3 bacteria were non-translational. The translating ΔcheY3 performed mainly runs with occasional stops, rather than B/F. Results show average % motility ± SEM. ## p = 0.003 % run compared to ΔcheY3; $$ p = 0.0026, $$$ p ≤ 0.0003 % B/F compared to ΔcheY3; ** p = 0.0057, *** p ≤ 0.006 % no translational motility compared to ΔcheY3. Statistics were performed using the Mann Whitney test. n ≥ 3 mice. D. The velocity of WT and ΔcheY3 was measured at 6h post-injection using MetaMorph. Results show average ± SEM. ***p<0.0001 compared to the non-translational counterparts; Statistics were performed using the Mann Whitney test. n ≥ 30 bacteria under each bar. Figure 6 Dissemination and persistence of ΔcheY3 in distant target tissues A. Groups of C57Bl/6 mice were injected intradermally in both ears with 1×106 GFP-expressing ΔcheY3 or WT. At the indicated times, ear skin-resident bacteria were visually assessed using confocal fluorescent microscopy. Visual data was processed manually to determine motility patterns. Results shown are mean ± SD. *p = 0.04, **p < 0.0069 ***p < 0.0001 compared to the WT; Results were analyzed using an unpaired t-test; n ≥ 3 mice under each bar. B. Groups of C57Bl/6 mice were injected intradermally in both ears with 5×104 GFP-expressing ΔcheY3 or WT B. burgdorferi. Both ears, back skin, heart, and both ankles tissues were harvested at the indicated time points for DNA isolation. qPCR analysis was performed to quantify spirochete burdens. Values represent the average of B. burgdorferi genomes (flaB) per 1000 copies of mouse genome (nidogen); values of zero were assigned as 0.0001 for representation on a log scale. *p<0.05, **p<0.01, ***p<0.0001 compared to WT; Results were analyzed using the Mann Whitney test. n ≥ 5 mice from two separate experiments. Figure 7 Infection with ΔcheY3 is sufficient to elicit B. burgdorferi-specific antibodies The mice used in the dissemination study (Figure 6) were sacrificed at the indicated time post-injection and sera were collected. B. burgdorferi-specific antibodies were quantified by ELISA analyses using bacterial sonicates as the capture antigen. A. Detection of IgM antibody levels in serum from WT-, ΔcheY3- and ΔmotB-infected mice. ΔmotB spirochetes was used as a control as these bacteria are reported to be non-motile, fail to disseminate from the injection site, and are cleared from the host within 48-72h after injection. **p < 0.01, ***p < 0.0001 as shown in the graph. # p < 0.05 and ### p < 0.0001 as compared to the ΔcheY3. Results were analyzed using the ANOVA test followed by the Tukey-Kramer Multiple Comparisons test. B. Detection of IgG antibody levels in serum from WT-, ΔcheY3- and ΔmotB-infected mice. **p < 0.01, ***p < 0.0001 as shown in the graph. # p < 0.05 as compared to WT. Statistical analyses were performed using ANOVA test followed by Tukey-Kramer Multiple Comparisons test. Figure 8 Model of the B. burgdorferi chemotaxis/motility system during the enzootic cycle A simplistic chemotaxis signaling pathway of B. burgdorferi (wave-like red shapes) is depicted here and is described in the Discussion. We postulate that CheY2 or CheY1 controls some non-chemotactic cellular processes such as serving as virulence determinants rather than acting as a classical chemotaxis response regulator, like CheY3. Tick illustrations were kindly provided by the Medical Entomology and Zoonoses Ecology team at Public Health, England. Table 1 ΔcheY3 lacks infectivity in mice via needle injectiona Strain Dosage (spirochetes/mouse) Number of Mice Infected WT 5 × 103 5/5 ΔcheY3 5 × 103 0/5 1 × 106 0/5 cheY3+ 5 × 104 2/2 a C3H/HeN mice were injected intradermally using the indicated in vitro-grown spirochete clones. Mice were sacrificed four weeks post injection, and infectivity was determined by reisolation or detection of B. burgdorferi genomes by PCR from the murine ear skin, tibiotarsal joint, and bladder tissue samples. Doses shown are the actual number of spirochetes injected in each mouse. Antunez-Lamas M Cabrera-Ordonez E Lopez-Solanilla E Raposo R Trelles-Salazar O Rodriguez-Moreno A Rodriguez-Palenzuela P 2009 Role of motility and chemotaxis in the pathogenesis of Dickeya dadantii 3937 (ex Erwinia chrysanthemi 3937) Microbiology 155 434 442 19202091 Armitage JP 1999 Bacterial tactic responses Advances in microbial physiology 41 229 289 10500847 Barthold SW Feng S Bockenstedt LK Fikrig E Feen K 1997 Protective and arthritis-resolving activity in sera of mice infected with Borrelia burgdorferi Clinical infectious diseases: an official publication of the Infectious Diseases Society of America 25 Suppl 1 S9 17 9233658 Battisti JM Bono JL Rosa PA Schrumpf ME Schwan TG Policastro PF 2008 Outer surface protein A protects Lyme disease spirochetes from acquired host immunity in the tick vector Infection and immunity 76 5228 5237 18779341 Berleman JE Bauer CE 2005 Involvement of a Che-like signal transduction cascade in regulating cyst cell development in Rhodospirillum centenum Molecular microbiology 56 1457 1466 15916598 Bourret RB Silversmith RE 2010 Two-component signal transduction Current opinion in microbiology 13 113 115 20219418 Bren A Eisenbach M 2000 How signals are heard during bacterial chemotaxis: protein-protein interactions in sensory signal propagation Journal of bacteriology 182 6865 6873 11092844 Brisson D Drecktrah D Eggers CH Samuels DS 2012 Genetics of Borrelia burgdorferi Annual review of genetics 46 515 536 Brown CR Reiner SL 1999a Development of lyme arthritis in mice deficient in inducible nitric oxide synthase The Journal of infectious diseases 179 1573 1576 10228086 Brown JP Zachary JF Teuscher C Weis JJ Wooten RM 1999b Dual role of interleukin-10 in murine Lyme disease: regulation of arthritis severity and host defense Infection and immunity 67 5142 5150 10496888 Burgdorfer W Barbour AG Hayes SF Benach JL Grunwaldt E Davis JP 1982 Lyme disease-a tick-borne spirochetosis? Science 216 1317 1319 7043737 Butler SM Camilli A 2005 Going against the grain: chemotaxis and infection in Vibrio cholerae Nature reviews. Microbiology 3 611 620 16012515 Casjens S Palmer N van Vugt R Huang WM Stevenson B Rosa P 2000 A bacterial genome in flux: the twelve linear and nine circular extrachromosomal DNAs in an infectious isolate of the Lyme disease spirochete Borrelia burgdorferi Molecular microbiology 35 490 516 10672174 Charon NW Cockburn A Li C Liu J Miller KA Miller MR 2012 The unique paradigm of spirochete motility and chemotaxis Annual review of microbiology 66 349 370 Djordjevic S Stock AM 1998 Structural analysis of bacterial chemotaxis proteins: components of a dynamic signaling system Journal of structural biology 124 189 200 10049806 Dunham-Ems SM Caimano MJ Pal U Wolgemuth CW Eggers CH Balic A Radolf JD 2009 Live imaging reveals a biphasic mode of dissemination of Borrelia burgdorferi within ticks The Journal of clinical investigation 119 3652 3665 19920352 Elias AF Stewart PE Grimm D Caimano MJ Eggers CH Tilly K 2002 Clonal polymorphism of Borrelia burgdorferi strain B31 MI: implications for mutagenesis in an infectious strain background Infection and immunity 70 2139 2150 11895980 Elsner RA Hastey CJ Olsen KJ Baumgarth N 2015 Suppression of Long-Lived Humoral Immunity Following Borrelia burgdorferi Infection PLoS pathogens 11 e1004976 26136236 Exner MM Wu X Blanco DR Miller JN Lovett MA 2000 Protection elicited by native outer membrane protein Oms66 (p66) against host-adapted Borrelia burgdorferi: conformational nature of bactericidal epitopes Infection and immunity 68 2647 2654 10768956 Falke JJ Bass RB Butler SL Chervitz SA Danielson MA 1997 The two-component signaling pathway of bacterial chemotaxis: a molecular view of signal transduction by receptors, kinases, and adaptation enzymes Annual review of cell and developmental biology 13 457 512 Fikrig E Bockenstedt LK Barthold SW Chen M Tao H Ali-Salaam P 1994 Sera from patients with chronic Lyme disease protect mice from Lyme borreliosis The Journal of infectious diseases 169 568 574 8158028 Fikrig E Feng W Barthold SW Telford SR 3rd Flavell RA 2000 Arthropod- and host-specific Borrelia burgdorferi bbk32 expression and the inhibition of spirochete transmission Journal of immunology 164 5344 5351 Floden AM Gonzalez T Gaultney RA Brissette CA 2013 Evaluation of RevA, a fibronectin-binding protein of Borrelia burgdorferi, as a potential vaccine candidate for lyme disease Clinical and vaccine immunology: CVI 20 892 899 23595502 Frank KL Bundle SF Kresge ME Eggers CH Samuels DS 2003 aadA confers streptomycin resistance in Borrelia burgdorferi Journal of bacteriology 185 6723 6727 14594849 Fraser CM Casjens S Huang WM Sutton GG Clayton R Lathigra R 1997 Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi Nature 390 580 586 9403685 Grimm D Elias AF Tilly K Rosa PA 2003 Plasmid stability during in vitro propagation of Borrelia burgdorferi assessed at a clonal level Infection and immunity 71 3138 3145 12761092 Grimm D Tilly K Byram R Stewart PE Krum JG Bueschel DM 2004 Outer-surface protein C of the Lyme disease spirochete: a protein induced in ticks for infection of mammals Proceedings of the National Academy of Sciences of the United States of America 101 3142 3147 14970347 Hagman KE Yang X Wikel SK Schoeler GB Caimano MJ Radolf JD Norgard MV 2000 Decorin-binding protein A (DbpA) of Borrelia burgdorferi is not protective when immunized mice are challenged via tick infestation and correlates with the lack of DbpA expression by B. burgdorferi in ticks Infection and immunity 68 4759 4764 10899883 Hajdusek O Sima R Ayllon N Jalovecka M Perner J de la Fuente J Kopacek P 2013 Interaction of the tick immune system with transmitted pathogens Frontiers in cellular and infection microbiology 3 26 23875177 Hanson MS Cassatt DR Guo BP Patel NK McCarthy MP Dorward DW Hook M 1998 Active and passive immunity against Borrelia burgdorferi decorin binding protein A (DbpA) protects against infection Infection and immunity 66 2143 2153 9573101 Harman MW Dunham-Ems SM Caimano MJ Belperron AA Bockenstedt LK Fu HC 2012 The heterogeneous motility of the Lyme disease spirochete in gelatin mimics dissemination through tissue Proceedings of the National Academy of Sciences of the United States of America 109 3059 3064 22315410 Hastey CJ Elsner RA Barthold SW Baumgarth N 2012 Delays and diversions mark the development of B cell responses to Borrelia burgdorferi infection Journal of immunology 188 5612 5622 Hazelbauer GL 2012 Bacterial chemotaxis: the early years of molecular studies Annual review of microbiology 66 285 303 He K Bauer CE 2014 Chemosensory signaling systems that control bacterial survival Trends in microbiology 22 389 398 24794732 Jarvelainen H Sainio A Koulu M Wight TN Penttinen R 2009 Extracellular matrix molecules: potential targets in pharmacotherapy Pharmacological reviews 61 198 223 19549927 Jewett MW Lawrence KA Bestor A Byram R Gherardini F Rosa PA 2009 GuaA and GuaB are essential for Borrelia burgdorferi survival in the tick-mouse infection cycle Journal of bacteriology 191 6231 6241 19666713 Jin H Subbarao K 2015 Live attenuated influenza vaccine Current topics in microbiology and immunology 386 181 204 25059893 Johnson RC Kodner C Russell M 1986 Active immunization of hamsters against experimental infection with Borrelia burgdorferi Infection and immunity 54 897 898 3781630 Kalish RA McHugh G Granquist J Shea B Ruthazer R Steere AC 2001 Persistence of immunoglobulin M or immunoglobulin G antibody responses to Borrelia burgdorferi 10–20 years after active Lyme disease Clinical infectious diseases: an official publication of the Infectious Diseases Society of America 33 780 785 11512082 Kato J Nakamura T Kuroda A Ohtake H 1999 Cloning and characterization of chemotaxis genes in Pseudomonas aeruginosa Bioscience, biotechnology, and biochemistry 63 155 161 Kern A Collin E Barthel C Michel C Jaulhac B Boulanger N 2011 Tick saliva represses innate immunity and cutaneous inflammation in a murine model of Lyme disease Vector borne and zoonotic diseases 11 1343 1350 21612525 Lane RS Piesman J Burgdorfer W 1991 Lyme borreliosis: relation of its causative agent to its vectors and hosts in North America and Europe Annual review of entomology 36 587 609 Leong JM Robbins D Rosenfeld L Lahiri B Parveen N 1998 Structural requirements for glycosaminoglycan recognition by the Lyme disease spirochete, Borrelia burgdorferi Infection and immunity 66 6045 6048 9826395 Lertsethtakarn P Ottemann KM Hendrixson DR 2011 Motility and chemotaxis in Campylobacter and Helicobacter Annual review of microbiology 65 389 410 Levine JF Wilson ML Spielman A 1985 Mice as reservoirs of the Lyme disease spirochete The American journal of tropical medicine and hygiene 34 355 360 3985277 Li C Bakker RG Motaleb MA Sartakova ML Cabello FC Charon NW 2002 Asymmetrical flagellar rotation in Borrelia burgdorferi nonchemotactic mutants Proceedings of the National Academy of Sciences of the United States of America 99 6169 6174 11983908 Lin T Gao L Zhang C Odeh E Jacobs MB Coutte L 2012 Analysis of an ordered, comprehensive STM mutant library in infectious Borrelia burgdorferi: insights into the genes required for mouse infectivity PloS one 7 e47532 23133514 Ma Y Seiler KP Eichwald EJ Weis JH Teuscher C Weis JJ 1998 Distinct characteristics of resistance to Borrelia burgdorferi-induced arthritis in C57BL/6N mice Infection and immunity 66 161 168 9423853 Mc KB 1948 B.C.G. vaccine in the prevention of tuberculosis Canadian Medical Association journal 58 575 577 18862255 McGee DJ Langford ML Watson EL Carter JE Chen YT Ottemann KM 2005 Colonization and inflammation deficiencies in Mongolian gerbils infected by Helicobacter pylori chemotaxis mutants Infection and immunity 73 1820 1827 15731083 Mead PS 2015 Epidemiology of Lyme disease Infectious disease clinics of North America 29 187 210 25999219 Mekalanos JJ 1994 Live bacterial vaccines: environmental aspects Current Opinion in Biotechnology 5 312 319 7765009 Moon KH Hobbs G Motaleb MA 2016 Borrelia burgdorferi CheD promotes various functions in chemotaxis and pathogenic life cycle of the spirochete Infection and immunity 84 6 Moriarty TJ Norman MU Colarusso P Bankhead T Kubes P Chaconas G 2008 Real-time high resolution 3D imaging of the lyme disease spirochete adhering to and escaping from the vasculature of a living host PLoS pathogens 4 e1000090 18566656 Morrison TB Ma Y Weis JH Weis JJ 1999 Rapid and sensitive quantification of Borrelia burgdorferi-infected mouse tissues by continuous fluorescent monitoring of PCR Journal of clinical microbiology 37 987 992 10074514 Motaleb MA Corum L Bono JL Elias AF Rosa P Samuels DS Charon NW 2000 Borrelia burgdorferi periplasmic flagella have both skeletal and motility functions Proceedings of the National Academy of Sciences of the United States of America 97 10899 10904 10995478 Motaleb MA Liu J Wooten RM 2015 Spirochetal motility and chemotaxis in the natural enzootic cycle and development of Lyme disease Current opinion in microbiology 28 106 113 26519910 Motaleb MA Miller MR Bakker RG Li C Charon NW 2007 Isolation and characterization of chemotaxis mutants of the Lyme disease Spirochete Borrelia burgdorferi using allelic exchange mutagenesis, flow cytometry, and cell tracking Methods in enzymology 422 421 437 17628152 Motaleb MA Miller MR Li C Bakker RG Goldstein SF Silversmith RE 2005 CheX is a phosphorylated CheY phosphatase essential for Borrelia burgdorferi chemotaxis Journal of bacteriology 187 7963 7969 16291669 Motaleb MA Pitzer JE Sultan SZ Liu J 2011a A novel gene inactivation system reveals altered periplasmic flagellar orientation in a Borrelia burgdorferi fliL mutant Journal of bacteriology 193 3324 3331 21441522 Motaleb MA Sultan SZ Miller MR Li C Charon NW 2011b CheY3 of Borrelia burgdorferi is the key response regulator essential for chemotaxis and forms a long-lived phosphorylated intermediate Journal of bacteriology 193 3332 3341 21531807 Mulay VB Caimano MJ Iyer R Dunham-Ems S Liveris D Petzke MM 2009 Borrelia burgdorferi bba74 is expressed exclusively during tick feeding and is regulated by both arthropod- and mammalian host-specific signals Journal of bacteriology 191 2783 2794 19218390 Nogueira SV Smith AA Qin JH Pal U 2012 A surface enolase participates in Borrelia burgdorferi-plasminogen interaction and contributes to pathogen survival within feeding ticks Infection and immunity 80 82 90 22025510 Norman MU Moriarty TJ Dresser AR Millen B Kubes P Chaconas G 2008 Molecular mechanisms involved in vascular interactions of the Lyme disease pathogen in a living host PLoS pathogens 4 e1000169 18833295 Novak EA Sultan SZ Motaleb MA 2014 The cyclic-di-GMP signaling pathway in the Lyme disease spirochete, Borrelia burgdorferi Frontiers in cellular and infection microbiology 4 56 24822172 Pal UaFE 2010 Tick Interactions Borrelia: molecular biology, host interaction, and pathogenesis Samuels DS Norfolk, United Kingdom Caister Academic Press 279 298 Parveen N Leong JM 2000 Identification of a candidate glycosaminoglycan-binding adhesin of the Lyme disease spirochete Borrelia burgdorferi Molecular microbiology 35 1220 1234 10712702 Patton TG Dietrich G Dolan MC Piesman J Carroll JA Gilmore RD Jr 2011 Functional analysis of the Borrelia burgdorferi bba64 gene product in murine infection via tick infestation PloS one 6 e19536 21559293 Pazy Y Motaleb MA Guarnieri MT Charon NW Zhao R Silversmith RE 2010 Identical phosphatase mechanisms achieved through distinct modes of binding phosphoprotein substrate Proceedings of the National Academy of Sciences of the United States of America 107 1924 1929 20080618 Pitzer JE Sultan SZ Hayakawa Y Hobbs G Miller MR Motaleb MA 2011 Analysis of the Borrelia burgdorferi cyclic-di-GMP-binding protein PlzA reveals a role in motility and virulence Infection and immunity 79 1815 1825 21357718 Policastro PF Schwan TG 2003 Experimental infection of Ixodes scapularis larvae (Acari: Ixodidae) by immersion in low passage cultures of Borrelia burgdorferi Journal of medical entomology 40 364 370 12943118 Radolf JD Caimano MJ Stevenson B Hu LT 2012 Of ticks, mice and men: understanding the dual-host lifestyle of Lyme disease spirochaetes Nature reviews. Microbiology 10 87 99 22230951 Ribeiro JM Mather TN Piesman J Spielman A 1987 Dissemination and salivary delivery of Lyme disease spirochetes in vector ticks (Acari: Ixodidae) Journal of medical entomology 24 201 205 3585913 Ribeiro JM Weis JJ Telford SR 1990 Saliva of the tick Ixodes dammini inhibits neutrophil function Experimental parasitology 70 382 388 2157607 Sadziene A Thompson PA Barbour AG 1996 A flagella-less mutant of Borrelia burgdorferi as a live attenuated vaccine in the murine model of Lyme disease The Journal of infectious diseases 173 1184 1193 8627071 Sarkar MK Paul K Blair D 2010 Chemotaxis signaling protein CheY binds to the rotor protein FliN to control the direction of flagellar rotation in Escherichia coli Proceedings of the National Academy of Sciences of the United States of America 107 9370 9375 20439729 Schwan TG Piesman J 2000 Temporal changes in outer surface proteins A and C of the lyme disease-associated spirochete, Borrelia burgdorferi, during the chain of infection in ticks and mice Journal of clinical microbiology 38 382 388 10618120 Skerker JM Prasol MS Perchuk BS Biondi EG Laub MT 2005 Two-component signal transduction pathways regulating growth and cell cycle progression in a bacterium: a system-level analysis PLoS biology 3 e334 16176121 Small CM Ajithdoss DK Rodrigues Hoffmann A Mwangi W Esteve-Gassent MD 2014 Immunization with a Borrelia burgdorferi BB0172-derived peptide protects mice against lyme disease PloS one 9 e88245 24505447 Smith JG Latiolais JA Guanga GP Citineni S Silversmith RE Bourret RB 2003 Investigation of the role of electrostatic charge in activation of the Escherichia coli response regulator CheY Journal of bacteriology 185 6385 6391 14563873 Stewart PE Bestor A Cullen JN Rosa PA 2008a A tightly regulated surface protein of Borrelia burgdorferi is not essential to the mouse-tick infectious cycle Infection and immunity 76 1970 1978 18332210 Stewart PE Rosa PA 2008b Transposon mutagenesis of the lyme disease agent Borrelia burgdorferi Methods in molecular biology 431 85 95 18287749 Sultan SZ Manne A Stewart PE Bestor A Rosa PA Charon NW Motaleb MA 2013 Motility is crucial for the infectious life cycle of Borrelia burgdorferi Infection and immunity 81 2012 2021 23529620 Sultan SZ Pitzer JE Boquoi T Hobbs G Miller MR Motaleb MA 2011 Analysis of the HD-GYP domain cyclic dimeric GMP phosphodiesterase reveals a role in motility and the enzootic life cycle of Borrelia burgdorferi Infection and immunity 79 3273 3283 21670168 Sultan SZ Pitzer JE Miller MR Motaleb MA 2010 Analysis of a Borrelia burgdorferi phosphodiesterase demonstrates a role for cyclic-di-guanosine monophosphate in motility and virulence Molecular microbiology 77 128 142 20444101 Sultan SZ Sekar P Zhao X Manne A Liu J Wooten RM Motaleb MA 2015 Motor rotation is essential for the formation of the periplasmic flagellar ribbon, cellular morphology, and Borrelia burgdorferi persistence within Ixodes scapularis tick and murine hosts Infection and immunity 83 1765 1777 25690096 Sze CW Zhang K Kariu T Pal U Li C 2012 Borrelia burgdorferi needs chemotaxis to establish infection in mammals and to accomplish its enzootic cycle Infection and immunity 80 2485 2492 22508862 Szurmant H Ordal GW 2004 Diversity in chemotaxis mechanisms among the bacteria and archaea Microbiology and molecular biology reviews: MMBR 68 301 319 15187186 Toker AS Macnab RM 1997 Distinct regions of bacterial flagellar switch protein FliM interact with FliG, FliN and CheY Journal of molecular biology 273 623 634 9356251 Tsao JI 2009 Reviewing molecular adaptations of Lyme borreliosis spirochetes in the context of reproductive fitness in natural transmission cycles Veterinary research 40 36 19368764 Wadhams GH Armitage JP 2004 Making sense of it all: bacterial chemotaxis Nature reviews. Molecular cell biology 5 1024 1037 15573139 Weis JJ McCracken BA Ma Y Fairbairn D Roper RJ Morrison TB 1999 Identification of quantitative trait loci governing arthritis severity and humoral responses in the murine model of Lyme disease Journal of immunology 162 948 956 Welch M Oosawa K Aizawa S Eisenbach M 1993 Phosphorylation-dependent binding of a signal molecule to the flagellar switch of bacteria Proceedings of the National Academy of Sciences of the United States of America 90 8787 8791 8415608 Whitchurch CB Leech AJ Young MD Kennedy D Sargent JL Bertrand JJ 2004 Characterization of a complex chemosensory signal transduction system which controls twitching motility in Pseudomonas aeruginosa Molecular microbiology 52 873 893 15101991 Wolanin PM Thomason PA Stock JB 2002 Histidine protein kinases: key signal transducers outside the animal kingdom Genome biology 3 REVIEWS3013 12372152 Wolgemuth CW 2015 Flagellar motility of the pathogenic spirochetes Seminars in cell & developmental biology 46 104 112 26481969 Yang XF Pal U Alani SM Fikrig E Norgard MV 2004 Essential role for OspA/B in the life cycle of the Lyme disease spirochete The Journal of experimental medicine 199 641 648 14981112 Zhang K Liu J Tu Y Xu H Charon NW Li C 2012 Two CheW coupling proteins are essential in a chemosensory pathway of Borrelia burgdorferi Molecular microbiology 85 782 794 22780444 Zhang X Yang X Kumar M Pal U 2009 BB0323 function is essential for Borrelia burgdorferi virulence and persistence through tick-rodent transmission cycle The Journal of infectious diseases 200 1318 1330 19754308

PMC005xxxxxx/PMC5116912.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0370512 457 Am J Psychiatry Am J Psychiatry The American journal of psychiatry 0002-953X 1535-7228 26046337 5116912 10.1176/appi.ajp.2015.14091185 NIHMS830486 Article Pediatric-Onset and Adult-Onset Separation Anxiety Disorder Across Countries in the World Mental Health Survey Silove Derrick M.D., Ph.D. Alonso Jordi M.D., Ph.D. Bromet Evelyn Ph.D. Gruber Mike M.S. Sampson Nancy B.A. Scott Kate Ph.D. Andrade Laura M.D., Ph.D. Benjet Corina Ph.D. de Almeida Jose Miguel Caldas M.D., Ph.D. De Girolamo Giovanni M.D. de Jonge Peter Ph.D. Demyttenaere Koen M.D., Ph.D. Fiestas Fabian M.D. Florescu Silvia M.D., Ph.D. Gureje Oye Ph.D. He Yanling M.D. Karam Elie M.D. Lepine Jean-Pierre Ph.D. Murphy Sam Ph.D. Villa-Posada Jose M.D. Zarkov Zahari M.D. Kessler Ronald C. Ph.D. Psychiatry Research and Teaching Unit and Ingham Institute, School of Psychiatry, University of New South Wales, Randwick, NSW, Australia; IMIM-Hospital del Mar Research Institute, Parc de Salut Mar, Pompeu Fabra University (UPF) and CIBER en Epidemiología y Salud Pública (CIBERESP), Barcelona, Spain; Department of Psychiatry, Stony Brook University School of Medicine, Stony Brook, N.Y..; Department of Health Care Policy, Harvard Medical School, Boston; Department of Psychological Medicine, University of Otago, Otago, New Zealand; Department/Institute of Psychiatry, University of São Paulo Medical School, São Paulo, Brazil; Department of Epidemiologic and Psychosocial Research, National Institute of Psychiatry Ramón de la Fuente, Mexico City, Mexico; Chronic Diseases Research Center (CEDOC) and Department of Mental Health, Faculdade de Ciências Médicas, Universidade Nova de Lisboa, Lisbon, Portgual; IRCCS St. John of God Clinical Research Centre and IRCCS Centro S. Giovanni di Dio Fatebenefratelli, Brescia, Italy; University of Groningen, University Medical Center Groningen, Groningen, the Netherlands; Department of Psychiatry, University Hospital Gasthuisberg, Katholieke Universiteit Leuven, Leuven, Belgium; Unit of Analysis and Generation of Evidence for Public Health, Peruvian National Institutes of Health, Lima, Peru; National School of Public Health, Management and Professional Development, Bucharest, Romania; Department of Psychiatry, University College Hospital, Ibadan, Nigeria; Shanghai Mental Health Center, Shanghai, the People’s Republic of China; Department of Psychiatry and Clinical Psychology, Faculty of Medicine, Balamand University, Beirut, Lebanon; Department of Psychiatry and Clinical Psychology, St. George Hospital University Medical Center, Beirut, Lebanon; Institute for Development Research Advocacy and Applied Care (IDRAAC), Beirut, Lebanon; Hôpital Lariboisière Fernand Widal, Assistance Publique Hôpitaux de Paris, University Paris Diderot and Paris Descartes Paris, Paris; School of Psychology, University of Ulster, Northern Ireland; Colegio Mayor de Cundinamarva University, Bogota, Colombia; National Center of Public Health and Analyses, Sofia, Bulgaria. Address correspondence to Dr. Silove (d.silove@unsw.edu.au) 17 11 2016 05 6 2015 7 2015 21 11 2016 172 7 647656 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objective The age-at-onset criterion for separation anxiety disorder was removed in DSM-5, making it timely to examine the epidemiology of separation anxiety disorder as a disorder with onsets spanning the life course, using cross-country data. Method The sample included 38,993 adults in 18 countries in the World Health Organization (WHO) World Mental Health Surveys. The WHO Composite International Diagnostic Interview was used to assess a range of DSM-IV disorders that included an expanded definition of separation anxiety disorder allowing onsets in adulthood. Analyses focused on prevalence, age at onset, comorbidity, predictors of onset and persistence, and separation anxiety-related role impairment. Results Lifetime separation anxiety disorder prevalence averaged 4.8% across countries (interquartile range [25th–75th percentiles]=1.4%–6.4%), with 43.1% of lifetime onsets occurring after age 18. Significant time-lagged associations were found between earlier separation anxiety disorder and subsequent onset of internalizing and externalizing DSM-IV disorders and conversely between these disorders and subsequent onset of separation anxiety disorder. Other consistently significant predictors of lifetime separation anxiety disorder included female gender, retrospectively reported childhood adversities, and lifetime traumatic events. These predictors were largely comparable for separation anxiety disorder onsets in childhood, adolescence, and adulthood and across country income groups. Twelve-month separation anxiety disorder prevalence was considerably lower than lifetime prevalence (1.0% of the total sample; interquartile range=0.2%–1.2%). Severe separation anxiety-related 12-month role impairment was significantly more common in the presence (42.4%) than absence (18.3%) of 12-month comorbidity. Conclusions Separation anxiety disorder is a common and highly comorbid disorder that can have onset across the lifespan. Childhood adversity and lifetime trauma are important antecedents, and adverse effects on role function make it a significant target for treatment. Although separation anxiety disorder traditionally has been a diagnosis assigned to children and adolescents, DSM-5 removed the 18-year age-at-onset restriction on diagnosis because studies had found later onset to be common. It is timely therefore to examine key aspects of the epidemiology of separation anxiety disorder as a condition with onsets that span the life course. The importance of adult-onset separation anxiety disorder is indicated by the fact that 20%–40% of adult patients with mood and anxiety disorders have been found to have symptoms of the disorder, and between one-third and one-half of these patients reported onsets after 18 years of age (1–3). Patients with adult separation anxiety disorder experience high levels of functional impairment and show a poor response to conventional treatments used for other anxiety subtypes (4). There is a dearth of epidemiologic data focusing on separation anxiety disorder across the lifespan. A longitudinal study commencing in childhood recorded a 5% lifetime prevalence of separation anxiety disorder by the time the cohort reached early adulthood (5). The National Comorbidity Survey Replication in the United States found a lifetime separation anxiety disorder prevalence of 6.6% after the pediatric age-at-onset requirement was removed, with two-thirds of case subjects having onsets after 17 years of age (6), but comparable data in other samples or populations have yet to be reported. The relationship of separation anxiety disorder with other mental disorders also remains to be clarified. A longstanding theory has posited a specific developmental relationship between childhood-onset separation anxiety disorder and adult panic disorder and/or agoraphobia (7, 8), an association supported by findings from twin and laboratory studies (9, 10). Yet a meta-analysis of community studies has indicated that separation anxiety disorder may represent a generic risk factor for a range of anxiety disorders and other forms of psychopathology in adulthood (11) rather than primarily for panic disorder and/or agoraphobia. Consistent with the principles of attachment theory (12), a recent developmental model has suggested that separation anxiety symptoms may mediate the associations between early family adversity and trauma and subsequent onset of common adult mental disorders (13). Family dysfunction and exposure to major disasters appear to be associated with subsequent onset of separation anxiety in childhood (13–15). Overall, however, data are limited concerning associations involving early adversity, exposure to trauma, onset of separation anxiety disorder, and a range of later psychopathologic outcomes. In the present study, we analyzed data from the cross-national World Health Organization (WHO) World Mental Health surveys (16) to assess the following key aspects of separation anxiety disorder in the general population: 1) lifetime and 12-month prevalence overall and by gender and country income grouping; 2) the proportions of case subjects experiencing childhood and adult onsets; 3) patterns of comorbidity and temporal relationships with onset and persistence of other common DSM-IV disorders; 4) associations of other predictors (sociodemographic variables, childhood adversities, lifetime traumatic events) with separation anxiety disorder onset and persistence; and 5) severity of role impairment associated with 12-month separation anxiety disorder. METHOD Samples Separation anxiety disorder was assessed in 18 World Mental Health surveys: nine in high-income countries (Belgium, France, Germany, Italy, the Netherlands, Northern Ireland, Portugal, Spain, and the United States), five in upper-middle income countries (Brazil, Bulgaria, Lebanon, Mexico, and Romania), and four in low-/lower-middle income countries (Colombia, Nigeria, the People’s Republic of China, and Peru). A total of 38,993 respondents were assessed. All surveys used probability sampling based on multistage area clustered household survey designs with no substitution for nonparticipants. The majority of surveys were based on nationally representative samples, the remainder on samples representative of particular urban areas (Sao Paulo in Brazil, Beijing and Shanghai in the People’s Republic of China), all nonrural areas in the country (Colombia, Mexico), or major regions of the country (Nigeria). Response rates ranged from 45.9% to 90.2% and averaged 70.3%. More details on the World Mental Health Survey sampling are presented elsewhere (17). The World Mental Health Survey interview was administered in two parts to reduce respondent burden. All respondents completed part I, which assessed core disorders. All respondents with a part I disorder plus a probability subsample of other part I respondents were administered part II, which assessed additional disorders and correlates. The part I data were weighted for differential probabilities of selection and to match population distributions on socio-demographic and geographic variables. The part II data were additionally weighted to adjust for differential probabilities of selection from part I into part II. Separation anxiety disorder was typically assessed in part II but in some countries was included in part I. In one-half of the surveys, the assessment of separation anxiety disorder was restricted to respondents in the age range 18–39 or 18–44, while in the other surveys, there was no such age restriction. Measures Overview World Mental Health Survey interviews were conducted face-to-face by lay interviewers. Consistent translation, back-translation, and harmonization procedures were used in adapting the interview for local administration (18). Consistent interviewer training and field quality-control procedures were applied across sites (19). All respondents provided informed consent according to the requirements of local institutional review boards before being interviewed. Diagnostic assessment The diagnostic interview was the WHO Composite International Diagnostic Interview (CIDI) (20), a fully-structured interview that assessed lifetime and 12-month prevalence of DSM-IV mood (major depressive disorder and/or dysthymia, bipolar disorder), anxiety (panic disorder and/or agoraphobia, specific phobia, social phobia, generalized anxiety disorder, posttraumatic stress disorder [PTSD]), and externalizing (intermittent explosive disorder, alcohol and drug abuse with or without dependence) disorders. A special probing strategy shown to yield improved age-at-onset reports of individual disorders was used (21). A blinded clinical reappraisal study using the Structured Clinical Interview for DSM-IV (SCID) (22) as the gold standard found good diagnostic concordance between core CIDI and SCID diagnoses, although the module for separation anxiety disorder was not included (23). The CIDI module for separation anxiety disorder departed from DSM-IV in assessing lifetime onset of symptoms not only as of age 18 but also for onsets that occurred at ages 19 or later (6). A separation anxiety disorder diagnosis required endorsement of at least three of the eight criterion A DSM-IV symptoms, having symptoms for at least 1 month, and experiencing associated clinically significant distress or role impairment. Sociodemographic variables Sociodemographic variables considered here include age at interview (18–34, 35–49, 50–64, and ≥65 years), gender, education (student, and among nonstudents, low, low-average, average-high, and high-level of education based on country-specific distributions), and marital status. Functional Impairment A modified version of the Sheehan Disability Scales (24) was used to assess severity of role impairment associated with separation anxiety disorder in the previous year. Respondents were asked to quantify severity of role impairment in home management, work, social life, and personal relationships using a 0–10 self-anchoring scale with the following response categories: none (0), mild (1–3), moderate (4–6), severe (7–9), and very severe (10). Respondents rated the Sheehan Disability Scales for the month in the previous year when separation anxiety disorder was most severe. Scores for the Sheehan Disability Scales were dichotomized into severe (range 7–10 for any domain of the scales) or not severe (scores lower than 7–10 for all domains). Childhood family adversities World Mental Health Survey respondents were asked retrospectively about exposure to a wide range of childhood family adversities. As reported previously (25), exploratory factor analysis of responses found one factor for experiences indicative of childhood maladaptive family functioning (parental mental illness, substance misuse, criminal behavior, domestic violence, and child physical abuse, sexual abuse, and neglect), while other childhood adversities were combined into a second scale that included parental death, parental divorce, other long separations from a parent, serious illness of a close family member, and family economic adversity. Traumatic events The CIDI assessed 29 lifetime traumatic events aggregated into the seven domains of accidents and natural disasters, war events, intimate and sexual violence, death of a loved one, other interpersonal violence, network events, and other traumas participants decided not to disclose (26). Dichotomous measures were created for one or more events in each of these domains. Analysis Procedures Cross-tabulations were used to estimate lifetime and 12-month separation anxiety disorder prevalence separately for men and women in each survey. Age-at-onset reports were analyzed using the two-part actuarial method to estimate survival curves (27). The proportion of lifetime cases with onsets after age 18 was calculated for each country and country income group. Discrete-time survival analysis with person-year as the unit of analysis and a logistic link function (28) was used to examine cross-lagged associations of temporally primary separation anxiety disorder with subsequent first onset of other disorders and reciprocal associations of other temporally primary disorders with subsequent first onset of separation anxiety disorder. Survival coefficients and standard errors were exponentiated to generate odds-ratios with 95% confidence intervals. The same survival analysis approach was used to estimate associations of sociodemographic variables, childhood adversities, and lifetime traumatic events (28) with first onset of separation anxiety disorder and to examine variation in strength of prediction as a function of age at onset (childhood [up through age 12], adolescence [ages 13–17], early adulthood [ages 18–29] and later onsets [ages ≥30]). The associations of the same predictors with persistence of separation anxiety disorder, defined as 12-month prevalence among lifetime cases, were examined using person-level logistic regression analysis controlling for age at onset and time since onset. Finally, cross-tabulations were used to examine joint associations of country income level, separation anxiety disorder age at onset, and 12-month comorbidities with 12-month severe separation anxiety-related role impairments. Relative fit of additive and interactive models was evaluated using the Akaike information criterion and Bayesian information criterion (29). Standard errors of estimates were based on the Taylor series linearization method implemented with SUDAAN software (30, 31) to adjust for the weighting and geographic clustering of World Mental Health data. Multivariate significance tests were carried out with Wald chi-square tests based on Taylor series coefficient variance-covariance matrices. Statistical significance was evaluated using 0.05-level two-sided tests. RESULTS Prevalence Lifetime prevalence of the expanded CIDI definition of DSM-IV separation anxiety disorder that allowed adult onsets was 4.8% in the total sample but with a much wider interquartile range (25th–75th percentiles: 1.4–6.4) and range (0.2%–9.8%) across countries than previously found for most other DSM-IV/CIDI disorders (32) (Table 1). Lifetime prevalence was higher among women than men in 15 of the 18 countries and significantly so in the total sample (5.6% compared with 4.0%; χ2=4.0, df=1, p<0.05). Twelve-month prevalence was considerably lower than lifetime prevalence (1.0% in the total sample; interquartile range: 0.2%–1.2%; range: 0.0%–2.7%), higher among women than men in 14 of 18 countries and significantly higher among women than men in the total sample (1.3% compared with 0.8%; χ2=12.0, df=1, p<0.001). Age at Onset The age-at-onset distribution of separation anxiety disorder was quite similar across the three country income groups (Figure 1). Median age at onset was in the late teens in high- and upper-middle income countries and in the mid-20s in low-/lower-middle income countries. The interquartile range of the age-at-onset distribution was wider (15–35 years of age in low-/lower-middle income countries; 9–35 in high- and upper-middle income countries) than previously found for most other DSM-IV/CIDI disorders (32). A total of 43.1% of respondents with lifetime separation anxiety disorder had onsets in adulthood (ages ≥18). The proportion of lifetime cases with adult onsets was significantly higher in low-/lower-middle income countries (53.8%) than in upper-middle income countries (39.1%; χ2=98.0, df=1, p<0.001) or high- income countries (41.6%; χ2=52.8, df=1, p<0.001). Although there were only two low/low-middle income countries with sizable numbers of lifetime separation anxiety disorder cases (N=359 in Colombia; N=154 in Peru), the proportion of these cases with adult onsets was comparable in the two surveys (55.7% compared with 51.8%; χ2=1.5, df=1, p=0.22). Persistence Comparison of individual-level age at onset and recency reports showed that the majority of people with lifetime separation anxiety disorder remitted within a decade of onset (Figure 2). However, the recovery curves became much less steep after approximately 10 years. As with age at onset, the distribution of time to remission was quite consistent across country income groups. This relatively rapid remission of separation anxiety disorder is consistent with the low 12-month/lifetime prevalence ratio shown in Table 2. Comorbidity Lifetime separation anxiety disorder was significantly comorbid with 13 of the other 14 DSM-IV/CIDI disorders assessed in the World Mental Health surveys, the exception being substance dependence with abuse. Survival analysis pooled across countries showed that these associations were a result of 1) significant associations between temporally primary separation anxiety disorder and subsequent first onset of 10 other disorders (odds ratios in the range of 1.3–2.8) coupled with 2) significant associations of nine other disorders with subsequent first onset of separation anxiety disorder (odds ratios in the range of 1.3–2.1) (Table 2). The vast majority of cross-lagged odds ratio pairs between separation anxiety disorder and other disorders were reciprocal in significance and comparable in magnitude. The major exceptions were that temporally primary separation anxiety disorder was a significantly stronger predictor of subsequent attention deficit hyperactivity disorder (ADHD) (odds ratio=2.8) than ADHD was of subsequent separation anxiety disorder (odds ratio=1.1) and that there were no significant associations between temporally primary PTSD, generalized anxiety disorder, or agoraphobia and subsequent-onset separation anxiety disorder in contrast to the reciprocal relationships. Disaggregation by country income group (results available upon request from Silove) showed considerable consistency in these patterns, with the vast majority of significant odds ratios in the total sample also significant in at least two of the three country income groups. The associations of other temporally primary disorders with subsequent separation anxiety disorder persistence (i.e., 12-month prevalence among lifetime cases controlling for age at onset and time since onset) were not significant. The same was generally true for the associations of temporally primary separation anxiety disorder with subsequent persistence of other disorders, with the exceptions being that temporally primary separation anxiety disorder was a predictor of persistence of agoraphobia (odds ratio=2.6), and to a lesser extent, PTSD. As with the analyses of first onset, disaggregation by country income group (results available upon request from Silove) showed consistency in the nonsignificance of these reciprocal associations between lifetime comorbidity and persistence. Other Predictors of Separation Anxiety Disorder Onset and Persistence Controlling for lifetime comorbid DSM-IV/CIDI disorders, age, and country, survival analysis showed that lifetime separation anxiety disorder was significantly associated with being female (odds ratio=1.1–1.4), having low through high-average (compared with high) education (odds ratio=1.5–1.7), maladaptive family functioning childhood adversities (odds ratio=1.7–2.8), other childhood adversities (odds ratio=1.2–1.8), and a variety of lifetime traumatic events (odds ratio=1.3–1.6) (Table 3). It is noteworthy that the associations involving maladaptive family functioning childhood adversities and other lifetime traumatic events predicted not only pediatric-onset but also adult-onset separation anxiety disorder, while other childhood adversities predicted only childhood-onset separation anxiety disorder. Disaggregation by country income group (detailed results available upon request from Silove) showed considerable consistency in these patterns, with the vast majority of significant odds ratios in the total sample also significant in at least two of the three country income groups. As with the analysis of comorbidity, the predictors considered here were not significantly related to separation anxiety disorder persistence either in the total sample or in country income groups (detailed results available upon request from Silove). For example, the associations of maladaptive family functioning childhood adversities with separation anxiety disorder persistence were nonsignificant both in the total sample (χ2=2.5, df=5, p=0.78) and in each of the three country income groups (χ2=2.9–10.2, df=5, p=0.07–0.72). In addition, we found that age at onset (defined in categories of childhood <13 years old, adolescence 13–17 years old, young adulthood 18–29 years old, and later adulthood ≥30 years old) was not significantly related to persistence either in the total sample (χ2=4.3, df=3, p=0.23) or in any of the three country income groups (χ2=3.0–5.4, df=3, p=0.15–0.39). Role Impairment of 12-Month Separation Anxiety Disorder More than one-third (35.3%) of respondents with 12-month separation anxiety disorder reported severe separation anxiety-related role impairment in the year before interview (Table 4). The rate of severe role impairment was considerably lower in low-/lower-middle income countries (17.6%) than upper-middle or high-income countries (38.2%–41.4%; χ2=12.6–18.1, df=1, p<0.001) and considerably higher in the presence than absence of 12-month comorbid DSM-IV/CIDI disorders (42.4% compared with 18.3%; χ2=21.6, df=1, p<0.001). Separation anxiety-related role impairment did not differ markedly as a function of separation anxiety disorder age at onset (χ2=2.9, df=3, p=0.40). A model that assumed additive effects of country income level, comorbidity, and age at onset predicted severe separation anxiety-related role impairment with no evidence of significant interaction between the factors. DISCUSSION The findings concerning the prevalence of separation anxiety disorder across countries are summarized in Figure 3. The 4.8% overall lifetime prevalence estimate found in this study suggests that separation anxiety disorder is a relatively common lifetime disorder, although with a much wider range of prevalence across countries (0.2%–9.8%; interquartile range: 1.4–6.4) than found for most other DSM-IV/CIDI disorders assessed in the World Mental Health Surveys (32). Temporally prior separation anxiety disorder was found to be associated with significantly elevated odds of subsequent first onset of a wide range of other disorders, including not only internalizing disorders (major depression, bipolar disorder, specific and social phobias, panic disorder, and/or generalized anxiety disorder) but also externalizing disorders (ADHD, oppositional defiant disorder, and conduct disorder). This finding is consistent with a recent meta-analysis concluding that separation anxiety disorder represents a generic risk factor for a range of common mental disorders (11). Importantly, therefore, our data do not support the hypothesis of an exclusive link between separation anxiety disorder and panic disorder and/or agoraphobia (7, 8), although the existence of a significant association between temporally primary separation anxiety disorder and subsequent persistence of agoraphobia has potential prognostic importance in relation to the latter disorder. We also found reciprocal associations of a similarly wide range of temporally primary disorders with the subsequent onset of separation anxiety disorder, a result that is broadly consistent with the suggestion that these comorbidities are due to common causes more than to direct effects of particular early-onset disorders on particular later-onset disorders (33). Indeed, an earlier World Mental Health Survey analysis found that separation anxiety disorder had a high factor loading on the internalizing factor of a two-dimensional model and that controls for summary internalizing-externalizing dimensions accounted for the significant cross-lagged associations of separation anxiety disorder with most comorbid disorders (34). Questions therefore remain whether separation anxiety disorder has any specificity as a risk factor for onset of particular secondary disorders either in childhood or adulthood (13). Maladaptive family functioning childhood adversities and exposure to traumatic life events were found to be associated with separation anxiety disorder onset but not persistence, both in the total sample and, separately, in low-/lower-middle, upper-middle, and high-income countries. Importantly, these associations were found for separation anxiety disorder onsets across the entire life course. It is possible that more in-depth future analyses will find significant specifications in the associations of particular types of childhood adversities or traumatic events at specific life course stages with subsequent separation anxiety disorder onset. Nevertheless, in such analyses, it will be important to account for more general associations that may exist involving a wide range of childhood adversities and traumatic events as a backdrop against which more specific specifications are considered. The mechanisms responsible for the ongoing liability to separation anxiety disorder and other mental disorders arising from early adversity and trauma, including the possible neurobiological mediators of these effects, are the focus of ongoing inquiry (35, 36). Our findings suggest that separation anxiety disorder is more likely to be seriously impairing in higher-income than low-/lower-middle income countries. The explanation of this specification is unclear, although it is possible that a culture of individuality and independence in higher-income countries results in low personal and social tolerance of separation anxiety, whereas a collectivist culture in low-/lower-middle income countries accepts a degree of separation anxiety as normative or even adaptive. A related unanswered question is why the range of prevalence across countries is so large. Limitations of the study include variation in response rates across countries, use of a fully-structured diagnostic interview that did not allow clinical probing to confirm diagnoses, and a cross-sectional design in which lifetime prevalence, childhood adversities, and lifetime traumatic events were assessed retrospectively. Although we used the DSM-IV 1-month duration requirement rather than the DSM-5 6-month recommended (although not mandated) duration criterion for adults in defining separation anxiety disorder, overestimation of prevalence in relation to the latter diagnostic system is likely to be slight given that the median persistence of disorder was 4–8 years. Recall bias is a more serious concern, since this might have led to an underestimation of lifetime prevalence among remitted cases and an overestimation of persistence, even though the World Mental Health Surveys used special probing strategies designed to improve recall accuracy (21) and in some countries limited duration of recall by assessing separation anxiety disorder only among respondents younger than ages 40–45. If biases exist, however, this means that separation anxiety disorder is an even more prevalent lifetime disorder with an even lower persistence than suggested by the present results. Retrospectively reported age at onset was unrelated to 12-month persistence among lifetime cases, suggesting that there was no tendency for respondents to overstate onset of separation anxiety disorder in adulthood. The high prevalence of adult-onset cases therefore supports the decision to remove the separation anxiety disorder age-at-onset restriction in DSM-5. An issue worthy of further investigation is that the ratio of 12-month/lifetime separation anxiety disorder prevalence is much lower than that of most other World Mental Health disorders (1.0% 12-month prevalence; range: 0.0%–2.7%; interquartile range: 0.2%–1.2%). Our findings have important implications for clinical practice. The results challenge the long-established view that separation anxiety disorder should be reserved for diagnosis among children and adolescents by indicating that adult onset is prevalent across countries and that adult-onset separation anxiety disorder is equally persistent and impairing as the pediatric-onset form of this disorder. Clinicians should be alerted to the need to consider separation anxiety disorder in the differential diagnosis of patients of all ages presenting not only with anxiety but with a wide variety of internalizing and externalizing disorders. As yet, existing psychological and pharmacological treatments used for the other anxiety disorders have proven to be ineffective for adult separation anxiety disorder (4, 37). There is consequently an urgent need to devise and test novel treatments for this disorder, particularly as it manifests in adulthood. The World Health Organization (WHO) World Mental Health Survey Initiative is supported by NIMH (R01 MH070884, R13 MH066849, and R01 MH069864), the National Institute on Drug Abuse (R01 DA016558), the John D. and Catherine T. MacArthur Foundation, the Pfizer Foundation, the Fogarty International Center (FIRCA R03-TW006481), the Pan American Health Organization, Eli Lilly, Ortho-McNeil Pharmaceutical, GlaxoSmithKline, and Bristol-Myers Squibb. None of these funders had any role in the design, analysis, interpretation of results, or preparation of this article. (A complete list of World Mental Health publications is available online [http://www.hcp.med.harvard.edu/wmh/].) The 2007 Australian National Survey of Mental Health and Wellbeing was funded by the Australian Government Department of Health and Ageing. The São Paulo Megacity Mental Health Survey is supported by the State of São Paulo Research Foundation Thematic Project (grant 03/00204-3). The Bulgarian Epidemiological Study of common mental disorders is supported by the Ministry of Health and the National Center for Public Health Protection. The Colombian National Study of Mental Health is supported by the Ministry of Social Protection. The ESEMeD project is funded by the European Commission (contracts QLG5-1999-01042, SANCO 2004123, and EAHC 20081308), the Piedmont Region (Italy), Fondo de Investigación Sanitaria, Instituto de Salud Carlos III, Spain (FIS 00/0028), Ministerio de Ciencia y Tecnología, Spain (SAF 2000-158-CE), Departament de Salut, Generalitat de Catalunya, Spain, Instituto de Salud Carlos III (CIBER CB06/02/0046, RETICS RD06/0011 REM-TAP), and other local agencies and by an unrestricted educational grant from GlaxoSmithKline. The Lebanese National Mental Health Survey (L.E.B.A.N.O.N.) is supported by the Lebanese Ministry of Public Health, WHO (Lebanon), National Institutes of Health/Fogarty International Center (R03 TW006481-01), Sheikh Hamdan Bin Rashid Al Maktoum Award for Medical Sciences, anonymous private donations to IDRAAC, Lebanon, and research grants from AstraZeneca, Eli Lilly, GlaxoSmithKline, Lundbeck, Novartis, Roche, and Servier. The Nigerian Survey of Mental Health and Wellbeing is supported by WHO (Geneva), WHO (Nigeria), and the Federal Ministry of Health, Abuja, Nigeria. The Northern Ireland Study of Mental Health was funded by the Health and Social Care Research and Development Division of the Public Health Agency. The Chinese World Mental Health Survey Initiative is supported by the Pfizer Foundation. The Portuguese Mental Health Study was carried out by the Department of Mental Health, Faculty of Medical Sciences, NOVA University of Lisbon, with collaboration of the Portuguese Catholic University, and was funded by the Champalimaud Foundation, the Gulbenkian Foundation, the Foundation for Science and Technology and Ministry of Health. The Romania World Mental Health study projects “Policies in Mental Health Area” and “National Study regarding Mental Health and Services Use” were carried out by the National School of Public Health and Health Services Management (former National Institute for Research and Development in Health, present National School of Public Health Management and Professional Development, Bucharest), with technical support from Metro Media Transylvania, the National Institute of Statistics-National Centre for Training in Statistics, SC, Cheyenne Services SRL, Statistics Netherlands and were funded by the Ministry of Public Health (former Ministry of Health), with supplemental support from Eli Lilly Romania SRL. The U.S. National Comorbidity Survey Replication is supported by NIMH (U01-MH60220), with supplemental support from the National Institute of Drug Abuse, the Substance Abuse and Mental Health Services Administration, the Robert Wood Johnson Foundation (grant 044708), and the John W. Alden Trust. The de-identified survey data are stored and were analyzed on secure servers at the World Mental Health Data Analysis Coordination Center at Harvard Medical School, Boston. The authors thank Herbert Matschinger for contributions to the World Mental Health Surveys. Dr. Silove receives royalties from Little, Brown Book Group and Oxford University Press; and he has served as a consultant for Counterpart International. Dr. Demyttenaere serves on the speaker’s bureaus and/or advisory panels of AstraZeneca, Eli Lilly, Johnson and Johnson, Lundbeck, Neurex, Servier, Shire, and Takeda and has also received research grants from Eli Lilly and Fonds voor Wetenschappelijk onderzoek Vlaanderen. Dr. Fiestas is an employee of the Peruvian National Institutes of Health. Dr. Kessler has served as a consultant for Hoffmann-La Roche and Johnson and Johnson Wellness and Prevention; he has also served on advisory boards for Johnson and Johnson Services Lake Nona Life Project, the Mensante Corporation, and U.S. Preventive Medicine; and he is a shareholder with DataStat. FIGURE 1 Age at Onset for Respondents With Separation Anxiety Disorder by Country Income FIGURE 2 Speed of Recovery From Separation Anxiety Disorder by Country Income FIGURE 3 Lifetime and 12-Month Prevalence of Separation Anxiety Disorder a Significant difference between males and females for lifetime prevalence (χ2=32.0, p<0.001) and 12-month prevalence (χ2=8.8, p=0.003). TABLE 1 Lifetime and 12-Month Prevalence of Separation Anxiety Disorder Stratified by Gender and Country Income Lifetime Prevalence 12-Month Prevalence Total Male Female Total Male Female Sample Size (N) Country and Country Income % SE % SE % SE % SE % SE % SE Male Female All countries 4.8 0.1 4.0 0.2 5.6* 0.2 1.0 0.1 0.8 0.1 1.3* 0.1 16,869 22,124 Low-/lower-middle income 5.5 0.4 4.5 0.4 6.4* 0.5 1.3 0.2 0.9 0.2 1.7 0.4 2,622 3,472 Colombia 9.8 0.8 8.1 1.1 11.3* 1.2 2.7 0.5 1.9 0.5 3.3 0.9 885 1,496 Nigeria 0.2 0.1 0.2 0.1 0.2 0.2 0.0 0.0 0.0 0.0 0.0 0.0 614 675 People’s Republic of China-Beijing/ Shanghai 1.3 0.5 0.7 0.4 2.1 1.0 0.2 0.1 0.0 0.0 0.4 0.2 326 297 Peru 6.1 0.7 5.6 0.8 6.6 0.9 1.2 0.2 0.9 0.2 1.4 0.3 797 1,004 Upper-middle income 4.7 0.2 4.0 0.3 5.3* 0.3 1.2 0.1 1.0 0.2 1.4 0.2 5,334 7,271 Brazil 7.7 0.4 6.7 0.6 8.6* 0.6 2.0 0.3 1.7 0.4 2.3 0.3 2,187 2,850 Bulgaria 1.4 0.4 1.5 0.8 1.2 0.3 0.4 0.3 0.6 0.5 0.2 0.1 951 1,282 Lebanon 6.9 1.3 4.9 1.5 9.0 1.8 1.9 0.5 1.1 0.5 2.7 0.8 251 365 Mexico 4.5 0.4 3.7 0.8 5.2 0.5 0.9 0.2 0.7 0.2 1.1 0.2 853 1,509 Romania 0.9 0.3 0.9 0.4 0.9 0.3 0.3 0.1 0.2 0.1 0.3 0.2 1,092 1,265 High-income 4.7 0.2 3.7 0.2 5.6* 0.3 0.9 0.1 0.7 0.1 1.0 0.1 8,913 11,381 Belgium 1.4 0.3 0.7 0.3 2.1* 0.4 0.1 0.1 0.0 0.0 0.3 0.2 599 591 France 3.5 0.5 3.1 0.7 3.9 1.0 0.9 0.3 0.4 0.2 1.4 0.7 692 772 Germany 2.0 0.4 1.5 0.4 2.6 0.6 0.4 0.2 0.5 0.3 0.4 0.3 806 912 Italy 1.5 0.4 0.6 0.3 2.3* 0.5 0.0 0.0 0.0 0.0 0.1 0.1 1,171 1,209 The Netherlands 3.0 0.6 2.0 0.8 4.0 1.1 0.6 0.3 0.5 0.3 0.7 0.6 472 637 Northern Ireland 5.1 0.5 5.1 0.8 5.2 0.6 0.4 0.1 0.5 0.2 0.4 0.1 822 1,164 Portugal 6.4 0.8 5.7 1.5 7.2 0.7 1.2 0.3 0.8 0.4 1.5 0.4 759 1,301 Spain 1.2 0.3 1.1 0.4 1.4 0.4 0.3 0.1 0.1 0.1 0.4 0.2 1,210 1,485 United States 9.2 0.4 7.4 0.5 10.8* 0.6 1.9 0.2 1.7 0.3 2.1 0.2 2,382 3,310 * p < 0.05 (two-sided test). TABLE 2 Time-Lagged Associations (Odds Ratios) of Temporally Primary Composite International Diagnostic Interview (CIDI) Separation Anxiety Disorder With the Subsequent Onset and Persistence of Other DSM-IV/CIDI Disorders and of Temporally Primary Other Disorders With the Subsequent Onset and Persistence of Separation Anxiety Disorder Temporally Primary Separation Anxiety Disorder Predicting Subsequent Onset and Persistence of Other Disorders Temporally Primary Other Disorders Predicting Subsequent Onset and Persistence of Separation Anxiety Disorder Onseta Persistenceb Onsetc Persistenced Disorder Odds Ratio 95% CI Odds Ratio 95% CI Odds Ratio 95% CI Odds Ratio 95% CI Internalizing Major depressive disorder 1.4* 1.3–1.6 1.3* 1.1–1.7 1.7* 1.4–2.0 1.4 0.9–2.1 Bipolar disorder 1.8* 1.4–2.3 1.3 0.8–2.2 1.9* 1.4–2.4 1.6 0.9–2.7 Panic disorder without agoraphobia 1.3* 1.0–1.7 1.0 0.6–1.8 1.8* 1.4–2.2 1.4 0.8–2.5 Generalized anxiety disorder 1.5* 1.2–1.9 1.4 1.0–1.9 1.1 0.8–1.6 0.9 0.5–1.6 Posttraumatic stress disorder 1.6* 1.3–2.1 1.8* 1.2–2.6 0.9 0.7–1.2 0.9 0.5–1.6 Social phobia 1.6* 1.2–2.0 1.2 0.8–1.7 1.4* 1.2–1.7 1.3 0.9–2.0 Specific phobia 1.7* 1.2–2.2 1.3 0.7–2.5 2.1* 1.8–2.4 1.0 0.7–1.4 Agoraphobia with or without panic 1.2 0.9–1.6 2.6* 1.4–4.7 1.3 0.9–1.8 1.1 0.6–1.9 Externalizing Attention deficit hyperactivity disorder 2.8* 1.6–4.6 0.8 0.4–1.9 1.1 0.8–1.5 1.2 0.7–2.0 Oppositional defiant disorder 1.6* 1.1–2.6 1.0 0.4–2.3 1.7* 1.3–2.2 1.4 0.8–2.4 Conduct disorder 1.4* 1.0–1.8 0.7 0.3–1.4 1.4* 1.1–1.9 0.8 0.4–1.6 Intermittent explosive disorder 1.3 1.0–1.7 0.7 0.5–1.2 1.3* 1.0–1.7 0.8 0.5–1.4 Substance abuse with or without dependence 1.0 0.8–1.2 1.3 0.9–1.8 1.4* 1.0–1.9 0.8 0.5–1.5 Substance dependence with abuse 1.0 0.8–1.4 1.3 0.8–2.1 1.0 0.7–1.4 1.0 0.4–2.2 a Discrete time survival analysis controlling for demographic variables, prior traumas, and childhood adversities as of age at onset of separation anxiety disorder, country, and person-year. b The data indicate a discrete time survival analysis predicting the prevalence of the disorder controlling for demographic variables, prior lifetime disorders, prior traumas, and childhood adversities as of age at onset of the disorder, country, and person-year. c Logistic regression analysis predicting 12-month prevalence of separation anxiety disorder among lifetime cases controlling for age at onset and time since onset of separation anxiety disorder, prior lifetime disorders as of the age at onset of separation anxiety disorder, prior lifetime trauma as of the age at onset of separation anxiety disorder, and prior childhood adversities and country. d The data indicate a logistic regression analysis predicting 12-month prevalence of the disorder among lifetime cases controlling for age at onset and time since onset of the disorder, prior lifetime disorders as of the age at onset of the disorder, prior lifetime trauma as of the age at onset of the disorder, prior childhood adversities, and country. * p<0.05 (two-sided test). TABLE 3 Predictors of Lifetime DSM–5/Composite International Diagnostic Interview (CIDI) Separation Anxiety Disorder in the Total Sample and by Life Course Stagea Total Sample Childhood (Age 13) Adolescence (Ages 13–17) Early Adulthood (Ages 18–29) Later Adulthood (Ages ≥30) Variable Odds Ratio 95% CI Odds Ratio 95% CI Odds Ratio 95% CI Odds Ratio 95% CI Odds Ratio 95% CI Sex Female 1.3* 1.2–1.5 1.4* 1.1–1.7 1.1 0.8–1.5 1.4* 1.1–1.9 1.1 0.7–1.7 Male (reference) 1.0 – 1.0 – 1.0 – 1.0 – 1.0 – Education b Student 1.2 1.0–1.6 – – – – 1.0 0.6–1.5 Low 1.5* 1.2–2.0 – – – – 1.4 1.0–2.2 2.0* 1.3–3.2 Low-average 1.6* 1.2–2.0 – – – – 1.4 1.0–2.0 2.0* 1.2–3.3 High-average 1.7* 1.4 1.0–1.9 1.8* 1.2–2.7 High (reference) 1.0 – 1.0 – 1.0 – 1.0 – 1.0 – Marital Status c Never married 1.1 0.9–1.4 – – – – 0.9 0.7–1.1 1.2 0.8–1.8 Previously married 1.3 0.9–1.8 – – – – 1.6* 1.1–2.3 1.7* 1.1–2.5 Currently married (reference) 1.0 – 1.0 – 1.0 – 1.0 – 1.0 – Number of maladaptive family functioning childhood adversitiesd, e 1 1.7* 1.4–2.0 1.7* 1.3–2.2 1.2 0.8–1.8 1.6* 1.2–2.1 2.1* 1.2–3.7 2 2.0* 1.6–2.4 2.3* 1.7–3.1 1.4 0.8–2.4 1.7* 1.2–2.3 2.0* 1.1–3.4 3 2.3* 1.8–2.8 2.6* 1.8–3.7 2.1* 1.3–3.4 1.7* 1.2–2.5 2.0* 1.1–3.4 4 2.8* 2.0–4.0 3.0* 1.9–5.0 1.7 0.8–3.6 3.3* 1.6–6.8 1.9* 1.0–3.8 ≥5 2.8* 2.1–3.6 3.7* 2.3–6.0 1.0 0.4–2.5 2.5* 1.3–4.6 3.7* 1.5–9.1 None (reference) 1.0 – 1.0 – 1.0 – 1.0 – 1.0 – Number of other childhood adversities f 1 1.3* 1.1–1.5 1.5* 1.2–1.9 1.3 0.9–1.9 1.0 0.8–1.3 1.1 0.7–1.5 2 1.2 1.0–1.5 1.7* 1.2–2.3 1.2 0.7–2.1 0.8 0.6–1.2 1.0 0.6–1.7 ≥3 1.8* 1.3–2.4 2.8* 1.6–4.9 1.4 0.6–3.6 1.3 0.7–2.3 0.5 0.2–1.5 None (Reference) 1.0 – 1.0 – 1.0 – 1.0 – 1.0 – Lifetime trauma g War 1.1 0.9–1.3 1.6* 1.0–2.5 1.4 0.7–2.5 0.9 0.6–1.3 1.0 0.6–1.5 Violence 1.1 0.9–1.2 2.0* 1.2–3.3 0.8 0.5–1.2 1.2 0.9–1.5 0.8 0.5–1.1 Sexual violence 1.6* 1.3–1.9 2.8 0.6–12.3 3.0* 1.5–5.8 1.6* 1.2–2.1 1.5* 1.0–2.1 Accident 1.4* 1.1–1.6 1.1 0.7–1.7 1.8* 1.2–2.6 1.3* 1.0–1.7 1.5* 1.0–2.3 Family death 1.4* 1.2–1.6 1.7* 1.1–2.5 1.3 0.8–2.0 1.3* 1.0–1.6 1.6* 1.2–2.3 Network events 1.3* 1.1–1.6 1.6 0.9–2.9 1.5* 1.0–2.2 1.3* 1.0–1.7 1.1 0.8–1.5 Other 1.5* 1.2–1.9 1.9* 1.1–3.3 1.7* 1.0–3.00 1.2 0.8–1.8 2.0* 1.3–3.3 a Discrete time-survival analysis controlling for person-year, country, age at interview, and prior lifetime DSM-IV/CIDI disorders as of separation anxiety disorder age at onset. The total sample model was estimated in all person-years, while the models for the various life course stages were restricted to person-years in the ranges indicated. The measures of education, marital status, and lifetime traumas were dated and were treated as time-varying covariates (i.e., coded as being present only as of the respondent’s age when they occurred). b The chi-square test (df=3) statistics for the total sample, early adulthood, and later adulthood are 20.22, 4.01, and 12.60, respectively, with the results for the total sample and later adulthood reaching statistical significance. c The chi-square test (df=2) statistics for the total sample, early adulthood, and later adulthood are 3.13, 10.06, and 5.70, respectively, with the results for early adulthood reaching statistical significance. d Measured using the Maladaptive Family Functioning Scale, which records items for parental mental illness, substance misuse, criminal behavior, domestic violence, physical and sexual abuse, and neglect. e The chi-square test (df=5) statistics for the total sample, childhood, adolescence, early adulthood, and later adulthood are 98.30, 48.08, 10.62, 18.63, and 13.20, respectively, with the results for all age groups except adolescence reaching statistical significance. f The chi-square test (df=3) statistics for the total sample, childhood, adolescence, early adulthood, and later adulthood are 20.23, 23.42, 2.59, 1.89, and 2.01, respectively, with the results for the total sample and childhood reaching statistical significance. g The chi-square test (df=7) statistics for the total sample, childhood, adolescence, early adulthood, and later adulthood are 117.47, 35.25, 44.40, 35.65, and 42.56, respectively, with the results for all age groups reaching statistical significance. * p<0.05 (two-sided test). TABLE 4 Prevalence of Severe Role Impairment Among Respondents With 12-Month DSM–5/Composite International Diagnostic Interview (CIDI) Separation Anxiety Disorder as a Joint Function of Country Income, Separation Anxiety Disorder Age at Onset, and 12-Month Comorbidity With the DSM-IV/CIDI Disorders Assessed in the World Mental Health Surveysa Total With Comorbidity Without Comorbidity Country Income and Age at Onset % SE % SE % SE Total sample 4–12 years old 33.9 5.2 37.3 5.9 8.5 8.2 13–17 years old 42.2 7.5 46.5 8.5 23.7 10.8 18–29 years old 35.7 4.3 45.3 5.3 16.7 5.3 ≥30 years old 31.7 5.8 40.4 7.3 21.2 8.9 Total 35.3 2.5 42.4 2.9 18.3 4.3 Low-/lower-middle income countries 4–12 years old 24.0 12.2 24.7 12.7 0.0 0.0 13–17 years old 14.5 7.9 13.0 8.7 19.4 18.8 18–29 years old 19.8 5.8 32.2 8.5 8.5 5.7 ≥30 years old 9.0 5.4 19.5 10.5 0.0 0.0 Total 17.6 4.1 25.3 6.0 6.9 3.8 Upper-middle income countries 4–12 years old 38.4 10.6 38.5 11.3 36.8 26.3 13–17 years old 30.6 10.7 27.8 13.1 39.3 17.0 18–29 years old 44.6 8.0 47.9 10.5 35.0 12.2 ≥30 years old 35.4 9.1 46.8 10.5 26.0 13.9 Total 38.2 4.1 41.6 4.4 30.6 8.9 High-income countries 4–12 years old 33.6 6.1 40.5 7.0 0.0 0.0 13–17 years old 62.3 10.5 71.1 9.2 0.0 0.0 18–29 years old 39.8 6.4 49.1 7.6 14.7 7.7 ≥30 years old 38.1 10.4 41.6 12.5 28.9 16.9 Total 41.4 3.8 49.3 4.3 14.2 5.7 a Logistic regression models assuming additive associations of country income level, age at onset, and 12-month comorbidity with severe role impairment fit the data better (Akaike information criterion=434.1; Bayesian information criterion=468.1) than models that also included all two-way interactions (Akaike information criterion=438.0; Bayesian information criterion=531.6) or all two-way and three-way interactions (Akaike information criterion=458.1; Bayesian information criterion=561.3). All other authors report no financial relationships with commercial interests. REFERENCES 1 Silove DM Marnane CL Wagner R The prevalence and correlates of adult separation anxiety disorder in an anxiety clinic BMC Psychiatry 2010 10 21 20219138 2 Pini S Abelli M Shear KM Frequency and clinical correlates of adult separation anxiety in a sample of 508 outpatients with mood and anxiety disorders Acta Psychiatr Scand 2010 122 40 46 19824987 3 Bögels SM Knappe S Clark LA Adult separation anxiety disorder in DSM-5 Clin Psychol Rev 2013 33 663 674 23673209 4 Miniati M Calugi S Rucci P Predictors of response among patients with panic disorder treated with medications in a naturalistic follow-up: the role of adult separation anxiety J Affect Disord 2012 136 675 679 22134042 5 Copeland WE Angold A Shanahan L Longitudinal patterns of anxiety from childhood to adulthood: the Great Smoky Mountains Study J Am Acad Child Adolesc Psychiatry 2014 53 21 33 24342383 6 Shear K Jin R Ruscio AM Prevalence and correlates of estimated DSM-IV child and adult separation anxiety disorder in the National Comorbidity Survey Replication Am J Psychiatry 2006 163 1074 1083 16741209 7 Gittelman R Klein DF Relationship between separation anxiety and panic and agoraphobic disorders Psychopathology 1984 17 suppl 1 56 65 6369368 8 Silove D Harris M Morgan A Is early separation anxiety a specific precursor of panic disorder-agoraphobia? a community study Psychol Med 1995 25 405 411 7675927 9 Battaglia M Pesenti-Gritti P Medland SE A genetically informed study of the association between childhood separation anxiety, sensitivity to CO(2), panic disorder, and the effect of childhood parental loss Arch Gen Psychiatry 2009 66 64 71 19124689 10 Roberson-Nay R Eaves LJ Hettema JM Childhood separation anxiety disorder and adult onset panic attacks share a common genetic diathesis Depress Anxiety 2012 29 320 327 22461084 11 Kossowsky J Pfaltz MC Schneider S The separation anxiety hypothesis of panic disorder revisited: a meta-analysis Am J Psychiatry 2013 170 768 781 23680783 12 Bowlby J Separation anxiety: a critical review of the literature J Child Psychol Psychiatry 1960 1 251 269 13 Milrod B Markowitz JC Gerber AJ Childhood separation anxiety and the pathogenesis and treatment of adult anxiety Am J Psychiatry 2014 171 34 43 24129927 14 Hoven CW Duarte CS Lucas CP Psychopathology among New York City public school children 6 months after September 11 Arch Gen Psychiatry 2005 62 545 552 15867108 15 Scheeringa MS Zeanah CH Reconsideration of harm’s way: onsets and comorbidity patterns of disorders in preschool children and their caregivers following Hurricane Katrina J Clin Child Adolesc Psychol 2008 37 508 518 18645742 16 Kessler RC Aguilar-Gaxiola S Alonso J The global burden of mental disorders: an update from the WHO World Mental Health (WMH) surveys Epidemiol Psichiatr Soc 2009 18 23 33 19378696 17 Heeringa SG Wells JE Hubbard F Kessler RC Ustun TB Sample Designs and Sampling Procedures The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders 2008 New York Cambridge University Press 14 32 18 Harkness J Pennell BE Villa A Kessler RC Ustun TB Translation Procedures and Translation Assessment in the World Mental Health Survey Initiative The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders 2008 New York Cambridge University Press 91 113 19 Pennell B-E Mneimneh ZN Bowers A Kessler RC Ustun TB Implementation of the World Mental Health Surveys The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders 2008 New York Cambridge University Press 33 57 20 Kessler RC Ustun TB Kessler RC Ustun TB The World Health Organization International Diagnostic Interview The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders 2008 New York Cambridge University Press 58 90 21 Knäuper B Cannell CF Schwarz N Improving accuracy of major depression age-of-onset reports in the US National Comorbidity Survey Int J Methods Psychiatr Res 1999 8 39 48 22 First MB User’s Guide for the Structured Clinical Interview for DSM-IV-TR Axis I Disorders: SCID-I 2002 New York Biometrics Research Department, New York State Psychiatric Institute 23 Haro JM Arbabzadeh-Bouchez S Brugha TS Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health surveys Int J Methods Psychiatr Res 2006 15 167 180 17266013 24 Leon AC Olfson M Portera L Assessing psychiatric impairment in primary care with the Sheehan Disability Scale Int J Psychiatry Med 1997 27 93 105 9565717 25 Kessler RC McLaughlin KA Green JG Childhood adversities and adult psychopathology in the WHO World Mental Health Surveys Br J Psychiatry 2010 197 378 385 21037215 26 Karam EG Friedman MJ Hill ED Cumulative traumas and risk thresholds: 12-month PTSD in the World Mental Health (WMH) surveys Depress Anxiety 2014 31 130 142 23983056 27 Halli SS Rao KV Advanced Techniques of Population Analysis 1992 New York Plenum 28 Efron B Logistic regression, survival analysis, and the Kaplan-Meier curve J Am Stat Assoc 1988 83 414 425 29 Posada D Buckley TR Model selection and model averaging in phylogenetics: advantages of Akaike information criterion and Bayesian approaches over likelihood ratio tests Syst Biol 2004 53 793 808 15545256 30 Wolter K Introduction to Variance Estimation 2007 New York Springer 31 Research Triangle Institute SUDAAN (Release 10.0.1) 2009 Research Triangle Park, NC Research Triangle Institute [Computer Software] 32 Kessler RC Angermeyer M Anthony JC on behalf of the WHO World Mental Health Survey Consortium Lifetime prevalence and age-of-onset distributions of mental disorders in the World Health Organization’s World Mental Health Survey Initiative World Psychiatry 2007 6 168 176 18188442 33 Kessler RC Cox BJ Green JG The effects of latent variables in the development of comorbidity among common mental disorders Depress Anxiety 2011 28 29 39 21225850 34 Kessler RC Ormel J Petukhova M Development of lifetime comorbidity in the World Health Organization World Mental Health surveys Arch Gen Psychiatry 2011 68 90 100 21199968 35 Carr CP Martins CMS Stingel AM The role of early life stress in adult psychiatric disorders: a systematic review according to childhood trauma subtypes J Nerv Ment Dis 2013 201 1007 1020 24284634 36 Gonzalez A The impact of childhood maltreatment on biological systems: implications for clinical interventions Paediatr Child Health (Oxford) 2013 18 415 418 37 Kirsten LT Grenyer BF Wagner R Impact of separation anxiety on psychotherapy outcomes for adults with anxiety disorders Couns Psychother Res 2013 8 36 42

PMC005xxxxxx/PMC5116915.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101676030 44856 Cell Chem Biol Cell Chem Biol Cell chemical biology 2451-9456 27341433 5116915 10.1016/j.chembiol.2016.05.014 NIHMS829074 Article Immunization with outer membrane vesicles displaying designer glycotopes yields class-switched, glycan-specific antibodies Valentine Jenny L. 1# Chen Linxiao 1# Perregaux Emily C. 3 Weyant Kevin B. 1 Rosenthal Joseph A. 4 Heiss Christian 5 Azadi Parastoo 5 Fisher Adam C. 2 Putnam David 13 Moe Gregory R. 6 Merritt Judith H. 2 DeLisa Matthew P. 134* 1 School of Chemical and Biomolecular Engineering, Cornell University, Ithaca, NY 14853 USA 2 Glycobia Inc., 33 Thornwood Drive, Ithaca, NY 14850 USA 3 Comparative Biomedical Sciences, Cornell University, Ithaca, NY 14853 USA 4 Department of Biomedical Engineering, Cornell University, Ithaca, NY 14853 USA 5 Complex Carbohydrate Research Center, The University of Georgia, 315 Riverbend Road, Athens, GA 30602, USA 6 Centers for Cancer and Immunobiology and Vaccine Development, Children’s Hospital Oakland Research Institute, Oakland, CA 94609, USA * Address correspondence to: Matthew P. DeLisa, School of Chemical and Biomolecular Engineering, Cornell University, Ithaca, NY 14853. Tel: 607-254-8560; Fax: 607-255-9166; md255@cornell.edu # These authors contributed equally to this work. 11 11 2016 23 6 2016 21 11 2016 23 6 655665 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary The development of antibodies against specific glycan epitopes poses a significant challenge due to difficulties obtaining desired glycans at sufficient quantity and purity, and the fact that glycans are usually weakly immunogenic. To address this challenge, we leveraged the potent immunostimulatory activity of bacterial outer membrane vesicles (OMVs) to deliver designer glycan epitopes to the immune system. This approach involved heterologous expression of two clinically important glycans, namely polysialic acid (PSA) and Thomsen-Friedenreich antigen (T antigen) in hypervesiculating strains of non-pathogenic Escherichia coli. The resulting glycOMVs displayed structural mimics of PSA or T antigen on their surfaces, and induced high titers of glycan-specific IgG antibodies following immunization in mice. In the case of PSA glycOMVs, serum antibodies potently killed Neisseria meningitidis serogroup B (MenB), whose outer capsule is PSA, in a serum bactericidal assay. These findings demonstrate the potential of glycOMVs for inducing class-switched, humoral immune responses against glycan antigens. Introduction Complex carbohydrates, or glycans, are a ubiquitous feature on the surface of cells from all three domains of life. For example, capsular polysaccharides (CPS) or lipid-linked lipopolysaccharides (LPS) present on the surface of pathogenic bacteria are well known to mediate host-pathogen interactions (Comstock and Kasper, 2006). Alternatively, by displaying glycans that are structurally similar to those of their host, certain pathogens are able to avoid immune recognition (Comstock and Kasper, 2006). In eukaryotes, surface glycans participate in a variety of key biological processes including adhesion, cell-cell recognition, differentiation, and immune recognition (Varki et al., 2009), and are also known to feature prominently in disease (Ohtsubo and Marth, 2006). Indeed, glycans on the surfaces of tumor cells are commonly expressed at atypical levels or with altered structural attributes, and these aberrant structures serve as unambiguous markers of malignancy for a number of cancers (Pinho and Reis, 2015). At present, the study of glycans and their myriad roles remains a daunting task due in large part to their inherent structural complexity and the relative lack of tools for their biosynthesis, analysis, and recognition. Antibodies (Abs) specific for glycan epitopes (glycotopes) are particularly useful clarifying the functions of glycans. Glycan-targeting Abs can be elicited by immunization with carbohydrate antigens, and the resulting Abs can be used to probe the structure and function of glycans (Calarese et al., 2005; Nonaka et al., 2014) or target glycans therapeutically (Luo et al., 2010; Zhang et al., 2010). Nonetheless, the creation of glycan-specific Abs by immunization poses a significant challenge for several reasons. First, it is very difficult to isolate glycan-based immunogens from cells and tissues at purities and quantities that are sufficient for mAb isolation. Glycans and glycoconjugates are almost always a heterogeneous mixture of structures when isolated from natural sources (Raman et al., 2005), which dilutes any potential antigenic response. Total chemical synthesis and chemoenzymatic synthesis can often yield more uniform glycotopes (Wang and Lomino, 2012), however, these techniques are labor intensive, difficult to scale, and exist predominantly in the laboratories of a handful of experts. Second, glycans alone usually elicit weaker T-cell independent immune responses, which are short-lived and lack IgM-to-IgG class switching (Avci and Kasper, 2010). A common strategy for enhancing the immunogenicity of carbohydrates is to covalently couple a glycan to a T-cell dependent antigen. For example, conjugates composed of bacterial CPS or LPS chemically bound to an immunogenic carrier protein induce high-affinity, class-switched mAbs (Astronomo and Burton, 2010; Avci and Kasper, 2010). Unfortunately, production of traditional conjugate vaccines is a complex, multistep process that is expensive, time consuming, and low yielding (Frasch, 2009). A simplified alternative for generating glycoconjugates known as protein glycan coupling technology (PGCT) has been described recently (Cuccui and Wren, 2014; Terra et al., 2012). This approach leverages laboratory strains of Escherichia coli for the expression of recombinant bacterial polysaccharides (e.g., O-polysaccharide antigens), which are conjugated in vivo to a co-expressed carrier protein by the Campylobacter jejuni oligosaccharyltransferase PglB. However, while PGCT has been used to make several novel protein/glycan combinations, it is limited by variable glycan conjugation efficiency as observed for certain heterologous polysaccharide substrates (Cuccui et al., 2013; Ihssen et al., 2015; Ihssen et al., 2010) and a challenging purification of the product antigen. This is particularly pertinent in the context of producing glycoconjugates carrying mammalian-like glycans (Cuccui and Wren, 2014). Here, we sought to develop an efficient method for generating class-switched, anti-glycan Abs that overcomes many of the challenges discussed above. To this end, our approach combined custom glycan biosynthesis with outer membrane vesicle (OMV) formation in laboratory strains of E. coli. OMVs are naturally occurring nanospherical structures (~20–250 nm) produced constitutively by all Gram-negative bacteria. They are composed of proteins, lipids, and glycans derived from the outer membrane and periplasm, and have natural adjuvant properties that strongly stimulate the innate, and more importantly, the adaptive immune response (Alaniz et al., 2007; Baker et al., 2014; Ellis et al., 2010). To expand the immunostimulatory potential of OMVs, genetic engineering techniques have been used to load OMVs with foreign protein antigens by targeting expression to the outer membrane or to the periplasm of an OMV-producing host strain (Chen et al., 2010; Muralinath et al., 2011). These OMV-associated recombinant proteins elicited strong and specific antibody responses following immunization in mice. Building on these earlier observations, we engineered hypervesiculating strains of E. coli (Bernadac et al., 1998) to produce OMVs that displayed foreign glycans on their exteriors. This involved creation of two heterologous pathways for biosynthesis of structural mimics of clinically important carbohydrates, namely poly-α2,8-N-acetyl neuraminic acid (polysialic acid or PSA) and Galβ1-3GalNAcα1 (Thomsen-Friedenreich antigen or T antigen). The resulting glycosylated OMVs (glycOMVs), whose surfaces were remodeled with the custom-designed PSA or T antigen epitopes, induced strong glycan-specific IgG antibody titers following immunization in BALB/c mice. Taken together, our results show that engineered glycOMVs represent an effective strategy for generating functional Abs against structurally defined glycotopes of biomedical importance. Results A bottom-up engineered pathway for biosynthesis of T antigen on the surface of OMVs T antigen is one of many ‘self’ antigens expressed on a variety of malignancies including breast, colon, prostate, and stomach cancer, and Abs recognizing T antigen could have clinical benefit (Astronomo and Burton, 2010; Heimburg-Molinaro et al., 2011). However, the low intrinsic immunogenicity of T antigen poses a barrier to vaccination even after conjugation to a carrier protein (Adluri et al., 1995). Here, we hypothesized that the adjuvanticity of OMVs could be leveraged to overcome the weak immunogenicity of the T antigen epitope. Since LPS is a major component of released OMVs (Baker et al., 2014) and can be engineered to display foreign glycans (Ilg et al., 2010; Valderrama-Rincon et al., 2012), we attempted to remodel the carbohydrate component of LPS with T antigen-containing glycans. Such remodeling involved the lipid carrier undecaprenylpyrophosphate (Und-PP) as an acceptor of engineered glycans, which are flipped to the periplasmic side of the inner membrane and subsequently transferred to lipid A-core by the O-polysaccharide antigen ligase WaaL (Fig. 1a). In most laboratory strains of E. coli, Und-PP is primed with N-acetylglucosamine (GlcNAc) by the enzyme WecA. To elaborate the native Und-PP-GlcNAc with Galβ1-3GalNAc, we expressed two heterologous glycosyltransferases (GTases): the α1,3-GalNAc-transferase (PglA) from Campylobacter jejuni for transfer of GalNAc to Und-PP-GlcNAc (Glover et al., 2005); and the β1,3-galactosyltransferase (WbnJ) from E. coli O86 for stereospecific addition of the terminal galactose residue (Yi et al., 2005). Additionally, the UDP-GlcNAc 4 epimerase (Gne) from the same locus as C. jejuni PglA was added to supply the requisite UDP-GalNAc (Bernatchez et al., 2005). Formation of the T antigen epitope on inner membrane lipids was assessed by introducing plasmid pTF in E. coli K12 strain MC4100 lacking the waaL gene, which causes accumulation of UndPP-linked glycans in the cytoplasmic membrane. Lipid-linked oligosaccharides (LLOs) were extracted from these cells, and the glycan portion was released and purified as previously described (Valderrama-Rincon et al., 2012). MALDI-TOF mass spectrometry (MS) analysis of these glycans identified a major peak consistent with the expected Gal-terminal T antigen structure (m/z = 609) (Fig. 1b). Following treatment of the isolated glycans with β1,3-galactosidase, MS analysis revealed a major peak (m/z = 447) consistent with the removal of a single hexose residue (Fig. 1b), thereby corroborating the linkage of a terminal β1,3 Gal residue. Next, to determine whether the recombinant T antigen could be displayed on the exterior of OMVs, the hypervesiculating E. coli K12 strain JC8031, which overproduces OMVs due to deletion of tolRA (Bernadac et al., 1998), was transformed with the pTF plasmid. OMVs isolated from these cells were subjected to dot blot analysis whereby intact OMVs were spotted directly onto nitrocellulose membranes without any denaturation steps, and membranes were probed with peanut agglutinin (PNA), a lectin that binds the Galβ1-3GalNAc structure of the T antigen (Lotan et al., 1975). Consistent with the observation that outer membrane glycolipids are a major component of OMVs (Baker et al., 2014), we observed a strong signal from the non-denatured OMV fraction derived from JC8031 cells carrying pTF (Fig. 1c). To confirm that this signal was due to incorporation of the recombinant T antigen into LPS structures, we analyzed the OMV fraction from a knockout mutant of JC8031 that lacked waaL (hereafter JC8032), which encodes the O-polysaccharide antigen ligase responsible for transferring engineered Und-PP-linked glycans to lipid A-core (Feldman et al., 2005). As expected, the formation of T antigen on OMVs was blocked in JC8032 cells (Fig. 1c), confirming that display of engineered glycotopes on lipid A-core involved WaaL-dependent assembly. The incorporation of foreign glycotopes into E. coli LPS structures had no visible effect on vesicle nanostructure. For example, the spherical bilayered shape of T antigen-containing OMVs was indistinguishable from control OMVs as evidenced by transmission electron microscopy (TEM) microscopy (Supplementary Fig. 1a). Likewise, analysis by dynamic light scattering (DLS) revealed that the majority of the purified vesicles had a diameter of 20–60 nm (Supplementary Fig. 1b), consistent with the size of E. coli-derived OMVs that were characterized previously (Park et al., 2010). To determine whether recombinant T antigen detected in the pelleted supernatant was associated with intact vesicles, rather than with released outer membrane fragments or other cellular debris, the OMV-containing fraction isolated from JC8031 cells carrying pTF was separated by density gradient ultracentrifugation. Coomassie staining and Western blotting of the resulting fractions revealed that total OMV proteins, the outer membrane protein OmpA, and recombinant T antigen all co-migrated to denser fractions (Supplementary Fig. 2a–c), reminiscent of the gradient profiles seen previously for intact OMVs and OMV-associated proteins (Kim et al., 2008). A bottom-up engineered pathway for biosynthesis of PSA on the surface of OMVs PSA is a CPS that coats the surface of MenB and E. coli K1, and is also expressed in human tissues, most notably on neural cell adhesion molecule (NCAM) (Moe et al., 2009). As a result of this latter point, PSA has proven to be a particularly difficult target for antibody generation. Indeed, PSA is poorly immunogenic even when conjugated to a carrier protein (Krug et al., 2004), possibly because of the similarity to self-antigens. To overcome this barrier, we engineered a hypervesiculating E. coli K12 strain to produce OMVs decorated with recombinant PSA glycans. Since E. coli K12 strains do not produce PSA naturally, this first required the creation of an artificial pathway for PSA biosynthesis. To create the core onto which PSA could polymerize, we generated plasmid pPSA, which enabled heterologous expression of the GTases LgtB from Neisseria gonorrhoeae and CstII from Campylobacter jejuni. These GTases were predicted to catalyze the successive transfer of galactose and sialic acid (N-acetyl neuraminic acid; NeuNAc), respectively, to the Und-PP acceptor. Plasmid pPSA also encoded the neuBACS genes from E. coli K1, which collectively coordinate the formation of precursor CMP-NeuNAc and polymerization of NeuNAc (Fig. 2a). Finally, the neuD gene from E. coli K1, which promotes efficient sialic acid synthesis by enhancing the activity of other proteins (e.g., NeuBAC) in the sialic acid pathway (Daines et al., 2000), was cloned on a separate plasmid named pNeuD. These two plasmids were introduced into a ΔnanA derivative of the hypervesiculating strain JC8031 (hereafter JC8033), which is unable to catabolize free NeuNAc due to absence of the N-acetylneuraminate lyase enzyme, NanA (Priem et al., 2002). LLOs were extracted from JC8033 cells carrying pPSA and pNeuD, and the glycan portion was purified and permethylated. Positive-ion mode MALDI-TOF MS of the permethylated glycans identified a major peak (m/z = 791.4) corresponding to a NeuNAc disaccharide, as well as minor peaks corresponding to tri-, tetra-, and pentasaccharides of NeuNAc (Fig. 2b). A minor peak consistent with a NeuNAc-NeuGc structure was also identified. To determine whether PSA was incorporated in OMVs derived from these cells, membrane vesicles were isolated and subjected to dot blot analysis using SEAM 12, a murine Ab that is cross-reactive with PSA and exhibits potent complement-mediated bactericidal activity against MenB (Granoff et al., 1998). As with the engineered T antigen, we observed a strong signal from the non-denatured OMV fraction derived from JC8033 cells carrying both pPSA and pNeuD (Fig. 2c). This signal was comparable to that obtained by similarly probing intact EV36 cells, a K-12/K1 hybrid E. coli strain that natively expresses PSA on its surface (Fig. 2c) (Vimr et al., 1989). In contrast, no detectable signal was observed from OMV fractions derived from JC8033 cells without any plasmids or carrying either the pPSA and pNeuD plasmid individually (Fig. 2c), indicating that NeuD was required for engineered PSA biosynthesis. Likewise, the absence of LgtB or CstII, or both, resulted in a similar lack of signal in dot blots probed with the SEAM 12 antibody (Fig. 2c). It is noteworthy that nearly identical signals were observed using a commercial anti-polysialic acid-NCAM antibody (Millipore) that recognizes α2,8-linked PSA. Density gradient ultracentrifugation of the OMV fraction from JC8033 carrying pPSA and pNeuD revealed that PSA co-migrated with total OMV proteins and OmpA (Supplementary Fig. 2d–f), and thus appeared to be associated with intact vesicles. It is also noteworthy that incorporation of foreign PSA glycan into E. coli LPS structures had no visible effect on vesicle nanostructure (Supplementary Fig. 1a), with vesicle size again ranging from 20–60 nm in diameter (Supplementary Fig. 1b). To confirm whether PSA was produced on the Und-PP acceptor, we analyzed the OMV fraction from the waaL knockout mutant of JC8033 (hereafter JC8034) carrying the pPSA and pNeuD plasmids. Unlike the case of T antigen biosynthesis above, the display of PSA on the exterior of OMVs was not dependent on WaaL (Fig. 2d). Likewise, PSA display was still observed in JC8033 cells lacking wecA (hereafter JC8035), which transfers GlcNAc to Und-PP and forms the hypothesized Und-PP-GlcNAc acceptor for LgtB (Fig. 2d). In light of these results, we hypothesized an alternative mechanism for incorporation of recombinant PSA into LPS structures involving direct conjugation of foreign saccharides to lipid A-core structures (Ilg et al., 2010). The basis for this hypothesis stems from the following: in E. coli K-12 strains, lipid A-core contains glucose residues that might serve as substrates for heterologously expressed LgtB, a promiscuous biocatalyst that can transfer galactose to a variety of different glucose- and glucosamine-containing acceptors (Blixt et al., 2001). To test this hypothesis, we generated a hypervesiculating derivative of E. coli ClearColi, a K-12 strain that produces truncated LPS structures, called lipid IVA, that lack saccharide acceptors for our heterologously expressed GTases (e.g., LgtB) (Mamat et al., 2008). This was accomplished by deleting the nlpI gene that is known to increase vesiculation on par with tolRA mutants (Kim et al., 2008; McBroom et al., 2006), resulting in strain ClearColi-ves. OMVs produced from ClearColi-ves cells carrying pPSA and pNeuD were indeed blocked for PSA display (Fig. 2d). In contrast, OMVs derived from the parental strain MG1655, also lacking nlpI (MG1655-ves) and carrying the pPSA and pNeuD plasmids, were decorated with PSA at a level that rivaled JC8033 carrying the same PSA pathway plasmids (Fig. 2d). These results support the notion that PSA was incorporated in LPS structures by direct conjugation to saccharides in lipid A-core. Immunization with glycOMVs elicits glycan-specific antibodies We next sought to assess the immunological potential of glycOMVs displaying the T antigen and PSA epitopes. Specifically, BALB/c mice were immunized via subcutaneous (s.c.) injection with either T antigen- or PSA-containing glycOMVs, after which blood was collected at regular intervals. Controls included ‘empty’ OMVs from plasmid-free JC8031 or JC8033 cells, LOS extracted from MenB strain S3446 (NmBLOS), and phosphate buffered saline (PBS). To determine whether glycOMVs generated glycan-specific Abs, the total T antigen- and PSA-specific IgG titers at the endpoint were measured by ELISA using the model glycoprotein carrier protein scFv13-R4 with a C-terminal glycosylation motif (Valderrama-Rincon et al., 2012) bearing the T antigen or native LOS from MenB, respectively, as immobilized antigen. In the case of the T antigen epitope, glycOMVs elicited a significantly higher (p < 0.01) level of glycan-specific IgGs compared to both the empty OMV and PBS control groups (Fig. 3a). Similarly, the total PSA-specific IgG titers were significantly increased (p < 0.01) for the group immunized with PSA glycOMVs compared to all other immunized groups (Fig. 3b). It is particularly noteworthy that the IgG titers for the group immunized with native MenB LOS were not significantly different (p > 0.2) than those measured in the PBS group, consistent with the weak immunogenicity of glycans alone. Hence, the immunogenicity of engineered carbohydrates was boosted by display on the exterior of OMVs. IgG titers were further broken down by analysis of IgG1 and IgG2a titers, wherein mean IgG1 to IgG2a antibody ratios served as an indicator of a Th1- or Th2-biased immune response. Mice immunized with T antigen and PSA glycOMVs showed a significant (p < 0.05) increase in mean titers of glycan-specific IgG1 and IgG2a in comparison to all other groups (Supplementary Fig. 3a and b). The similar levels observed for IgG1 versus IgG2a titers suggested no measurable Th1/Th2 bias. T antigen-specific antibodies detect target antigen in Western blot format To determine the diagnostic potential of these glycan-specific Abs, we performed Western blot analysis using sera generated through glycOMV immunization. As expected, Abs generated by immunization with T antigen glycOMVs cross-reacted exclusively with the model carrier protein scFv13-R4 bearing the T antigen, generating a signal that was on par with that obtained using PNA lectin (Fig. 4a). In contrast, when the membrane was probed with Abs generated by immunization with empty OMVs, there was no visible binding to either glycosylated or aglycosylated scFv13-R4 (Fig 4a). PSA-specific antibodies exhibit complement-mediated bactericidal activity We next investigated whether the serum Abs produced by glycOMV immunization were immunologically relevant. For this, we performed a complement-mediated serum bactericidal activity (SBA) assay using the sera collected from mice immunized with PSA glycOMVs. SBA is an established method by which the activity of Abs against N. meningitidis is measured, and it correlates with protection for all serogroups of the pathogen (Martin et al., 2005). Here, we hypothesized that PSA-specific Abs generated by glycOMV immunization would bind to capsular PSA on the surface of MenB and, in the presence of components of the human complement system, would mediate bacteriolysis of the pathogen. In the group immunized with PSA glycOMVs, 50% SBA was observed at over 100-fold dilutions of the serum, a level that was on par with the anti-MenB antibody, SEAM 12 (Fig. 4b). In contrast, no killing was observed for sera collected from any of the control groups, or for the control anti-MenC antibody, over the dilutions tested (Fig. 4b). Complete killing was observed in immunized groups at dilutions as high as 10-fold, indicating that the Abs present in serum from glycOMV-immunized mice were immunologically functional. Discussion We have developed a new approach for generating class-switched, anti-glycan Abs that leverages the immunostimulatory properties of OMVs (Alaniz et al., 2007; Chen et al., 2010; Ellis et al., 2010; Sanders and Feavers, 2011) to boost the immune response to glycan epitopes, which are notoriously weak antigens (Astronomo and Burton, 2010; Avci and Kasper, 2010). An important first step involved converting laboratory strains of E. coli into factories for glycosylated OMV production by combining bacterial vesiculation with engineered pathways for designer glycan biosynthesis. We anticipate that this strategy could be generalized to create many other structurally diverse and biomedically relevant glycotopes on the exterior of OMVs for both diagnostic and therapeutic applications. It is worth noting that compared to the expression of protein antigens in OMVs, expression of glycans is a more elaborate undertaking. For protein antigens, expression in OMVs simply requires targeting the antigen of interest either to the periplasmic space by genetic fusion of an N-terminal export signal or to the cell surface by genetic fusion to an outer membrane carrier protein (Baker et al., 2014). Following vesiculation, the periplasmic- or outer membrane-targeted proteins become constituents of the OMV lumen or exterior, respectively. In contrast, display of carbohydrate antigens on OMVs requires the coordinated expression of multiple heterologous glycosyltransferases for directing the synthesis of desired glycans onto bacterial lipid carriers that subsequently localize to the outer membrane and become constituents of released OMVs. Despite the challenges, several groups including ours have used glycoengineering as a tool to remodel the bacterial outer membrane with mammalian glycotopes of interest including ganglioside GM3 (Ilg et al., 2010), Lewis Y (LeY) antigen (Yavuz et al., 2011), and trimannosyl core N-glycan (Valderrama-Rincon et al., 2012). Presumably, expression of these different cell surface glycans in a hypervesiculating host strain would yield uniquely glycosylated OMVs, although this remains to be shown. Importantly, once new glycan structures are created, however, production of glycOMV immunogens is significantly less complicated, less time consuming, less expensive, and more scalable than conventional approaches for producing glycoconjugates. It requires only one cultivation step to generate the final product, which can be easily and economically isolated by a single ultracentrifugation step (Chen et al., 2010). Moreover, the clinical translation of OMVs has also been established recently by Bexsero, a four component vaccine against N. meningitidis serogroup B that has been approved in the U.S. for preventing infection caused by this serogroup. The active components of this vaccine are three recombinant proteins identified by reverse vaccinology combined with detergent extracted OMVs prepared from a naturally occurring epidemic strain (Gorringe and Pajon, 2012). While engineered OMVs such as the ones we described herein have not yet found their way into the clinic, Bexsero represents a first important step in that direction. The ability of T antigen- and PSA-modified glycOMVs to elicit class-switched, glycan-specific IgGs in mice is significant in light of the low intrinsic immunogenicity that has been observed for these glycans, even when conjugated to a carrier protein (Adluri et al., 1995; Krug et al., 2004). The poor immunogenicity of these glycans has been attributed to immunologic tolerance that arises due to their resemblance with structures present in human and murine hosts (i.e., self antigens). Hence, our observation that glycOMVs triggered high titers of class-switched IgGs represents a significant advance in the pursuit of Abs against glycotopes of interest. While a molecular-level understanding of how OMVs boost the immune response to these glycans remains to be determined, we suspect that it stems from the potent adjuvanticity afforded by OMVs which: (i) are readily phagocytosed by professional antigen-presenting cells; (ii) carry pathogen-associated molecular patterns (PAMPs) within their structure that can stimulate both innate and adaptive immunity; and (iii) possess strong proinflammatory properties (Alaniz et al., 2007; Ellis et al., 2010; Sanders and Feavers, 2011). The fact that PSA and T antigen are self antigens of humans and a potential cause of immunopathology has hindered their development as vaccines. However, there are reasons to believe that many tumor-specific carbohydrate antigens including T antigen and PSA have a number of attributes that make them viable targets for vaccine development, most notably their widespread and high expression in several different cancers and their low or cryptic expression on normal cells (Astronomo and Burton, 2010; Heimburg-Molinaro et al., 2011). Interestingly, in the case of PSA, Miller and colleagues recently published an essay in which they reviewed the data on PSA as a self antigen and concluded that (2→8)-α-Neu5Ac conjugates will be as safe and effective as the polysaccharide protein conjugate vaccines for the other four meningococcal serogroups (Robbins et al., 2011). Aside from the potential immunotherapeutic applications enabled by glycOMVs, their ability to elicit glycan-specific Abs can be exploited to produce high-affinity reagents for glycobiology and glycomedicine. Currently, carbohydrate-binding lectins are the primary means to detect glycans in numerous analytical assays; however, lectins are limited by their poor sensitivity and binding affinity as well as lack of specificity towards less common glycan structures (Haab, 2012). Our demonstration that polyclonal serum from immunization with T antigen glycOMVs can be used for immunodetection of glycans illuminates the diagnostic potential of serum Abs elicited by glycOMVs and ensures that these Abs will find use even in cases where vaccine-induced autoimmunity proves to be an insurmountable obstacle. The potent bactericidal activity of the PSA glycOMV-stimulated serum Abs against N. meningitidis serogroup B in the presence of human complement further confirmed the authenticity of the engineered glycotope mimics as well as the full functionality of the Abs they elicited. To our surprise, vaccinations using empty OMV controls elicited measurably higher IgG titers in mice as compared to titers measured for the PBS control mice; however, none of these serum Abs were bactericidal. It should also be noted that the PSA-specific IgG titers for the empty OMV-immunized groups were still significantly (p < 0.01) less than those generated from the PSA glycOMV-immunized groups. Nonetheless, we attribute these unexpected ELISA signals to the presence of serum Abs against other components in the OMVs, which cross-reacted with similar components present in the NmBLOS. Indeed, when ELISA plates were instead coated with a synthetic PSA-ADH derivative (Granoff et al., 1998), the signal from the empty OMV group was notably lower. Overall, the results of this study reveal that glycOMVs are a reliable and robust method for generating class-switched Ab responses to glycan epitopes of interest. Compared to current approaches for achieving the same goal, glycOMVs represent a solution that is considerably less complicated and significantly more scalable. By rewiring the glycan biosynthetic pathway, it should be possible to generate glycOMVs displaying a wide array of biomedically relevant glycotopes found on the surfaces of bacteria and human cells, as we demonstrated here with T antigen and PSA. Moreover, the ability of OMVs to overcome immunologic tolerance by eliciting strong immune responses to glycans characterized as self antigens should further expand the palette of glycans that can be targeted by this approach. There also exist opportunities to couple glycOMVs with emerging techniques for Ab discovery such as immune repertoire mining (Lavinder et al., 2015), which could provide unprecedented access to a renewable source of high-quality, glycan-binding affinity reagents for interrogating the glycomes of living organisms or treating human disease. And since immunization with PSA glycOMVs yielded Abs that were fully functional (i.e., bactericidal), it stands to reason that glycOMVs themselves might eventually find use in a therapeutic context. Experimental Procedures Bacterial strains and plasmids A description of all bacterial strains and plasmids used in this study, including those that were constructed herein, is provided in the Supplementary Methods, along with a complete list in Supplementary Table 1. Briefly, unless otherwise stated, most strains used herein are based on E. coli strain JC8031, a tolRA mutant strain that is known to hypervesiculate (kindly provided by Roland Lloubes, Centre national de la Recherche Scientifique) (Bernadac et al., 1998). Cell growth and OMV preparation OMVs were prepared as described previously (Chen et al., 2010). Briefly, cells were freshly transformed with plasmids for T antigen or PSA biosynthesis and selected on medium supplemented with the appropriate antibiotic. An overnight culture of a single colony was subcultured into 100–200 mL of Luria-Bertani (LB) medium. The culture was grown to mid-log phase, at which time protein expression was induced with L-arabinose (0.2%) and/or IPTG (0.1 mM), if necessary. Cell-free culture supernatants were collected 16–20 h post-induction and filtered through a 0.2 μm filter. Vesicles were isolated by ultracentrifugation (Beckman-Coulter; TiSW28 rotor; 141,000xg; 3 h; 4°C) and resuspended in PBS. OMVs were quantified by the bicinchoninic-acid assay (BCA Protein Assay; Pierce) using BSA as the protein standard. OMVs were further separated by density-gradient ultracentrifugation as previously described (Kim et al., 2008). Briefly, OMVs were prepared as described above but resuspended in a 50 mM HEPES (pH 6.8) solution. This solution was adjusted to 45% (v/v) Optiprep (Sigma) in 1.5 mL. All other Optiprep layers were prepared using the same 50-mM HEPES (pH 6.8) solution. Optiprep/HEPES gradient layers were added to a 12-mL ultracentrifuge tube as follows: 0.33 mL of 10%, 0.33 mL of 15%, 0.66 mL of 20%, 0.66 mL of 25%, 0.9 mL of 30%, 0.9 mL of 35%, 1.5 mL of 45% containing the prepared OMVs, and enough 60% to nearly fill the tube. Gradients were centrifuged (Beckman-Coulter; TiSW41 rotor; 180,000xg; 3 h; 4°C), then, a total of ten fractions of 0.5 mL each were removed sequentially from the top of the gradient. These fractions were analyzed by Western blot and dot blot analyses as described below. Glycoprotein expression and purification E. coli MC4100 ΔwaaL::kan was co-transformed with plasmid pTrc99A-ssDsbA-scFv13-R4DQNAT (Valderrama-Rincon et al., 2012) and either pTF or empty vector control pMW07. An overnight culture was used to inoculate 100 mL of LB medium containing ampicillin and chloramphenicol. Cultures were grown to an OD600 of ~ 2.0, and induced overnight with 0.2% arabinose and 0.1 mM IPTG. Cells were harvested after which glycosylated or aglycosylated scFv13-R4 proteins were purified using Ni-NTA spin columns (Qiagen) according to manufacturer’s instructions. Proteins were buffer exchanged to PBS and the concentration adjusted to 1 mg/mL. Western blot and dot blot analysis OMV and LPS samples were prepared for SDS-PAGE analysis by boiling for 15 min and cooling to room temperature in the presence of loading buffer containing β-mercaptoethanol. OMV samples were normalized by total protein concentration, which was quantified using the BCA method as detailed above. Samples were run on Any kD polyacrylamide gels (BioRad, Mini-PROTEAN® TGX) and transferred to a PVDF membrane. After blocking with Carbo-Free blocking solution (Vector Labs), membranes with T antigen samples were probed first with biotinylated peanut agglutinin (Vector Labs) and then with streptavidin-HRP (Abcam). Similarly, after blocking with a 5% milk solution, membranes with PSA samples were probed first with SEAM 12 primary antibody specific against N. meningitidis B CPS (Granoff et al., 1998) and then with the corresponding anti-mouse HRP-conjugated secondary antibody (Promega). Membranes from density-gradient samples were also probed with an OmpA-specific antibody (kindly provided by Wilfred Chen, University of Delaware) and then with anti-mouse HRP-conjugated secondary antibody (Promega). Signal was visualized using HRP substrate and either an X-ray film developer or a ChemiDoc Imaging System (BioRad). Western blot analysis of glycosylated and aglycosylated scFv13-R4 was performed according to standard protocols. Briefly, protein-containing samples were separated by SDS-PAGE and transferred to PVDF membranes. Membranes were then probed with either: biotinylated PNA (Vector labs) and then with streptavidin-HRP (Vector Labs); immunized sera (from either empty OMV or T antigen glycOMV groups) at a concentration of 1:1000, followed by anti-mouse IgG HRP (VWR); or anti-His-HRP (Sigma). Signal was visualized using HRP substrate and imaged using a ChemiDoc Imaging System (BioRad). For dot blot analysis, OMV samples were normalized by total protein concentration, which was quantified using the BCA method as detailed above, and spotted directly onto a nitrocellulose membrane. Alternatively, OMVs were boiled for 10 min and cooled to room temperature prior to spotting on the membrane. After blocking with a 5% milk solution, membranes with PSA samples were probed first with SEAM 12 and then with HRP-conjugated anti-mouse IgG. For the T antigen, membranes were blocked with Carbo-Free blocking solution (Vector Labs), and then probed with biotinylated peanut agglutinin (Vector Labs) and subsequently streptavidin-HRP (Abcam). Signal was visualized using HRP substrate and either an X-ray film developer or a ChemiDoc Imaging System (BioRad). Electron microscopy Structural analysis of vesicles was performed via transmission electron microscopy as previously described (Chen et al., 2010). Briefly, vesicles were negatively stained with 2% uranyl acetate and deposited on 400-mesh Formvar carbon-coated copper grids. Imaging was performed using a FEI Tecnai F20 transmission electron microscope. MALDI-TOF MS analysis For structural characterization of T antigen, LLOs were extracted from MC4100 ΔwaaL::kan cells carrying pTF. An overnight culture of a single colony was subcultured into LB medium. Cultures were grown at 30°C and induced when optical density at 600 nm (OD600) reached ~2.0. Cultures were then harvested after 20 h. Cell pellets were collected, resuspended in methanol and lysed via sonication. Material was then dried at 60°C and subsequently resuspended in 2:1 chloroform:methanol solution (v/v, CM) via sonication and washed two times with the CM solution. The pellet was then washed in water. Lipids were extracted with 10:10:3 chloroform:methanol:water (v/v/v, CMW) followed by methanol. Extracts were then loaded into a DEAE cellulose column and eluted with 300 mM NH4OAc in CMW. The LLOs were extracted with chloroform and dried. Glycans released from LLOs were dried and resuspended in dH2O. 10-μL β-1,3-galactosidase (NEB) reactions were prepared using supplied buffer and incubated at 37°C. Reaction products were monitored by MALDI/TOF-MS. For structural characterization of PSA, the carbohydrate was isolated from the cells following a published procedure (Willis et al., 2013), except that the cells were disrupted by French press and that the gel-filtration step was omitted. The yield of carbohydrate was about 1 mg. Next, the samples were permethylated as described (Anumula and Taylor, 1992) and analyzed by MALDI/TOF-MS in reflector positive-ion mode on a ABISciex 5800 MALDI/TOF-TOF using α-dihyroxybenzoic acid (DHBA, 20 mg/mL solution in 50% methanol:water) as matrix. Mouse immunizations Three groups of four or five BALB/c female mice aged six- to eight-weeks old (The Jackson Laboratory) were each immunized s.c. with either PBS alone (control) or 100 μL of PBS containing: 10 μg of OMVs from JC8031 cells carrying no plasmid (empty OMVs) or 10 μg of OMVs from JC8031 cells harboring pTF (T antigen glycOMVs). Separately, four groups of five or six BALB/c female mice aged six- to eight-weeks old (The Jackson Laboratory) were each immunized s.c. with either PBS alone (control) or 100 μL of PBS containing: 2 μg of native LOS from N. meningitidis serogroup B (NmBLOS), 10 μg of OMVs from JC8033 cells carrying no plasmid (empty OMVs), or 10 μg of OMVs from JC8033 cells harboring pPSA and pNeuD (PSA glycOMVs). NmBLOS was prepared from MenB strain S3446 identically as described previously (Apicella et al., 1997). PSA content of the PSA glycOMV and NmBLOS doses were similar and determined via reactivity to the SEAM 12 antibody (Granoff et al., 1998). All PBS used was at pH 7.4. Each group of mice was boosted with an identical dosage of antigen 28 days and 56 days after the priming dose. Blood was collected from each mouse from the mandibular sinus immediately before and 14 days after the first immunization, immediately before and 14 days after the first boosting dose, immediately before the second boosting dose, and at 14 days and 28 days after the second boosting dose. The protocol number for the animal studies (2009-0096) was approved by the Institutional Animal Care and Use Committee at Cornell University. Statistical analysis was performed using one-way analysis of variance (ANOVA) followed by the post-hoc Tukey-Kramer test for multiple comparisons. Enzyme-linked immunosorbant assay (ELISA) Glycan-specific Abs produced in immunized mice were measured via indirect ELISA using a modification of a previously described protocol (Chen et al., 2010). Briefly, sera were isolated from the collected blood draws after centrifugation at 2,200xg for 10 min. 96-well plates (Maxisorp; Nunc Nalgene) were coated with E. coli-derived LPS containing T antigen or NmBLOS (2 μg/mL in PBS pH 7.4) and incubated overnight at 4°C. For comparison, ELISAs were also performed using a synthetic PSA-adipic acid dihydrazide (ADH) derivative, which was prepared as described (Granoff et al., 1998) and used to coat microtiter plates at a concentration of 20 μg/ml in PBS (pH 7.4). The next day, plates were washed 3 times with PBST (PBS, 0.05% Tween-20, 0.3% BSA) and blocked overnight at 4°C with 5% nonfat dry milk (Carnation) in PBS. Samples were serially diluted, in triplicate, between 1:100-1:12,800,000 in blocking buffer and added to the plate for 2 h at 37°C. Plates were washed 3 times with PBST and incubated for 1 h at 37°C in the presence of one of the following horseradish peroxidase-conjugated Abs: goat anti-mouse IgG (1:5000; Abcam), anti-mouse IgG1 (1:5000; Abcam), or anti-mouse IgG2a (1:5000; Abcam). After 3 additional washes with PBST, 3,3′-5,5′-tetramethylbenzidine substrate (1-Step Ultra TMB-ELISA; Thermo Scientific) was added and the plate was incubated at room temperature for 30 min. The reaction was halted with 2M H2SO4. Absorbance was quantified via microplate spectrophotometer (Molecular Devices) at a wavelength of 450 nm. Serum antibody titers were determined by measuring the lowest dilution that resulted in signal three standard deviations above background. Statistical significance was determined using Tukey-Kramer post hoc honest significant difference test and compared against the PBS control case. Complement-mediated bactericidal assay Bactericidal assays were conducted similar to a previously published protocol (Moe et al., 1999). Briefly, MenB strain H44/76 was grown overnight on chocolate agar plates (Remel). Single colonies were used to inoculate a culture in Franz media starting at an OD620 ~0.15 and grown at 37°C, 5% CO2 to OD620 ~0.6. The bacteria were diluted in Dulbeccos PBS with Ca2+ and Mg2+ (DPBS), containing 1% human serum albumin (Sigma-Aldrich). Approximately 300–400 CFU meningococci were incubated with 20% human serum (from a healthy adult with no detectable anticapsular antibody to group B polysaccharide) that had been depleted of IgG with a Protein G column and serum collected from mouse immunizations. Percent survival was calculated as the CFU/mL after 60 min incubation of bacteria compared to baseline CFU/ml at time zero determined by average bacterial growth in buffer alone, with heat-inactivated complement, with active complement, or a mixture of heat-inactivated and active complement. Murine antibodies SEAM 12 and anti-MenC were used as positive and negative controls, respectively. Supplementary Material Supplemental We thank Roland Lloubes for strain JC8031, Eric Vimr for strain EV36, and Wilfred Chen for anti-OmpA antibody used in this work. This material is based upon work supported in part by The Jennifer Leigh Wells Family (G.R.M.), NSF Grants CBET 1159581 and CBET 1264701 (both to M.P.D.), an NSF GK-12 “Grass Roots” Fellowship (to L.C.), and the following NIH Grants: T32GM008500 (to E.C.P.), GM088905-01 (to M.P.D.), EB005669-01 (to D.P. and M.P.D.), R56AI114793-01 (to D.P. and M.P.D.), R43AI091336-01 (to A.C.F.), R43GM093483 (to A.C.F.), R44GM093483-02 (to J.H.M.), and P41GM10349010 (to P.A.). We also acknowledge the New York State Office of Science, Technology and Academic Research (NYSTAR) Distinguished Faculty Award (to M.P.D.), the Cornell Center for Materials Research Shared Facilities, which are supported through the NSF MRSEC program (DMR-1120296), and the Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, U.S. Department of Energy grant DE-FG02-93ER20097 (to P.A.). Figure 1 Expression of recombinant T antigen glycotope on OMVs (a) Schematic of recombinant T antigen biosynthesis, which begins with glycan assembly on endogenous Und-PP-GlcNAc structure in the inner membrane and involves the coordinated action of heterologously expressed GalE epimerase and the glycosyltransferases PglA and WbnJ (shown in red). Trisaccharide glycan is flipped into the periplasmic space by Wzx where it is subsequently transferred onto lipid A core by WaaL and translocated to the outer membrane. (b) MALDI-MS profile of glycans released from LLOs by acid hydrolysis. LLOs were extracted from E. coli MC4100 ΔwaaL::kan cells carrying plasmid pTF. Glyans released from LLOs were dried and resuspended in dH2O. The major signal at m/z 609 corresponded to HexHexNAc2. Following digestion of glycans with β-1,3-galactosidase at 37°C, the major signal at m/z 447 corresponded to HexNAc2. (c) Dot blot analysis of OMV fractions, generated from plasmid-free JC8031 cells (empty), JC8031 cells carrying pTF, and JC8032 cells, which lacked waaL, carrying pTF. Blots were probed with peanut agglutinin to confirm presence or absence of T antigen on exterior of OMVs. Fetuin and asialofetuin served as negative and positive controls, respectively. Figure 2 Expression of recombinant PSA glycotope on OMVs (a) Schematic of PSA biosynthesis pathway, which involves NeuABCD for the formation of CMP-NeuNAc and NeuS for its polymerization into PSA. (b) Positive-ion MALDI-TOF spectrum of permethylated PSA. Glycolipids were extracted from JC8033 cells carrying plasmids pPSA and pNeuD. The major signal at m/z 791.4, with additional significant peaks at m/z 1152.6, 1513.7, and 1874.9 corresponding to tri, tetra, and penta NeuNAc structures. The predominance of the dimer seen here is consistent with preliminary NMR analysis of the same material. (c) Dot blot analysis of non-denatured OMV fractions generated from plasmid-free JC8033 cells (−) or JC8033 cells carrying either pNeuD, pPSA, or pPSA/pNeuD together. Also shown are OMV fractions from JC8033 cells carrying PSA/pNeuD where the pPSA plasmid lacked cstII (pPΔC), lgtB (pPΔL), or both (pPΔCL). Strain EV36 served as a positive control. (d) Dot blot analysis of non-denatured OMV fractions generated from plasmid-free JC8033 cells (−) or the following strains each carrying pPSA/pNeuD: JC8033; JC8034, which lacked waaL; and JC8035, which lacked wecA. Also shown are OMV fractions from hypervesiculating versions of MG1655 and ClearColi (MG1655-ves and ClearColi-ves, respectively) carrying pPSA/pNeuD. Blots were probed with SEAM 12 antibody to confirm the presence or absence of PSA on exterior of OMVs. Figure 3 GlycOMVs boost production of glycan-specific IgG antibodies Median antigen-specific IgG titers of individual mice immunized with (a) T antigen glycOMVs or (b) PSA glycOMVs in endpoint (day 54 and 84, respectively) serum. For the T antigen epitope, three groups of BALB/c mice were immunized s.c. with: 10 μg OMVs from plasmid-free JC8031 cells (empty), 10 μg OMVs from JC8031 cells carrying pTF (T antigen glycOMVs); or PBS. Glycosylated scFv13-R4 bearing T antigen was used as immobilized antigen. For the PSA epitope, four groups of BALB/c mice immunized s.c. with: 10 μg OMVs from plasmid-free JC8031 ΔnanA cells (empty); 10 μg OMVs from JC8031 ΔnanA cells carrying pPSA and pNeuD (PSA glycOMVs); 2 μg MenB LOS; or PBS. NmBLOS was used as immobilized antigen. Mice were boosted on day 28 and 56 with same doses. Asterisk (*) represents statistical significance (p < 0.01; Tukey-Kramer Post-Hoc HSD) versus all other groups. Figure 4 Diagnostic and therapeutic potential of glycan-specific antibodies (a) Western blot analysis of scFv13-R4 glycosylated with T antigen (+) or aglycosylated scFv13-R4 (−). Blots were probed with PNA, polyclonal sera from groups immunized with T antigen glycOMVs or empty OMVs, or anti-His-HRP antibody. (b) Representative killing activity of antibodies in the serum of mice immunized with: PBS, empty OMVs, and PSA glycOMVs. Survival data is derived from standard serum bactericidal assay (SBA) where dilutions of serum from immunized mice were tested against MenB strain H44/76 in the presence of human complement. Murine antibodies against MenB (SEAM 12) and MenC (anti-MenC) served as positive and negative controls, respectively. The SBA curves for PBS and empty OMVs are representative of all mice in the group (n = 6) and three of six PSA glycOMV that had the highest response to the vaccine. Author Contributions. J.L.V. and L.C. designed research, performed research, analyzed data, and wrote the paper. E.C.P., K.B.W., and J.A.R. performed research and analyzed data. C.H. and P.A. performed structural analysis, analyzed the corresponding data, and wrote the paper. A.C.F., D.P., and G.R.M. wrote the paper. J.H.M. and M.P.D. conceptualized project, designed research, analyzed data, and wrote the paper. Competing financial interests. J.H.M. is an employee of Glycobia, Inc. A.C.F., J.H.M. and M.P.D. have a financial interest in Glycobia, Inc. Adluri S Helling F Ogata S Zhang S Itzkowitz SH Lloyd KO Livingston PO 1995 Immunogenicity of synthetic TF-KLH (keyhole limpet hemocyanin) and sTn-KLH conjugates in colorectal carcinoma patients Cancer Immunol Immunother 41 185 192 7553688 Alaniz RC Deatherage BL Lara JC Cookson BT 2007 Membrane vesicles are immunogenic facsimiles of Salmonella typhimurium that potently activate dendritic cells, prime B and T cell responses, and stimulate protective immunity in vivo J Immunol 179 7692 7701 18025215 Anumula KR Taylor PB 1992 A comprehensive procedure for preparation of partially methylated alditol acetates from glycoprotein carbohydrates Anal Biochem 203 101 108 1524204 Apicella MA Griffis JM Schneider H 1997 Isolation and characterization of lipopolysaccharides, lipooligosaccharides, and lipid A Bacterial pathogenesis Bavoil VLCaPM San Diego, CA Academic Press, Inc 123 133 Astronomo RD Burton DR 2010 Carbohydrate vaccines: developing sweet solutions to sticky situations? Nat Rev Drug Discov 9 308 324 20357803 Avci FY Kasper DL 2010 How bacterial carbohydrates influence the adaptive immune system Annu Rev Immunol 28 107 130 19968562 Baker JL Chen L Rosenthal JA Putnam D DeLisa MP 2014 Microbial biosynthesis of designer outer membrane vesicles Curr Opin Biotechnol 29C 76 84 Bernadac A Gavioli M Lazzaroni JC Raina S Lloubes R 1998 Escherichia coli tol-pal mutants form outer membrane vesicles J Bacteriol 180 4872 4878 9733690 Bernatchez S Szymanski CM Ishiyama N Li J Jarrell HC Lau PC Berghuis AM Young NM Wakarchuk WW 2005 A single bifunctional UDP-GlcNAc/Glc 4-epimerase supports the synthesis of three cell surface glycoconjugates in Campylobacter jejuni J Biol Chem 280 4792 4802 15509570 Blixt O Brown J Schur MJ Wakarchuk W Paulson JC 2001 Efficient preparation of natural and synthetic galactosides with a recombinant beta-1,4-galactosyltransferase-/UDP-4′-gal epimerase fusion protein J Org Chem 66 2442 2448 11281786 Calarese DA Lee HK Huang CY Best MD Astronomo RD Stanfield RL Katinger H Burton DR Wong CH Wilson IA 2005 Dissection of the carbohydrate specificity of the broadly neutralizing anti-HIV-1 antibody 2G12 Proc Natl Acad Sci U S A 102 13372 13377 16174734 Chen DJ Osterrieder N Metzger SM Buckles E Doody AM DeLisa MP Putnam D 2010 Delivery of foreign antigens by engineered outer membrane vesicle vaccines Proc Natl Acad Sci U S A 107 3099 3104 20133740 Comstock LE Kasper DL 2006 Bacterial glycans: key mediators of diverse host immune responses Cell 126 847 850 16959564 Cuccui J Thomas RM Moule MG D’Elia RV Laws TR Mills DC Williamson D Atkins TP Prior JL Wren BW 2013 Exploitation of bacterial N-linked glycosylation to develop a novel recombinant glycoconjugate vaccine against Francisella tularensis Open Biol 3 130002 23697804 Cuccui J Wren B 2014 Hijacking bacterial glycosylation for the production of glycoconjugates, from vaccines to humanised glycoproteins J Pharm Pharmacol Daines DA Wright LF Chaffin DO Rubens CE Silver RP 2000 NeuD plays a role in the synthesis of sialic acid in Escherichia coli K1 FEMS Microbiol Lett 189 281 284 10930752 Ellis TN Leiman SA Kuehn MJ 2010 Naturally produced outer membrane vesicles from Pseudomonas aeruginosa elicit a potent innate immune response via combined sensing of both lipopolysaccharide and protein components Infect Immun 78 3822 3831 20605984 Feldman MF Wacker M Hernandez M Hitchen PG Marolda CL Kowarik M Morris HR Dell A Valvano MA Aebi M 2005 Engineering N-linked protein glycosylation with diverse O antigen lipopolysaccharide structures in Escherichia coli Proc Natl Acad Sci U S A 102 3016 3021 15703289 Frasch CE 2009 Preparation of bacterial polysaccharide-protein conjugates: analytical and manufacturing challenges Vaccine 27 6468 6470 19555714 Glover KJ Weerapana E Imperiali B 2005 In vitro assembly of the undecaprenylpyrophosphate-linked heptasaccharide for prokaryotic N-linked glycosylation Proc Natl Acad Sci U S A 102 14255 14259 16186480 Gorringe AR Pajon R 2012 Bexsero: a multicomponent vaccine for prevention of meningococcal disease Hum Vaccin Immunother 8 174 183 22426368 Granoff DM Bartoloni A Ricci S Gallo E Rosa D Ravenscroft N Guarnieri V Seid RC Shan A Usinger WR 1998 Bactericidal monoclonal antibodies that define unique meningococcal B polysaccharide epitopes that do not cross-react with human polysialic acid J Immunol 160 5028 5036 9590252 Haab BB 2012 Using lectins in biomarker research: addressing the limitations of sensitivity and availability Proteomics Clin Appl 6 346 350 22927350 Heimburg-Molinaro J Lum M Vijay G Jain M Almogren A Rittenhouse-Olson K 2011 Cancer vaccines and carbohydrate epitopes Vaccine 29 8802 8826 21964054 Ihssen J Haas J Kowarik M Wiesli L Wacker M Schwede T Thony-Meyer L 2015 Increased efficiency of Campylobacter jejuni N-oligosaccharyltransferase PglB by structure-guided engineering Open Biol 5 140227 25833378 Ihssen J Kowarik M Dilettoso S Tanner C Wacker M Thony-Meyer L 2010 Production of glycoprotein vaccines in Escherichia coli Microb Cell Fact 9 61 20701771 Ilg K Yavuz E Maffioli C Priem B Aebi M 2010 Glycomimicry: display of the GM3 sugar epitope on Escherichia coli and Salmonella enterica sv Typhimurium Glycobiology 20 1289 1297 20574043 Kim JY Doody AM Chen DJ Cremona GH Shuler ML Putnam D DeLisa MP 2008 Engineered bacterial outer membrane vesicles with enhanced functionality J Mol Biol 380 51 66 18511069 Krug LM Ragupathi G Ng KK Hood C Jennings HJ Guo Z Kris MG Miller V Pizzo B Tyson L 2004 Vaccination of small cell lung cancer patients with polysialic acid or N-propionylated polysialic acid conjugated to keyhole limpet hemocyanin Clin Cancer Res 10 916 923 14871967 Lavinder JJ Horton AP Georgiou G Ippolito GC 2015 Next-generation sequencing and protein mass spectrometry for the comprehensive analysis of human cellular and serum antibody repertoires Curr Opin Chem Biol 24 112 120 25461729 Lotan R Skutelsky E Danon D Sharon N 1975 The purification, composition, and specificity of the anti-T lectin from peanut (Arachis hypogaea) J Biol Chem 250 8518 8523 811657 Luo XM Lei MY Feidi RA West AP Jr Balazs AB Bjorkman PJ Yang L Baltimore D 2010 Dimeric 2G12 as a potent protection against HIV–1 PLoS Pathog 6 e1001225 21187894 Mamat U Meredith TC Aggarwal P Kuhl A Kirchhoff P Lindner B Hanuszkiewicz A Sun J Holst O Woodard RW 2008 Single amino acid substitutions in either YhjD or MsbA confer viability to 3-deoxy-d-manno-oct-2-ulosonic acid-depleted Escherichia coli Mol Microbiol 67 633 648 18093093 Martin D McCallum L Glennie A Ruijne N Blatchford P O’Hallahan J Oster P 2005 Validation of the serum bactericidal assay for measurement of functional antibodies against group B meningococci associated with vaccine trials Vaccine 23 2218 2221 15755599 McBroom AJ Johnson AP Vemulapalli S Kuehn MJ 2006 Outer membrane vesicle production by Escherichia coli is independent of membrane instability Journal Of Bacteriology 188 5385 5392 16855227 Moe GR Bhandari TS Flitter BA 2009 Vaccines containing de-N-acetyl sialic acid elicit antibodies protective against neisseria meningitidis groups B and C J Immunol 182 6610 6617 19414816 Moe GR Tan S Granoff DM 1999 Differences in surface expression of NspA among Neisseria meningitidis group B strains Infection and immunity 67 5664 5675 10531214 Muralinath M Kuehn MJ Roland KL Curtiss R 3rd 2011 Immunization with Salmonella enterica serovar Typhimurium-derived outer membrane vesicles delivering the pneumococcal protein PspA confers protection against challenge with Streptococcus pneumoniae Infect Immun 79 887 894 21115718 Nonaka M Fukuda MN Gao C Li Z Zhang H Greene MI Peehl DM Feizi T Fukuda M 2014 Determination of carbohydrate structure recognized by prostate-specific F77 monoclonal antibody through expression analysis of glycosyltransferase genes J Biol Chem 289 16478 16486 24753248 Ohtsubo K Marth JD 2006 Glycosylation in cellular mechanisms of health and disease Cell 126 855 867 16959566 Park KS Choi KH Kim YS Hong BS Kim OY Kim JH Yoon CM Koh GY Kim YK Gho YS 2010 Outer membrane vesicles derived from Escherichia coli induce systemic inflammatory response syndrome PLoS One 5 e11334 20596524 Pinho SS Reis CA 2015 Glycosylation in cancer: mechanisms and clinical implications Nat Rev Cancer 15 540 555 26289314 Priem B Gilbert M Wakarchuk WW Heyraud A Samain E 2002 A new fermentation process allows large-scale production of human milk oligosaccharides by metabolically engineered bacteria Glycobiology 12 235 240 12042246 Raman R Raguram S Venkataraman G Paulson JC Sasisekharan R 2005 Glycomics: an integrated systems approach to structure-function relationships of glycans Nat Methods 2 817 824 16278650 Robbins JB Schneerson R Xie G Hanson LA Miller MA 2011 Capsular polysaccharide vaccine for Group B Neisseria meningitidis, Escherichia coli K1, and Pasteurella haemolytica A2 Proc Natl Acad Sci U S A 108 17871 17875 22025709 Sanders H Feavers IM 2011 Adjuvant properties of meningococcal outer membrane vesicles and the use of adjuvants in Neisseria meningitidis protein vaccines Expert Rev Vaccines 10 323 334 21434800 Terra VS Mills DC Yates LE Abouelhadid S Cuccui J Wren BW 2012 Recent developments in bacterial protein glycan coupling technology and glycoconjugate vaccine design J Med Microbiol 61 919 926 22516134 Valderrama-Rincon JD Fisher AC Merritt JH Fan YY Reading CA Chhiba K Heiss C Azadi P Aebi M DeLisa MP 2012 An engineered eukaryotic protein glycosylation pathway in Escherichia coli Nat Chem Biol 8 434 436 22446837 Varki A Cummings RD Esko JD Freeze HH Stanley P Bertozzi CR Hart GW Etzler ME 2009 Essentials of Glycobiology 2 Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press Vimr ER Aaronson W Silver RP 1989 Genetic analysis of chromosomal mutations in the polysialic acid gene cluster of Escherichia coli K1 J Bacteriol 171 1106 1117 2644224 Wang LX Lomino JV 2012 Emerging technologies for making glycan-defined glycoproteins ACS Chem Biol 7 110 122 22141574 Willis LM Stupak J Richards MR Lowary TL Li J Whitfield C 2013 Conserved glycolipid termini in capsular polysaccharides synthesized by ATP-binding cassette transporter-dependent pathways in Gram-negative pathogens Proc Natl Acad Sci U S A 110 7868 7873 23610430 Yavuz E Maffioli C Ilg K Aebi M Priem B 2011 Glycomimicry: display of fucosylation on the lipo-oligosaccharide of recombinant Escherichia coli K12 Glycoconj J 28 39 47 21286806 Yi W Shao J Zhu L Li M Singh M Lu Y Lin S Li H Ryu K Shen J 2005 Escherichia coli O86 O-antigen biosynthetic gene cluster and stepwise enzymatic synthesis of human blood group B antigen tetrasaccharide J Am Chem Soc 127 2040 2041 15713070 Zhang G Zhang H Wang Q Lal P Carroll AM de la Llera-Moya M Xu X Greene MI 2010 Suppression of human prostate tumor growth by a unique prostate-specific monoclonal antibody F77 targeting a glycolipid marker Proc Natl Acad Sci U S A 107 732 737 20080743

PMC005xxxxxx/PMC5117188.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8809954 8282 J Evol Biol J. Evol. Biol. Journal of evolutionary biology 1010-061X 1420-9101 27500505 5117188 10.1111/jeb.12939 NIHMS801610 Article Rise and Fall of Vector Infectivity During Sequential Strain Displacements by Mosquito-Borne Dengue Virus Andrade Christy C. 13* Young Katherine I. 1 Johnson William L. 14 Villa Maria E. 1 Buraczyk Cynthia A. 1 Messer William B. 2 Hanley Kathryn A. 1 1 Department of Biology, New Mexico State University, Las Cruces, NM, USA 2 Department of Molecular Microbiology and Immunology and Division of Infectious Diseases, Department of Medicine, Oregon Health and Sciences University, Portland, OR, USA 3 Current address: Biology Department, Gonzaga University, Spokane, WA, USA 4 Current address: Colorado School of Public Health, Colorado State University, Fort Collins, CO, USA * Corresponding author contact information: Biology Department, Gonzaga University, 502 E. Boone Avenue, Spokane, WA 99258; phone: (509) 313-6638; fax: (509) 313-5804; andradec@gonzaga.edu 21 7 2016 8 8 2016 11 2016 01 11 2017 29 11 22052218 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Each of the four serotypes of mosquito-borne dengue virus (DENV-1-4) comprises multiple, genetically-distinct strains. Competitive displacement between strains within a serotype is a common feature of DENV epidemiology and can trigger outbreaks of dengue disease. We investigated the mechanisms underlying two sequential displacements by DENV-3 strains in Sri Lanka that each coincided with abrupt increases in dengue hemorrhagic fever (DHF) incidence. First, the post-DHF strain, displaced the pre-DHF strain in the 1980s. We have previously shown that post-DHF is more infectious than pre-DHF for the major DENV vector, Aedes aegypti. Then, the ultra-DHF strain evolved in situ from post-DHF and displaced its ancestor in the 2000s. We predicted that ultra-DHF would be more infectious for Ae. aegypti than post-DHF but found that ultra-DHF infected a significantly lower percentage of mosquitoes than post-DHF. We therefore hypothesized that ultra-DHF had effected displacement by disseminating in Ae. aegypti more rapidly than post-DHF, but this was not borne out by a timecourse of mosquito infection. To elucidate the mechanisms that shape these virus-vector interactions, we tested the impact of RNA interference, the principal mosquito defense against DENV, on replication of each of the three DENV strains. Replication of all strains was similar in mosquito cells with dysfunctional RNAi, but in cells with functional RNAi, replication of pre-DHF was significantly suppressed relative to the other two strains. Thus differences in susceptibility to RNAi may account for the differences in mosquito infectivity between pre-DHF and post-DHF, but other mechanisms underlie the difference between post-DHF and ultra-DHF. dengue virus competitive displacement Aedes mosquito vector trade-off RNA interference INTRODUCTION Mosquito-borne dengue virus (DENV, genus Flavivirus) is the agent of dengue disease, which ranges in severity from subclinical infection to classical dengue fever to severe dengue disease characterized by intravascular leakage, hemorrhage and hypovolemic shock (WHO, 2009), referred to as dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS) (Simmons et al., 2012). DENV is composed of four antigenically- and genetically-distinct serotypes (DENV-1-4), each serotype in turn comprises multiple genotypes, which differ by > 6% nucleotide identity, and genotypes include multiple, distinct lineages that differ by ≤ 6% nucleotide identity, which have been termed strains, clades, lineages or groups (Rico-Hesse, 2003, Weaver & Vasilakis, 2009), and here will be called strains. In the decades since World War II, the geographic range and incidence of all four DENV serotypes has increased steadily (Guzman et al., 2010); at present an estimated 390 million people are infected with DENV annually across the tropical and subtropical zones annually, of whom 96 million experience some form of dengue disease (Bhatt et al., 2013). While a dengue vaccine (Dengvaxia®), was recently approved in Mexico, Brazil, El Salvador, and the Philippines, it has limited efficacy and will not be given to children aged <9 years old (Hadinegoro et al., 2015). There are no licensed antiviral therapies to curb transmission of DENV. DENV genetic diversity shapes epidemiology and disease risk in a complex fashion. Infection with a given serotype is thought to result in homotypic protection against infection with all genotypes in that serotype (Rothman, 2011, Snow et al., 2014), but also primes the host for higher levels of viral replication (Pozo-Aguilar et al., 2014) and severe disease (DHF/DSS) upon heterotypic infection with a different serotype (Guzman et al., 2013). Moreover, strains and genotypes within serotypes may differ significantly in their tendency to cause DHF/DSS (Rico-Hesse, 2010). For example, the Southeast Asian genotype of DENV-2 is responsible for multiple dengue disease outbreaks characterized by high levels of DHF, whereas the American DENV-2 genotype has rarely been implicated in a case of DHF (Rico-Hesse, 2007). As a consequence of the increasing global mobility of humans, DENV strains are often moved from the site in which they originally evolved to new locations (Allicock et al., 2012). An increasingly common feature of DENV epidemiology is the invasion of a region by a genotype or strain of a given serotype, either through introduction from another region or in situ evolution, and replacement of the native genotype or strain of that same serotype (dos Santos et al., 2011, Carrillo-Valenzo et al., 2010, Drumond et al., 2012, Duong et al., 2013b, Duong et al., 2013a). Worryingly, it is often strains with a predilection for causing DHF that successfully invade, displacing strains with lower virulence (i.e. a lower tendency to cause DHF, for the purposes of this study) (Rico-Hesse, 2003, Rico-Hesse, 2010). Although some within-serotype strain replacements appear to be the result of passive, stochastic turnover during population bottlenecks (Teoh et al., 2013, Myat Thu et al., 2005), there is strong evidence that many others are due to active competitive displacement. As DENV replicates in two hosts, the mosquito and the human, competitive advantage may be achieved by enhanced infection of either or both (Althouse & Hanley, 2015, Lourenco & Recker, 2010). Investigations of DENV displacement (summarized in Table 1) have typically shown that the displacing genotype or strain was significantly more infectious for Ae. aegypti than the displaced genotype or strain when mosquitoes fed on artificial infectious bloodmeals at similar titers. In the one exception to this pattern, there was no difference between the displacing and displaced strain in mosquito infection, but the displacing strain produced higher viral titers in patient blood samples compared to the displaced strain. In previous work as well as the current study, we have sought to elucidate the mechanisms driving a series of DENV strain displacement events that occurred in Sri Lanka where, despite co-circulation of all four DENV serotypes, severe dengue disease was uncommon until 1989, when the nation experienced a sudden increase in DHF cases (Messer et al., 2003, Messer et al., 2002). Phylogenetic analysis demonstrated that this spike in DHF was associated with invasion of DENV-3 genotype III B strain and displacement of the native DENV-3 genotype III A strain (Messer et al., 2003); for simplicity we have subsequently renamed these strains the post-DHF and pre-DHF strains, respectively (Hanley et al., 2008). We have previously shown that the post-DHF strain replicates to higher titers and disseminates more efficiently in Ae. aegypti than pre-DHF (Table 1), providing one mechanism for the competitive displacement. In 2000, Sri Lanka experienced yet another abrupt surge in DHF incidence, and phylogenetic analysis revealed that this rise in DHF coincided with the appearance and spread of a new DENV-3 genotype III strain: strain SL post-2000, which we here term the ultra-DHF strain (Kanakaratne et al., 2009). The ultra-DHF strain evolved from and replaced the post-DHF strain (Kanakaratne et al., 2009). The rapidity of this replacement, and the absence of evidence for any interruption of DENV transmission that might have facilitated a population bottleneck, suggests that the post-DHF to ultra-DHF transition was the result of competitive displacement rather than drift. Here we tested the prediction that ultra-DHF would, like most other successful DENV invaders (Table 1), have an advantage over the post-DHF strain and pre-DHF strain in mosquito infection. Using multiple isolates from each of the competing DENV strains, we investigated potential differences in transmission efficacy, extrinsic incubation period (EIP; the time between acquisition of an infectious bloodmeal and ability to transmit that bloodmeal (Christofferson & Mores, 2011)), and response to RNA interference, an important mediator of the mosquito immune response (Blair, 2011). MATERIALS AND METHODS Viruses and Cell lines Three isolates of the pre-DHF strain (DENV-3-3002, DENV-3-3009, DENV-3-3011, (Hanley et al., 2008)), 3 isolates of the post-DHF strain (DENV-3-3001, DENV-3-3006, DENV-3-3010, (Hanley et al., 2008)) and 5 isolates of the ultra-DHF strain (DENV-3-3050, DENV-3-3053, DENV-3-3054, DENV-3-3055, DENV-3-3060) were generously provided to the Hanley laboratory by the de Silva laboratory at the University of North Carolina, Chapel Hill. Details on passage history and the nomenclature used in previously published studies of these viruses are provided in Table 2. Isolation of viral genomes, reverse transcription of amplicons including the C, prM and E genes (nucleotides 95 - 2413) and sequencing were conducted, essentially as previously described (Durbin et al., 2001), for each of the isolates listed above. Two cell lines derived from Ae. albopictus, a secondary vector of DENV, were utilized to dissect the impact of RNAi on virus replication kinetics: C6/36 cells lack a functional RNAi response (Brackney et al., 2010) while U4.4 cells possess a functional RNAi pathway (Schnettler et al., 2012, Leger et al., 2013). Additionally, C6/36 cells were used for generation of virus stocks and for titrations of stock viruses as well as infected mosquitoes. C6/36 cells were maintained in minimum essential medium (MEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 2mM L-glutamine, 2mM non-essential amino acids (Gibco, Life Technologies, Grand Island, NY), and 0.05 mg/ml gentamycin (Invitrogen, Life Technologies, Grand Island, NY). U4.4 cells were maintained in Mitsuhashi & Maramorosch medium (Himedia, VWR, Sugar Land, TX) supplemented with 20% FBS (Gibco, Life Technologies, Grand Island, NY), 0.13% sodium bicarbonate (Gibco, Life Technologies, Grand Island, NY), and 0.05 mg/ml gentamycin (Invitrogen, Life Technologies, Grand Island, NY). Both cell types were maintained at 32°C, 80% relative humidity (RH) with 5% CO2. Quantification of viral titer Virus titer was determined using serial dilutions of virus cell culture supernatant or homogenized mosquito samples in either C6/36 or U4.4 cells as previously described (Hanley et al., 2008). Briefly, confluent monolayers of designated cells in 24-well plates were inoculated with serial ten-fold dilutions of sample in cell culture medium and incubated at 32°C for 2 hours. Cells were then overlaid with 1% methylcellulose in OptiMem media (Invitrogen) supplemented with 2% FBS, 2mM L-glutamine (Gibco, Life Technologies, Grand Island, NY), and 0.05 mg/ml gentamycin (Invitrogen, Life Technologies, Grand Island, NY) and incubated 5 days at 32°C. When mosquito body and head samples were titered, the overlay was additionally supplemented with 5 μg/ml Fungizone (Gibco, Life Technologies, Grand Island, NY). Viral infection foci (hereafter termed plaques as in (Hanley et al., 2008)) were visualized by immunostaining with DENV-3-specific antibody. Mosquitoes and Mosquito Infection Eggs of Ae. aegypti from the National Institutes of Health (NIH) colony, described in Hanley et al. (2008), were hatched following standard procedures (Pal, 1967, Wattam & Christensen, 1992) and larvae were transferred to water-filled pans and maintained at 27°C with 80% RH. Larvae were fed daily on dry commercial cat food (Special Kitty Original Formula, Bentonville, AR). Pupae were relocated to water cups within a mesh-screened cage at 27°C with 80% RH. Adults were offered 10% sucrose solution ad libitum. Bloodmeals were prepared as previously described (Hanley et al., 2008). Briefly, 2 ml of washed red blood cells prepared from fresh, defibrinated rabbit blood (Hemostat, Dixon, CA) supplemented with 10% sucrose was spiked with 1mL of freshly-thawed stock virus. Sealed cartons of mosquitoes starved of sugar for 24-48 hours were placed under water-jacketed parafilm-covered membrane feeders (Lillie Glass, Smyrna, GA) for 20 minutes. Samples of the infectious bloodmeal were collected after feeding to quantify titer on C6/36 cells as described. Mosquitoes were then cold-anesthetized for 2 minutes at −20°C; engorged mosquitoes were separated on a chilled tray, transferred into new cartons, incubated for a specified time period at 27°C, 80% RH on a 12:12 hour light:dark cycle with ad libitum access to 10% sucrose. At the end of the incubation period, mosquitoes were cold-killed and stored at −80°C until dissection and homogenization. After a competent Ae. aegypti has imbibed an infectious bloodmeal, dengue virus infects the midgut. Midgut infection can occur as early as 2 days post engorgement, though 5 -10 days is more common (Salazar et al., 2007, Hardy et al., 1983). Viruses then replicate in the midgut and disseminate to and infect other tissues in the mosquito including the salivary glands. It typically takes 10-14 days post engorgement for virus to reach the salivary gland, however this can happen in as few as 3 days (Salazar et al., 2007). To quantify infection and dissemination, bodies (as a measure of infection) were separated from heads (as a measure of dissemination) and each was separately homogenized using motorized pestle in 250μl Hanks Balanced Salt Solution (Gibco) supplemented with 10% FBS, 250 μg/ml amphotericin B (Gibco), 1% ciprofloxacin, 150 μg/ml clindamycin, and stored at −80°C until virus titer was determined as described with C6/36 cells. In the first infection experiment, mosquitoes were fed on bloodmeals containing one of multiple virus isolates from each of the post-DHF and ultra-DHF strains (Table 2) and incubated for 14 days. To assess the time course of mosquito infection for each viral strain, mosquito infections were carried out as described above and subsets of mosquitoes were cold-killed on days 2, 6, 10 and 14 post-engorgement. Mosquitoes were then assayed for infection and dissemination as described above. Viral replication kinetics in cell culture To assess the impact of RNAi on viral replication, the replication curve of each of the 11 viral isolates in Table 2 was measured in C6/36 and U4.4 cells as previously described (Troyer et al., 2001). Replication curves are a common and sensitive measure of viral fitness in a cultured cell line (Andrade et al., 2011, Ciota et al., 2007). Briefly, triplicate 25 cm2 flasks of a particular cell line were infected with each virus isolate at a multiplicity of infection (MOI, ratio of plaque-forming units to cells) of 0.01 based on viral titer determined in the matching cell line. Diluted virus was added in a total volume of 1 ml and cells were incubated for 2 hours at 32°C with gentle rocking every 20 minutes. Cells were then washed twice with 1X PBS (Gibco) to remove unadsorbed virus and replenished with 5 ml of cell culture media. One ml of cell culture supernatant was harvested immediately post-wash (time 0) and 1 ml of appropriate medium was added back to each flask. Collected supernatants were stabilized in 1X SPG (2.18 mM sucrose, 38 mM potassium phosphate [monobasic], 72 mM potassium phosphate [dibasic], 60 mM L-glutamic acid), clarified by centrifugation, aliquoted and stored at −80°C. Samples were collected daily for a total of 7 days following the same procedure. Viral titer was quantified as described above in the same cell type in which the replication kinetics assays were performed. RESULTS Endpoint infectivity of post-DHF and ultra-DHF DENV-3 for Ae. aegypti To test the hypothesis that ultra-DHF displaced post-DHF because it is more infectious for Ae. aegypti than post-DHF, we fed mosquitoes on artificial bloodmeals containing one of 3 isolates of the former or 5 isolates of the latter and assayed bodies for virus, as a measure of infection, and heads, as a measure of dissemination, 14 days post-feeding. Percentage infection data were arcsin-square root transformed to render them normal; all transformed data were checked to ensure that they did conform to the normal distribution. Contra our hypothesis, the ultra-DHF strain of DENV-3 infected, on average, a significantly lower percentage of Ae. aegypti (mean ± 1 SE: 16% ± 4%) than the post-DHF strain (37%± 3%) (Fig. 1 and Table 3; t6 = 3.29, P = 0.017). There was no significant difference between the two strains in the mean virus titer in infected bodies (Table 3, t6 = 2.32, P = 0.06), total dissemination (Table 3, t6 = 0.94, P = 0.38), dissemination when the analysis was limited to only infected mosquitoes (Table 3, t6 =1.57, P = 0.17) or the mean titer of virus in infected heads (Table 3, t6 = 1.66, P = 0.15). Average bloodmeal titer did not differ significantly between the two strains either before or after feeding (P > 0.12 for both comparisons). Bloodmeal titers did decline over the course of feeding, and although most mosquitoes fed within the first few minutes after the bloodmeal was offered, we nonetheless conducted a logistic regression that included post-feeding bloodmeal titer as an independent variable as the most conservative comparison between the strains. This analysis, which used virus isolate nested within virus strain and post-feeding bloodmeal titer of each virus isolate as independent variables and infection or dissemination of each mosquito as the dependent variable did not show a significant effect of virus nested within strain (χ25 = 9.81, P = 0.08), but a significant effect of post-feeding bloodmeal titer (χ21 = 13.46, P = 0.0002) on the percentage of mosquitoes infected. This analysis found no effect of either virus strain (χ25 = 5.47, P = 0.36) or bloodmeal titer (χ25 = 2.65, P = 0.10) on percent disseminated infections. Infection time course of displaced and displacing dengue virus strains in Ae. aegypti To test the hypothesis that the ultra-DHF strain displaced post-DHF due to a shorter EIP by the displacing strain, mosquitoes were fed on bloodmeals containing isolates of both strains, as well as the pre-DHF strain, and infection was measured at days 2, 6, 10 and 14 post-feeding. Titer data from day 2 post-feeding were sporadically elevated and this was most likely due to contamination from undigested bloodmeal, as has been previously observed by others (Turell, 1988); consequently analysis was limited to days 6, 10 and 14 post-feeding (Table 4, Table 5, and Fig. 2). Bloodmeal titers (Table 4) were not significantly different among the three strains prior to feeding (F2= 1.8, P = 0.23) or after feeding (F2= 2.8, P = 0.12). We first compared each pair of displacing-displaced strains (post-DHF versus pre-DHF and ultra-DHF versus post-DHF) at day 14 post-feeding to determine whether the differences detected previously, i.e. greater dissemination in post-DHF compared to pre-DHF and greater infectivity in post-DHF compared to ultra-DHF, were repeatable. As we showed previously (Hanley et al., 2008), the pre-DHF strain disseminated in a lower percentage of Ae. aegypti than the post-DHF strain (respective mean ± 1 SE : 19% ± 16%, 62% ± 16%), however in this study at day 14 post-infection this difference was not significant (t4 = 1.92, P = 0.13). The percent dissemination by each strain did not differ significantly between the two studies (P > 0.3 for each comparison). We attribute the lack of significance to the higher dissemination by pre-DHF strain 3009 in the current study compared to the previous one. For the 3002 and 3011 pre-DHF strains, percent dissemination was remarkably similar between the two studies (0% and 5% for 3002 and 3% and 8% for 3011), but for 3009 percent dissemination increased from 25% in our previous study to 50% in the current one. Consistent with the patterns described in the previous section of this paper, in the timecourse experiment ultra-DHF infected a lower percentage (25% ± 9%) of mosquitoes than post-DHF (64% ± 17%); this difference was significant by a one-tailed t-test (t6 = −2.11, P = 0.04). To detect differences in EIP, we compared the infection dynamics in the three strains with a two-factor ANOVA including strain and day post-feeding as factors; a significant interaction between day and strain would indicate differences in infection dynamics. However there was no significant interaction between day post-feeding and virus strain on the percentage of mosquitoes infected (Table 4 and Fig. 2; F4,24 = 0.68, P = 0.61). Additionally, there was no significant main effect of virus strain on the percentage of mosquitoes infected (F2,24 = 2.62, P = 0.09), or day post feeding (6, 10 and 14), on the percentage of mosquitoes infected (F2,24 = 0.44, P = 0.65). Similarly, there was no significant interaction between day post-feeding and virus strain on the percentage of mosquitoes with a disseminated infection (Table 5 and Fig. 2, F4,24 = 1.39, P = 0.27). Neither virus strain (F2,24 = 3.00, P = 0.07) or day post-feeding (F2,24 = 2.97, P = 0.07) had a significant effect on the total percentage of mosquitoes with a disseminated infection. It is noteworthy that in analysis of both infection and dissemination, the impact of virus strain approached significance and the trend for the post-DHF strain to reach higher percentages of infection and dissemination than the other two strains was consistent with the previous experiments in this paper and previous studies (Hanley et al., 2008). Small sample sizes precluded the analysis of virus titer in the bodies and heads (Table 4 and Table 5). Impact of RNAi on pre-DHF, post-DHF and ultra-DHF DENV-3 To test the effect of RNAi on the three DENV-3 strains, we quantified their replication kinetics in one Ae. albopictus cell line that has a functional RNAi response (U.4.4) and one that has a dysfunctional RNAi pathway (C6/36) (Fig. 3). Repeated measures ANOVA revealed a significant time × strain interaction among the three strains in C6/36 cells (Fig. 3 (top); F14 = 2.4, P = 0.004) but a Tukey-Kramer post-hoc test showed that none of the strains differed from each other in overall replication (P>0.05 for all pairwise comparisons). In U4.4 cells (Fig. 3 (top)), in contrast, there was a highly significant interaction of time × strain (F14= 9.2, P < 0.0001) and a Tukey-Kramer post-hoc analysis showed that the pre-DHF strain replicated to a significantly lower level than the post-DHF or ultra-DHF strain (P < 0.05), whereas the post- DHF and ultra-DHF strains did not differ from each other (P > 0.05). We have repeated the experiment in U4.4 cells infected with all three strains for a period of 5 days and find qualitatively similar results (data not shown). Sequence variation in the structural proteins of pre-DHF, post-DHF and ultra-DHF DENV-3 DENV possesses a positive-sense, single-segment RNA genome that encodes three structural proteins (capsid (C), pre-membrane (prM) and envelope (E)) which together form the virion and seven non-structural proteins that facilitate infection (Hanley & Weaver, 2010). Table 6 shows those nucleotide and amino acid variants that distinguished viral strains, i.e. those variants that were conserved within each strain but differed between at least one strain and the other two. The single non-synonymous variant was a serine at amino acid 124 of the envelope gene (numbering from the start of the protein) in the pre-DHF strain but a proline at the corresponding position in post-DHF and ultra-DHF isolates. DISCUSSION Competitive displacement among strains of DENV (Table 1) as well as other arthropod-borne viruses (arboviruses) (Moudy et al., 2007, Schuh et al., 2014) is frequently mediated by differences in vector infectivity between the displacing and displaced strain. We have previously shown that the post-DHF strain of DENV-3, which invaded Sri Lanka and displaced the native pre-DHF DENV-3 strain was more infectious for Ae. aegypti than the pre-DHF strain (Hanley et al., 2008). Subsequently, Kanakaratne et al. (2009) reported that the post-DHF strain had been displaced by its evolutionary offshoot, the ultra-DHF strain. We predicted that this displacement too would be associated with greater mosquito infectivity, a prediction that we tested in the current study. To our surprise, we found instead that the ultra-DHF strain was significantly less infectious than the post-DHF strain. This apparent contradiction could be resolved if the loss in total infection by the ultra-DHF strain were coupled to a gain in the rapidity with which the virus reached this endpoint, i.e. a reduction in the extrinsic incubation period (EIP). Vectorial capacity is negatively related to duration of EIP, and thus a reduction in EIP confers a large fitness gain for arboviruses (Ciota & Kramer, 2013, Christofferson & Mores, 2011). For example, the displacement of the NY99 strain of West Nile virus in North America by the WN02 strain has been attributed to the difference between the two strains in EIP, which is as much as 4 days shorter in WN02 than NY99 (Moudy et al., 2007). Similarly, Quiner et al. (2014) reported that a DENV-2 clade displacement in Nicaragua was associated with a replicative advantage of the invading strain in native mosquitoes that was only observed at 7 days post feeding but not at 14 or 21 days. However in the current study we did not find a difference in the time course of infection among the three strains. All three strains showed relatively low mean percent infection and dissemination on day 6 post feeding; these averages rose substantially for the post-DHF strain but not for the pre-DHF or ultra-DHF strains on day 10 post-feeding. Thus there is no evidence that the displacement of the post-DHF strain by the ultra-DHF strain is due to a shorter EIP in the latter. The decline in vector infectivity by the ultra-DHF strain may be attributable to other evolutionary trade-offs that ultimately enhance transmission but are more difficult to quantify in the laboratory. First, there may be a trade-off between infectivity and pathogenicity of DENV for its vector. By conventional evolutionary wisdom, arthropod-borne pathogens should be benign for their vectors, on which they depend for transmission ((Ewald, 1993), but see (Elliot et al., 2003)). In particular, vector-borne pathogens should not reduce the longevity of their vectors, which must exceed the EIP if the pathogen is to be transmitted at all. The evidence for an impact of DENV on longevity of Ae. aegypti is mixed. Carrington et al. (2015) reported that Ae. aegypti that acquired a dengue infection by feeding on hospitalized dengue patients, the most natural route of infection, did not show a difference in survival compared to mosquitoes that were not infected for a period of 12 days post-feeding. In contrast, a meta-analysis by found that horizontally-transmitted arboviruses, including DENV, did reduce survival of their vectors. However this analysis included studies in which DENV infection was established via intra-thoracic inoculation of mosquitoes, and the relevance of this route of infection is questionable. Second, efficacy of horizontal transmission, as measured by infection after feeding on a bloodmeal, may be inversely related to efficacy of vertical transmission to eggs. While vertical transmission of DENV certainly occurs, recent reviews of the literature determined that it is unlikely to be a significant factor in the epidemiology of the virus ((Lequime & Lambrechts, 2014, Grunnill & Boots, 2015), see also (Adams & Boots, 2010)). However, in situations where a high proportion of the population is immune, it remains possible that vertical transmission could rescue a viral lineage from extinction until a sufficiently large susceptible population had accumulated. Third, the reduced mosquito infectivity of the ultra-DHF strain of DENV-3 may reflect a trade-off between decreased replication in the vector with increased replication in the human host. It has long been hypothesized that the arbovirus cycle of alternation between vertebrate hosts and arthropod vectors imposes trade-offs that prevent optimization of fitness in either host (Ciota & Kramer, 2010, Coffey et al., 2013). Variation among DENV strains in viremia levels in the human host may be mediated by variation in fundamental rates of virus replication or by variation in the proclivity for antibody-mediated enhancement during secondary infection (OhAinle et al., 2011); either mechanism could confer a competitive advantage on the strain with greater replication. In Vietnam the invasive Asian 1 DENV-2 strain caused significantly higher viremia in pediatric patients than the displaced Asian/American strain, but the two strains showed no differences in infectivity for mosquitoes when spiked at constant titer into artificial bloodmeals (Table 1). DENV generally infects more Ae. aegypti as virus titer in human blood increases (Carrington & Simmons, 2014, Duong et al., 2015), though Duong et al. have recently reported that, for any particular level of viremia, people with asymptomatic or presymptomatic dengue infections are significantly more infectious to feeding mosquitoes than people with a symptomatic infection (Duong et al., 2015, Nguyet et al., 2013, Whitehorn et al., 2015), revealing that the relationship between viremia and transmission is more complex than previously thought. Thus higher replication of ultra-DHF than post-DHF in the human host could confer a transmission advantage even if it came at the cost of decreased mosquito infectivity sensu stricto – that is decreased infectivity of ultra-DHF compared to post-DHF when virus titer is held constant. Finally, the decrease in ultra-DHF infectivity could have been driven by mutations that enabled immune escape at the cost of decreased vector infectivity. The paradigm for DENV immunity has long been that infection with any strain in a particular serotype generates cross-immunity to all other strains within that serotype (Katzelnick et al., 2016), whereas antibodies against that strain will not neutralize strains in the other three serotypes and often enhance their replication. However recent studies suggest that there is variation in cross-immunity among DENV strains within a serotype (Wahala et al., 2010, Flipse & Smit, 2015, Katzelnick et al., 2015, Katzelnick et al., 2016, Waggoner et al., 2016, Forshey et al., 2016). Thus it was reasonable to hypothesize that the decrease in ultra-DHF mosquito infectivity could have been due to mutations in the structural proteins that enabled escape from antibodies raised against other strains of DENV-3 or against neutralization by heterologous antibody generated by other serotypes. During invasion in Vietnam, the Asian 1 DENV-2 strain acquired mutations at amino acid 160 in the envelope protein from K to either Q or M. Wang et al. (2015) tested whether the mutations at this site provided protection from heterologous antibodies (those raised against other DENV serotypes). They found, to their surprise, that these mutations actually increased neutralization by heterologous antibodies. In the current study, we found a change from serine to proline at amino acid 124 in the envelope protein between the pre-DHF and post-DHF strains, suggesting that the two strains may differ in their susceptibility to either homologous or heterologous immunity. This position has previously been identified as an informative (i.e. variable) site in DENV-3 (Wahala et al., 2010) and this region of E has been implicated in differential infectivity in tissue culture ((Lee et al., 2006), for example). However, we found no consistent difference in the amino acid sequence of the structural proteins between the post-DHF and ultra-DHF strains, thus ruling out immune escape as an explanation for the ascendancy of the ultra-DHF strain. Generally, the strains showed a high ratio of synonymous to non-synonymous variants (23:1), typical of arboviruses generally and dengue virus in particular, signaling the action of strong purifying selection on these viruses (Hanley & Weaver, 2008). In addition to documenting the pattern of mosquito infectivity in this system, we sought to discern mechanisms by which these strains of DENV may shape their interaction with their mosquito vector. Recent studies have shown that viruses can self-regulate tropism for specific host tissues and levels of replication within those tissues through interaction with RNAi (Guo & Steitz, 2014). RNAi is triggered by the presence of double-stranded RNA (dsRNA) within cells (Kingsolver et al., 2013) and serves as the major defense for mosquitoes against arboviruses (Blair, 2011). As a rule arboviruses possess RNA genomes that produce long dsRNA during replication (Westaway et al., 1999) as well as secondary structures, particularly in the untranslated regions (UTR) (Romero et al., 2006), that form regions of dsRNA. Changes in the genome sequence could potentially modify secondary structures and thereby alter stimulation of RNAi. Moreover, DENV possesses mechanisms to suppress RNAi. Schnettler et al. (2012) have shown that the DENV sfRNA, an approximately 400 nt subgenomic fragment encompassing the 3’ terminus of the 3’ UTR (Schnettler et al., 2012), downregulates RNAi. Additionally, Kakumani et al. (2013) have recently reported that the DENV non-structural 4B (NS4B) is a viral suppressor of RNAi. Thus mutations in the nucleotide or amino acid sequence of the 3’ UTR or NS4B, or elsewhere in the genome, could alter viral regulation of RNAi. We found that while the pre-DHF, post-DHF and ultra-DHF all replicate with similar dynamics in mosquito cells that lack RNAi, the replication of pre-DHF was significantly suppressed, relative to the other two strains, in mosquito cells with a functional RNAi pathway. These findings indicate that pre-DHF either stimulates RNAi more strongly or suppresses it less effectively than the other two strains, a distinction currently being investigated in our lab. This difference likely explains the greater mosquito infectivity of post-DHF compared to pre-DHF. However, post-DHF and ultra-DHF did not differ in their interaction with RNAi, suggesting that the difference in mosquito infectivity between these two strains was mediated by mechanisms other than RNAi. There is an important caveat to the conclusion that the DENV-3 strains studied here underwent a rise and fall in vector infectivity over the course of sequential competitive displacements. We utilized a colonized strain of Ae. aegypti rather than local Sri Lankan mosquitoes, because we were unable to obtain mosquitoes from Sri Lanka. It is well known that rates of mosquito infection by DENV are determined by interactions between host genotype and virus genotype (Lambrechts et al., 2013, Lambrechts, 2011, Lambrechts et al., 2009). However local adaptation of DENV strains to co-occurring mosquito populations has not been demonstrated, suggesting that colonized strains may serve as a useful model for wild populations. Moreover we also detected a significant difference in replication in the pre-DHF and post-DHF strains in U4.4 cultured cells derived from an entirely different vector species, Ae. albopictus, suggesting that the differences observed between the two in infection of NIH colony Ae. aegypti can be generalized. A more trivial limitation to the study is the use of artificial bloodmeals to infect mosquitoes. While arboviruses are less infectious when vectors imbibe an artificial meal rather than blood from a living host (Althouse & Hanley, 2015), there is no reason to suppose that the use of artificial bloodmeals, as opposed to human bloodmeals, influences relative infectivity of different arbovirus strains. In sum, the studies reported here have revealed that the displacement of the pre-DHF strain of DENV-3 by the post-DHF strain was likely mediated by greater infectivity of the latter for Ae. aegypti, that this difference in infectivity is associated with a difference in the interaction of the two strains with RNAi, the major mosquito defense against arboviruses, but that the two strains do not differ in EIP. In contrast, the ultra-DHF strain displaced the post-DHF strain despite a lower infectivity for Ae. aegypti. EIP and interaction with RNAi were similar between the two strains; moreover there were no differences in the coding sequence for the structural proteins between the two, indicating that the success of the ultra-DHF strain was not attributable to immune escape. This suggests that the ultra-DHF strain gained its competitive advantage over post-DHF from differences in the effects of the two strains on mosquito longevity or differences in replication in humans. We have reported that the replication of pre-DHF and post-DHF DENV-3 do not differ in human hepatoma cells (Hanley et al., 2008), but this finding cannot be generalized to replication in humans. Unfortunately there are at present no animal models that adequately recapitulate DENV replication and dengue disease (Cassetti et al., 2010). Thus it will be necessary in future to sample viremic humans and mosquitoes in dengue-endemic areas and employ whole-genome sequencing and phylogenetic analysis to further probe the role of hosts and vectors, as well as the trade-off between the two, in displacements among strains of this important pathogen. Acknowledgments The authors have no conflict of interest to declare. This work was supported by grants from the National Center for Research Resources (5P20RR016480-12) and the National Institute of General Medical Sciences (8 P20 GM103451-12) of the NIH, by NIH grant 1R15AI113628-01, and by the Howard Hughes Medical Institute Undergraduate Research Scholars program at NMSU (grants 52006932 and 52008103). We are extremely grateful to the lab of Aravinda de Silva at University of North Carolina for providing the virus isolates used in this research and to Dr. Stephen Whitehead at the National Institute of Allergy and Infectious Disease, NIH for providing antibodies, other reagents, and advice. Figure 1 Isolates of the post-DHF DENV-3 strain (black, solid) infected a significantly higher percentage of Aedes aegypti mosquitoes than isolates of the ultra-DHF DENV-3 strain (red, hatched) 14 days after engorgement with bloodmeals containing designated dengue virus isolates. Sample size (N) is reported in parentheses below isolate name. See text for statistical analysis. Figure 2 Mean percentage (plus or minus one standard error) of Aedes aegypti mosquitoes infected with (top panel) and generating a disseminated infection of (bottom panel) pre-DHF (blue squares), post-DHF (black circles), and ultra-DHF (red triangles) strains of DENV-3 at days 6, 10, and 14 after feeding on bloodmeals containing isolates of each strain. See text for statistical analysis. Figure 3 Replication kinetics of pre-DHF (solid blue square, circle and upright triangle), post-DHF (solid black diamond, crossed lines, and inverted triangle), and ultra-DHF (open, red symbols) strains in C6/36 (top panel) and U4.4 (bottom panel) cells. Note that y-axes are at different scales to enhance clarity. See text for statistical analysis. Table 1 Summary of studies to date that have investigated the relative fitness of displaced and displacing dengue virus strains in either mosquito vectors or human hosts following documented competitive displacements among strains of the same serotype. Geographic Region DENV Serotype Displacing genotype or strain Displaced strain Impact of displacement on disease Relative infectivity for Aedes aegypti Differences in other phenotypes Reference Thailand 1 genotype I New strain genotype Early strain None New > Early Not investigated (Lambrechts et al., 2012) South America 2 Southeast Asian genotype American genotype Increase in DHF Southeast Asian > American Replication of Southeast Asian genotype > American genotype in human monocytes and dendritic cells (Anderson & Rico-Hesse, 2006, Armstrong & Rico-Hesse, 2001, Armstrong & Rico-Hesse, 2003, Cologna et al., 2005, Rico-Hesse, 2010) Nicaragua 2 Southeast Asian genotype strain NI-2B Southeast Asian genotype strain NI-I Increase in severe disease NI-2B > NI-I Replication of NI-2B > NI-I in Ae. albopictus (C6/36) cells and human dendritic cells (Quiner et al., 2014, OhAinle et al., 2011) Vietnam, Cambodia and Thailand 2 Asian I genotype Asian/American genotype None Asian 1 = Asian/American Asian I associated with higher viremia than Asian/American in patient blood samples (Vu et al., 2010) Sri Lanka 3 genotype III post-DHF strain genotype III pre-DHF strain Increase in DHF post-DHF > pre-DHF Replication of post-DHF = pre-DHF in African green monkey kidney (Vero) cells and Ae. albopictus (C6/36) cells (Hanley et al., 2008) Table 2 Description of virus isolates utilized in the study DENV-3 isolates used in this study Isolate name used in previous studies Genotype/Clade Strain Passage History‡ 3002 83SriLan2*,SK0087† III/A Pre-DHF C6/36 (p5) 3009 89SriLan2*,SK0396† III/A Pre-DHF C6/36 (p4, p5) 3011 85SriLan*,073† III/A Pre-DHF C6/36 (p4) 3001 89SriLan1*,SK0389† III/B Post-DHF C6/36 (p4, p5) 3006 97SriLan1 III/B Post-DHF C6/36 (p6) 3010 93SriLan1*,SK0693† III/B Post-DHF C6/36 (p6) 3050 Not applicable III/SL-Post2000 Ultra-DHF C6/36 (p4) 3053 Not applicable III/SL-Post2000 Ultra-DHF C6/36 (p4) 3054 Not applicable III/SL-Post2000 Ultra-DHF C6/36 (p4) 3055 Not applicable III/SL-Post2000 Ultra-DHF C6/36 (p4) 3060 Not applicable III/SL-Post2000 Ultra-DHF C6/36 (p4) * Messer et al. (2003) † CDC ‡ in cases in which two passages were used, the passage in italics was used for the initial comparison of post-DHF and ultra-DHF and the remaining experiments utilized the other passage. Table 3 Infection and dissemination of designated isolates of the post-DHF and ultra-DHF strains of dengue virus serotype 3 in Aedes aegypti mosquitoes 14 days after engorgement on an artificial bloodmeal containing designated titers of virus. Strain Isolate No. Fed Bloodmeal titer pre-feeding* Bloodmeal titer post-feeding* % (No.) Infected Mean titer ± 1 SE† in infected bodies % (No.) Total dissemination % (No.) dissemination from infected bodies only Mean titer ± 1 SE† in infected heads Post-DHF 3001 24 7.5 7.0 38 (9) 2.6±0.3 8 (2) 22 (2) 1.9±0.8 3006 27 7.0 7.3 41 (11) 3.3±0.3 30 (8) 73 (8) 1.7±0.3 3010 25 7.2 7.0 32 (8) 2.8±0.3 20 (5) 63 (5) 2.6±0.2 Ultra-DHF 3050 25 7.0 6.0 8 (2) 3.0±0.1 8 (2) 100 (2) 2.2±0.3 3053 25 7.0 6.3 8 (2) 4.1±0.2 4 (1) 50 (1) 4.1±0.0 3054 18 7.3 6.7 22 (4) 3.6±0.4 11 (2) 50 (2) 3.3±0.6 3055 14 7.5 6.9 29 (4) 3.7±0.2 29 (4) 100 (4) 3.0±0.6 3060 27 7.0 7.1 11 (3) 3.4±0.6 11 (3) 100 (3) 2.1±0.4 * log10 PFU/ml † log10 PFU/mosquito Table 4 Infection of designated isolates of pre-DHF (blue text), post-DHF (black text) and ultra-DHF (red text) strains of dengue virus serotype 3 in Aedes aegypti mosquitoes at 6, 10 and 14 days after engorgement on an artificial bloodmeal containing designated titers of virus. 6 Days post feeding 10 Days post feeding 14 Days post feeding Strain Isolate Bloodmeal titer pre-feeding* Bloodmeal titer post-feeding* No. fed % (No.) Infected Mean titer ± 1 SE† in infected bodies No. fed % (No.) Infected Mean titer ± 1 SE† in infected bodies No. fed % (No.) Infected Mean titer ± 1 SE† in infected bodies Pre-DHF 3002 7.0 6.3 4 0 (0) - 5 0 (0) - 9 0 (0) - 3009 7.3 7.2 26 62 (16) 2.9 ± 0.2 33 61 (2) 3.0 ± 0.2 50 56 (28) 3.5 ± 0.2 3011 7.2 6.8 11 27 (3) 2.8 ± 0.6 12 8 (1) 4.0 ± 0.0 13 8 (1) 2.9 ± 0.0 Post-DHF 3001 7.1 7.0 4 50 (2) 3.0 ± 0.1 16 56 (9) 3.5 ± 0.4 16 63 (10) 4.2 ± 0.2 3006 7.4 7.0 6 67 (4) 2.3 ± 0.3 7 71 (5) 3.3 ± 0.5 11 36 (4) 3.0 ± 0.2 3010 7.5 7.2 6 0 (0) - 9 89 (8) 3.5 ± 0.3 18 94 (17) 3.6 ± 0.1 Ultra-DHF 3050 6.8 6.7 8 25 (2) 3.0 ± 1.0 8 50 (4) 2.9 ± 0.9 11 0 (0) - 3053 6.5 6.2 5 0 (0) - 7 43 (3) 2.6 ± 0.2 11 9 (1) 3.5 ± 0.0 3054 7.6 6.8 6 33 (2) 2.4 ± 0.4 7 29 (2) 2.1 ± 0.5 8 38 (3) 3.7 ± 0.3 3055 7.3 6.6 5 0 (0) - 5 20 (1) 3.2 ± 0.0 11 36 (4) 3.8 ± 0.5 3060 6.8 6.6 7 71 (5) 3.5 ± 0.6 6 33 (2) 3.0 ± 0.8 9 44 (4) 3.9 ± 0.3 * log10 PFU/ml † log10 PFU/mosquito Table 5 Dissemination of designated isolates of the pre-DHF (blue text), post-DHF (black text) and ultra-DHF (red text) strains of dengue virus serotype 3 in Aedes aegypti mosquitoes at 6, 10 and 14 days after engorgement on an artificial bloodmeal containing designated titers of virus. 6 Days post feeding 10 Days post feeding 14 Days post feeding Strain Isolate Bloodmeal titer pre-feeding* Bloodmeal titer* No. fed % (No.) Total dissemination Mean titer ± 1 SE† in infected heads No. fed % (No.) Total dissemination Mean titer ± 1 SE† in infected heads No. fed % (No.) Total Dissemination Mean titer ± 1 SE† in infected heads Pre-DHF 3002 7.0 6.3 4 0 (0) - 5 0 (0) - 9 0 (0) - 3009 7.3 7.2 26 31 (8) 1.9 ± 0.3 33 36 (12) 2.2 ± 0.2 50 50 (25) 2.4 ± 0.2 3011 7.2 6.8 11 18 (2) 1.4 ± 0.8 12 8 (1) 2.6 ± 0.0 13 8 (1) 2.0 ± 0.0 Post-DHF 3001 7.1 7.0 4 0 (0) - 16 38 (6) 2.3 ± 0.5 16 56 (9) 2.9 ± 0.3 3006 7.4 7.0 6 33 (2) 1.0 ± 0.1 7 57 (4) 2.8 ± 0.3 11 36 (4) 2.2 ± 0.6 3010 7.5 7.2 6 0 (0) - 9 78 (7) 2.5 ± 0.3 18 94 (17) 2.5 ± 0.2 Ultra-DHF 3050 6.8 6.7 8 13 (1) 1.6 ± 0.0 8 25 (2) 2.5 ± 1.5 11 0 (0) - 3053 6.5 6.2 5 0 (0) - 7 0 (0) - 11 9 (1) 3.1 ± 0.0 3054 7.6 6.8 6 0 (0) - 7 14 (1) 1.4 ± 0.0 8 68 (3) 2.8 ± 0.8 3055 7.3 6.6 5 0 (0) - 5 20 (1) 1.4 ± 0.0 11 36 (4) 2.8 ± 0.9 3060 6.8 6.6 7 57 (4) 2.2 ± 0.3 6 17 (1) 2.0 ± 0.0 9 44 (4) 3.3 ± 0.4 * log10 PFU/ml † log10 PFU/mosquito Table 6 Nucleotide and amino acid variants in each of the three structural genes of the dengue virus genome that distinguish the pre-DHF, post-DHF and ultra-DHF strains of dengue virus serotype 3; viral strain, i.e. those variants that were conserved within each strain but differed between at least one strain and the other two. Gene Nucleotide Position* Nucleotide Difference Amino acid Difference† Pre-DHF Post-DHF Ultra-DHF Capsid 243 a g g None pre-Membrane 408 t c c None 471 a g g None 558 g a a None 600 c t t None 654 t c c None 771 t c c None 784 t t c None 807 c t c None Envelope 883 t c c None 915 t c c None 924 t g g None 1083 c t t None 1210 t c c Ser – Pro 1371 t a a None 1620 a c c None 1803 a g g None 1842 t t c None 1845 c c t None 2055 a a g None 2091 c t t None 2130 g a a None 2217 c t c None 2307 c t t None * Nucleotides numbered from the start of the genome † Amino acids numbered from the start of the specified protein References Adams B Boots M How important is vertical transmission in mosquitoes for the persistence of dengue? Insights from a mathematical model Epidemics 2010 2 1 10 Allicock OM Lemey P Tatem AJ Pybus OG Bennett SN Mueller BA Phylogeography and population dynamics of dengue viruses in the Americas. Mol Biol Evol 2012 29 1533 43 22319149 Althouse BM Hanley KA The tortoise or the hare? Impacts of within-host dynamics on transmission success of arthropod-borne viruses. Philos Trans R Soc Lond B Biol Sci 2015 370 Anderson JR Rico-Hesse R Aedes aegypti vectorial capacity is determined by the infecting genotype of dengue virus. Am J Trop Med Hyg 2006 75 886 92 17123982 Andrade CC Maharaj PD Reisen WK Brault AC North American West Nile virus genotype isolates demonstrate differential replicative capacities in response to temperature. J Gen Virol 2011 92 2523 33 21775581 Armstrong PM Rico-Hesse R Differential susceptibility of Aedes aegypti to infection by the American and Southeast Asian genotypes of dengue type 2 virus. Vector Borne Zoonotic Dis 2001 1 159 68 12680353 Armstrong PM Rico-Hesse R Efficiency of dengue serotype 2 virus strains to infect and disseminate in Aedes aegypti. Am J Trop Med Hyg 2003 68 539 44 12812340 Bhatt S Gething PW Brady OJ Messina JP Farlow AW Moyes CL The global distribution and burden of dengue. Nature 2013 496 504 7 23563266 Blair CD Mosquito RNAi is the major innate immune pathway controlling arbovirus infection and transmission. Future Microbiol 2011 6 265 77 21449839 Brackney DE Scott JC Sagawa F Woodward JE Miller NA Schilkey FD C6/36 Aedes albopictus cells have a dysfunctional antiviral RNA interference response. PLoS Negl Trop Dis 2010 4 e856 21049065 Carrillo-Valenzo E Danis-Lozano R Velasco-Hernandez JX Sanchez-Burgos G Alpuche C Lopez I Evolution of dengue virus in Mexico is characterized by frequent lineage replacement. Arch Virol 2010 155 1401 12 20549264 Carrington LB Nguyen HL Nguyen NM Duong TH Tuan TV Giang NT Naturally-acquired dengue virus infections do not reduce short-term survival of infected Aedes aegypti from Ho Chi Minh City, Vietnam. Am J Trop Med Hyg 2015 92 492 6 25561566 Carrington LB Simmons CP Human to mosquito transmission of dengue viruses. Front Immunol 2014 5 290 24987394 Cassetti MC Durbin A Harris E Rico-Hesse R Roehrig J Rothman A Report of an NIAID workshop on dengue animal models. Vaccine 2010 28 4229 34 20434551 Christofferson RC Mores CN Estimating the magnitude and direction of altered arbovirus transmission due to viral phenotype. PLoS One 2011 6 e16298 21298018 Ciota AT Kramer LD Insights into arbovirus evolution and adaptation from experimental studies. Viruses 2010 2 2594 617 21994633 Ciota AT Kramer LD Vector-virus interactions and transmission dynamics of West Nile virus. Viruses 2013 5 3021 47 24351794 Ciota AT Lovelace AO Ngo KA Le AN Maffei JG Franke MA Cell-specific adaptation of two flaviviruses following serial passage in mosquito cell culture. Virology 2007 357 165 74 16963095 Coffey LL Forrester N Tsetsarkin K Vasilakis N Weaver SC Factors shaping the adaptive landscape for arboviruses: implications for the emergence of disease. Future Microbiol 2013 8 155 76 23374123 Cologna R Armstrong PM Rico-Hesse R Selection for virulent dengue viruses occurs in humans and mosquitoes. J Virol 2005 79 853 9 15613313 dos Santos FB Nogueira FB Castro MG Nunes PC de Filippis AM Faria NR First report of multiple lineages of dengue viruses type 1 in Rio de Janeiro, Brazil. Virol J 2011 8 387 21813012 Drumond BP Mondini A Schmidt DJ Bosch I Nogueira ML Population dynamics of DENV-1 genotype V in Brazil is characterized by co-circulation and strain/lineage replacement. Arch Virol 2012 157 2061 73 22777179 Duong V Henn MR Simmons C Ngan C Y B Gavotte L Complex dynamic of dengue virus serotypes 2 and 3 in Cambodia following series of climate disasters. Infect Genet Evol 2013a 15 77 86 22677620 Duong V Lambrechts L Paul RE Ly S Lay RS Long KC Asymptomatic humans transmit dengue virus to mosquitoes. Proc Natl Acad Sci U S A 2015 112 14688 93 26553981 Duong V Simmons C Gavotte L Viari A Ong S Chantha N Genetic diversity and lineage dynamic of dengue virus serotype 1 (DENV-1) in Cambodia. Infect Genet Evol 2013b 15 59 68 21757030 Durbin AP Karron RA Sun W Vaughn DW Reynolds MJ Perreault JR Attenuation and immunogenicity in humans of a live dengue virus type-4 vaccine candidate with a 30 nucleotide deletion in its 3′-untranslated region. Am J Trop Med Hyg 2001 65 405 13 11716091 Elliot SL Adler FR Sabelis MW How virulent should a parasite be to its vector? . Ecology 2003 84 2568 2574 Ewald PW Vectors, vertical transmission, and the evolution of virulence. Evolution of infectious disease 1993 35 56 Oxford University Press Oxford, UK. Flipse J Smit JM The Complexity of a Dengue Vaccine: A Review of the Human Antibody Response. PLoS Negl Trop Dis 2015 9 e0003749 26065421 Forshey BM Reiner RC Olkowski S Morrison AC Espinoza A Long KC Incomplete Protection against Dengue Virus Type 2 Re-infection in Peru. PLoS Negl Trop Dis 2016 10 e0004398 26848841 Grunnill M Boots M How Important is Vertical Transmission of Dengue Viruses by Mosquitoes (Diptera: Culicidae)? J Med Entomol 2015 Guo YE Steitz JA Virus meets host microRNA: the destroyer, the booster, the hijacker. Mol Cell Biol 2014 34 3780 7 25047834 Guzman MG Alvarez M Halstead SB Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: an historical perspective and role of antibody-dependent enhancement of infection. Arch Virol 2013 158 1445 59 23471635 Guzman MG Halstead SB Artsob H Buchy P Farrar J Gubler DJ Dengue: a continuing global threat. Nat Rev Microbiol 2010 8 S7 16 21079655 Hadinegoro SR Arredondo-Garcia JL Capeding MR Deseda C Chotpitayasunondh T Dietze R Efficacy and Long-Term Safety of a Dengue Vaccine in Regions of Endemic Disease. N Engl J Med 2015 373 1195 206 26214039 Hanley KA Nelson JT Schirtzinger EE Whitehead SS Hanson CT Superior infectivity for mosquito vectors contributes to competitive displacement among strains of dengue virus. BMC Ecol 2008 8 1 18269771 Hanley KA Weaver SC Domingo E Parrish C Holland JF Arbovirus evolution. Origin and evolution of viruses 2008 351 393 Elsevier St. Louis, MO. Hanley KA Weaver SC Frontiers in Dengue Virus Research 2010 Caister Academic Press Hardy JL Houk EJ Kramer LD Reeves WC Intrinsic factors affecting vector competence of mosquitoes for arboviruses. Annu Rev Entomol 1983 28 229 62 6131642 Kakumani PK Ponia SS S RK Sood V Chinnappan M Banerjea AC Role of RNA interference (RNAi) in dengue virus replication and identification of NS4B as an RNAi suppressor. J Virol 2013 87 8870 83 23741001 Kanakaratne N Wahala WM Messer WB Tissera HA Shahani A Abeysinghe N Severe dengue epidemics in Sri Lanka, 2003-2006. Emerg Infect Dis 2009 15 192 9 19193262 Katzelnick LC Fonville JM Gromowski GD Bustos Arriaga J Green A James SL Dengue viruses cluster antigenically but not as discrete serotypes. Science 2015 349 1338 43 26383952 Katzelnick LC Montoya M Gresh L Balmaseda A Harris E Neutralizing antibody titers against dengue virus correlate with protection from symptomatic infection in a longitudinal cohort. Proc Natl Acad Sci U S A 2016 113 728 33 26729879 Kingsolver MB Huang Z Hardy RW Insect antiviral innate immunity: pathways, effectors, and connections. J Mol Biol 2013 425 4921 36 24120681 Lambrechts L Quantitative genetics of Aedes aegypti vector competence for dengue viruses: towards a new paradigm? Trends Parasitol 2011 27 111 4 21215699 Lambrechts L Chevillon C Albright RG Thaisomboonsuk B Richardson JH Jarman RG Genetic specificity and potential for local adaptation between dengue viruses and mosquito vectors. BMC Evol Biol 2009 9 160 19589156 Lambrechts L Fansiri T Pongsiri A Thaisomboonsuk B Klungthong C Richardson JH Dengue-1 virus clade replacement in Thailand associated with enhanced mosquito transmission. J Virol 2012 86 1853 61 22130539 Lambrechts L Quillery E Noel V Richardson JH Jarman RG Scott TW Specificity of resistance to dengue virus isolates is associated with genotypes of the mosquito antiviral gene Dicer-2. Proc Biol Sci 2013 280 20122437 23193131 Lee E Wright PJ Davidson A Lobigs M Virulence attenuation of Dengue virus due to augmented glycosaminoglycan-binding affinity and restriction in extraneural dissemination. J Gen Virol 2006 87 2791 801 16963737 Leger P Lara E Jagla B Sismeiro O Mansuroglu Z Coppee JY Dicer-2- and Piwi-mediated RNA interference in Rift Valley fever virus-infected mosquito cells. J Virol 2013 87 1631 48 23175368 Lequime S Lambrechts L Vertical transmission of arboviruses in mosquitoes: a historical perspective. Infect Genet Evol 2014 28 681 90 25077992 Lourenco J Recker M Viral and epidemiological determinants of the invasion dynamics of novel dengue genotypes. PLoS Negl Trop Dis 2010 4 e894 21124880 Messer WB Gubler DJ Harris E Sivananthan K de Silva AM Emergence and global spread of a dengue serotype 3, subtype III virus. Emerg Infect Dis 2003 9 800 9 12899133 Messer WB Vitarana UT Sivananthan K Elvtigala J Preethimala LD Ramesh R Epidemiology of dengue in Sri Lanka before and after the emergence of epidemic dengue hemorrhagic fever. Am J Trop Med Hyg 2002 66 765 73 12224589 Moudy RM Meola MA Morin LL Ebel GD Kramer LD A newly emergent genotype of West Nile virus is transmitted earlier and more efficiently by Culex mosquitoes. Am J Trop Med Hyg 2007 77 365 70 17690414 Myat Thu H Lowry K Jiang L Hlaing T Holmes EC Aaskov J Lineage extinction and replacement in dengue type 1 virus populations are due to stochastic events rather than to natural selection. Virology 2005 336 163 72 15892958 Nguyet MN Duong TH Trung VT Nguyen TH Tran CN Long VT Host and viral features of human dengue cases shape the population of infected and infectious Aedes aegypti mosquitoes. Proc Natl Acad Sci U S A 2013 110 9072 7 23674683 OhAinle M Balmaseda A Macalalad AR Tellez Y Zody MC Saborio S Dynamics of dengue disease severity determined by the interplay between viral genetics and serotype-specific immunity. Sci Transl Med 2011 3 114ra128 Pal R The establishment of reference and marker strains and their shipment. Bull World Health Organ 1967 36 583 5 5299456 Pozo-Aguilar JO Monroy-Martinez V Diaz D Barrios-Palacios J Ramos C Ulloa-Garcia A Evaluation of host and viral factors associated with severe dengue based on the 2009 WHO classification. Parasit Vectors 2014 7 590 25500154 Quiner CA Parameswaran P Ciota AT Ehrbar DJ Dodson BL Schlesinger S Increased replicative fitness of a dengue virus 2 clade in native mosquitoes: potential contribution to a clade replacement event in Nicaragua. J Virol 2014 88 13125 34 25187539 Rico-Hesse R Microevolution and virulence of dengue viruses. Adv Virus Res 2003 59 315 41 14696333 Rico-Hesse R Dengue virus evolution and virulence models. Clin Infect Dis 2007 44 1462 6 17479944 Rico-Hesse R Dengue virus virulence and transmission determinants. Curr Top Microbiol Immunol 2010 338 45 55 19802577 Romero TA Tumban E Jun J Lott WB Hanley KA Secondary structure of dengue virus type 4 3′ untranslated region: impact of deletion and substitution mutations. J Gen Virol 2006 87 3291 6 17030863 Rothman AL Immunity to dengue virus: a tale of original antigenic sin and tropical cytokine storms. Nat Rev Immunol 2011 11 532 43 21760609 Salazar MI Richardson JH Sanchez-Vargas I Olson KE Beaty BJ Dengue virus type 2: replication and tropisms in orally infected Aedes aegypti mosquitoes. BMC Microbiol 2007 7 9 17263893 Schnettler E Sterken MG Leung JY Metz SW Geertsema C Goldbach RW Noncoding flavivirus RNA displays RNA interference suppressor activity in insect and Mammalian cells. J Virol 2012 86 13486 500 23035235 Schuh AJ Ward MJ Leigh Brown AJ Barrett AD Dynamics of the emergence and establishment of a newly dominant genotype of Japanese encephalitis virus throughout Asia. J Virol 2014 88 4522 32 24501419 Simmons CP Farrar JJ Nguyen v, V. Wills B Dengue. N Engl J Med 2012 366 1423 32 22494122 Snow GE Haaland B Ooi EE Gubler DJ Review article: Research on dengue during World War II revisited. Am J Trop Med Hyg 2014 91 1203 17 25311700 Teoh BT Sam SS Tan KK Johari J Shu MH Danlami MB Dengue virus type 1 clade replacement in recurring homotypic outbreaks. BMC Evol Biol 2013 13 213 24073945 Troyer JM Hanley KA Whitehead SS Strickman D Karron RA Durbin AP A live attenuated recombinant dengue-4 virus vaccine candidate with restricted capacity for dissemination in mosquitoes and lack of transmission from vaccinees to mosquitoes. Am J Trop Med Hyg 2001 65 414 9 11716092 Turell MJ Reduced Rift Valley fever virus infection rates in mosquitoes associated with pledget feedings. Am J Trop Med Hyg 1988 39 597 602 3207178 Vu TT Holmes EC Duong V Nguyen TQ Tran TH Quail M Emergence of the Asian 1 genotype of dengue virus serotype 2 in viet nam: in vivo fitness advantage and lineage replacement in South-East Asia. PLoS Negl Trop Dis 2010 4 e757 20651932 Waggoner JJ Balmaseda A Gresh L Sahoo MK Montoya M Wang C Homotypic Dengue Virus Reinfections in Nicaraguan Children. J Infect Dis 2016 Wahala WM Donaldson EF de Alwis R Accavitti-Loper MA Baric RS de Silva AM Natural strain variation and antibody neutralization of dengue serotype 3 viruses. PLoS Pathog 2010 6 e1000821 20333252 Wang C Katzelnick LC Montoya M Hue KD Simmons CP Harris E Evolutionarily Successful Asian 1 Dengue Virus 2 Lineages Contain One Substitution in Envelope That Increases Sensitivity to Polyclonal Antibody Neutralization. J Infect Dis 2015 Wattam AR Christensen BM Variation in Aedes aegypti mRNA populations related to strain, sex, and development. Am J Trop Med Hyg 1992 47 702 7 1449211 Weaver SC Vasilakis N Molecular evolution of dengue viruses: contributions of phylogenetics to understanding the history and epidemiology of the preeminent arboviral disease. Infect Genet Evol 2009 9 523 40 19460319 Westaway EG Khromykh AA Mackenzie JM Nascent flavivirus RNA colocalized in situ with double-stranded RNA in stable replication complexes. Virology 1999 258 108 17 10329573 Whitehorn J Kien DT Nguyen NM Nguyen HL Kyrylos PP Carrington LB Comparative Susceptibility of Aedes albopictus and Aedes aegypti to Dengue Virus Infection After Feeding on Blood of Viremic Humans: Implications for Public Health. J Infect Dis 2015 212 1182 90 25784733 WHO Dengue: guidelines for diagnosis, treatment, prevention and contro 2009 World Health Organization

PMC005xxxxxx/PMC5117189.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985193R 5098 J Org Chem J. Org. Chem. The Journal of organic chemistry 0022-3263 1520-6904 26360634 5117189 10.1021/acs.joc.5b01703 NIHMS819262 Article Synthesis of Naamidine A and Selective Access to N2-Acyl-2-aminoimidazole Analogues Gibbons Joseph B. † Salvant Justin M. † Vaden Rachel M. † Kwon Ki-Hyeok † Welm Bryan E. ‡ Looper Ryan E. *† † Department of Chemistry, University of Utah, 315 South 1400 East, Salt Lake City, Utah 84112, United States ‡ Immunobiology and Cancer Program, Oklahoma Medical Research Foundation, 825 Northeast 13th Street, Oklahoma City, Oklahoma 73104, United States * Corresponding Author: r.looper@utah.edu 2 11 2016 24 9 2015 16 10 2015 21 11 2016 80 20 1007610085 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. A short and scalable synthesis of naamidine A, a marine alkaloid with a selective ability to inhibit epidermal growth factor receptor (EGFR)-dependent cellular proliferation, has been achieved. A key achievement in this synthesis was the development of a regioselective hydroamination of a monoprotected propargylguanidine to deliver N3-protected cyclic ene-guanidines. This permits the extension of this methodology to prepare N2-Acyl analogues in a fashion that obviates the troublesome acylation of the free 2-aminoimidazoles, which typically yields mixtures of N2- and N2,N2-diacylated products. Graphical abstract INTRODUCTION Marine sponges from the Leucetta family have produced a wealth of natural products comprising highly functionalized 2-aminoimidazoles (2-AIs).1 This family of alkaloids effects a number of diverse biological activities (Figure 1). Naamine D (1), for example, has been shown to be a moderate inhibitor of iNOS, an isozyme scrutinized for its involvement in a number of diseases.2 Naamine D was also shown to be active against the opportunistic pathogen in AIDS patients, Cryptococcus neoformas (MIC = 6.25 μg/mL). N,N-Dimethylnaamine D (2) was active against an antimicrobial panel consisting of Escherichia coli, Staphylococcus aureus, Bacillus subtilis and Candida albicans.3 Kealiinine B (3) was recently reported to show antiproliferative activity (IC50 ~ 10 μM) against the breast cancer cell line T47D, while other kealiinine analogues have displayed modest activity against MCF-7 proliferation.4,5 Isonaamine C (4) was found to be cytotoxic to a variety of cell lines,6 while leucettamine A was found to be a leukotriene B4 (LTB4) antagonist.7 These examples clearly demonstrate that the 2-aminoimidazole, bearing a variety of substitution patterns, serves as an important heterocyclic scaffold for small-molecule drug discovery. Our interest in this family stems from the selective cytotoxicity of naamidine A (6). Studies by Ireland and co-workers determined 6 to be a selective inhibitor for EGF-mediated growth in epidermal growth factor receptor (EGFR) transfected NIH3T3 cells (IC50 = 11.3 μM) yet displayed a 21-fold decrease in potency against insulin-mediated growth (IC50 = 242 μM).8 This particular selectivity prompted in vivo studies, where nude mice xenografts of EGF-overexpressing A431 epidermal carcinoma displayed 87.4% tumor growth inhibition when treated with 6 at 25 mg/kg. Although many compounds affect EGFR signaling, 6 is the first known example to stimulate phosphotransferase activity of extracellular regulated kinases ERK1/2.9 This sustained increase in MAPK activity has been shown to be a result of naamidine A-induced expression of p21, leading to inhibition of cyclin-dependent kinase activity and activation of caspases 3, 8, and 9.10 Since the EGFR signaling pathway is overexpressed in many human tumors, the ability to selectively inhibit EGFR-mediated proliferation represents an important strategy for new chemotherapeutics. Herein, we report the synthesis of 6, as well as related analogues via regioselective construction of cyclic ene-guanidines. RESULTS AND DISCUSSION The structural novelty of 6 and other highly substituted 2-AI scaffolds has generated interest in several synthetic laboratories.11–16 We previously reported the synthesis of naamine A (8) via an addition–hydroamination–isomerization sequence utilizing the propargylcyanamide 7 (Scheme 1).17 Analogous to the syntheses of 6 by Ohta and Watson, we were able to add the N-Me-dehydrohydantoin selectively to N2 via silylated N-methylparabanic acid (Scheme 1).11,12 However, the transamination reaction of the piperidinone to the free 2-aminoimidazole proved problematic on larger scales. We had also simultaneously discovered the tandem addition–hydroamination sequence that was reported by Van der Eycken employing N,N-di-Boc guanidines. Removal of the Boc groups with TFA in this sequence presented problems, as cleavage of electron-rich groups at N1 was quite facile under acidic conditions (e.g., those needed for the synthesis of isonaamine C). Furthermore, while trying to access simplified naamidine A analogues, exemplified by the reaction of 9 with 2-fluorobenzoyl chloride, we obtained an unfavorable 1:2 mixture of mono-acylated and diacylated N2 products (10/11). A recent report by Jiang and co-workers identified the same problem, requiring forcing conditions or extra protecting group manipulations to obtain the monoacyl-2-aminoimidazoles in low to moderate yields, reinforcing the need for a high-yielding and selective strategy to access monosubstituted N2-Acyl-2-aminoimidazoles.18 These shortcomings necessitated a revised synthesis of 6 that would allow for (a) reproducible and scalable procedures, (b) the presence of acid labile groups, and (c) differential protection of N2/N3 for selective functionalization. Our attempts to address these issues are presented in the synthesis of naamidine A (Scheme 2). Cu(I)-mediated A3-coupling of the required amine, alkyne, and aldehyde gave 12 (Scheme 2).19 Deallylation with Pd(0) gave the secondary propargylamine (13) in good yield.10 Instead of installing the di-Boc guanidine,14 we prepared the monoacylguanidine 14 using the activated Cbz-cyanamide potassium salt guanylation conditions previously developed in our laboratory.20,21 It is important to note that four pathways are operable in the cyclization of 14: N3- versus N2-cyclization and 5-exo-dig versus 6-endo-dig cyclization. We knew that monoacylguandines prefer the tautomeric form in which the imino tautomer is directly conjugated with the acyl group and the other nitrogen forms a hydrogen bond to the carbonyl (as depicted in 10). This would suggest that the unconjugated nonbonding lone pair on N3 would initiate cyclization. We also knew that the Ag(I)-catalyzed cyclization proceeds preferentially in a 5-exo-dig fashion; however, with an electron-rich alkyne substituent, selectivity can be significantly diminished. For example, when unsubstituted at C5, p-MeOPh-substituted propargylguanidines cyclize with only modest 5-exo-dig selectivity of ~2:1.14 To our delight, treatment of 14 with AgNO3 in CH2Cl2 provided a single isomer (15) in 87% isolated yield. The regioselectivity of this process was ultimately supported by X-ray crystallography of the intermediate 18b (Figure 2). Importantly, this leaves N2 open for subsequent functionalization. The Cbz group is readily cleaved under standard hydrogenolysis conditions. Fortunately, isomerization of the exocyclic alkene provides the 2-aminoimidazole nucleus before it can be reduced. The benzyl ether is also cleaved during this step to provided naamine A in quantitative yield. Again the N-Me-hydantoin can be installed by Ohta’s method to provide naamidine A in good yield. This sequence has proven to be robust and scalable, delivering gram quantities of naamidine A in six steps and 33% overall yield. With the ability to control the regioselectivity of the monoacylpropargylguanidine cyclization, we returned our attention to generating N2-substituted analogues (Scheme 3). We envisioned 15 as an ideal intermediate for selective N2-Acylation, as N3 is protected and the imino tautomer is forced C2=N2 and should yield only monoacylation products. Indeed, both electron-rich and electron-poor aryl chlorides gave the N2-monoacylguanidines 16a–d in excellent yields (Scheme 3). Alkanoyl chlorides are also reactive to give 16e,f. Most notably, hydrogenation conditions that cleaved the Cbz group, isomerized the ene-guanidine, and cleaved the phenolic benzylether in the preparation of 8 resulted in no reaction in the conversion of 16a → 17a. More forcing conditions (elevated H2 pressures) could initiate reductive cleavage of the Cbz group and isomerization, but the benzyl ether was surprisingly difficult to cleave. Ultimately, a nonsupported Pd(II) catalyst was successful, providing the fully deprotected targets 17a–f in excellent overall yields. To complement our focused library, the same methodology was involved in the construction of C5-phenyl and C4-benzyl analogues (Scheme 4). The same synthetic sequence accessing 15 was also employed to prepare substrates 18a and 18b.19 Acylation of these intermediates gave N2-substituted precursors 19a–h in excellent yields. Hydrogenation over palladium on carbon, with 60 psi H2, was sufficient to effect the deprotection with isomerization and deliver the 2-aminoimidazole analogues 20a–h. Confirmation of N2-selective acylation was confirmed by X-ray crystallography of 20h. Again, 18b was characterized by X-ray crystallography, confirming that the initial hydroamination proceeds to give the N3-protected intermediates (Figure 2). The fact that the acylation/deprotection with isomerization sequence yields the mono-N2-substituted 2-aminoimidazoles was ultimately confirmed by X-ray crystallography of product 20h. This structure shows that even in the now aromatized amino-imidazole nucleus, the exocyclic N2-imino tautomer is preferred with H bonding between N3 and the N2-Acyl group with a C2–N3 imino bond length of 1.33 Å. Studies to evaluate the cytotoxicity of 17a–e and 20a–i revealed that 20h was effective against metastatic tumor cells derived from a chemoresistant breast cancer patient (PE1007070 cells) with an EC50 = 8.8 μM.22 Moreover, 20h did not significantly affect the viability of immortalized, nontumorogenic mammary tissue (hTERT-HMEC cells), suggesting a cancer-specific mechanism of action. The effect of 20h on cell viability was also measured in a breast cancer cell line (MCF-7) and an untransformed mammary epithelial cell line (MCF-10A) (Figure 3). As with the patient-derived cells, 20h was found to significantly reduce the viability of MCF-7 cells (EC50 = 1.4 μM) while having no significant effect on the untransformed mammary cell line. Despite the reported selectivity of naamidine A (6) to inhibit proliferation in EGFR transfected NIH3T3 cells, no selectivity was observed in the antiproliferative activity of MCF-7 versus MCF-10A cells (EC50 = 5.9 and 8.1 μM, respectively). Taken together, these results suggest that the natural-product-inspired N2-Acyl-2-aminoimidazoles can exploit cancer-selective mechanisms to cause cell death. Studies to further understand this mechanism and evaluate its therapeutic potential are underway. CONCLUSION In summary, we have shown that monoacylguanidines preferentially adopt the N-acylimino tautomer, and that this can be reliably used to predict reactivity. Thus, the hydroamination of monoacylpropargylguanidines can be effected regioselectively to generate N3-acyl-2-aminoimidazoles and subsequently free 2-aminoimidazoles after deprotection with isomerization of the Cbz-protected variants. This strategy further exploits the confined N2-imino tautomer to allow selective N2-Acylation and deliver these analogues without contamination from the diacylated derivatives. The effectiveness of this approach was demonstrated by completing a gram-scale synthesis of both naamine A and naamidine A. The discovery of 20h as a more selective antiproliferative agent than naamidine A highlights the necessity to efficiently prepare mono-N2-Acyl-2-aminoimidazoles to further study this selectivity. EXPERIMENTAL SECTION General Considerations All reactions requiring anhydrous conditions were performed under a positive pressure of nitrogen using flame-dried glassware. Acetonitrile (MeCN), dichloromethane (CH2Cl2), and toluene (PhMe) were degassed with nitrogen and passed through activated alumina. Methanol (MeOH) and triethylamine (Et3N) were distilled from CaH2 immediately prior to use. Reactions were monitored to completion by TLC and visualized by a dual short-/long-wave UV lamp and stained with an aqueous solution of potassium permanganate and/or organic solution of phosphomolybdic acid. Flash chromatography was performed on silica gel Siliaflash P60 (40–63 μm). Infrared spectra were recorded as thin films, and absorptions are reported in cm−1 relative to polystyrene (1601 cm−1); HRMS mass spectra were determined by ESI/APCI-TOF. 1H NMR and spectra were recorded on 500 and 300 MHz spectrometers as indicated. The chemical shifts (δ) of proton resonances were reported relative to the deuterated solvent peak (7.26 ppm for CDCl3, 3.31 for CD3OD, and 2.50 ppm for DMSO-d6) using the following format: chemical shift [multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet), coupling constant(s) (J in Hz), integral; 13C NMR spectra were recorded at 125 and 75 MHz. The chemical shifts (δ) of carbon resonances were reported relative to the deuterated solvent peak (77.2 ppm for CDCl3 and 39.5 for DMSO-d6). Procedures for the Synthesis of Naamidine A N-Allyl-1-(4-(benzyloxy)phenyl)-4-(4-methoxyphenyl)-N-methylbut-3-yn-2-amine (12) To a 500 mL pressure flask equipped with a stir bar were added 4-methoxyphenylacetylene (5.35 mL, 40.5 mmol), N-allylmethylamine (3.46 mL, 36.4 mmol), p-OBn-phenylacetaldehyde (9.2 g, 40.5 mmol), CuBr (0.52 g, 3.6 mmol), acetonitrile (140 mL), and 1 g of oven-dried 4 Å molecular sieves. The flask was heated at 80 °C for 24 h and then allowed to cool to room temperature. The mixture was filtered through Celite and rinsed with EtOAc (500 mL). The organic layer was washed with aqueous solutions of saturated NaHCO3 (500 mL) and brine (500 mL). The organic layer was dried over Na2SO4. After filtration, the organic layer was concentrated and purified via flash chromatography using 4:1 hexanes/EtOAc to give 8 as a dark red oil (10.8 g, 65%): Rf = 0.35 (4:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.37−7.27 (m, 4H), 7.24 (d, J = 8.8 Hz, 2H), 7.24 (overlapped, 1H), 7.15 (d, J = 8.3 Hz, 2H), 6.84 (d, J = 8.8 Hz, 2H), 6.73 (d, J = 8.8 Hz, 2H), 5.78 (ddt, J = 6.4, 10.7, 17.1 Hz, 1H), 5.14 (dd, J = 1.5, 17.1 Hz, 1H), 5.05 (dd, J = 2.0, 10.3 Hz, 1H), 4.95 (s, 2H), 3.72 (dd, J = 6.4, 8.8 Hz, 1H), 3.70 (s, 3H), 3.15 (dd, J = 5.9, 13.7 Hz, 1H), 3.03 (dd, J = 7.3, 13.5 Hz, 1H), 2.27 (s, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 159.2, 157.4, 137.2, 136.0, 133.0 131.2, 130.4, 128.5, 127.8, 127.4, 117.6, 115.5, 114.5, 113.8, 88.5, 85.0, 69.9, 58.4, 58.2, 55.2, 39.5, 37.7 ppm; IR (thin film) 2954, 1606, 1508, 1454, 1420, 1381, 1289, 1243, 1173, 1106, 1026, 921, 831, 807, 791, 732, 696 cm−1; HRMS (ESI+) calcd for C28H30NO2 m/z (M + H) 412.2277, found 412.2278. 1-(4-(Benzyloxy)phenyl)-4-(4-methoxyphenyl)-N-methylbut-3-yn-2-amine (13) To a 500 mL round-bottom flask equipped with a stir bar were added 12 (10.7 g, 26.0 mmol), thiosalicylic acid (8.0 g, 52 mmol), Pd(PPh3)4 (0.6 g, 0.5 mmol), and CH2Cl2 (260 mL). The reaction was allowed to stir at room temperature under N2 overnight. The reaction mixture was concentrated and redissolved in EtOAc (200 mL). The organic layer was washed with saturated NaHCO3 (200 mL) and brine (200 mL). The organic layer was dried over Na2SO4. After filtration, the organic layer was concentrated and purified via flash chromatography using 100% EtOAc (with 0.5% Et3N) to give 13 as an orange oil (6.6 g, 91%): Rf = 0.35 (100% EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.45 (d, J = 7.3, 2H), 7.40 (t, J = 6.8 Hz, 2H), 7.34 (d, J = 8.8 Hz, 3H), 7.27 (d, J = 8.3 Hz, 2H), 6.95 (d, J = 8.8 Hz, 2H), 6.83 (d, J = 8.8 Hz, 2H), 5.06 (s, 2H), 3.80 (s, 3H), 3.72 (t, J = 6.4 Hz, 1H), 2.98 (dd, J = 2.4, 9.4 Hz, 2H), 2.55 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 159.4, 157.7, 137.2, 133.0, 130.8, 128.7, 128.0, 127.6, 115.5, 114.7, 88.7, 84.6, 70.1, 55.3, 53.9, 41.3, 34.2 ppm; IR (thin film) 2933, 1606, 1508, 1454, 1441, 1380, 1289, 1244, 1173, 1107, 1027, 831, 737, 697, 668 cm−1; HRMS (ESI+) calcd for C25H26NO2 m/z (M + H) 372.1964, found 372.1966. N-Cbz-1-(1-(4-(benzyloxy)phenyl)-4-(4-methoxyphenyl)but-3-yn-2-yl)-1-methylguanidine (14) To a 250 mL round-bottom flask equipped with a stir bar were added TMSCl (1.65 mL, 13.0 mmol), benzyloxycarbonylcyanamide potassium salt (2.58 g, 12.0 mmol) and 50 mL of acetonitrile. The reaction mixture was allowed to stir for 10 min under N2. A solution of 13 (4.8 g, 13 mmol) in acetonitrile (15 mL) was added to the suspension, and the reaction was allowed to stir for 1 h. The reaction mixture was concentrated to approximately one-quarter of the original volume and then diluted with EtOAc (100 mL). The organic layer was washed with aqueous solutions of saturated Na2CO3 (100 mL) and brine (100 mL). The organic layer was dried over Na2SO4. After filtration, the organic layer was concentrated and purified via flash chromatography using 1:1 hexanes/EtOAc to give 14 as a yellow foam (5.9 g, 90%): Rf = 0.42 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.44 (d, J = 7.3 Hz, 4H), 7.42−7.27 (m, 8H), 7.20 (d, J = 8.8 Hz, 2H), 6.90 (d, J = 8.3 Hz, 2H), 6.83 (d, J = 8.8 Hz, 2H), 6.02 (br s, 2H), 5.16 (d, J = 2.4 Hz, 2H), 5.03 (s, 2H), 3.80 (s, 3H), 3.04 (dd, J = 7.3, 13.2 Hz, 1H), 2.95 (dd, J = 6.4, 13.2 Hz, 1H), 2.90 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 173.1, 164.0, 160.7, 159.8, 157.9, 137.8, 137.1, 133.2, 130.7, 129.1, 128.7, 128.4, 128.0, 127.9, 127.7, 114.8, 114.0, 86.1, 84.9, 70.1, 66.8, 55.4, 50.2, 39.7 ppm; IR (thin film) 2934, 1642, 1589, 1536, 1508, 1440, 1378, 1280, 1244, 1172, 1152, 1107, 1026, 909, 831, 799, 732, 696 cm−1; HRMS (ESI+) calcd for C34H34N3O4 m/z (M + H) 548.2549, found 548.2556. Benzyl (Z)-4-(4-(Benzyloxy)benzyl)-2-imino-5-(4-methoxybenzylidene)-3-methylimidazolidine-1-carboxylate (15) To a 25 mL round-bottom flask equipped with a stir bar were added 14 (0.51 g, 0.91 mmol), AgNO3 (0.02 g, 0.09 mmol), and dichloromethane (9.1 mL). The flask was wrapped with aluminum foil, and the reaction was allowed to stir at room temperature under N2 overnight. The reaction mixture was concentrated and purified via flash chromatography using 5% MeOH in CH2Cl2 to give 15 as a light yellow foam (0.43 g, 87%): Rf = 0.28 (5% MeOH in CH2Cl2); 1H NMR (CDCl3, 300 MHz) δ 7.46−7.20 (m, 8H), 6.97 (d, J = 8.7 Hz, 2H), 6.94−6.87 (m, 4H), 6.74 (d, J = 4.3 Hz, 2H), 6.71 (d, J = 4.0 Hz, 2H), 5.39 (s, 1H), 4.99 (s, 2H), 4.92 (d, J = 11.5 Hz, 1H), 4.29 (d, J = 12.0 Hz, 1H), 4.08 (dd, J = 4.2, 6.6 Hz, 1H), 3.77 (s, 3H), 3.08 (s, 3H), 2.99 (dd, J = 4.2, 13.7 Hz, 1H), 2.73 (dd, J = 7.3, 13.7 Hz, 1H) ppm; 13C NMR (CDCl3, 125 MHz) δ 158.7, 157.9, 154.0, 151.2, 137.1, 134.2, 131.1, 129.5, 128.8, 128.7, 128.5, 128.4, 128.3, 128.1, 127.5, 114.8, 113.8, 113.4, 70.0, 68.6, 65.0, 55.4, 37.8 ppm; IR (thin film) 2923, 2851, 1734, 1607, 1510, 1454, 1382, 1299, 1247, 1178, 1033, 830, 738, 698 cm−1; HRMS (ESI+) calcd for C34H34N3O4 m/z (M + H) 548.2549, found 548.2555. 4-((2-Amino-4-(4-methoxybenzyl)-1-methyl-1H-imidazol-5-yl)-methyl)phenol (Naamine A, 8) To a 10 mL round-bottom flask equipped with a stir bar were added 15 (0.25 g, 0.46 mmol), Pd(OH)2 on carbon (20 wt %, 0.032 g, 0.046 mmol), and MeOH (4.6 mL). A H2 balloon was attached, and the reaction was allowed to stir overnight. The reaction mixture was filtered through Celite and rinsed with dichloromethane. The reaction mixture was concentrated to a pale yellow solid (0.12 g, 84%, mp = 182 °C) and used without further purification to give 8 as naamine A: 1H NMR (CD3OD, 500 MHz) δ 7.08 (d, J = 8.3 Hz, 2H), 6.84 (d, J = 8.3 Hz, 2H), 6.76 (d, J = 8.3 Hz, 2H), 6.64 (d, J = 8.8 Hz, 2H), 3.76 (s, 2H), 3.72 (s, 3H), 3.69 (s, 2H), 3.08 (s, 3H) ppm; 13C NMR (DMSO-d6, 125 MHz) δ 168.5, 157.8, 156.2, 148.9, 134.5, 132.4, 130.6, 130.1, 129.5, 120.3, 115.8, 114.0, 55.6, 32.7, 29.4, 28.6 ppm; IR (thin film) 2923, 2852, 1610, 1511, 1457, 1369, 1245, 1175, 1035, 814, 773, 668, 652 cm−1; HRMS (ESI+) calcd for C19H22N3O2 m/z (M + H) 324.1712, found 324.1714. Preparation of Naamidine A (6) To a 50 mL round-bottom, two-neck flask equipped with a stir bar and reflux condenser were added 1-methylparabanic acid (2.04 g, 15.9 mmol) and acetonitrile (14.5 mL). Bis(trimethylsilyl)acetamide (4.9 mL, 20.0 mmol) was added via syringe, and the reaction mixture was allowed to reflux for 2 h. Without exposing the reaction flask to the open atmosphere, the solvent was removed under reduced pressure. The reaction mixture was placed under N2 and diluted with PhMe (10.5 mL). The solution was transferred via cannula to a 50 mL round-bottom, two-neck flask equipped with a stir bar and reflux condenser containing 8 (1.03 g, 3.2 mmol, mp = 188 °C) under a N2 atmosphere. The reaction mixture was allowed to reflux for 16 h. The reaction was allowed to cool to room temperature, diluted with EtOAc (10 mL), then transferred to a 100 mL round-bottom flash to be concentrated. The mixture was purified via flash chromatography using 85:15 PhMe/MeOH with 1% Et3N to give 6 as a bright yellow solid (1.10 g, 76%): Rf = 0.4 (85:15 PhMe/MeOH with 1% NEt3); 1H NMR (CDCl3, 500 MHz) δ 7.11 (d, J = 8.8 Hz, 2H), 6.83 (d, J = 7.8 Hz, 2H), 6.81 (d, J = 8.8 Hz, 2H), 6.73 (d, J = 8.8 Hz, 2H), 3.87 (s, 4H), 3.77 (s, 3H), 3.40 (s, 3H), 3.17 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 163.3, 158.5, 157.0, 155.2, 146.7, 134.7, 131.3, 129.5, 129.3, 128.7, 127.0, 115.0, 114.3, 55.5, 32.1, 30.0, 28.8, 25.0 ppm; IR (thin film) 3335, 1789, 1736, 1665, 1612, 1569, 1512, 1486, 1445, 1392, 1303, 1247, 1174, 1153, 1035, 821, 776, 727, 606 cm−1; HRMS (ESI+) calcd for C23H24N5O4 m/z (M + H) 434.1828, found 434.1840. General Procedure A: Acylation of 15 To Give 16a–f Benzyl-2-(benzoylimino)-4-(4-(benzyloxy)benzyl)-5-((Z)-4-methoxybenzylidene)-3-methylimidazolidine-1-carboxylate (16a) To a 25 mL round-bottom flask equipped with a stir bar were added 15 (498 mg, 0.91 mmol), Et3N (0.25 mL, 1.8 mmol, 2.0 equiv), benzoyl chloride (0.16 mL, 1.4 mmol, 1.5 equiv), and dichloromethane (9.1 mL). The reaction was allowed to stir for 1 h. The reaction mixture was concentrated and purified via flash chromatography using 1:1 hexanes/EtOAc to give 16a as a light yellow foam (545 mg, 92%): Rf = 0.43 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 8.12 (d, J = 7.0 Hz, 2H), 7.51−7.11 (m, 13H), 7.08 (d, J = 8.8 Hz, 2H), 6.98 (d, J = 8.7 Hz, 2H), 6.78 (d, J = 8.7 Hz, 2H), 6.75 (d, J = 8.9 Hz, 2H), 5.45 (s, 1H), 5.01 (s, 2H), 4.80 (d, J = 19.8 Hz, 1H), 4.42 (d, J = 19.8 Hz, 1H), 4.08 (dd, J = 4.2, 6.6 Hz, 1H), 3.77 (s, 3H), 3.14 (s, 3H), 3.03 (dd, J = 4.2, 13.5 Hz, 1H), 2.78 (dd, J = 7.6, 13.5 Hz, 1H) ppm; 13C NMR (CDCl3, 75 MHz) δ 175,6, 158.7, 157.9, 151.9, 149.1, 137.3, 137.0, 134.5, 131.4, 131.1, 129.7, 129.3, 128,7, 128.6, 128.2, 128.1, 128.0, 127.8, 127.5, 127.1, 117.4, 114.8, 113.7, 70.0, 68.7, 64.6, 55.3, 37.9, 31.0 ppm; IR (thin film) 3033, 2933, 1746, 1647, 1607, 1511, 1455, 1379, 1315, 1282, 1248, 1178, 1075, 1037, 1024, 866, 826, 739, 713, 697 cm−1; HRMS (ESI+) calcd for C41H37N3O5Na m/z (M + Na) 674.2631, found 674.2632. Benzyl-4-(4-(benzyloxy)benzyl)-2-((2-fluorobenzoyl)imino)-5-((Z)-4-methoxybenzylidene)-3-methylimidazolidine-1-carboxylate (16b) Prepared according to the general procedure A with 2-fluorobenzoyl chloride, with purification on silica gel eluting with 1:1 hexanes/EtOAc to give 16b as a yellow oil (540 mg, 89% yield): Rf = 0.42 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.91 (dt, J = 2.0, 7.8 Hz, 1H), 7.44−7.01 (m, 11H), 6.97 (d, J = 8.8 Hz, 4H), 6.85 (t, J = 8.8 Hz, 2H), 6.77 (d, J = 8.8 Hz, 2H), 6.66 (d, J = 8.8 Hz, 2H), 5.46 (s, 1H), 5.00 (s, 2H), 4.82 (d, J = 19.5 Hz, 1H), 4.33 (d, J = 19.5 Hz, 1H), 4.11 (dd, J = 3.4, 7.5 Hz, 1H), 3.76 (s, 3H), 3.15 (s, 3H), 3.00 (dd, J = 4.4, 13.7 Hz, 1H), 2.78 (dd, J = 7.5, 13.7 Hz, 1H) ppm; 13C NMR (CDCl3, 75 MHz) δ 172.7, 161.3 (d, JCF = 253.8 Hz), 158.7, 158.0, 151.8, 149.1, 137.0, 134.4, 132.6, 132.3 (d, JCF = 9.0 Hz), 131.0, 129.5, 128.7, 128.7, 128.2, 128.2, 127.7, 127.5, 127.0, 123.6 (d, JCF = 3.5 Hz), 117.3, 116.4 (d, JCF = 23.0 Hz), 114.9, 113.6, 70.1, 68.9, 64.6, 55.3, 37.9, 31.0 ppm; IR (thin film) 3033, 2930, 1743, 1598, 1510, 1483, 1452, 1407, 1379, 1314, 1282, 1246, 1177, 1116, 1029, 909, 862, 817, 756, 733, 696 cm−1; HRMS (ESI+) calcd for C41H36N3O5FNa m/z (M + Na) 692.2537, found 692.2545. Benzyl-4-(4-(benzyloxy)benzyl)-2-((4-methoxybenzoyl)imino)-5-((Z)-4-methoxybenzylidene)-3-methylimidazolidine-1-carboxylate (16c) Prepared according to the general procedure A with 4-methoxybenzoyl chloride, with purification on silica gel eluting with 2:1 hexanes/EtOAc to give 16c as a yellow oil (78 mg, 88% yield): Rf = 0.48 (2:1 EtOAc/hexanes); 1H NMR (CDCl3, 300 MHz) δ 8.08 (d, J = 8.7 Hz, 2H), 7.44−7.28 (m, 5H), 7.24−7.04 (m, 5H), 6.98 (d, J = 8.4 Hz, 2H), 6.89 (d, J = 9.0 Hz, 2H), 6.83−6.71 (m, 6H), 5.44 (s, 1H), 5.00 (s, 2H), 4.80 (d, J = 20.0 Hz, 1H), 4.43 (d, J = 20.0 Hz, 1H), 4.06 (dd, J = 4.1, 7.1 Hz, 1H), 3.84 (s, 3H), 3.76 (s, 3H), 3.12 (s, 3H), 3.02 (dd, J = 4.1, 13.5 Hz, 1H), 2.77 (dd, J = 7.1, 13.5 Hz, 1H); 13C NMR (CDCl3, 75 MHz) δ 175.3, 162.5, 158.9, 158.0, 151.6, 149.3, 137.1, 134.7, 133.7, 131.8, 131.2, 130.2, 129.8, 129.5, 128.8, 128.7, 128.3, 128.2, 128.0, 127.6, 127.4, 117.4, 114.9, 113.8, 70.1, 68.6, 64.7, 55.6, 55.4, 38.1, 31.1; IR (thin film) 3033, 2933, 2837, 1743, 1598, 1509, 1454, 1378, 1281, 1236, 1176, 1163, 1110, 1074, 1027, 907, 861, 844, 826, 726, 696 cm−1; HRMS (ESI+) calcd for C42H39N3O6Na m/z (M + Na) 704.2737, found 704.2742. Benzyl-4-(4-(benzyloxy)benzyl)-5-((Z)-4-methoxybenzylidene)-3-methyl-2-((3-(trifluoromethyl)benzoyl)imino)imidazolidine-1-carboxylate (16d) Prepared according to the general procedure A with 3-trifluoromethylbenzoyl chloride, with purification on silica gel eluting with 2:1 EtOAc/hexanes to give 16d as a yellow oil (61 mg, 94% yield): Rf = 0.66 (2:1 EtOAc/hexanes); 1H NMR (CDCl3, 300 MHz) δ 8.41 (s, 2H), 8.35 (d, J = 7.5 Hz, 1H), 8.30 (d, J = 8.0 Hz, 1H), 7.96 (d, J = 8.0 Hz, 1H), 7.72 (t, J = 7.5 Hz, 2H), 7.50 (t, J = 8.0 Hz, 1H), 7.42−7.29 (m, 4H), 7.22−7.11 (m, 4H), 7.00 (d, J = 8.5 Hz, 2H), 6.80 (d, J = 9.0 Hz, 2H), 6.77 (d, J = 8.0 Hz, 2H), 5.53 (s, 1H), 5.00 (s, 2H), 4.78 (d, J = 12.0 Hz, 1H), 4.41 (d, J = 12.0 Hz, 1H), 4.13 (dd, J = 4.5, 7.3 Hz, 1H), 3.77 (s, 3H), 3.18 (s, 3H), 3.04 (dd, J = 4.5, 13.8 Hz, 1H), 2.84 (dd, J = 7.3, 13.8 Hz, 1H); 13C NMR (CDCl3, 75 MHz) δ 173.9, 160.9, 158.9, 158.0, 152.9, 149.0, 138.0, 137.0, 134.2, 133.8, 132.9, 131.9 (q, JCF = 33.4 Hz), 131.3 (q, JCF = 3.8 Hz), 131.0, 130.4, 130.2 129.9, 129.5, 129.7 (q, JCF = 3.8 Hz), 129.2, 128.7, 128.5, 128.2, 128.1 127.5, 126.8, 125.4, 124.4, 123.2, 122.3, 117.7, 114.8, 113.7, 70.0, 68.9, 64.6, 55.2, 37.8, 30.9; IR (thin film) 2935. 1797, 1743, 1606, 1511, 1455, 1379, 1332, 1313, 1300, 1275, 1249, 1226, 1167, 1124, 1070, 1033, 996, 908, 858, 818, 789, 729, 693 cm−1; HRMS (ESI+) calcd for C42H37N3O5F3 m/z (M + H) 720.2685, found 720.2689. Benzyl-4-(4-(benzyloxy)benzyl)-2-(isobutyrylimino)-5-((Z)-4-methoxybenzylidene)-3-methylimidazolidine-1-carboxylate (16e) Prepared according to the general procedure A with isobutyryl chloride, with purification on silica gel eluting with 1:1 EtOAc/hexanes to give 16e as a yellow oil (49 mg, 87% yield): Rf = 0.32 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.64−7.37 (m, 10H), 7.28 (d, J = 8.8 Hz, 2H), 7.15 (d, J = 8.8 Hz, 2H), 7.08 (d, J = 7.8 Hz, 2H), 6.96 (d, J = 8.8 Hz, 2H), 5.60 (s, 1H), 5.21 (s, 2H), 5.07 (d, J = 19.5 Hz, 1H), 4.71 (d, J = 19.5 Hz, 1H), 4.20 (dd, J = 4.2, 7.1 Hz, 1H), 3.97 (s, 3H), 3.24 (s, 3H), 3.18 (dd, J = 4.2, 13.5 Hz, 1H), 2.93 (dd, J = 7.1, 13.5 Hz, 1H), 2.84 (sept, J = 6.8 Hz, 1H), 1.44 (d, J = 6.8 Hz, 3H), 1.39 (d, J = 6.8 Hz, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 188.0, 158.7, 157.9, 150.3, 149.2, 137.0, 134.6, 131.0, 129.5, 129.4, 128.7, 128.6, 128.2, 128.0, 127.9, 127.4, 127.2, 117.0, 114.8, 113.7, 70.0, 68.6, 64.6, 55.2, 38.6, 37.9, 30.8, 20.1 ppm; IR (thin film) 3033, 2964, 2929, 1745, 1663, 1607, 1455, 1379, 1273, 1249, 1179, 1123, 1077, 1037, 923, 864, 826, 738, 697 cm−1; HRMS (ESI+) calcd for C38H39N3O5Na m/z (M + Na) 640.2787, found 640.2775. Benzyl-4-(4-(benzyloxy)benzyl)-5-((Z)-4-methoxybenzylidene)-3-methyl-2-((2-methylbutanoyl)imino)imidazolidine-1-carboxylate (16f) Prepared according to the general procedure A with 2-methylbutyryl chloride, with purification on silica gel eluting with 1:1 EtOAc/hexanes to give 16f as a yellow oil (28 mg, 85% yield): Rf = 0.44 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.44−7.17 (m, 8H), 7.08 (d, J = 8.7 Hz, 2H), 6.94 (d, J = 8.5 Hz, 2H), 6.86 (d, J = 7.3 Hz, 2H), 6.75 (d, J = 8.4 Hz, 2H), 5.39 (d, J = 5.8 Hz, 1H), 5.00 (s, 2H), 4.88 (dd, J = 4.0 Hz, 20.5 Hz, 1H), 4.51 (dd, J = 4.0 Hz, 20.5 Hz, 1H), 3.99 (p, J = 3.7, 1H), 3.75 (s, 3H), 3.03 (s, 3H), 2.97 (dd, J = 4.4, 13.5 Hz, 1H), 2.72 (m, 1H), 2.46 (ddq, J = 6.9, 5.6, 7.0 Hz, 1H), 1.83 (m, 1H), 1.51 (m, 1H), 1.18 (dd, J = 6.9, 7.0, 3H), 0.97 (dt, J = 4.0, 7.6 Hz, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 187.5, 158.9, 158.0, 151.1, 150.7, 149.4, 137.2, 134.7, 131.2, 129.7, 128.8, 128.7, 128.4, 128.2, 128.0, 127.6, 127.4, 127.3, 117.2, 114.9, 113.8, 70.1, 68.8, 64.8, 55.4, 45.9, 45.5, 38.1, 31.1, 27.7, 27.5, 17.2, 16.4, 12.2, 12.0; IR (thin film) 2963, 2931, 2873, 1746, 1653, 1607, 1511, 1456, 1378, 1249, 1179, 1119, 1077, 1039, 827, 741, 696 cm−1; HRMS (ESI+) calcd for C39H41N3O5Na m/z (M + Na) 654.2944, found 654.2941. General Procedure B: Deprotection with Isomerization of 16 to 17 N-(5-(4-Hydroxybenzyl)-4-(4-methoxybenzyl)-1-methyl-1,3-dihydro-2H-imidazol-2-ylidene)benzamide (17a) To a 5 mL round-bottom flask equipped with a stir bar were added 16a (52 mg, 0.08 mmol), PdCl2 (25 mg, 0.18 mmol), and methanol (0.9 mL). The reaction was allowed to stir until completion under a H2 atmosphere balloon. The reaction mixture was filtered through 0.45 μM PTFE syringe filter and rinsed with additional methanol and CH2Cl2. The solvent was removed, and the product was triturated with diethyl ether. The solid was isolated to give 17a as an off-white solid (34 mg, 92%, mp = 135 °C): Rf = 0.40 (5% MeOH in CH2Cl2); 1H NMR (DMSO-d6, 500 MHz) δ 9.41 (s, 1H), 8.10 (d, J = 7.3 Hz, 2H), 7.63 (t, J = 7.33 Hz, 1H), 7.53 (t, J = 7.8 Hz, 2H), 7.20 (d, J = 8.8 Hz, 2H), 6.88 (d, J = 8.3 Hz, 2H), 6.86 (d, J = 8.8 Hz, 2H), 6.69 (d, J = 6.4 Hz, 2H), 4.03 (s, 2H), 4.00 (s, 2H), 3.69 (s, 3H), 3.15 (s, 3H) ppm; 13C NMR (DMSO-d6, 125 MHz) δ 158.5, 156.7, 133.3, 130.4, 130.0, 129.5, 129.0, 128.9, 127.0, 116.0, 114.4, 55.6, 49.0, 31.7, 28.7, 27.4 ppm; IR (thin film) 3926, 2932, 1688 1612, 1510, 1474, 1453, 1408, 1363, 1301, 1246, 1174, 1104, 1033, 908, 818, 731, 706 cm−1; HRMS (ESI+) calcd for C26H26N3O3 m/z (M + H) 428.1974, found 428.1973. 2-Fluoro-N-(5-(4-hydroxybenzyl)-4-(4-methoxybenzyl)-1-methyl-1,3-dihydro-2H-imidazol-2-ylidene)benzamide (17b) Prepared according to the general procedure B with 16b, with purification via trituration with diethyl ether to give 17b as a waxy solid (23 mg, 88% yield): Rf = 0.30 (1:1 hexanes/EtOAc); 1H NMR (DMSO-d6, 300 MHz) δ 9.29 (s, 1H), 7.80 (s, 1H), 7.43 (s, 1H), 7.19 (d, J = 8.1 Hz, 4H), 6.89 (d, J = 8.1 Hz, 2H), 6.84 (d, J = 8.4 Hz, 2H), 6.67 (d, J = 8.1 Hz, 2H), 3.89 (s, 4H), 3.71 (s, 3H), 3.20 (s, 3H) ppm; 13C NMR (DMSO-d6, 125 MHz) δ 157.8, 155.9, 131.3, 129.9, 129.7, 129.4, 128.3, 116.7 (d, JCF = 22.8 Hz), 115.8, 114.2, 55.4, 29.5, 27.7 ppm; IR (thin film) 1686, 1581, 1512, 1478, 1441, 1305, 1247, 1173, 1156, 1105, 904 cm−1; HRMS (ESI+) calcd for C26H24N3O3FNa m/z (M + Na) 468.1699, found 468.1700. N-(5-(4-Hydroxybenzyl)-4-(4-methoxybenzyl)-1-methyl-1,3-dihydro-2H-imidazol-2-ylidene)-4-methoxybenzamide (17c) Prepared according to the general procedure B with 16c, with purification via trituration with diethyl ether to give 17c as an off-white solid (50 mg, 95% yield, mp = 178 °C): Rf = 0.30 (1:1 hexanes/EtOAc); 1H NMR (DMSO-d6, 300 MHz) δ 9.44 (s, 1H), 8.07 (d, J = 9.3 Hz, 2H), 7.21 (d, J = 8.8 Hz, 2H), 7.10 (d, J = 8.8 Hz, 2H), 6.89 (d, J = 8.3 Hz, 2H), 6.87 (d, J = 8.3 Hz, 2H), 6.69 (d, J = 8.3 Hz, 2H), 4.06 (s, 2H), 4.01 (s, 2H), 3.85 (s, 3H), 3.72 (s, 3H), 3.43 (s, 3H) ppm; 13C NMR (DMSO-d6, 75 MHz) δ 162.8, 157.8, 155.9, 130.4, 129.6, 129.2, 128.8, 126.2, 115.2, 113.7, 64.6, 55.3, 54.8, 31.5, 27.8, 26.6, 14.9 ppm; IR (thin film) 2929, 1605, 1585, 1569, 1510, 1465, 1367, 1303, 1248, 1166, 1101, 1030, 906, 843, 815, 770, 728, 692, 668 cm−1; HRMS (ESI+) calcd for C27H28N3O4 m/z (M + H) 458.2080, found 458.2083. N-(5-(4-Hydroxybenzyl)-4-(4-methoxybenzyl)-1-methyl-1,3-dihydro-2H-imidazol-2-ylidene)-3-(trifluoromethyl)benzamide (17d) Prepared according to the general procedure B with 16d, with purification via trituration with diethyl ether to give 17d as an off-white solid (58 mg, 92% yield, mp = 237 °C): Rf = 0.40 (2:1 EtOAc/hexanes); 1H NMR (DMSO-d6, 300 MHz) δ 9.32 (s, 2H), 8.38 (s, 2H), 7.79 (d, J = 7.8 Hz, 1H), 7.64 (t, J = 7.5 Hz, 2H), 7.21 (d, J = 8.7 Hz, 2H), 6.91 (d, J = 8.4 Hz, 2H), 6.85 (d, J = 8.7 Hz, 2H), 6.69 (d, J = 8.4 Hz, 2H), 3.93 (s, 4H), 3.71 (s, 3H), 3.29 (s, 3H) ppm; 13C NMR (DMSO-d6, 75 MHz) δ 157.8, 155.9, 132.2, 129.5, 129.0, 124.5, 115.4, 113.9, 55.1, 28.9, 27.1 ppm; IR (thin film) 1564, 1532, 1512, 1483, 1383, 1322, 1277, 1248, 1170, 1153, 1113, 916 cm−1; HRMS (ESI+) calcd for C27H25N3O3F3 m/z (M + H) 496.1848, found 496.1850. N-(5-(4-Hydroxybenzyl)-4-(4-methoxybenzyl)-1-methyl-1,3-dihydro-2H-imidazol-2-ylidene)isobutyramide (17e) Prepared according to the general procedure B with 16e, with purification via trituration with diethyl ether to give 17e as an off-white solid (20 mg, 98% yield, mp = 196 °C): Rf = 0.17 (1:1 hexanes/EtOAc); 1H NMR (DMSO-d6, 300 MHz) δ 7.09 (d, J = 8.8 Hz, 2H), 6.82 (d, J = 8.3 Hz, 2H), 6.75 (d, J = 8.8 Hz, 2H), 6.64 (d, J = 8.3 Hz, 2H), 3.86 (s, 2H), 3.80 (s, 2H), 3.72 (s, 3H), 3.16 (s, 3H), 2.63 (sep, J = 6.8 Hz, 1H), 1.19 (d, J = 6.8 Hz, 6H) ppm; 13C NMR (DMSO-d6, 75 MHz) δ 176.3, 157.6, 155.8, 136.3, 130.0, 129.1, 128.7, 126.3, 125.5, 115.1, 113.5, 54.7, 33.5, 31.0, 28.0, 26.5, 18.6, ppm; IR (thin film) 3274, 2968, 2472, 1670, 1611, 1510, 1465, 1404, 1301, 1244, 1174, 1102, 1032, 973, 816 cm−1; HRMS (ESI+) calcd for C23H28N3O3 m/z (M + H) 394.2147, found 394.2138. N-(5-(4-Hydroxybenzyl)-4-(4-methoxybenzyl)-1-methyl-1,3-dihydro-2H-imidazol-2-ylidene)-2-methylbutanamide (17f) Prepared according to the general procedure B with 16f, with purification via trituration with diethyl ether to give 17f as an off-white solid (16 mg, 99%, mp = 102 °C): Rf = 0.33 (1:1 hexanes/EtOAc); 1H NMR (DMSO-d6, 300 MHz) δ 9.40 (s, 1H), 7.17 (d, J = 8.7 Hz), 6.87 (d, J = 8.2 Hz, 2H), 6.84 (d, J = 8.7 Hz, 2H), 6.67 (d, J = 8.5 Hz, 2H), 3.98 (s, 2H), 3.94 (s, 2H), 3.71 (s, 3H), 3.36 (s, 3H), 2.89 (sextet, J = 6.6 Hz, 1H), 1.61 (m, J = 7.4 Hz, 1H), 1.43 (m, J = 6.8 Hz, 1H), 1.10 (d, J = 6.8 Hz, 3H), 0.88 (t, J = 7.4 Hz, 3H) ppm; 13C NMR (DMSO-d6, 75 MHz) δ 176.1, 157.9, 156.1, 136.7, 130.5, 129.5, 129.1, 126.8, 125.9, 115.4, 113.9, 55.1, 31.3, 28.5, 26.9, 26.4, 16.9, 11.5 ppm; IR (thin film) 3357, 2965, 2483, 2076, 1670, 1653, 1635, 1612, 1558, 1510, 1458, 1405, 1301, 1245, 1175, 1118, 1033, 971, 816 cm−1; HRMS (ESI+) calcd for C24H30N3O3 m/z (M + H) 408.2303, found 408.2286. Preparation of Compound 18a N-(1,3-Diphenylprop-2-yn-1-yl)-N-methylprop-2-en-1-amine (S1a) In a 250 mL high-pressure flask containing a magnetic stir bar were added benzaldehyde (3.1 g, 29.0 mmol), phenylacetylene (2.95 g, 28.9 mmol), N-allylmethylamine (1.88 g, 26.3 mmol), oven-dried molecular sieves (grade 564, 3 Å, 8–12 mesh) (ca. 2 g), and acetonitrile (200 mL). The flask was sealed and placed in a preheated 80 °C oil bath for 24 h. The reaction flask was removed from the oil bath and allowed to cool to room temperature. CuBr (0.38 g, 2.6 mmol) was added, and the flask was sealed and returned to the preheated 80 °C oil bath for 48 h. The reaction tube was removed from the oil bath and allowed to cool to room temperature. The reaction mixture was filtered through Celite and rinsed with EtOAc (50 mL). The filtrate was concentrated under reduced pressure. The crude product was purified via flash chromatography, eluting with 9:1 hexanes/EtOAc to give S1a as a dark orange oil (5.3 g, 88%): Rf = 0.78 (2:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.65 (d, J = 6.9 Hz, 2H), 7.56−7.53 (m, 2H), 7.40−7.26 (m, 6H), 5.93 (ddt, J = 6.6 Hz, 10.4 Hz, 17.1 Hz, 1H), 5.32 (dd, J = 17.1 Hz, 1.5 Hz, 1H), 5.18 (dd, J = 10.4 Hz, 1.2 Hz, 1H), 4.99 (s, 1H), 3.89 (s, 3H), 3.19 (d, J = 5.1 Hz, 2H), 2.23 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 138.9, 136.3, 131.9, 128.5, 128.4, 128.3, 127.7, 123.4, 117.8, 88.5, 84.9, 59.8, 57.9 ppm; IR (thin film): 3061, 3030, 2978, 2945, 2844, 2788, 1598, 1489, 1448, 1324, 1273, 1196, 1155, 1127, 1070, 1023, 994, 963, 917, 754, 726, 689 cm−1; HRMS (ESI+) calcd for C19H19N m/z 262.1590 (M + H), found 262.1572. N-Methyl-1,3-diphenylprop-2-yn-1-amine (S2a) In a 250 mL round-bottom flask containing a magnetic stir bar were added Pd(PPh3)4 (1.3 g, 1.1 mmol), thiosalicylic acid (7.0 g, 45.0 mmol), and CH2Cl2 (100 mL). A solution of S1a (5.3 g, 22.7 mmol) in 15 mL of CH2Cl2 was added, and the reaction mixture was allowed to stir at room temperature under N2 for 12 h. The solvent was then removed under reduced pressure, and the crude product was redissolved in Et2O (10 mL). The organic layer was washed with aqueous solutions of saturated NaHCO3 (50 mL) and brine (50 mL) and then dried and filtered over Na2SO4. The crude product was purified via flash chromatography, eluting with 4:1 hexanes/EtOAc to give S2a as a dark orange oil (3.1 g, 73%): Rf = 0.22 (2:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.61−7.31 (m, 10H), 4.76 (s, 1H), 2.57 (s, 3H), 1.47 (s, 1H) ppm; 13C NMR (CDCl3, 75 MHz) δ 140.3, 131.8, 128.6, 128.4, 128.2, 127.9, 127.7, 123.2, 89.1, 85.7, 56.4, 33.9 ppm; IR (thin film) 3060, 3029, 2933, 2850, 2793, 1653, 1598, 1559, 1540, 1489, 1473, 1449, 1306, 1214, 1177, 1098, 1071, 1027, 915, 755, 691 cm−1; HRMS (ESI+) calcd for C16H16N m/z 222.1259 (M + H), found 222.1288. Benzyl (Z)-5-Benzylidene-2-imino-3-methyl-4-phenylimidazolidine-1-carboxylate (S3a) In a 100 mL round-bottom flask containing a magnetic stir bar were added potassium benzyloxycarbonylcyanamide (0.65 g, 3.0 mmol), TMSCl (0.34 g, 3.1 mmol), and acetonitrile (15 mL). The solution was stirred at room temperature for 10 min. A solution of S2a (0.46 g, 2.4 mmol) in acetonitrile (3.5 mL) was then added, and the reaction mixture was allowed to stir at room temperature for 1 h. The solvent was removed under reduced pressure, and the crude product was dissolved in EtOAc (150 mL). The organic layer was washed with aqueous solutions of saturated NaHCO3 (50 mL) and brine (50 mL) and then dried and filtered over Na2SO4. The crude product was purified via flash chromatography, eluting with 1:1 hexanes/EtOAc to give S3a as a dark brown oil (0.75 g, 85%): Rf = 0.48 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.58−7.23 (m, 15H), 5.18 (s, 2H), 2.83 (s, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 164.3, 161.3, 158.0, 137.2, 132.1, 128.9, 128,8, 128.6, 128.3, 128.2, 127.9, 127.6, 122.7, 87.0, 85.1, 67.1, 51.4, 30.1 ppm; IR (thin film) 3331, 3031, 2939, 1736, 1646, 1596, 1534, 1491, 1450, 1379, 1153, 1050, 1028, 801, 757, 696 cm−1; HRMS (ESI+) calcd for C25H24N3O2 m/z 398.1869 (M + H), found 398.1877. (Z)-Benzyl 5-Benzylidene-2-imino-3-methyl-4-phenylimidazolidine-1-carboxylate (18a) In a 50 mL foil-wrapped round-bottom flask containing a magnetic stir bar were added S3a (0.75, 2.0 mmol), AgNO3 (35 mg, 0.20 mmol), and CH2Cl2 (20 mL). The solution was stirred at room temperature for 6 h. The solvent was then removed under reduced pressure, and the crude product was redissolved in EtOAc (50 mL). The organic layer was washed with aqueous solutions of saturated NaHCO3 (15 mL) and brine (15 mL) and then dried -and filtered over Na2SO4. The crude product was purified via flash chromatography, eluting with 1:1 hexanes/EtOAc to give 18a as a dark brown oil (0.53 g, 71%): Rf = 0.31 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.43−7.12 (m, 11H), 7.08 (d, J = 7.0 Hz, 2H), 6.90 (d, J = 7.5 Hz, 2H), 5.52 (d, J = 3.0 Hz, 1H), 5.00 (d, J = 3.0 Hz, 1H), 4.79 (d, J = 19.5 Hz, 1H), 4.37 (d, J = 19.5 Hz, 1H), 2.81 (s, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 154.1, 151.4, 137.6, 136.3, 34.4, 129.2, 129.1, 128.7, 128.5, 128.4, 128.2, 127.4, 127.1, 113.4, 68.4, 67.6, 30.2 ppm; IR (thin film) 3346, 3031, 1734, 1684, 1652, 1495, 1426, 1386, 1303, 1249, 1197, 1161, 1047, 1026, 957, 797, 696 cm−1; HRMS (ESI+) calcd for C25H24N3O2 m/z 398.1869 (M + H), found 398.1876. Preparation of Compound 18b N-(1-(4-Methoxyphenyl)-3-phenylprop-2-yn-1-yl)-N-methylprop-2-en-1-amine (S1b) Prepared according to the A3-coupling procedure of S1a using p-anisaldehyde, n-allylmethylamine, and phenylacetylene, with purification on silica gel eluting with 2:1 hexanes/EtOAc to give a dark orange oil (12.7 g, 65%): Rf = 0.78 (2:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.59−7.53 (m, 4H), 7.37−7.26 (m, 3H), 6.49 (d, J = 8.7 Hz, 2H), 5.92 (ddt, J = 6.6 Hz, 10.5 Hz, 17.4 Hz, 1H), 5.33 (dd, J = 17.4 Hz, 2.0 Hz, 1H), 5.19 (dd, J = 9.3 Hz, 2.0 Hz, 1H), 4.94 (s, 1H), 3.83 (s, 3H), 3.19 (d, J = 6.6 Hz, 2H), 2.24 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 159.1, 136.3, 131.9, 131.1, 129.7, 128.4, 128.2, 123.4, 117.7, 113.6, 88.3, 85.3, 59.3, 57.8, 55.4, 37.8 ppm; IR (thin film) 2948, 2834, 2786, 1642, 1609, 1583, 1507, 1488, 1441, 1301, 1244, 1169, 1126, 1107, 1033, 994, 962, 916, 850, 807, 778, 754, 689, 583, 524 cm−1; HRMS (ESI+) calcd for C20H21NO m/z 292.1701 (M + H), found 292.1699. 1-(4-Methoxyphenyl)-N-methyl-3-phenylprop-2-yn-1-amine (S2b) Prepared according to the Pd(0)-deallylation procedure with S1b, with purification on silica gel eluting with 2:1 hexanes/EtOAc to give S2b a dark orange oil (2.1 g, 44%): Rf = 0.22 (2:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.54−7.48 (m, 4H), 7.33−7.31 (m, 3H), 6.9 (d, J = 8.7 Hz, 2H), 5.18 (s, 1H), 3.81 (s, 3H), 2.56 (s, 3H), 1.81 (s, 1H) ppm; 13C NMR (CDCl3, 75 MHz) δ 159.2, 132.4, 131.7, 128.8, 128.3, 128.1, 123.1, 113.8, 89.2, 85.5, 55.6, 55.3, 33.7 ppm; IR (thin film) 2953, 2834, 2790, 1609, 1584, 1508, 1488, 1462, 1440, 1301, 1243, 1171, 1095, 1031, 956, 913, 829, 754, 727, 703, 689, 573, 547, 524 cm−1; HRMS (ESI+) calcd for C17H17NONa m/z 274.1208 (M + Na), found 274.1213. Benzyl (Z)-5-Benzylidene-2-imino-4-(4-methoxyphenyl)-3-methylimidazolidine-1-carboxylate (S3b) Prepared according to the guanylation procedure of S2b, with purification on silica gel eluting with 1:1 hexanes/EtOAc to give a dark orange oil (2.97 g, 82%): Rf = 0.48 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.51−7.43 (m, 6H), 7.36−7.25 (m, 7H), 6.9 (d, J = 6.3 Hz, 2H), 5.18 (s, 2H), 3.80 (s, 3H), 2.80 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 164.1, 160.9, 159.4, 137.6, 131.9, 129.0, 128.7, 128.6, 128.4, 128.0, 127.7, 122.2, 113.9, 86.6, 85.2, 66.9, 55.3, 50.6, 29.7 ppm; IR (thin film) 3403, 2932, 1646, 1584, 1532, 1508, 1488, 1440, 1376, 1273, 1246, 1121, 1150, 1110, 1027, 908, 845, 799, 775, 755, 729, 690, 647, 586, 552 cm−1; HRMS (ESI+) calcd for C26H26N3O3 m/z 428.1974 (M + Na), found 428.1979. (Z)-Benzyl 5-Benzylidene-2-imino-4-(4-methoxyphenyl)-3-methylimidazolidine-1-carboxylate (18b) Prepared according to the Ag(I) cyclization procedure, with purification by silica gel eluting with 1:1 hexanes/EtOAc to give a dark brown oil (1.2 g, 87%): Rf = 0.18 (1:1 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.28−7.16 (m, 9H), 7.10−7.08 (m, 2H), 6.92−6.88 (m, 4H), 5.47 (d, J = 2.1 Hz, 1H), 4.92 (d, J = 2.1 Hz, 1H), 4.82 (d, J = 19.5 Hz, 2H), 4.33 (d, J = 19.5 Hz, 2H), 3.81 (s, 3H), 2.75 (s, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 106.1, 153.5, 151.4, 136.5, 135.1, 134.4, 129.7, 129.6, 128.7, 128.4, 127.4, 127.0, 114.5, 113.0, 62.2, 67.0, 55.4, 30.1 ppm; IR (thin film) 3404, 2932, 1646, 1548, 1532, 1508, 1488, 1440, 1376, 1273, 1246, 1171, 1150, 1110, 1027, 908, 845, 799, 779, 755, 728, 690, 647, 586, 552 cm−1; HRMS (ESI+) calcd for C26H26N3O3 m/z 428.1974 (M + Na), found 428.1979. General Procedure C: Acylation of 18 To Give 19 Benzyl-2-(benzoylimino)-5-((Z)-benzylidene)-3-methyl-4-phenylimidazolidine-1-carboxylate (19a) In a 10 mL round-bottomed flask containing a magnetic stir bar were added 18a (73 mg, 0.18 mmol), benzoyl chloride (0.032 mL, 0.28 mmol, 1.5 equiv), triethylamine (0.051 mL, 0.37 mmol, 2.0 equiv), and dichloromethane (2 mL) under N2. The reaction was stirred at room temperature for 2 h. The solution was concentrated under reduced pressure, and the crude material was dissolved in EtOAc (20 mL). The organic layer was washed with aqueous solutions of saturated NaHCO3 (15 mL) and brine (15 mL). The organic layer was dried over Na2SO4, and the resulting material was purified via flash chromatography (3:2 hexanes/EtOAc) to yield 19a as a light brown foam (86 mg, 93%): Rf = 0.22 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 8.21−8.18 (m, 2H), 7.51−7.32 (m, 8H), 7.25−7.11 (m, 8H), 6.8−6.78 (m, 2H), 5.77 (d J = 2.0 Hz, 1H), 5.16 (d, J = 2.0 Hz, 1H), 4.70 (d, J = 12.0 Hz, 2H), 4.63 (d, J = 12.0 Hz, 2H), 2.93 (s, 3H) ppm; 13C NMR (CDCl3, 75 MHz) δ 178.8, 151.8, 149.3, 137.1, 136.8, 135.4, 134.5, 133.9, 131.6, 129.7, 129.4, 129.3, 127.8, 127.5, 116.9, 68.8, 67.0, 30.6 ppm; IR (thin film) 3060, 3029, 1744, 1557, 1494, 1448, 1404, 1377, 1315, 1277, 1226, 1173, 1144, 1080, 1036, 1020, 976, 909, 856, 794, 752, 727, 696, 668 cm−1; HRMS (ESI+) calcd for C32H27N3NaO3 m/z (M + Na) 524.1950, found 524.1963. Benzyl 2-(Benzoylimino)-5-((Z)-benzylidene)-4-(4-methoxyphenyl)-3-methylimidazolidine-1-carboxylate (19b) Prepared according to general procedure C using 18b and benzoyl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19b as a light brown foam (96 mg, 80%): Rf = 0.22 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.18 (d, J = 8.0 Hz, 2H), δ 7.50−7.21 (m, 14H), δ 6.91 (d, J = 8.5 Hz, 2H), δ 6.80 (d, J = 7.0 Hz, 2H), δ 5.74 (d, J = 1.8 Hz, 1H), δ 5.13 (d, J = 1.8 Hz, 1H), δ 4.72 (d, J = 12.0 Hz, 1H), δ 4.24 (d, J = 12.0 Hz, 1H), δ 3.82 (s, 3H), δ 2.90 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 175.2, 160.5, 151.9, 149.4, 137.3, 135.6, 134.7, 134.6, 134.4, 131.7, 129.8, 129.4, 128.7, 128.5, 128.4, 128.3, 128.2, 127.6, 116.8, 114.9, 69.0, 66.8, 55.6, 30.7 ppm; IR (thin film) 3404, 2932, 1646, 1548, 1532, 1508, 1488, 1440, 1376, 1273, 1246, 1171, 1150, 1110, 1027, 908, 845, 799, 779, 755, 728, 690 cm−1; HRMS (ESI+) calcd for C33H29N3NaO4 m/z (M + Na) 554.2056, found 554.2066. Benzyl 5-((Z)-Benzylidene)-2-((4-methoxybenzoyl)imino)-3-methyl-4-phenylimidazolidine-1-carboxylate (19c) Prepared according to general procedure C using 18a and 4-methoxybenzoyl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19c as a light brown foam (0.12 g, 95%): Rf = 0.19 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.16 (d, J = 9.0 Hz, 2H), 7.41−7.37 (m, 3H), 7.33−7.30 (m, 2H), 7.26−7.12 (m, 8H), 6.93 (d, J = 9.0 Hz, 2H), 6.81 (d, J = 7.5 Hz, 2H), 5.72 (d, J = 2.0 Hz, 1H), 5.13 (s, 1H), 4.71 (d, J = 12.3 Hz), 4.65 (d, J = 12.3 Hz), 3.86 (s, 3H), 2.92 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 175.0, 162.6, 151.5, 149.5, 137.1, 135.6, 134.8, 134.2, 131.8, 130.1, 129.5, 129.4, 1283, 127.9, 127.6, 116.9, 113.4, 68.9, 67.2, 55.6, 30.8 ppm; IR (thin film) 3058, 2951, 1745, 1652, 1597, 1507, 1456, 1427, 1249, 1227, 1177, 1162, 1022, 974, 863, 843, 731, 693 cm−1; HRMS (ESI+) calcd for C33H29N3NaO4 m/z (M + Na) 554.2056, found 554.2061. Benzyl 5-((Z)-Benzylidene)-3-methyl-4-phenyl-2-((3-(trifluoromethyl)benzoyl)imino)imidazolidine-1-carboxylate (19d) Prepared according to general procedure C using 18a and 3-trifluoromethylbenzoyl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19d as a light brown foam (0.12 g, 87%): Rf = 0.31 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.47 (s, 1H), 8.37 (d, J = 8.0 Hz, 1H), 7.74 (d, J = 8.0 Hz, 1H), 7.55 (t, J = 7.5 Hz, 1H), 7.42 (m, 4H), 7.34 (m, 2H), 7.26 (m, 3H), 7.19 (d, J = 7.0 Hz, 2H), 7.15 (t, J = 7.5 Hz, 2H), 6.78 (d, J = 7.0 Hz, 2H), 5.80 (s, 1H), 5.20 (s, 1H), 4.69 (d, J = 11.8 Hz), 4.64 (d, J = 11.8 Hz), 2.95 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 173.6, 155.1, 144.3, 138.1, 136.7, 135.4, 134.4, 133.9, 133.1, 130.6 (q, JCF = 32.4 Hz), 128.8, 128.7, 128.5, 128.5, 128.4, 128.3, 128.2 (q, JCF = 2.8 Hz), 128.0, 127.7, 126.8 (q, JCF = 3.6 Hz), 124.3 (q, JCF = 270.5 Hz), 117.4, 69.2, 67.3, 30.8 ppm; IR (thin film) 1699, 1652, 1616, 1325, 1259, 1166, 1121, 1070, 998, 920, 855, 817, 758, 692 cm−1; HRMS (ESI+) calcd for C33H26F3N3NaO3 m/z (M + Na) 592.1824, found 592.1821. Benzyl 5-((Z)-Benzylidene)-2-(isobutyrylimino)-3-methyl-4-phenylimidazolidine-1-carboxylate (19e) Prepared according to general procedure C using 18a and isobutyryl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19e as a light brown foam (38 mg, 92%): Rf = 0.34 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.42−7.35 (m, 3H), 7.29−7.15 (m, 10H), 6.86 (d, J = 7.0 Hz, 2H), 5.72 (d, J = 2.0 Hz, 1H), 5.07 (d, J = 2.0 Hz, 1H), 4.74 (d, J = 12.3 Hz), 4.69 (d, J = 12.3 Hz), 2.81 (s, 3H), 2.71 (m, 1H), 1.26 (d, J = 7.0 Hz, 6H) ppm; 13C NMR (CDCl3, 125 MHz) δ 187.6, 150.2, 149.5, 137.1, 135.6, 134.7, 134.1, 129.5, 128.6, 128.4, 128.3, 127.8, 127.6, 116.6, 68.8, 67.1, 38.8, 30.6, 20.0 ppm; IR (thin film) 3030, 2966, 2360, 2340, 1743, 1653, 1598, 1494, 1455, 1403, 1378, 1345, 1261, 1175, 1121, 1080, 1023, 977, 919, 847, 820, 752, 730, 695, 668, 634, 598, 557 cm−1; HRMS (ESI+) calcd for C29H29N3NaO3 m/z (M + Na) 490.2107, found 490.2103 (M + Na). Benzyl 5-((Z)-Benzylidene)-3-methyl-2-((2-methylbutanoyl)-imino)-4-phenylimidazolidine-1-carboxylate (19f) Prepared according to general procedure C using 18a and 2-methylbutyryl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19f as a light brown foam (57 mg, 70%): Rf = 0.39 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.42−7.12 (m, 13H), δ 6.86 (d, J = 7.5 Hz, 2H), 5.72 (d, J = 2.0 Hz, 1H), 5.08 (d, J = 2.0 Hz, 1H), 4.77−4.66 (m, 2H), 2.81 (s, 3H), 2.53 (m, 1H), 1.87 (m, 1H), 1.55 (m, 1H), 1.23 (d, J = 7 Hz, 3H), 1.01 (t, J = 6.5 Hz, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 186.8, 150.6, 149.6, 137.3, 137.1, 135.6, 134.7, 134.3, 134.1, 129.5, 129.4, 128.5, 128.4, 128.3, 127.9, 127.8, 127.5, 116.4, 68.8, 45.8, 30.7, 27.6, 16.8, 12.1 ppm; IR (thin film) 3031, 2963, 2931, 2873, 1744, 1653, 1597, 1494, 1456, 1403, 1375, 1264, 1175, 1113, 1080, 1039, 978, 908, 752, 730, 695, 668, 633, 588 cm−1; HRMS (ESI+) calcd for C30H31N3NaO3 m/z (M + Na) 504.2263, found 504.2275. Benzyl 5-((Z)-Benzylidene)-2-((2-fluorobenzoyl)imino)-3-methyl-4-phenylimidazolidine-1-carboxylate (19g) Prepared according to general procedure C using 18a and 2-fluorobenzoyl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19g as a light brown foam (120 mg, 95%): Rf = 0.28 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.99 (t, J = 8.0 Hz, 1H), 7.50−7.34 (m, 4H), 7.34−7.29 (m, 2H), 7.21−7.07 (m, 8H), 6.99 (d, J = 8.0 Hz, 2H), 6.84 (d, J = 7.0 Hz), 5.71 (d, J = 2.0 Hz, 1H), 5.17 (d, J = 2.0 Hz, 1H), 4.75 (d, J = 11.8 Hz), 4.54 (d, J = 11.8 Hz), 2.93 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 172.3, 161.3 (d, JCF = 423.0 Hz), 151.5, 149.3, 136.7, 135.2, 134.4, 133.9, 132.6, 132.5, 134.4, 129.4, 128.7, 128.4, 128.3, 128.0, 127.9, 127.5, 125.9 (d, JCF = 16.6 Hz), 123.8 (d, JCF = 6.6 Hz), 117.2, 116.5 (d, JCF = 38.4 Hz), 69.0, 67.9, 30.5 ppm; IR (thin film) 3031, 1745, 1596, 1483, 1404, 1378, 1316, 1280, 1263, 1223, 1179, 1157, 1111, 1081, 1023, 974, 909, 866, 782, 755, 732, 696, 655 cm−1; HRMS (ESI+) calcd for C32H26FN3NaO3 m/z (M + Na) 542.1856, found 542.1865. Benzyl 5-((Z)-Benzylidene)-2-((2-fluorobenzoyl)imino)-4-(4-methoxyphenyl)-3-methylimidazolidine-1-carboxylate (19h) Prepared according to general procedure C using 18b and 2-fluorobenzoyl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19h a light brown foam (0.98 g, 83%): Rf = 0.25 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 7.99 (t, J = 6 Hz, 1H), 7.44 (m, 1H), 7.24−7.05 (m, 10H), 6.99−6.83 (m, 5H), 5.69 (d, J = 1.8 Hz, 1H), 5.15 (d, J = 1.8 Hz, 1H), 4.77 (d J = 19.8 Hz), 4.64 (d, J = 19.8 Hz), 3.82 (s, 3H), 2.89 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 161.3 (d, JCF = 253.4 Hz), 160.4, 151.5, 149.4, 135.3, 134.4, 134.3, 132.7, 132.6, 129.4, 128.8, 128.4, 128.3, 128.0, 127.5, 123.8 (d, JCF = 4.0 Hz), 117.1, 116.5 (d, JCF = 23.0 Hz), 114.8, 69.0, 66.6, 55.4, 30.4 ppm; IR (thin film) 2933, 2834, 2790, 109, 1584, 1508, 1488, 1462, 1440, 1301, 1243, 1171, 1095, 1031, 956, 913, 829, 783, 754, 727, 689, 660, 634, 618, 573, 547, 524 cm−1; HRMS (ESI+) calcd for C33H28FN3NaO4 m/z (M + Na) 572.1962, found 572.1980. Benzyl 5((Z)-benzylidene)-4-(4-methoxyphenyl)-3-methyl-2-((3-(trifluoromethyl)benzoyl)imino)imidazolidine-1-carboxylate (19i) Prepared according to general procedure C using 18b and 3-trifluoromethylbenzoyl chloride, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 19i as a light brown foam (95%): Rf = 0.25 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 1 MHz) δ 8.46 (s, 1H), 8.36 (d, J = 9.5 Hz, 1H), 7.73 (d, J = 9.0 Hz, 1H), 7.55 (t, J = 10.0 Hz, 1H), 7.27−7.10 (m, 10), 6.93 (d, J = 10.5 Hz, 2H), 6.78 (d, J = 9.5 Hz, 2H), 5.77 (s, 1H), 5.18 (s, 1H), 4.70 (d, J = 14.5 Hz), 4.62 (d, J = 14.5 Hz), 3.89 (s, 3H), 2.92 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 173.3, 160.4, 152.8, 149.1, 137.9, 135.3, 134.0, 132.8, 130.3 (q, JCF = 24.7 Hz), 129.3, 128.5, 128.3, 128.3, 128.2, 128.1, 127.9, 127.5, 126.5 (q, JCF = 2.9 Hz), 124.2 (q, JCF = 203.2 Hz), 117.0, 114.7, 68.9, 66.7, 55.4, 30.4 ppm; IR (thin film) 1775, 1739, 1670, 1608, 1514, 1383, 1323, 1252, 1172, 1127, 1072, 1030, 770 cm−1; HRMS (ESI+) calcd for C34H28F3N3NaO4 m/z (M + Na) 622.1930, found 622.1927. General Procedure D: Deprotection with Isomerization of 19 to 20 N-(4-Benzyl-1-methyl-5-phenyl-1H-imidazol-2(3H)-ylidene)benzamide (20a) In a 5 mL test tube containing a magnetic stir bar were added 19a (84 mg, 0.17 mmol), Pd/C (10% w/w, 9 mg), and distilled MeOH (2 mL) under a stream of N2. The reaction tube was then sealed in a pressure vessel and purged with H2 three times. The pressure vessel was then charged with H2 at 60 psi, and the reaction was stirred at room temperature for 24 h. After releasing the H2 from the pressure vessel, the solution was filtered with a nonpolar syringe filter followed by addition of 5 mL of hot methanol to wash the filter. The filtrate was concentrated via rotary evaporation under reduced pressure, and the resulting material was purified via flash chromatography (3:2 hexanes/EtOAc) to yield 20a as a light brown foam (44 mg, 72%): Rf = 0.47 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.28 (d, J = 8.5 Hz, 2H), 7.51−7.46 (m, 8H), 7.36−7.26 (m, 2H), 7.18−7.09 (m, 3H), 3.80 (s, 2H), 3.50 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 170.9, 137.9, 137.2, 132.8, 132.6, 130.6, 130.2, 129.7, 129.4, 129.2, 128.9, 128.6, 128.5, 127, 34.6, 31 ppm; IR (thin film) 1695, 1653, 1601, 1560, 1494, 1472, 1452, 1379, 1314, 1269, 1025, 765, 742, 700, 658 cm−1; HRMS (ESI+) calcd for C24H22N3O m/z (M + H) 368.1763, found 368.1768. N-(4-Benzyl-5-(4-methoxyphenyl)-1-methyl-1H-imidazol-2(3H)-ylidene)benzamide (20b) Prepared according to general procedure D using 19b, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 20b as a light brown foam (9.7 mg, 84%): Rf = 0.47 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.27 (d, J = 7.0 Hz, 2H), 7.45−7.40 (m, 3H), 7.32−7.27 (m, 4H), 7.22 (m, 1H), 7.15 (d, J = 7.5 Hz, 2H), 7.01 (d, J = 9.0 Hz, 2H), 3.86 (s, 3H), 3.83 (s, 2H), 3.49 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 160.5, 138.7, 131.9, 130.8, 129.2, 128.9, 128.4, 128.1, 127.2, 124.5, 120.0, 114.8, 55.7, 32.4, 31.0 ppm; IR (thin film) 3061, 2933, 1675, 1636, 1566, 1541, 1494, 1464, 1453, 199, 1350, 1288, 1246, 1174, 1108, 1025, 1004, 906, 832, 718, 709, 645, 593 cm−1; HRMS (ESI+) calcd for C25H23N3NaO2 m/z (M + Na) 420.1688, found 420.1698. N-(4-Benzyl-1-methyl-5-phenyl-1,3-dihydro-2H-imidazol-2-ylidene)-4-methoxybenzamide (20c) Prepared according to general procedure D using 19c, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 20c as a light brown foam (46 mg, 62%): Rf = 0.29 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.30 (d, J = 9.0 Hz, 2H), 7.48−7.44 (m, 3H), 7.31−7.27 (m, 2H), 7.10−7.02 (m, 3H), 7.00−6.95 (m, 4H), 3.85 (s, 3H), 3.60 (s, 2H), 3.48 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 163.3 137.8, 131.4, 130.5, 129.7, 129.3, 128.9, 128.4, 127.3, 126.9, 113.9, 55.7, 32.9, 30.8 ppm; IR (thin film) 2858, 1678, 1603, 1573, 1514, 1494, 1453, 1401, 1348, 1311, 1176, 1027, 846, 766 cm−1; HRMS (ESI+) calcd for C25H23N3NaO2 m/z (M + Na) 420.1688, found 420.1688. N-(4-Benzyl-1-methyl-5-phenyl-1,3-dihydro-2H-imidazol-2-ylidene)-3-(trifluoromethyl)benzamide (20d) Prepared according to general procedure D using 19d, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 20d a light brown foam (123 mg, 87%): Rf = 0.76 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.55 (s, 1H), 8.43 (d, J = 7.5 Hz, 1H), 7.68 (d, J = 7.5 Hz, 1H), 7.53−7.48 (m, 4H), 7.40−7.37 (m, 2H), 7.32−7.29 (m, 2H), 7.26−7.24 (m, 1H), 7.15 (d, J = 8.0 Hz, 2H), 3.87 (s, 2H), 3.54 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 173.3, 150.8, 139.5, 137.3, 132.1, 120.5, 129.5, 129.3, 129.2, 128.5, 128.4, 127.7, 127.3, 127.2 (q, JCF = 3.8 Hz), 125.9 (q, JCF = 3.8 Hz), 124.8, 120.6, 95.0, 30.8, 30.3 ppm; IR (thin film) 3062, 1598, 1568, 1471, 1362, 1315, 1276, 1216, 1162, 1117, 1084, 1067, 907, 795, 763, 726 cm−1; HRMS (ESI+) calcd for C25H21N3OF3 m/z (M + H) 436.1637, found 436.1639. N-(4-Benzyl-1-methyl-5-phenyl-1H-imidazol-2(3H)-ylidene)-isobutyramide (20e) Prepared according to general procedure D using 19e, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 20e as a light brown foam (24 mg, 74%): Rf = 0.50 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.42−7.39 (m, 3H), 7.33 (d, J = 8.0 Hz, 2H), 7.24−7.20 (m, 3H), 7.13 (d, J = 8.0 Hz 2H), 3.83 (s, 2H), 3.31 (s, 3H), 2.49 (m, 1H), 1.13 (d, J = 7.0 Hz, 6H) ppm; 13C NMR (CDCl3, 125 MHz) δ 140.3, 130.2, 129.7, 129.0, 128.6, 128.5, 128.4, 126.2, 35.7, 32.9, 31.8, 19.8 ppm; IR (thin film) 3028, 2968, 2873, 1653, 1602, 1540, 1506, 1494, 1466, 1456, 1437, 1399, 1383, 1312, 1221, 1190, 1156, 1098, 1014, 950, 910, 867, 725, 697 cm−1; HRMS (ESI+) calcd for C21H23N3O m/z (M + Na) 356.1739, found 356.1743 (M + H). N-(4-Benzyl-1-methyl-5-phenyl-1H-imidazol-2(3H)-ylidene)-2-methylbutanamide (20f) Prepared according to general procedure D using 19f, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give 20f as a light brown foam (33 mg, 80%): Rf = 0.44 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 7.42−7.39 (m, 3H), 7.29 (d, J = 8.0 Hz, 2H), 7.24−7.20 (m, 3H), 7.05 (d, J = 8.0 Hz 2H), 3.77 (s, 2H), 3.30 (s, 3H), 2.43 (m, 1H), 1.69 (m, 1H), 1.41 (m, 1H), 1.10 (d, 7.0 Hz, 3H), 0.87 (t, J = 7.0 Hz, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 139.9, 130.3, 129.1, 128.8, 128.6, 128.5, 42.8, 32.3, 27.2, 17.7, 11.1 ppm; IR (thin film) 2835, 1609, 1583, 1508, 1488, 1442, 1419, 1301, 1244, 1169, 1126, 1107, 1069, 1033, 994, 962, 917, 850, 807, 778, 754, 690, 584 cm−1; HRMS (ESI+) calcd for C22H25N3O m/z (M + H) 348.2076, found 348.2082 (M + H). N-(4-Benzyl-1-methyl-5-phenyl-1H-imidazol-2(3H)-ylidene)-2-fluorobenzamide (20g) Prepared according to general procedure D using 19g, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give a light brown (7.1 mg, 89%): Rf = 0.82 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.08 (dt, J = 2.0, 8.0 Hz, 1H), 7.49 (m, 3H), 7.38 (m, 3H), 7.28 (m, 2H), 7.21 (m, 2H), 7.16 (m, 2H), 7.10 (m, 1H), 3.85 (s, 2H), 3.49 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 161.7 (d, JCF = 252.4 Hz), 137.8, 132.4 (d, JCF = 8.1 Hz), 131.9, 130.5, 129.4, 129.3, 129.1, 128.5, 128.0, 127.1, 125.6, 123.9 (d, JCF = 3.6 Hz), 116.7 (d, JCF = 23.2 Hz), 113.3, 31.2, 30.8 ppm; IR (thin film) 3029, 1683, 1560, 1494, 1452, 1350, 1286, 1259, 1222, 1135, 1127, 1075, 1054, 1030, 1014, 967, 817, 755, 725, 696, 643 cm−1; HRMS (ESI+) calcd for C24H21FN3O m/z (M + H) 386.1669, found 386.1677. N-(4-Benzyl-5-(4-methoxyphenyl)-1-methyl-1H-imidazol-2(3H)-ylidene)-2-fluorobenzamide (20h) Prepared according to general procedure D using 19h, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give a light brown foam (13 mg, 81%): Rf = 0.41 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 500 MHz) δ 8.07 (t, J = 8.0 Hz, 2H), 7.34 (m, 1H), 7.32−7.27 (m, 4H), 7.21 (t, J = 8.5 Hz, 1H), 7.19−7.16 (m, 3H), 7.08 (t, J = 9.5 Hz, 1H), 7.00 (d, J = 8.5 Hz, 2H), 3.86 (s, 3H), 3.81 (s, 2H), 3.44 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 171.8, 162.5 (d, JCF = 253.4 Hz), 160.2, 148.7, 137.8, 131.8, 131.7 (d, JCF = 1.9 Hz), 131.6, 128.8, 128.2, 126.8, 126.4, 124.8, 123.6 (d, JCF = 3.8 Hz), 119.9, 116.5 (d, JCF = 22.9 Hz), 114.5, 55.4, 31.0, 30.1 ppm; IR (thin film) 2929, 2360, 2340, 1684, 1569, 1511, 1494, 1455, 1401, 1339, 1290, 1248, 1176, 1032, 834, 815, 757, 731, 696, 667 cm−1; HRMS (ESI+) calcd for C25H22FN3NaO2 m/z (M + Na) 438.1594, found 438.1601. N-(4-Benzyl-5-(4-methoxyphenyl)-1-methyl-1H-imidazol-2(3H)-ylidene)-3-(trifluoromethyl)benzamide (20i) Prepared according to general procedure D using 19i, with purification using silica gel eluting with 3:2 hexanes/EtOAc to give a light brown foam (99 mg, 53% yield): Rf = 0.76 (3:2 hexanes/EtOAc); 1H NMR (CDCl3, 300 MHz) δ 8.54 (s, 1H), 8.42 (d, J = 7.8 Hz, 1H), 7.67 (d, J = 7.2 Hz, 1H), 7.50 (t, J = 8.4 Hz, 1H), 7.32−7.23 (m, 5H), 7.15 (d, J = 6.9 Hz, 2H), 7.03 (d, J = 9.0 Hz, 2H), 3.87 (s, 3H), 3.84 (s, 2H), 3.50 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 173.0, 160.3, 150.5, 139.4, 137.3, 131.9, 131.6, 130.1 (q, JCF = 32.2 Hz), 129.0, 128.2, 128.1, 127.0, 126.9 (q, JCF = 3.8 Hz), 125.7 (q, JCF = 3.8 Hz), 124.3 (q, JCF = 270.4 Hz), 120.0, 119.4, 114.6, 55.4, 30.5, 29.9 ppm; IR (thin film) 1569, 1512, 1466, 1363, 1317, 1278, 1249, 1217, 1165, 1121, 1069, 1034, 906, 834, 768, 725 cm−1; HRMS (ESI+) calcd for C26H22F3N3NaO2 m/z (M + Na) 488.1562, found 488.1559. Supplementary Material SI R.E.L. thanks the NIH, General Medical Sciences (R01 GM090082, P41 GM08915), Cu̅rza, Amgen, and Eli Lilly for financial support. B.E.W. thanks the NIH, National Cancer Institute (R01 CA140296), for funding. We thank Dr. Atta Arif (U. of U. Chemistry) for help with X-ray crystallography studies. We thank John Sullivan and Richard Nkansah for exploratory work on this project. Figure 1 Representative Leucetta alkaloids. Figure 2 X-ray structures of 18b and 20h. Figure 3 Antiproliferative effects of naamidine A and 20h. Scheme 1 First Generation Synthesis of Naamidine A and Analogues Scheme 2 Synthesis of Naamidine A (6) Scheme 3 Generation of N2-Acyl Naamidine A Analogues Scheme 4 Generation of N2-Acyl-2-aminoimidazoles Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.joc.5b01703. X-ray data for 18b (CIF) X-ray data for 20h (CIF) X-ray crystallography data for compounds 18b and 20h. 1H and 13C NMR spectra for all compounds (PDF) Elemental composition report (PDF) Notes The authors declare no competing financial interest. 1 (a) Sullivan JD Giles RL Looper RE Curr Bioact Compd 2009 5 39 78 (b) Koswatta PB Lovely CJ Nat Prod Rep 2011 28 511 528 20981389 (c) Roué M Quévrain E Domart-Coulon Bourguet-Kondracki M-L Nat Prod Rep 2012 29 739 751 22660834 2 Dunbar DC Rimoldi JM Clark AM Kelly M Hamann MT Tetrahedron 2000 56 8795 8798 3 Crews P Clark DP Tenney K J Nat Prod 2003 66 2 177 182 12608847 4 Gibbons JB Gligorich KM Welm BE Looper RE Org Lett 2012 14 4734 4737 22966873 5 Das J Bhan A Mandal SS Lovely CJ Bioorg Med Chem Lett 2013 23 6183 6187 24076171 6 Gross H Kehraus S Konig GM Woerheide G Wright AD J Nat Prod 2002 65 1190 1193 12193030 7 Chan GW Mong S Hemling ME Freyer AJ Offen PH DeBrosse CW Sarau HM Westley JW J Nat Prod 1993 56 116 121 8383730 8 Copp BR Fairchild CR Cornell L Casazza AM Robinson S Ireland CM J Med Chem 1998 41 3909 3911 9748366 9 James RD Jones DA Aalbersberg W Ireland CM Mol Cancer Ther 2003 2 747 751 12939464 10 LaBarbera DV Modzelewska K Glazar AI Gray PD Kaur M Liu T Grossman D Harper MK Kuwada SK Moghal N Ireland CM Anti-Cancer Drugs 2010 20 425 436 11 Ohta S Tsuno N Nakamura S Taguchi N Yamashita M Kawasaki I Fujieda M Heterocycles 2000 53 1939 1955 12 (a) Aberle NS Lessene G Watson KG Org Lett 2006 8 419 421 16435849 (b) Aberle N Catimel J Nice EC Watson KG Bioorg Med Chem Lett 2007 17 3741 3748 17462892 13 Ermolat’ev DS Bariwal JB Steenackers HPL De Keersmaecker SCJ Van der Eycken EV Angew Chem, Int Ed 2010 49 9465 4968 14 Gainer MJ Bennett NR Takahashi Y Looper RE Angew Chem, Int Ed 2011 50 684 687 15 Das J Koswatta PB Jones JD Yousufuddin M Lovely CJ Org Lett 2012 14 6210 6213 23215346 16 Su Z Peng L Melander C Tetrahedron Lett 2012 53 1204 1206 17 Giles RL Sullivan JD Steiner AM Looper RE Angew Chem, Int Ed 2009 48 3116 3120 18 Zhang N Zhang Z Wong ILK Wan S Chow LMC Jiang T Eur J Med Chem 2014 83 74 83 24952376 19 For leading references, see: (a) Wei C Li Z Li CJ Synlett 2004 1472 1483 (b) Zani L Bolm C Chem Commun 2006 4263 4275 (c) Li CJ Acc Chem Res 2010 43 581 590 20095650 (d) Koradin C Polborn K Knochel P Angew Chem, Int Ed 2002 41 2535 2538 (e) Gommermann N Koradin C Polborn K Knochel P Angew Chem, Int Ed 2003 42 5763 5766 20 Genêt JP Blart E Savignac M Lemeune S Lemaire-Audoire S Bernard JM Synlett 1993 1993 680 682 21 (a) Looper RE Haussener TJ Mack JBC J Org Chem 2011 76 6967 6971 21732649 (b) Kwon K Haussener TJ Looper RE Org Synth 2015 92 91 102 27812231 22 Gligorich KM Vaden RM Shelton DN Wang G Matsen CB Looper RE Sigman MS Welm BE Breast Cancer Res 2013 15 4 R58 23879992

PMC005xxxxxx/PMC5117190.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8610640 104 J Bone Miner Res J. Bone Miner. Res. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 0884-0431 1523-4681 24839202 5117190 10.1002/jbmr.2282 NIHMS788576 Article IGFBP-2 directly stimulates osteoblast differentiation Xi Gang Ph.D. 1 Wai Christine B.S. 1 DeMambro Victoria M.S. 2 Rosen Clifford J M.D. 2 Clemmons David R M.D. 1* 1 Department of Medicine, University of North Carolina at Chapel Hill 2 Maine Medical Center Research Institute * Address all correspondence and requests for reprints to: David R. Clemmons, M.D, CB7170, 8024 Burnett Womack, University of North Carolina, Chapel Hill, North Carolina 27599-7170, david_clemmons@med.unc.edu. Tel: (919) 966-4735, Facsimile: (919) 966-6025 6 7 2016 11 2014 21 11 2016 29 11 24272438 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Insulin like growth factor binding protein two (IGFBP-2) is important for acquisition of normal bone mass in mice; however, the mechanism by which IGFBP-2 functions is not defined. These studies investigated the role of IGFBP-2 in stimulating osteoblast differentiation. MC-3T3 preosteoblasts expressed IGFBP-2, and IGFBP-2 knockdown resulted in a substantial delay in osteoblast differentiation, reduced osteocalcin expression and Alizarin red staining. These findings were replicated in primary calvarial osteoblasts obtained from IGFBP-2 −/− mice and addition of IGFBP-2 rescued the differentiation program. In contrast, overexpression of IGFBP-2 accelerated the time course of differentiation as well as increasing the total number of differentiating cells. By day 6 IGFBP-2 overexpressing cells expressed twice as much osteocalcin as control cultures and this difference persisted. To determine the mechanism by which IGFBP-2 functions, the interaction between IGFBP-2 and receptor tyrosine phosphatase β (RPTPβ) was examined. Disruption of this interaction inhibited the ability of IGFBP-2 to stimulate AKT activation and osteoblast differentiation. Knockdown of RPTPβ enhanced osteoblast differentiation whereas overexpression of RPTPβ was inhibitory. Adding back IGFBP-2 to RPTPβ overexpressing cells was able to rescue cell differentiation via enhancement of AKT activation. To determine the region of IGFBP-2 that mediated this effect an IGFBP-2 mutant that contained substitutions of key amino acids in the heparin binding domain-1 (HBD-1) was prepared. This mutant had a major reduction in its ability to stimulate differentiation of calvarial osteoblasts from IGFBP-2 −/− mice. Addition of a synthetic peptide that contained the HBD-1 sequence to calvarial osteoblasts from IGFBP-2 −/− mice rescued differentiation and osteocalcin expression. In summary, the results clearly demonstrate that IGFBP-2 stimulates osteoblast differentiation and that this effect is mediated through its heparin binding domain-1 interacting with RPTPβ. The results suggest that stimulation of differentiation is an important mechanism by which IGFBP-2 regulates the acquisition of normal bone mass in mice. IGFBP-2 Osteoblast differentiation pAKT PTEN RPTPβ Introduction IGFBP-2 is a member of a family of six IGF binding proteins. Although a major function of this class of proteins is to transport the IGFs through the circulation and extracellular fluids, thereby restricting their access to receptors, each form of binding protein has been found to have distinct actions.(1) Initial studies showed that IGFBP-2 can enhance the effect of IGF-II to stimulate alkaline phosphatase in bone cell cultures.(2) When IGFBP-2 gene expression was deleted in mice, tibial bone volume was reduced and both micro CT and pQCT analysis showed diminished trabecular number and volume.(3) A subsequent study defined a 13 amino acid region of IGFBP-2 (termed HBD-1) that was required for biologic activity. Substitution of key residues within this region resulted in loss of the ability of IGFBP-2 to stimulate osteoblast proliferation in vitro.(4) Importantly when a synthetic peptide containing the HBD-1 sequence was injected into the IGFBP-2 −/− mice, micro CT analysis showed that trabecular volume and density could be rescued.(4) Furthermore this peptide was shown to stimulate osteoblast proliferation in vivo. Prior studies have suggested that IGF-I and IGFBP-2 play a role in osteoblast differentiation. Cell type specific deletion of IGF-I in osteoblasts resulted in decreased femoral BMD and decreased bone formation rate.(5) Some studies have suggested a correlation between IGFBP-2 and osteoblast differentiation. During induction of differentiation in the osteoblast sheets, analysis of gene expression profiles showed that IGFBP-2 is one of the genes that showed the greater increase.(6) PTH increases IGFBP-2 expression in differentiated osteoblasts.(7) Finally mesenchymal stromal cells can be made to further differentiate into osteoblasts with dexamethasone and this requires the interaction of the α5 integrin and IGFBP-2.(8) Although recent studies have documented the importance of the HBD-1 domain of IGFBP-2 for osteoblast growth, the relative importance of this domain for osteoblast differentiation has not been determined. A recent study demonstrated that the HBD-1 region bound directly to a cell surface receptor termed receptor tyrosine phosphatase β and that RPTP-β was expressed by MC-3T3 cells.(9) It further demonstrated that IGFBP-2 binding to this receptor induced RPTPβ dimerization which inhibited its phosphatase activity. Since the primary substrate of this phosphatase was shown to be PTEN, subsequent analysis showed that engagement of this receptor on osteoblast surfaces resulted in enhanced tyrosine phosphorylation of PTEN which inhibited its activity. This was associated with increased AKT activation. These data imply that IGFBP-2 may be functioning directly to augment constitutive AKT activation in osteoblasts. Since AKT activation has been linked to osteoblast differentiation(10), the current studies were undertaken to determine if IGFBP-2 could directly stimulate osteoblast differentiation, if it was functioning through interaction with RPTPβ and if this interaction was mediated through the HBD-1 domain. Materials and Methods Human IGF-I was a gift from Genentech (South San Francisco, CA). Immobilon-P membrane, LY294002 and PD98059 were purchased from EMDmillipore Corp. (Billerica, MA). α-MEM, streptomycin and penicillin were purchased from Life Technologies (Grand Island, NY). Anti-RPTPβ antibody was purchased from BD Bioscience (San Diego, CA). Antibodies against phospho-AKT (S473), pErk1/2, cleaved caspase-3 and PTEN were purchased from Cell Signaling Technology Inc. (Beverly, MA). Anti-phospho-tyrosine (PY99), osteocalcin, β-actin antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). PQ401 was purchased from TOCRIS bioscience (Bristol, United Kindom). IGFBP-2 antiserum was prepared as previously described.(11) The horseradish peroxidase-conjugated mouse anti-rabbit, goat anti-mouse, and mouse anti-rabbit light chain-specific antibodies were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA). All other reagents were obtained from Sigma unless otherwise stated. The synthetic peptide containing the linker located heparin-binding domain of IGFBP-2 (188KHLSLEEPKKLRP200) (referred to as HBD-1 peptide) and a scrambled HBD peptide (CKPLRLSKEEHPLK) (referred to as HBD control peptide), were synthesized by the Protein Chemistry Core Facility at the University of North Carolina at Chapel Hill. Purity and the sequences were confirmed by mass spectrometry. Mice Generation of the original mixed background strain B6;129-Igfbp2<tm1Jep>, which we refer to as Igfbp2−/− mice, has been described previously.(3) The original mice were backcrossed onto C57BL/6J background for 10 generations. Igfbp2+/+ mice were C57BL/6J controls. All of the experimental studies were performed with male mice. All of the animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of University of North Carolina at Chapel Hill. Cell culture MC-3T3 E1 clone 4 (CL4) cells were obtained from ATCC (Manassas, VA). Cells were cultured in α-MEM (glucose 1000mg/L) containing 10% fetal bovine serum (Thermo Fishers Scientific, Pittsburgh, PA). After confluency, culture medium was changed to differentiation medium (DM) which contained 10% fetal bovine serum plus 50 ug/ml ascorbic acid and 4 mM β-glycerol phosphate. Fresh DM was applied every 72 hr. IGFBP-2 (1 ug/ml), HBD-1 (1ug/ml or as stated), control peptide (1ug/ml) or HBD mutant IGFBP-2 (1 ug/ml) were added to the differentiation medium, and replaced every 72 hr. Neonatal calvarial osteoblasts were isolated from 3–5-day-old mice. Briefly, calvariae were digested five times with collagenase type 2 (250 unit/ml) and trypsin (0.05%) plus EDTA (0.02%) in the PBS. The cells released from digests 2–5 were collected as primary calvarial osteoblasts and maintained in DMEM (glucose 1000 mg/L) supplemented with 10% FBS and nonessential amino acids. Construction of cDNAs and establishment of MC3T3 cells expressing wild type IGFBP-2, RPTPβ and LacZ Mouse IGFBP-2 cDNA was amplified from mouse pCMV-SPORT6 (ATCC, Manassas, VA) using a 5′ primer sequence corresponding to nucleotides 89 to 110 of mouse IGFBP-2 (5′-ATGCTGCCGAGATTGGGCGGCC-3′) and a 3′ primer sequence complementary to nucleotides 981 to 1003 (5′-GGGCCCATGCCCAAAGTGTGCAG-3′). After DNA sequencing to confirm that the correct sequence had been amplified, the PCR product was subcloned into pENTR/D-TOPO vector and subsequently transferred into the pLenti6-V5 DEST expression vector using the LR Clonase reaction and following the manufacturer's instructions (Life Technologies, Grand Island, NY). Constructions of RPTPβ and LacZ have been described previously.(9) The constructs contained the correct sequences was verified by DNA sequencing. 293FT cells (Life Technologies, Grand Island, NY) were prepared for generation of virus stocks and CL4 expressing IGFBP-2, RPTPβ and LacZ were established using procedures that have been described previously.(12) Construction of cDNAs and establishment of IGFBP-2 Si, RPTPβ Si and LacZ Si cells Based on Life Technologies’ website design tools, a sequence containing 21 oligonucleotides (GGAAAGAGACCAACACTGAGC) was used to construct the shRNA template plasmid to inhibit the translation of mouse IGFBP-2 mRNA. GCCAATGCATACAGCAGTAAT was used to construct the shRNA template to knock down mouse RPTPβ. The oligonucleotides were synthesized by Nucleic Acids Core Facility at UNC, annealed and ligated into BLOCK-iT™ U6 RNAi Entry Vector (Cat# K4945-00, Life Technologies, Grand Island, NY) following manufacturer’s instructions. The complete sequence was verified by DNA sequencing. The expression vector was generated using the Gateway LR recombination reaction between the Entry Vector and BLOCK-iT™ Lentiviral RNAi Gateway® Vector (Cat# K4943-00, Life Technologies, Grand Island, NY). A sequence targeting LacZ was used as a control. After confirmation of the sequence, plasmid DNA was prepared using a Plasmid Midi Kit (Promega, Madison, WI). 293FT cells (Life Technologies, Grand Island, NY) were transfected and used to prepare for generation of virus stocks.(12) CL4 cells expressing small hairpin RNA sequence targeting IGFBP-2 (IGFBP-2 Si), RPTPβ (RPTPβ Si) and corresponding control CL4 expressing small hairpin RNA sequence targeting LacZ (LacZ Si) were established using procedures described previously.(12) Immunoprecipitation and Immunoblotting The cell monolayers were lysed in a modified radioimmunoprecipitation assay (RIPA) buffer as previously described.(12) Immunoprecipitation was performed by incubating 0.5 mg of cell lysate protein with 1 ug of each of the following antibodies: anti-IGFBP-2 and PY99 at 4°C overnight. Immunoblotting was performed as previously described(12) using a dilution 1:1000 for anti-pAKT (Ser473), PTEN and β-actin antibodies, a dilution 1:500 for anti-RPTPβ antibody, a dilution 1:200 for anti-osteocalcin antibody and a dilution 1:10000 for anti-IGFBP-2 antibody. The proteins were visualized using enhanced chemiluminescence (Thermo Fisher Scientific, Rockford, IL). Total cellular protein in the lysates was determined using BCA (Thermo Fisher Scientific, Rockford, IL). Alizarin Red staining Cells were washed with PBS twice before were fixed with 10% formalin. After 10 min fixation, 1% Alizarin Red (pH 4.2) was applied and incubated for another 10 min before it was removed. Cells were washed with ddH2O twice and drying. Images were captured using Leica M420 Microscope. RNA isolation and quantitative real-time PCR Total RNA was prepared using RNeasy plus mini kit (Qiagen, Valencia, CA, USA) for cellular extracts. cDNA was then generated from 500 ng of RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) per the manufacturer’s instructions. Quantitative real-time expression analysis was run on the CFX384 Real-time System using the iQ SYBR Green Supermix and C1000 thermal Cycler (Bio-Rad, Hercules, CA, USA). Relative expression of mRNA was determined after normalization to Hprt levels using the ΔCt method. Primers were designed, sequenced and validated to be 95% to 100% efficient by Primer Design Ltd (Southampton, UK). All primer sequences are listed in Supplementary Table 1. Cell proliferation and apoptosis assay Calvarial osteoblasts isolated from IGFBP-2 −/− mice were seeded in 6 well plate. After reaching confluency the culture medium was changed to DM or DM plus HBD-1 or DM plus IGFBP-2. Control cells and IGFBP-2 overexpressing cells were plated in 24 well plates using the same plating density. After reaching confluency the culture medium was changed to DM. Fresh DM was applied every 72 hr. After cells were exposed to DM for indicated days, the cells were released with 0.05% Trypsin-EDTA and counted. To quantify apoptosis, calvarial osteoblasts isolated from IGFBP-2 −/− mice were exposed to DM alone or DM plus the different concentrations of HBD-1 peptide for 21 days. Cell lysates were harvested as described previously and immunoblotted using an anti-cleaved caspase-3 antibody. Statistical analysis Densitometry results are expressed as the mean ± standard deviation (SD). All experiments were replicated at least three times to assure reproducibility. The results were analyzed for statistically significant differences using Student’s t-test or analysis of variance followed by Bonferroni multiple comparison post hoc test. Statistical significance was set at p<0.05. Results IGFBP-2 stimulates osteoblast differentiation Since we had shown that IGFBP-2 enhances AKT activation in osteoblasts,(9) we determined if IGFBP-2 regulates osteoblast differentiation. MC-3T3 cells have been shown to secrete IGFBP-2 and its secretion increases significantly between day 6 and 9 following the addition of differentiation medium. RNAi was used to determine the significance of these changes and if inhibiting IGFBP-2 synthesis would alter differentiation (Fig 1A). Compared to control cultures the differentiation of MC-3T3 cells in which IGFBP-2 synthesis had been inhibited was significantly attenuated. Both osteocalcin expression (Fig 1B) and the number of alizarin red positive cells were reduced (Fig 1C). To confirm the importance of IGFBP-2 for differentiation of preosteoblasts, calvarial pre-osteoblasts, isolated from IGFBP-2 −/− mice were analyzed. These cells showed impaired osteocalcin expression and differentiation compared to cells from control littermates (Fig 1D) (e.g., a 2.3 ± 0.1 fold greater level of osteocalcin in control cells on day 21, compared to IGFBP-2 −/− cells, p<0.01). The time course of differentiation was prolonged in cultures from the IGFBP-2 −/− mice and only 1.8 ± 0.2% cells had completed differentiation by day 21 (Fig 1E). The addition of IGFBP-2 to these cultures restored differentiation (Fig 1E). In contrast, overexpression of IGFBP-2 in MC-3T3 cells significantly enhanced the osteocalcin expression (Fig 1G) (e.g., a 2.5 ± 0.1 fold greater level of osteocalcin on day 6 compared to control cultures, p<0.05). In addition, in IGFBP-2 overexpressing cells osteocalcin was detected on day 3 whereas it was detected on day 6 following the addition of differentiation medium in control cells (Fig 1G). Major differences in osteocalcin expression were detected at each time point and persisted up to day 21. In addition alizarin red positive cells were detected on day 6 in IGFBP-2 overexpressing cells whereas they were not detected until day 15 in control cells (Fig 1H). To confirm these results, we analyzed the expression of several genes that are important for osteoblast differentiation. As shown in supplemental figures 1 and 2, osteocalcin, alkaline phosphatase and Wnt10b were induced significantly in the IGFBP-2 overexpressing cells compared to control cells on days 6 and 9, thereby reflecting the acceleration of differentiation. Osteopontin was significantly increased at day 9. Runx2 and osterix whose expression peaks early in differentiation declined between days 3 and 9 in the control cells and were significantly decreased in the IGFBP-2 overexpression cells whereas following IGFBP-2 knockdown, their expression was increased compared to control cultures, suggesting that differentiation is delayed (Supplemental Fig 1 and 2). Overexpression of IGFBP-2 reduced the amount of serum supplementation that was necessary to induce differentiation. For example the addition of 5% serum containing differentiation medium to IGFBP-2 transfected cells induced a similar level of cell differentiation compared to control cells exposed to 10% serum (Supplemental Fig 3A). IGF-I stimulated differentiation of control and IGFBP-2 overexpressing cells but the percentage of cells that differentiated remained significantly greater in the IGFBP-2 overexpressing cells (Supplemental Figure 3B and C). These results strongly suggest that IGFBP-2 is able to stimulate preosteoblast differentiation and that it is one of the factors that is present in 10% FBS that induces these changes. AKT and PI-3 kinase activation are required for osteoblast differentiation Several sub clones of MC-3T3 cells were originally derived from mouse calvarial osteoblasts.(13) Among them: clone 4 cells (CL4 cells) were found to differentiate in the appropriate medium. Since previous studies have shown the importance of AKT activation for osteoblast differentiation(14), we determined the IGF-I-stimulated AKT activation response during differentiation. The results showed that AKT Ser473 phosphorylation was stimulated by IGF-I during differentiation phase in MC-3T3, CL4 cells (Fig 2A). A clone of MC-3T3 cells derived from the same parental cell line (CL24) is unable to differentiate. When these cells were analyzed neither basal nor IGF-I-stimulated AKT Ser473 phosphorylation could be detected after several days in differentiation medium (Fig 2A). Importantly, overexpression of IGFBP-2 enhanced IGF-I stimulated AKT activation (Fig 2A) In order to determine the time point in the differentiation cycle wherein PI-3 kinase activation was required to induce preosteoblast differentiation, the PI-3 kinase inhibitor, LY 294002, was utilized. The results show that inhibition of PI-3 kinase completely prevented differentiation when the inhibitor was applied on day 3 following the addition of differentiation medium (Fig 2B). It also significantly suppressed differentiation when it was applied on day 6. However if it was applied after day six the inhibitory effect was minimal at day 9 and completely lost by day 12 or later (Fig 2B). When this experiment was repeated using cells overexpressing IGFBP-2, differentiation was detectible and was attenuated when the inhibitor was added on day 1 or day 3 but it was not altered if the inhibitor was added at day 6 or later (Fig 2C). Since inhibition of PI-3 kinase activation prevented differentiation, we examined the time course of AKT phosphorylation at different time points during differentiation in control and IGFBP-2 overexpressing cells. AKT activation was minimal at day 3 in control cells but it was increased significantly in the overexpressing cells (Fig. 2D). (e.g., to a level that was 7.5 fold greater at day 3, p<0.001). These results are consistent with the differences in osteocalcin expression shown in Fig 1G. When AKT activation was analyzed in primary osteoblasts that were obtained from IGFBP-2 −/− mice there was a significant reduction in constitutive pAKT expression compared to cells from control +/+ animals on days 3, 6 and 9 (Fig 2E) (e.g., to a level that was 3.1 ± 0.4 fold less than osteoblasts from +/+ mice at day 9, p<0.01). Importantly the differences in constitutive AKT activation correlated with those detected in osteocalcin expression (Fig 1D). IGFBP-2 enhances AKT activation via suppressing RPTPβ dephosphorylation of PTEN Our previous study in smooth muscle cells showed that IGFBP-2 enhances IGF-I-stimulated AKT activation via direct binding of IGFBP-2 to RPTPβ which catalyzes its polymerization and thereby inhibits its ability to dephosphorylate PTEN.(9) That study also showed that MC-3T3 cells expressed RTPTβ and that IGFBP-2 exposure increased PTEN tyrosine phosphorylation. Since tyrosine phosphorylation of PTEN attenuates its ability to inhibit AKT activation, this results in an enhancement of constitutive and IGF-I-stimulated AKT phosphorylation. RPTPβ expression increases during osteoblast differentiation(15), therefore we hypothesized that IGFBP-2 enhanced AKT activation during osteoblast differentiation through this same mechanism. To test this hypothesis we first examined whether IGFBP-2 overexpression and IGF-I addition stimulated IGFBP-2/RPTPβ association during differentiation. The results showed that the formation of the IGFBP-2/RPTPβ complex was detected on day 6 in control cultures (Fig 3A). This is consistent with the level of constitutive AKT activation (Fig 2D). Following IGF-I stimulation, complex formation increased in control cells and, in cells overexpressing IGFBP-2, there was an increase in basal and IGF-I stimulated IGFBP-2/RPTPβ association on days 3 and 6 compared to control cells (Fig 3A) (e.g., 3.6 ± 0.9 fold and 2.6 ± 0.1 fold increases in IGF-I stimulated complex formation in IGFBP-2 overexpressing cells compared to LacZ cells on days 3 and 6, p<0.05, respectively). Correspondingly PTEN tyrosine phosphorylation was increased in the IGFBP-2 overexpressing cells compared to control cells on days 3 and 6 (Fig 3B). To directly determine the role of RPTPβ on osteoblast differentiation, we manipulated the RPTPβ level in MC-3T3 cells. Knockdown of RPTPβ during the differentiation phase enhanced basal and IGF-I stimulated PTEN tyrosine phosphorylation as well as AKT activation (Fig 3C) (e.g., a 2.3 ± 0.2 fold greater level of pAKT expression in RPTPβ Si cells compared to control, p<0.05). Further analysis showed that osteocalcin expression was enhanced on days 9, 12, and 21 (Fig 3D) as well as cell differentiation on days 15, 18 and 21 compared to control cultures (Fig 3E). In contrast overexpression of RPTPβ inhibited PTEN tyrosine phosphorylation (Fig 3F) and impaired osteoblast differentiation (Fig 3G). We have shown previously that IGFBP-2 binding to RPTPβ induces polymerization which inhibits its phosphatase activity. When IGFBP-2 was added back to the RPTPβ overexpressing cells there was enhanced osteocalcin expression at day 12 and this also rescued differentiation (Fig 3H). Correspondingly the addition of IGFBP-2 enhanced basal (e.g., a 3.9 ± 0.4 fold greater compared to control, p<0.01) and IGF-I-stimulated AKT phosphorylation (e.g., a 1.7 ± 0.2 fold greater compared to no IGFBP-2 treatment, p<0.05) (Fig 3I). These results clearly show that IGFBP-2 regulation of RPTPβ activity plays an important role in preosteoblast differentiation and that RPTPβ regulates osteoblast differentiation through modulation of PTEN tyrosine phosphorylation. Disruption of IGFBP-2/RPTPβ interaction impairs IGF-I-stimulated AKT activation and osteoblast differentiation To confirm the importance of the IGFBP-2/RPTPβ interaction, MC-3T3 cells overexpressing IGFBP-2 were analyzed. Following the addition of IGF-I there was a significant increase in IGFBP-2/RPTPβ association (Fig 4A) (e.g., a 2.5 ± 0.2 fold greater level of complex formation in IGFBP-2 overexpressing cells compared to control cells p< 0.05). To inhibit this interaction we utilized an anti-RPTPβ blocking antibody and determined its effect on the IGFBP-2/RPTPβ interaction, downstream signaling and differentiation. The antibody inhibited IGF-I stimulated IGFBP-2/RPTPβ association in control and IGFBP-2 overexpressing cells (Fig 4A). Disruption of their interaction was functionally significant since it inhibited AKT activation (Fig 4B) and osteocalcin expression (Fig 4C) (e.g., 74 ± 8% reduction in osteocalcin with 500ng/ml, p<0.01). Since IGF-I stimulates IGFBP-2/RPTPβ association that is critical for AKT activation(9) and osteoblast differentiation, we determined whether the requirement for IGF-I was changed when IGFBP-2 was overexpressed. To block IGF-I signaling, PQ401, a specific inhibitor of IGF-I receptor tyrosine kinase was used. The results show that PQ401 treatment significantly impaired osteocalcin expression and cell differentiation in IGFBP-2 overexpressing cells (Fig 4D and E). HBD-1 peptide mediates the IGFBP-2 effect on osteoblast differentiation The HBD-1 domain of IGFBP-2 mediates its stimulatory effect on osteoblast proliferation.(4) To investigate the importance of the HBD-1 domain for osteoblast differentiation, we utilized an IGFBP-2 mutant in which the charged amino acids within the HBD-1 sequence were changed to alanine and an 13 amino acid synthetic peptide that contained this sequence and examined their abilities to alter the differentiation of calvarial preosteoblasts isolated from IGFBP-2 null mice. The results show that unlike wild type IGFBP-2 when the HBD-1 mutant form of IGFBP-2 was added, its ability to stimulate preosteoblast differentiation was significantly impaired (Fig 5A). To further investigate the function of HBD-1 domain, an HBD-1 peptide was added to differentiation medium. The peptide was able to rescue IGFBP-2 −/− cell differentiation (Fig 5A). When the results were quantified the differences were significant (Fig 5A). When increasing concentrations of the HBD-1 peptide were added a substantial increase in the percentage of cells that differentiated was noted at 500 ng/ml and it increased further with 1000 ng/ml (Fig 5B). When osteocalcin expression was analyzed this effect was confirmed and there was an incremental increase between 250 and 1000 ng/ml (Fig 5C). To determine whether HBD-1-enhanced cell differentiation was due to change of cell survival, we measured the cleaved caspase-3, an indicator for cell apoptosis. The results showed that the HBD-1 peptide had no effect on osteoblast apoptosis (Supplemental Figure 4A). Discussion Although IGFBP-2 functions with IGF-II to increase bone mass(16), and IGFBP-2 knockout mice have decreased cortical and trabecular bone(3–4), the specific role of IGFBP-2 in modifying osteoblast differentiation has not been reported. Prior studies showed that IGFBP-2 enhances the effect of both IGF-I and IGF-II in stimulating bone accretion in vivo.(4,16) Subsequently we showed that a peptide containing the HBD-1 sequence rescues the normal bone phenotype in IGFBP-2 −/− mice and that this peptide stimulated osteoblast proliferation.(4) These studies extend those findings to demonstrate that the HBD-1 peptide as well as intact IGFBP-2 stimulates osteoblast differentiation. The results clearly demonstrate that overexpression of IGFBP-2 results in acceleration of the differentiation program as well as increasing the total number of cells reaching the stage of mature osteoblast formation. Proteins that are markers of differentiation, such as osteocalcin, are increased in response to IGFBP-2 and they are expressed earlier in the differentiation program following IGFBP-2 stimulation. The results show that the expression of Wnt10b, alkaline phosphatase and osteopontin were increased in a similar manner. The effect of IGFBP-2 is mediated through the cell surface receptor RPTPβ since addition of an antibody which inhibited its binding to this receptor significantly attenuated its ability to stimulate signaling events that are linked to osteoblast differentiation. Moreover, knockdown of this receptor inhibited the ability of IGFBP-2 to stimulate differentiation. That the HBD-1 domain was important for signaling within the intact protein was confirmed using site directed mutagenesis. Specifically addition a mutant with an altered HBD-1 sequence resulted in attenuated differentiation. Both the time course and the absolute number of cells as well as expression of osteocalcin were deceased. Our prior studies showed that an 13 amino acid peptide containing the HBD-1 sequence stimulated trabecular bone formation in IGFBP-2 −/− mice.(4) Keipe et al(17) demonstrated that a 117 amino acid carboxy terminal fragment of IGFBP-2 that would have contained the HBD-1 sequence exerted a strong mitogenic effect on growth plate chondrocytes and the effect was equal to intact IGFBP-2. These studies extend those observations to show that a peptide encompassing the sequence of HBD-1 is sufficient to stimulate osteoblast differentiation through its interaction with RPTPβ. Our prior study showed that RPTPβ was present on the surface of MC-3T3 cells and that IGFBP-2 could interact with this protein to alter PTEN tyrosine phosphorylation.(9) IGFBP-2 stimulated RPTPβ polymerization thereby attenuating its phosphatase activity. Since PTEN is a RPTPβ substrate, this resulted in increased tyrosine phosphorylation of PTEN which attenuated PTEN enzymatic activity thereby leading to increased AKT phosphorylation. These studies extend those observations showing that enhancement of constitutive AKT phosphorylation occurs concomitantly with earlier differentiation in cells that overexpress IGFBP-2 and that these responses are attenuated when IGFBP-2 expression is diminished. That these changes are mediated through RPTPβ was proven by demonstrating that inhibition of IGFBP-2 binding to RPTPβ could block enhanced AKT activation and differentiation, and that knocking down of RPTPβ resulted in escape from its ability to inhibit AKT activation as well as a reduction in cell responsiveness to intact IGFBP-2. The studies also demonstrated that constituently synthesized IGFBP-2 is important for osteoblast differentiation. Knockdown of constituently synthesized IGFBP-2 resulted in attenuation of differentiation and expression of osteocalcin as well as constitutive AKT phosphorylation. Addition of an antibody that inhibits the binding of IGFBP-2 to RPTPβ could attenuate these responses in non-transfected MC-3T3 cells. Therefore, these results further support the conclusion that IGFBP-2 functions by attenuating RPTPβ mediating PTEN dephosphorylation and stimulating AKT activation, leading to enhanced osteoblast differentiation. The importance of AKT activation for differentiation was also shown by inhibiting PI-3 kinase, which is upstream of AKT. The addition of a PI-3 kinase inhibitor, after 3 day exposure to differentiation medium, completely prevented osteoblast differentiation. However, to obtain similar level of inhibition in IGFBP-2 overexpressing cells the inhibitor needed to be added at an earlier time point. Since previous studies have also shown that suppression of MAP kinase activation stimulated osteoblast differentiation,(14,18–19) we also analyzed MAP kinase activation using a similar experimental paradigm. Consistent with prior reports, our results showed that inhibition of MAP kinase activation significantly stimulated osteoblast differentiation, however, this stimulation was only detected when inhibitor was added at the early stage of cell differentiation, such as on day 3 and day 6 (Supplemental Fig 4B). Importantly, over-expression of IGFBP-2 did not significantly alter MAP kinase activation, compared to control cells (Supplemental Fig 4C), indicating that MAP kinase pathway did not play an important role in mediating the stimulatory effect of IGFBP-2 on osteoblast differentiation. Consistently, exogenous addition of a peptide containing HBD-1 sequence or IGFBP-2 or overexpression of IGFBP-2 had no significant effect on cell proliferation in the differentiation medium (Supplemental Fig 4D and E). Other studies have suggested that expression of IGFBP-2 correlates with changes in osteoblast differentiation although they have not shown the direct causal links reported herein. Specifically Kawase et al induced chondrocytes sheets to differentiate into osteoblasts and showed increased secretion of both IGF-I and IGFBP-2 that occurred during deposition of osteoid and mineralized tissue.(6) Similarly, Hamidouche et al. demonstrated that induction of osteoblast differentiation from mesenchymal stromal cells was accompanied by an increase in the synthesis of IGF-II, IGFBP-2 and the α5 integrin subunit.(8) Both IGF-II and IGFBP-2 were shown to increase the expression of phenotypic markers as well as the in vitro osteogenic capacity of the cells. They also demonstrated that downregulation of the α5 subunit decreased IGF-II and IGFBP-2 expression and that their expression was dependent on constituitive integrin α5 activation suggesting a link between increased IGFBP-2 synthesis and differentiation. Lee et al showed that treatment of mesenchymal stem cells during osteoblast induction with parathyroid hormone resulted in increased expression of IGF-I, IGF-II and IGFBP-2 whereas PTH treatment of cord blood derived the stem cells that did not differentiate into osteoblasts did not show these changes.(7) These findings have been extended to human osteoblasts wherein it was demonstrated that IGFBP-2 and IGF-I or II expression are upregulated during in vitro induction of differentiation and that this correlated negatively with proliferation.(20) Palermo et al showed that IGF-II stimulated IGFBP-2 synthesis in tibial osteoblast cultures during differentiation and that IGFBP-2 was the most abundant form of IGFBP that was induced.(2) Thraikill et al reported that MC-3T3 cells increased IGFBP-2 expression between days 10 and 14 of differentiation: concomitant with the onset of osteocalcin expression.(21) Several factors that have been shown to stimulate IGFBP-2 expression by MC-3T3 cells specifically phorbol esters(22) and FGF(23). A more recent study by Yerges et al demonstrated that IGFBP-2 SNPs were associated with lumbar volumetric bone mineral density in humans and that only 7 genes were found to be this tightly associated.(24) Other cell types have also been analyzed to determine the role of IGFBP-2 in differentiation. In hematopoetic stem cells IGFBP-2 supports stem cell expansion but no specific mechanism by which it stimulates differentiation in this cell type has been defined.(25–26) Knockdown of IGFBP-2 in zebrafish embryos resulted in disruption of cardiac development and impaired differentiation of cardiomyocytes.(27) Additionally there were vessel sprouting defects which suggested a role in angiogenesis and endothelial cell differentiation. Our studies have demonstrated that IGFBP-2 expression is required for osteoclast differentiation.(28) Cells derived from IGFBP-2 −/− mice showed minimal osteoclast differentiation which could be rescued with exogenous addition of IGFBP-2. The defect appeared to be an inability to form mature osteoclasts that retain full bone resorbing activity. The role of IGFBP-2 in differentiation has been intensively studied in skeletal myoblasts wherein it has been demonstrated that the addition of differentiation medium to myoblasts in culture results in a major increase in expression of IGFBP-2 and inhibition of IGFBP-2 using neutralizing antibodies inhibits myoblast differentiation.(29) Therefore it appears that IGFBP-2 coordinately regulates the ability of IGF-I and IGF-II stimulate differentiation in several cell types. Numerous studies have shown a positive effect of IGF-I on bone formation in vivo and on osteoblast differentiation in vitro. Addition of IGF-I to culture medium with BMP-2 enhanced osteoblast differentiation and this effect was believed to be mediated through AKT.(30) Yeh et al. demonstrated that BMP-2 and IGF-I induced a synergistic increase in osteoblast differentiation and that this response could be inhibited by a protein kinase D inhibitor.(31) IGF-I also mediates chondrocyte differentiation and mature chondrocytes can differentiate into osteoblasts therefore this indirectly alters osteoblast differentiation.(32) Similarly IGF-II has been shown to enhance osteogenic differentiation and it directly potentiates the effects of BMP 9 on alkaline phosphatase activity.(33) These effects are inhibited by PI-3 kinase inhibitors suggesting the AKT pathway that is the primary mediator. Matrix IGF-I has been shown to maintain bone mass by enhancing osteoblast differentiation. This effect required concomitant injection of IGFBP-3 which improved bone matrix localization.(10) Our current study also showed that blockage of IGF-I signaling significantly impaired osteoblast differentiation even though IGFBP-2 was overexpressed. We conclude that IGF-I and IGFBP-2 function coordinately to stimulate differentiation and both peptides are required for an optimal stimulation. Transgenic mice that overexpress IGF-I in osteoblasts have increased trabecular bone and increased bone formation.(34) Conditional IGF-I receptor null mice showed decreased osteoblast number and reduced trabecular volume and impaired differentiation and calcification(35) and locally produced skeletal IGF-I plays an important role in trabecular bone integrity.(36) Since IGFBP-2 stimulates trabecular bone formation and osteoblast differentiation, our results suggest that the two proteins are functioning coordinately. This conclusion is supported by the observation that PTH is a potent stimulant of not only IGF-I synthesis in bone but it also induces IGFBP-2. In conclusion, the results of our studies suggest that IGFBP-2 functions to enhance the ability of IGF-I to stimulate osteoblast differentiation and that this effect is specific for IGFBP-2. Since IGF-I can stimulate both osteoblast proliferation and differentiation, our findings suggests that IGFBP-2 may function directly to coordinate these responses and independently of its transport capacity for the IGFs. Supplementary Material Supp Info The authors wish to thank Ms. Laura Lindsey, University of North Carolina at Chapel Hill, for her help in preparing the manuscript. This work was supported by a grant (AR-06114) from the National Institute of Health. Figure 1 IGFBP-2 stimulates osteoblast differentiation (A) Equal amounts of cell lysate from MC-3T3 cells expressing a shRNA sequence targeting LacZ (Ctrl Si) or IGFBP-2 (IGFBP-2 Si) were immunoblotted with indicated antibody. β-actin was immunoblotted as a loading control. (B) Cell lysate from Ctrl Si or IGFBP-2 Si cells on the indicated day after differentiation medium (DM) exposure were immunoblotted with the indicated antibody. (C) Ctrl Si or IGFBP-2 Si expressing cells were stained by Alizarin Red following the procedure described in “Materials and Methods” on indicated day after DM exposure. (D) Cell lysates from IGFBP-2−/− or IGFBP-2 +/+ derived calvarial osteoblasts prepared on indicated day after DM exposure were immunoblotted with indicated antibody. The bar graph shows the ratio of scanning densitometry units of osteocalcin/β-actin obtained from three individual experiments. (E) Calvarial osteoblasts isolated from IGFBP-2 +/+ or IGFBP-2 −/− mice were exposed to DM alone or DM plus IGFBP-2 (1 ug/ml) and stained with Alizarin Red on day 21. (F) Lysates from cells expressing LacZ or IGFBP-2 were immunoblotted with indicated antibody. (G) Lysates from cells expressing LacZ or IGFBP-2 on indicated day after DM exposure were immunoblotted with indicated antibody. β-actin was immunoblotted as a loading control. The bar graphs show the ratio of scanning densitometry units of osteocalcin/β-actin obtained from three individual experiments. (H) Cells expressing LacZ or IGFBP-2 were stained with Alizarin Red on indicated day after DM exposure. Figure 2 AKT activation is required for osteoblast differentiation (A) Lysates obtained from MC-3T3 cells, clone 24 (CL24) or clone 4 (CL4) on day 6 after DM exposure were immunoblotted with indicated antibody. Lysates from quiescent cells expressing LacZ or IGFBP-2 on day 6 after DM exposure were immunoblotted with the indicated antibody. Wild type MC-3T3 cells (B) or IGFBP-2 overexpressing cells (C) were stained by Alizarin Red on day 21 after differentiation medium (DM) alone or DM plus LY294002 which was added on the indicated day. The medium was changed every 72hr. (D) Lysates obtained from cells overexpressing IGFBP-2 or LacZ after indicated day of DM exposure were immunblotted with indicated antibody. The bar graph shows the ratio of scanning densitometry units of pAKT/β-actin obtained from three individual experiments. (E) Lysates from IGFBP-2−/− or IGFBP-2 +/+ derived calvarial osteoblasts obtained on indicated day after DM exposure were immunoblotted with indicated antibody. The bar graph shows the ratio of scanning densitometry units of pAKT/β-actin obtained from three individual experiments. Figure 3 IGFBP-2 enhances AKT activation via suppressing RPTPβ dephosphorylation of PTEN (A) Cell lysates from quiescent LacZ or IGFBP-2 overexpressing cells were immunoprecipitated with an anti-IGFBP-2 antibody and immunoblotted with an anti-RPTPβ antibody. β-actin was immunoblotted as a loading control. (B) Lysates from LacZ or IGFBP-2 overexpressing cells on indicated day after differentiation medium (DM) exposure were immunoprecipated with an anti-PY99 antibody and immunoblotted with an anti-PTEN antibody. PTEN was immunoblotted as an input control. (C, D) Lysates from MT-3C3 cells expressing shRNA sequence targeting LacZ (Ctrl Si) or RPTPβ (RPTPβ Si) were immunoblotted with the indicated antibodies. (E) Cells expressing Ctrl Si or RPTPβ Si were stained by Alizarin Red on indicated day after DM exposure. (F) Lysates from LacZ or RPTPβ overexpressing cells were immunoblotted with anti-HA and β-actin antibodies. The same cell lysates were immunoprecipitated with an anti-PY99 antibody and immuoblotted with an anti-PTEN antibody. PTEN was immunoblotted as an input control. (G) Cells expressing LacZ or RPTPβ were stained by Alizarin Red on indicated day after DM exposure. (H) Cells expressing RPTPβ were stained by Alizarin Red on day 21 after DM alone or DM plus IGFBP-2 exposure. Lysates from the same RPTPβ overexpressing cultures were immunoblotted with anti-osteocalcin and β-actin antibodies. (I) Lysates from quiescent RPTPβ overexpressing cells obtained on day 6 after IGF-I alone or IGFBP-2 alone (1ug/ml) or IGF-I plus IGFBP-2 (1 ug/ml) were immunoblotted with anti-pAKT and β-actin antibodies. Figure 4 Disruption of IGFBP-2/RPTPβ interaction impairs IGF-I-stimulated AKT activation and osteoblast differentiation (A) Lysates from quiescent LacZ or IGFBP-2 overexpressing cells obtained on day 6 after differentiation medium (DM) exposure treated with or without IGF-I alone (10 min) or IGF-I following a 4 hr exposure to anti-fibronectin domain (FN3) antibody were immunoprecipitated using an anti-IGFBP-2 antibody and immunoblotted with an anti-RPTPβ antibody (B) Lysates from quiescent IGFBP-2 overexpressing cells that received the same treatments as in panel A were obtained on Day 6 after DM exposure and were immunoblotted with indicated antibody. (C) Lysates from LacZ overexpressing cells obtained on day 21 after DM exposure following incubation with the indicated concentration of anti-fibronectin antibody (FN3) were immunoblotted with anti-osteocalcin and anti-β-actin antibodies. (D) Lysates from IGFBP-2 overexpressing cells obtained on day 9 and 12 after DM exposure following incubation with PQ401 (10 uM) were immunoblotted with anti-osteocalcin and anti-β-actin antibodies. (E) Cells expressing IGFBP-2 that had been incubated with or without PQ401 were stained with Alizarin Red on day 18 after DM exposure. Figure 5 The heparin binding domain-1 (HBD-1) mediates the IGFBP-2 effect on osteoblast differentiation (A) Calvarial osteoblasts isolated from IGFBP-2 +/+ or IGFBP-2 −/− mice were exposed to differentiation medium (DM) alone, DM plus IGFBP-2 (1 ug/ml), DM plus control peptide (Ctrl Pep, 1 ug/ml), DM plus HBD-1 (1 ug/ml) or DM plus the HBD-1 IGFBP-2 mutant protein (IGFBP-2 MP, 1 ug/ml) then stained with Alizarin Red on day 21. The bar graph shows the percentage of stained area that was quantified using NIH Image J (1.47n). (B, C) Calvarial osteoblasts isolated from IGFBP-2 −/− mice were exposed to DM alone (Ctrl) or DM plus the indicated concentration of HBD-1 peptide for 21 days and stained with Alizarin Red (B). Cell lysates were immunoblotted with anti-osteocalcin and β-actin antibodies (C). Disclosures The authors state that they have no conflicts of interest. Author’s roles: Study design: GX, CR and DC; Data collection: GX, CW and VD; Data analysis and interpretation: GX, CR and DC. Drafting and reviewing manuscript: GX, CR and DC. References 1 Firth SM Baxter RC Cellular actions of the insulin-like growth factor binding proteins Endocr Rev 2002 23 6 824 854 12466191 2 Palermo C Manduca P Gazzerro E Foppiani L Segat D Barreca A Potentiating role of IGFBP-2 on IGF-II-stimulated alkaline phosphatase activity in differentiating osteoblasts Am J Physiol Endocrinol Metab 2004 286 4 E648 E657 14665441 3 DeMambro VE Clemmons DR Horton LG Bouxsein ML Wood TL Beamer WG Canalis E Rosen CJ Gender-specific changes in bone turnover and skeletal architecture in igfbp-2-null mice Endocrinology 2008 149 5 2051 2061 18276763 4 Kawai M Breggia AC DeMambro VE Shen X Canalis E Bouxsein ML Beamer WG Clemmons DR Rosen CJ The heparin-binding domain of IGFBP-2 has insulin-like growth factor binding-independent biologic activity in the growing skeleton J Biol Chem 2011 286 16 14670 14680 21372140 5 Govoni KE Lee SK Chung YS Behringer RR Wergedal JE Baylink DJ Mohan S Disruption of insulin-like growth factor-I expression in type IIalphaI collagen-expressing cells reduces bone length and width in mice Physiol Genomics 2007 30 3 354 362 17519362 6 Kawase T Okuda K Kogami H Nakayama H Nagata M Nakata K Yoshie H Characterization of human cultured periosteal sheets expressing bone-forming potential: in vitro and in vivo animal studies J Tissue Eng Regen Med 2009 3 3 218 229 19266490 7 Lee JH Hwang KJ Kim MY Lim YJ Seol IJ Jin HJ Jang YK Choi SJ Oh W Cho YH Lee YH Human parathyroid hormone increases the mRNA expression of the IGF system and hematopoietic growth factors in osteoblasts, but does not influence expression in mesenchymal stem cells J Pediatr Hematol Oncol 2012 34 7 491 496 23007338 8 Hamidouche Z Fromigue O Ringe J Haupl T Marie PJ Crosstalks between integrin alpha 5 and IGF2/IGFBP2 signalling trigger human bone marrow-derived mesenchymal stromal osteogenic differentiation BMC Cell Biol 2010 11 44 20573191 9 Shen X Xi G Maile LA Wai C Rosen CJ Clemmons DR Insulin-like growth factor (IGF) binding protein 2 functions coordinately with receptor protein tyrosine phosphatase beta and the IGF-I receptor to regulate IGF-I-stimulated signaling Mol Cell Biol 2012 32 20 4116 4130 22869525 10 Xian L Wu X Pang L Lou M Rosen CJ Qiu T Crane J Frassica F Zhang L Rodriguez JP Xiaofeng J Shoshana Y Shouhong X Argiris E Mei W Xu C Matrix IGF-1 maintains bone mass by activation of mTOR in mesenchymal stem cells Nat Med 2012 18 7 1095 1101 22729283 11 Cohick WS Clemmons DR Regulation of insulin-like growth factor binding protein synthesis and secretion in a bovine epithelial cell line Endocrinology 1991 129 3 1347 1354 1714831 12 Xi G Shen X Clemmons DR p66shc negatively regulates insulin-like growth factor I signal transduction via inhibition of p52shc binding to Src homology 2 domain-containing protein tyrosine phosphatase substrate-1 leading to impaired growth factor receptor-bound protein-2 membrane recruitment Mol Endocrinol 2008 22 9 2162 2175 18606861 13 Wang D Christensen K Chawla K Xiao G Krebsbach PH Franceschi RT Isolation and characterization of MC3T3-E1 preosteoblast subclones with distinct in vitro and in vivo differentiation/mineralization potential J Bone Miner Res 1999 14 6 893 903 10352097 14 Raucci A Bellosta P Grassi R Basilico C Mansukhani A Osteoblast proliferation or differentiation is regulated by relative strengths of opposing signaling pathways J Cell Physiol 2008 215 2 442 451 17960591 15 Schinke T Gebauer M Schilling AF Lamprianou S Priemel M Mueldner C Neunaber C Streichert T Ignatius A Harroch S Amling M The protein tyrosine phosphatase Rptpzeta is expressed in differentiated osteoblasts and affects bone formation in mice Bone 2008 42 3 524 534 18178537 16 Conover CA Johnstone EW Turner RT Evans GL John Ballard FJ Doran PM Khosla S Subcutaneous administration of insulin-like growth factor (IGF)-II/IGF binding protein-2 complex stimulates bone formation and prevents loss of bone mineral density in a rat model of disuse osteoporosis Growth Horm IGF Res 2002 12 3 178 183 12162999 17 Kiepe D Van Der Pas A Ciarmatori S Standker L Schutt B Hoeflich A Hugel U Oh J Tonshoff B Defined carboxy-terminal fragments of insulin-like growth factor (IGF) binding protein-2 exert similar mitogenic activity on cultured rat growth plate chondrocytes as IGF-I Endocrinology 2008 149 10 4901 4911 18556354 18 Higuchi C Myoui A Hashimoto N Kuriyama K Yoshioka K Yoshikawa H Itoh K Continuous inhibition of MAPK signaling promotes the early osteoblastic differentiation and mineralization of the extracellular matrix J Bone Miner Res 2002 17 10 1785 1794 12369782 19 Nakayama K Tamura Y Suzawa M Harada S Fukumoto S Kato M Miyazono K Rodan GA Takeuchi Y Fujita T Receptor tyrosine kinases inhibit bone morphogenetic protein-Smad responsive promoter activity and differentiation of murine MC3T3-E1 osteoblast-like cells J Bone Miner Res 2003 18 5 827 835 12733721 20 Viereck V Siggelkow H Pannem R Braulke T Scharf JG Kubler B Alteration of the insulin-like growth factor axis during in vitro differentiation of the human osteosarcoma cell line HOS 58 J Cell Biochem 2007 102 1 28 40 17372931 21 Thrailkill KM Siddhanti SR Fowlkes JL Quarles LD Differentiation of MC3T3-E1 osteoblasts is associated with temporal changes in the expression of IGF-I and IGFBPs Bone 1995 17 3 307 313 8541146 22 Hakeda Y Yoshizawa K Hurley M Kawaguchi H Tezuka K Tanaka K Satoh T Kumegawa M Stimulatory effect of a phorbol ester on expression of insulin-like growth factor (IGF) binding protein-2 and level of IGF-I receptors in mouse osteoblastic MC3T3-E1 cells J Cell Physiol 1994 158 3 444 450 7510294 23 Hurley MM Abreu C Hakeda Y Basic fibroblast growth factor regulates IGF-I binding proteins in the clonal osteoblastic cell line MC3T3-E1 J Bone Miner Res 1995 10 2 222 230 7538725 24 Yerges LM Klei L Cauley JA Roeder K Kammerer CM Ensrud KE Nestlerode CS Lewis C Lang TF Barrett-Connor E Moffett SP Hoffman AR Ferrell RE Orwoll ES Zmuda JM Candidate gene analysis of femoral neck trabecular and cortical volumetric bone mineral density in older men J Bone Miner Res 2010 25 2 330 338 19619005 25 Huynh H Zheng J Umikawa M Zhang C Silvany R Iizuka S Holzenberger M Zhang W Zhang CC IGF binding protein 2 supports the survival and cycling of hematopoietic stem cells Blood 2011 118 12 3236 3243 21821709 26 Celebi B Mantovani D Pineault N Insulin-like growth factor binding protein-2 and neurotrophin 3 synergize together to promote the expansion of hematopoietic cells ex vivo Cytokine 2012 58 3 327 331 22459634 27 Wood AW Schlueter PJ Duan C Targeted knockdown of insulin-like growth factor binding protein-2 disrupts cardiovascular development in zebrafish embryos Mol Endocrinol 2005 19 4 1024 1034 15618288 28 DeMambro VE Maile L Wai C Kawai M Cascella T Rosen CJ Clemmons D Insulin-like growth factor-binding protein-2 is required for osteoclast differentiation J Bone Miner Res 2012 27 2 390 400 22006816 29 Sharples AP Al-Shanti N Hughes DC Lewis MP Stewart CE The role of insulin-like-growth factor binding protein 2 (IGFBP2) and phosphatase and tensin homologue (PTEN) in the regulation of myoblast differentiation and hypertrophy Growth Horm IGF Res 2013 23 3 53 61 23583027 30 Mukherjee A Rotwein P Akt promotes BMP2-mediated osteoblast differentiation and bone development J Cell Sci 2009 122 Pt 5 716 726 19208758 31 Yeh LC Ma X Matheny RW Adamo ML Lee JC Protein kinase D mediates the synergistic effects of BMP-7 and IGF-I on osteoblastic cell differentiation Growth Factors 2010 28 5 318 328 20380591 32 Longobardi L Granero-Molto F O'Rear L Myers TJ Li T Kregor PJ Spagnoli A Subcellular localization of IRS-1 in IGF-I-mediated chondrogenic proliferation, differentiation and hypertrophy of bone marrow mesenchymal stem cells Growth Factors 2009 27 5 309 320 19639489 33 Chen L Jiang W Huang J He BC Zuo GW Zhang W Luo Q Shi Q Zhang BQ Wagner ER Luo J Tang M Wietholt C Luo X Bi Y Su Y Liu B Kim SH He CJ Hu Y Shen J Rastegar F Huang E Gao Y Gao JL Zhou JZ Reid RR Luu HH Haydon RC He TC Deng ZL Insulin-like growth factor 2 (IGF-2) potentiates BMP-9-induced osteogenic differentiation and bone formation J Bone Miner Res 2010 25 11 2447 2459 20499340 34 Zhao G Monier-Faugere MC Langub MC Geng Z Nakayama T Pike JW Chernausek SD Rosen CJ Donahue LR Malluche HH Fagin JA Clemens TL Targeted overexpression of insulin-like growth factor I to osteoblasts of transgenic mice: increased trabecular bone volume without increased osteoblast proliferation Endocrinology 2000 141 7 2674 2682 10875273 35 Zhang M Xuan S Bouxsein ML von Stechow D Akeno N Faugere MC Malluche H Zhao G Rosen CJ Efstratiadis A Clemens TL Osteoblast-specific knockout of the insulin-like growth factor (IGF) receptor gene reveals an essential role of IGF signaling in bone matrix mineralization J Biol Chem 2002 277 46 44005 44012 12215457 36 Kesavan C Wergedal JE Lau KH Mohan S Conditional disruption of IGF-I gene in type 1alpha collagen-expressing cells shows an essential role of IGF-I in skeletal anabolic response to loading Am J Physiol Endocrinol Metab 2011 301 6 E1191 E1197 21878662

PMC005xxxxxx/PMC5117427.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9421564 8584 Shock Shock Shock (Augusta, Ga.) 1073-2322 1540-0514 27755473 5117427 10.1097/SHK.0000000000000735 NIHMS810373 Article WHAT’S NEW IN SHOCK, NOVEMBER 2016? Efron Philip A. MD Departments of Surgery, Anesthesiology, Aging and Geriatric Research, and Molecular Genetics and Microbiology 18 8 2016 11 2016 01 11 2017 46 5 465467 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The journal Shock is unique in that it reports on all aspects of inflammation, injury and sepsis – in a sense, “Shock is where the action is!” This month’s journal only serves to emphasize how Shock remains at the forefront of international clinical and scientific research. Be it a review article or basic science manuscript, the November 2016 issue of Shock is a must read for all individuals interested in the field. This month’s journal begins with a very thorough and complete review of plasma transfusion from our colleagues on the west coast of the United States (1). Understanding the benefits versus the detriments of transfusion has become a key aspect of daily bedside medicine. In fact, transfusion is now tracked as a ‘best practice measure’ by many national organizations. Watson, et al’s essential review of the medical use of plasma, from its origins to its ongoing research (1) is both opportune and beneficial. In addition to summarizing all of the key literature and studies regarding the use of plasma for shock, it discusses the various formulations of plasma that are currently being used or in development (1). Certainly this article will be in the library of every critical care resident for the next decade. As Shock is the official journal of a consortium of international societies, it is of no surprise that the next article illustrates some exceptional work from South America. Dr. Filho, et al have produced some very timely work regarding blood lactate levels and sepsis (2). The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) has defined septic shock as the requirement for a vasopressor (in the absence of hypovolemia) to maintain a minimum mean arterial pressure of 65mmHg and a serum lactate level greater than 2mmol/L (>18mg/dL) (3). With this in mind, abnormal lactate levels above a level of 2mmol/L are now more relevant than ever, as presented by Filho and colleagues, who were able to use an Emergency Room admission lactate cutoff of 2.5mmol/L to predict 28 day mortality, and this had a receiver operator curve of 0.70 (2). Although Filho et al.’s work is a retrospective analysis, this real world study presents a simple, clinically relevant test that can be utilized for predicting mortality in septic patients with organ dysfunction or pressor requirements (2). Not unexpectedly, Dr. Herndon’s group continues to lead the way in demonstrating the effectiveness of interventional therapies that are United States Food and Drug Administration approved as well as inexpensive. It goes against the better judgement of any clinician to give a medication that potentially lowers blood pressure in the patient in the acute phase of an inflammatory response. However, Dr. Wurzer, et al have demonstrated that giving propanolol in burned children is not only safe but potentially helpful (4). The medication reduced cardiogenic stress without adversely affecting other multiple important parameters (4). This includes, but is not limited to: peripheral oxygen delivery, events of lactic acidosis, wound healing, and mortality (4). Clearly, Dr. Wurzer and colleagues are part of a select few investigators with the remarkable ability to see beyond many clinicians’ trepidation to utilize agents such as propanolol in the initial phase of inflammation and shock to help improve outcomes (5, 6). Keeping with the theme of top notch work accomplished in both the burn population as well as around the world, Dr. Lopez-Rodriguez, et al have continued their analysis of their single-center, prospective, randomized, double-blind clinical trial on the effect of selective decontamination of the digestive tract (SDD) in severe burn patients (7). ‘No organ is an island,’ and this is certainly true of the intestines in the acute and sub-acute phases of shock, trauma and burn. After previously demonstrating that SDD can improve mortality after burns, the authors have illustrated that this simple intervention was independently associated with a reduction in organ dysfunction in severely burned patients, which also suggests a modulatory role of SDD on the inflammatory response (7). And what was able to induce this huge impact? The answer is the early and temporary use of non-proprietary antibiotics, such as polymyxin, amphotericin B and cefotaxime (7). Again, an inexpensive and safe intervention is demonstrated to have an impact, this time influencing the host as well as the microbiome of these very sick patients (8). Not to be superseded, French investigators have also analyzed an inexpensive drug that has been debated for decades in sepsis. I am, of course, referring to steroids! Dr. Herve, et al conducted a multicenter, prospective, randomized, double-blind, pilot study comparing two low dose regimens of hydrocortisone: 200mg versus 300mg (9). Their results have illustrated, like many of the recent sepsis prospective trials in sepsis, that it’s not so much the specifics of the intervention, such as the total dose, but the early recognition of the need for a therapy along with its timely institution that is most important to the patient’s outcome (9). Thus, the authors found no difference in 28 day mortality between the groups – however, the analysis may lead to important further weight based steroid experimentation, as well reviews of sepsis persistence and etomidate use in these patients (9). Leelahavanichkul, et al have made further progress in the field of practical predictive medicine (10). This work from Thailand looked at serum (1→3)-b-D-glucan (BG) in not just septic mice, but septic humans (10). BG is a key structural polysaccharide of the cell wall of most fungi and the authors have revealed that it can be used as a marker of gastrointestinal leakage and sepsis, even during/after bacterial infections (10). As clinicians are moving towards precision medicine (11), this type of biomarker will prove vital as scientists tailor therapy only to those individuals that require intervention. It’s impossible to investigate inflammation without considering toll-like receptors (TLR), and when you combine Dr. Billiar’s laboratory and the journal Shock, outstanding work is always revealed. That tradition continues this month as Korff, Scott, Billiar, et al use clever investigative basic science, including knock out and chimera mice and a clinically relevant model of hemorrhagic shock (12), to further our understanding of the role of TLR in inflammation. Their work illustrated that TLR2 regulates both bone marrow and non-bone marrow derived cells’ ability to affect inflammation, and more importantly, organ failure (12). Interestingly, the mechanism of this was in conjunction with TLR4 (12). It is only through clinically relevant work like this that progress will be made to enable future physicians to immunomodulate the human patient after trauma and hemorrhagic shock. As a global journal, revealing and analyzing cutting edge technology has always been a vital component of the journal Shock. Thus, in this November’s issue the editors are very proud to publish the work of Dr. Lane Smith and the collaborative team at Wake Forrest. Photoacoustic (PA) imaging, is a technology that evaluates both tissue structure and function through ultrasound and laser energy (13). The investigators have been able to illustrate that PA is able to measure, in real-time, oxygen saturation in the macro and microcirculation during acute hypoxia (13). Anyone involved in research regarding shock will need to be familiar with this technology, as its future use, in combination with other agents, will likely become common place in the critically ill (13). Clinical relevance does not mean abandoning basic science at its most fundamental level. Kidney injury is a key driver of poor patient outcomes - much more than previously realized. Dr. Wang and colleagues from China have performed some very elegant basic research regarding myofibrillogenesis regulator 1 (MR-1) in a clinically relevant murine model of renal ischemia/reperfusion injury (I/R) (14). MR-1 is a mitochondrial-targeted protein, and the researchers demonstrated that through the recruitment of phosphatidylinositol 3 kinase-dependent phosporylated-Akt to the mitochondria, MR-1 was able to protect the kidney from I/R injury by inhibiting mitochondrial permeability transition pore opening and maintaining mitochondrial integrity (14). The various methodologies conducted by the investigators in this study are impressive, and this work has the capacity to introduce new therapeutics to the realm of shock research (14). Returning to the investigation of burn injury, Dr. Caldwell’s laboratory has revealed another important aspect to the treatment of depression (15). Although depression is not considered normally in the area of shock research, depression and medication for its treatment are fairly prevalent in the United States. Dr. Johnson, et al were able to determine that amitriptyline, a tricyclic antidepressant that inhibits acid sphingomyelinase, had significant effects on the immunity of host after burn injury (15). This included a reduction in lymphocyte precursors, lymphocyte numbers and neutrophil recruitment (15). Clearly, their work suggests that future precision medicine for inflammation, injury and infection will require consideration of the patient’s ‘medication profile.’ Which leads to the concept that precision medicine will also need to take into consideration a patient’s diet with the treatment of, let’s say, of cardiovascular disease. Although debated to some extent, patients whose lifestyles incorporate certain seeds or fish are much less likely to succumb to certain pathology, such as acute myocardial infarctions. However, the medical practitioner is not able to pre-emptively create such environment when their patient arrives at the emergency room. Interestingly, Burban et al have determined that one single intravenous omega-3 bolus before reperfusion in a clinically relevant rat model of I/R-induced shock was able to improve multiple parameters (16). This included an increased mean arterial pressure and carotid blood flow as well as decreased cardiac troponin levels (16). This improvement in blood pressure and vasoreactivity could have significant repercussions regarding the treatment of acute coronary syndromes, and again the work in this issue reveals possible therapies that are both inexpressive and safe (16)! The need to improve cardiac function and outcomes is not limited to myocardial infractions, though. Dr. Li, et al have investigated the well-known cardio-dysfunction induced by severe sepsis and septic shock. Translating work done in cardiac I/R, the authors were able to conduct research on intermedin, a calcitonin related peptide, using the cecal ligation and puncture murine sepsis model (17). Early and late administration of the compound, specifically intermedin 1–53, to septic mice improved heart function, tissue oxygenation/perfusion and survival (17). This was in part through the Rho kinase phosphorylation pathway and by increasing intracellular calcium (17). As a journal of world-wide importance, it is of no surprise that notable work is being published in this issue Shock regarding melioidosis. This is an infection with Burkholderia pseudomallei that is relatively common in Southeast Asia and Northern Australia with significant morbidity and mortality. Dr. Weehuiz has produced some impressive data with a collaborative international laboratory group regarding the use of a monoclonal anti-IL-1b antibody in this condition (18). Although treatment with such compounds in the past failed to improved outcomes, application of anti-IL1b antibodies could be effective in more select patient population, such as those suffering from melioidosis (18). This research was able to demonstrate that their intervention were able to affect the inflammasome and improve the host’s response to the bacterium B. pseudomallei(18). Taiwanese investigators have made some ‘in-roads’ regarding our understanding of the acute respiratory distress syndrome, as displayed by the work of Day, et al. Using a rat model of the Acute Respiratory Distress Syndrome (ARDS) combined with intraperitoneal lipopolysaccharide injection, the authors were able test the hypothesis that preactivated and disaggregated shape-changed platelets could attenuate lung injury (19). Their intervention was able to improve the outcomes of the host on multiple levels, including, but not limited to, histological, cellular and molecular results (19). This work has improved not only our understanding of the relationship of pathology of ARDS, but the anti-inflammatory and anti-oxidative properties of preactivated and disaggregated shape-changed platelets, which may be considered as a future therapeutic in a disease process that has few current interventions besides supportive care (19). Finally, in an article that combines some of the themes investigated by other scientists in this issue (sepsis, ARDS and a single injection of nutritional compound), Dr. Yeh and colleagues have eloquently demonstrated that glutamine can modify progenitor cell and lung injury in mice exposed to severe infection (20). Their basic science work revealed that an intravenous injection of glutamine after cecal ligation and puncture could promote the mobilization of endothelial progenitor cells, in part due to the release of C-X-C motif chemokine 1, vascular endothelial growth factor and nitric oxide (20). These effects are associated with improved vascular function, ameliorated inflammation and reduced damage of lung tissues (20). Although the use of glutamine in the critically ill is currently debated, the Shock journal has never shied away from contentious issues that require further investigation in order to elucidate appropriate answers for the scientific and medical community! Based on this issue and the ten issues before it in 2016, there can be no doubt that the journal Shock has led the way this year in reporting the best research, both laboratory and clinical, regarding inflammation, injury and sepsis. For several decades now, this journal has truly served its readers around the world. 1 Watson JJJ Pati S Schreiber MA Plasma transfusion: History, current realities, and novel improvements Shock 46 ____-____ 2016 2 Filho RR Rocha LL Corrêa TD Pessoa CMS Colombo G Assuncão MSC Blood lactate levels cutoff and mortality prediction in sepsis-time for a reappraisal? A retrospective cohort study Shock 46 ____-____ 2016 3 Singer M Deutschman CS Seymour CW Shankar-Hari M Annane D Bauer M Bellomo R Bernard GR Chiche JD Coopersmith CM Hotchkiss RS Levy MM Marshall JC Martin GS Opal SM Rubenfeld GD van der Poll T Vincent JL Angus DC The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) JAMA 315 80 10 2016 4 Wurzer P Branski LK Clayton RP Hundeshagen G Forbes AA Voigt CD Andersen CR Kamolz L-P Woodson LC Suman OE Finnerty CC Herndon DN Propranolol reduces cardiac index but does not adversely affect peripheral perfusion in severely burned children Shock 46 ____-____ 2016 5 Morelli A Ertmer C Westphal M Rehberg S Kampmeier T Ligges S Orecchioni A D’Egidio A D’Ippoliti F Raffone C Venditti M Guarracino F Girardis M Tritapepe L Pietropaoli P Mebazaa A Singer M Effect of heart rate control with esmolol on hemodynamic and clinical outcomes in patients with septic shock: a randomized clinical trial JAMA 310 1683 91 2013 24108526 6 Novotny NM1 Lahm T Markel TA Crisostomo PR Wang M Wang Y Ray R Tan J Al-Azzawi D Meldrum DR beta-Blockers in sepsis: reexamining the evidence Shock 31 113 9 2009 18636043 7 López-Rodríguez L de la Cal MA García-Hierro P Herrero R Martins J van Saene HKF Lorente JA Selective digestive decontamination attenuates organ dysfunction in critically ill burn patients Shock 46 ____-____ 2016 8 Krezalek MA1 DeFazio J Zaborina O Zaborin A Alverdy JC The Shift of an Intestinal “Microbiome” to a “Pathobiome” Governs the Course and Outcome of Sepsis Following Surgical Injury Shock 45 475 82 2016 26863118 9 Hervé H Rémy B Gentilhomme A Francois C-GJ Freche A Kaidomar M Bernard G Pradier C Dellamonica J Bernardin G Effects of increasing hydrocortisone to 300 mg per day in the treatment of septic shock: A pilot study Shock 46 ____-____ 2016 10 Leelahavanichkul A Worasilchai N Wannalerdsakun S Jutivorakool K Somparn P Issara-Amphorn J Tachaboon S Srisawat N Finkelman M Chindamporn A Gastrointestinal leakage detected by serum (1→3)-β-D-glucan in mouse models and a pilot study in patients with sepsis Shock 46 ____-____ 2016 11 Mathias B Lipori G Moldawer LL Efron PA Integrating “big data” into surgical practice Surgery 159 371 4 2016 26603852 12 Korff S Loughran P Cai C Fan J Elson G Shang L Pires SS Lee YS Guardado J Scott M Billiar TR TLR2 on bone marrow and non-bone marrow derived cells regulates inflammation and organ injury in cooperation with TLR4 during resuscitated hemorrhagic shock Shock 46 ____-____ 2016 13 Smitih L Varagic J Yamaleyeva L Photoacoustic imaging for the detection of hypoxia in the rat femoral artery and skeletal muscle microcirculation Shock 46 ____-____ 2016 14 Wang X-R Ding R Tao T-Q Mao H-M Liu M Xie Y-S Liu X-H Myofibrillogenesis regulator 1 rescues renal ischemia/reperfusion injury by recruitment of P13K –dependent P-AKT to mitochondria Shock 46 ____-____ 2016 15 Johnson BL III Rice TC Xia BT Boone KI Green EA Gulbins E Caldwell CC Amitriptyline usage exacerbates the immune suppression following burn injury Shock 46 ____-____ 2016 16 Burban M Meyer G Olland A Séverac F Yver B Toti F Schini-Kerth V Meziani F Boisramé-Helms J An intravenous bolus of EPA:DHA 6:1 protects against myocardial ischemia-reperfusion-induced shock Shock 46 ____-____ 2016 17 Li T Yu Z Wu H Wu Y Zhang J Peng X Zang J Xiang X Liu L Beneficial effect of intermedin 1–53 in septic shock rats: Contributions of Rho kinase and BKCA pathway-mediated improvement in cardiac function Shock 46 ____-____ 2016 18 Weehuizen TAF Lankelma JM De Jong HK De Boer OJ Roelofs JJTH Day NPJ Gram H De Vos AF Wiersinga WJ Therapeutic administration of a monoclonal anti-IL-1β antibody protects against experimental melioidosis Shock 46 ____-____ 2016 19 Day Y-K Chen K-H Chen Y-L Huang T-H Sung P-H Lee F-Y Chen C-H Chai H-T Yin T-C Chiang H-J Chung S-Y Chang H-W Yip H-K Preactivated and disaggregated shape-changed platelets protected against acute respiratory distress syndrome complicated by sepsis through inflammation suppression Shock 46 ____-____ 2016 20 Pai M-H Shih Y-M Shih J-M Yeh C-L Glutamine administration modulates endothelial progenitor cell and lung injury in septic mice Shock 46 ____-____ 2016

PMC005xxxxxx/PMC5117435.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9214969 2488 Methods Mol Biol Methods Mol. Biol. Methods in molecular biology (Clifton, N.J.) 1064-3745 1940-6029 27714613 5117435 10.1007/978-1-4939-6451-2_8 NIHMS810158 Article Protein chemical modification inside living cells using split inteins Borra Radhika 1 Camarero Julio A. 12* 1 Department of Pharmacology and Pharmaceutical Sciences, University of Southern California, Los Angeles, CA 90089-9121, USA 2 Department of Chemistry, University of Southern California, Los Angeles, CA 90089-9121, USA Phone: 323-442-1417, Fax: 323-224-7473, jcamarer@usc.edu 13 8 2016 2017 01 1 2018 1495 111130 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Methods to visualize, track, measure, and perturb or activate proteins in living cells are central to biomedical efforts to characterize and understand the spatial and temporal underpinnings of life inside cells. Although fluorescent proteins have proven to be extremely useful for in vivo studies of protein function, their utility is inherently limited because their spectral and structural characteristics are interdependent. These limitations have spurred the creation of alternative approaches for the chemical labeling of proteins. We describe in this protocol the use of fluorescence resonance emission transfer (FRET)-quenched DnaE split-inteins for the site-specific labeling and concomitant fluorescence activation of proteins in living cells. We have successfully employed this approach for the site-specific in-cell labeling of the DNA binding domain (DBD) of the transcription factor YY1 using several human cell lines. Moreover, we have shown that this approach can be also used for modifying proteins in order to control their cellular localization and potentially alter their biological activity. split-intein protein trans-splicing Npu intein protein labeling fluroescence 1. Introduction Understanding the roles of specific proteins in cellular processes is a fundamental goal of molecular biology [1–3]. Methods to label and visualize proteins inside living cells are extremely useful in the study of localization, movement, interactions, and microenvironments of proteins in living cells. Although fluorescent proteins have revolutionized such studies, they have numerous shortcomings, which have spurred the creation of alternative approaches to chemically label proteins in living cells. These next generation approaches combine the genetic targeting capabilities of fluorescent proteins with the diversity and environmental sensitivity of fluorescent small molecules and/or other biophysical probes. Most of the available techniques, however, provide only limited temporal resolution for labeling of biomolecules in living cells. Ideally these approaches should be modular, thus making possible the introduction of a wide variety of fluorophores or other type of biophysical probes. The kinetics of the labeling reaction should be fast enough to provide temporal resolution to satisfy the most time-sensitive biological assays. It should also allow spatial control during the in-cell labeling process. The labeling reaction should introduce minimal modifications on the target protein in order to preserve its original structure and biological function. Finally, it should make possible the simultaneous introduction of different probes onto multiple target proteins for simultaneous tracking purposes. One of the most promising approaches for in-cell protein labeling involves the use of intein-mediated protein trans-splicing (Fig. 1) [4]. Protein trans-splicing is a naturally occurring post-translational modification similar to protein splicing with the difference being that the intein self-processing domain is split in two fragments, called N-intein (IN) and C-intein (IC), respectively [5, 6]. These two intein fragments are inactive individually, however, they can bind each other with high specificity under appropriate conditions to form a functional protein-splicing domain. Split mini-inteins have been widely used by our group and others for the site-specific modification of proteins in vitro [7–10] and in living cells [11–13]. In-cell labeling of proteins can be easily accomplished by expressing the protein of interest fused to the IN fragment. The second half of the split intein can be chemically synthesized to contain any chemical probe at the C-extein moiety, and then introduced into the cells by using peptide transducing domains (PTD) [11, 13]. Intein-mediated labeling of proteins is highly modular allowing the covalent site-specific incorporation of a myriad of biophysical probes into proteins [8–10]. The kinetics of protein splicing is also relatively fast, with a number of split-inteins having reaction times in the order of several minutes [9, 14, 15]. Moreover, the recent development of conditional protein splicing, both through chemical and photochemical means, makes possible the chemical modification of proteins in living cells with temporal and spatial control [16–19]. One of the best-characterized naturally occurring split-inteins are α-subunit DNA polymerase III (DnaE) intein [20], with many known orthologs with high sequence homology in many cyanobacteria species (Fig. 2A) [14, 21]. The DnaE split-inteins are characterized for having IN and IC fragments with ≈120 and ≈30 residues, respectively. The relatively small size of the IC fragment facilitates its chemical synthesis thus allowing the use of synthetic IC fragments bearing different biophysical probes in the C-extein segment to be used for the chemical modification of proteins through protein trans-splicing [9, 13, 17, 18]. We will use in this protocol the Nostoc puntiforme PCC73102 (Npu) DnaE split-intein. This particular DnaE split-intein has one of the highest rate reported for protein trans-splicing (τ1/2 ≈ 60 s) [15] and a high splicing yield [15, 22]; and therefore is ideal for in-cell protein labeling purposes. The use of protein trans-splicing for the site-specific labeling of proteins with fluorogenic dyes for in-cell tracking purposes requires that the labeling process must be linked to the simultaneous activation of fluorescence (Fig. 1). This can be accomplished by making use of fluorescence resonance emission transfer (FRET)-quenched DnaE split-inteins for the site-specific labeling and concomitant fluorescence activation of proteins in living cells. In this protocol we use fluorescein and dabcyl as fluorescence donor and FRET-quencher, respectively (Fig. 2), but any other combination of donor and quencher could be also used. The fluorescein group is introduced at the C-terminus of the first four residues (Cys-Phe-Asn-Lys) of the C-extein, which are required for efficient trans-splicing (Fig. 2) [13] The dabcyl group (QM, Fig. 2) is introduced on residue 22 of the Npu DnaE IC polypeptide (peptide IC, Table 1). This position is in close proximity to the C-extein (≈17 Å, Fig. 2B) and provides an excellent FRET-quenching (>99% FRET-quenching) [13]. In this chapter we describe the protocol for in-cell C-terminal labeling of the DNA binding domain (DBD) of the transcription factor Yin Yang 1 (YY1) in live U2OS and HeLa cells. YY1 is a ubiquitously distributed multifunctional transcription factor belonging to the GLI-Kruppel class of zinc finger proteins [23]. The protein is involved in repressing and activating a diverse number of promoters including negative regulation of p53, thus making it of particular interest [24, 25]. In the example described in this chapter the DBD of YY1 was labeled at its C-terminal with a fluorophore (fluorescein) and a nuclear localization (NLS) signal peptide to demonstrate the potential of this technique to control the localization of and biological function of a protein/protein domain. The protocol described uses humans U2OS cells but it could be easily adapted to any other mammalian cell line. It is important to note that this approach is highly modular and can be used for in-cell labeling of proteins with other biophysical probes or peptide sequences required for activity. In addition, the use of different orthogonal split inteins should also make possible the simultaneous labeling of multiple proteins with different probes. Before performing the labeling experiment in-cell, it is advisable to test the labeling reaction in vitro first. This protocol also provides instructions on how to evaluate labeling by protein trans-splicing in vitro. 2. Materials All solutions were prepared using ultrapure water with a resistivity of 18 MΩ × cm at 25° C and analytical grade reagents. All reagents and solutions were stored at room temperature unless indicated otherwise. 2.1 Instruments Water bath able to operate at 42° C and 94°C. Table-top micro centrifuge capable of operating at 14,000 rpm. Microbiology incubator set at 37°C. Temperature controlled incubator Shaker. Orbital shaker. Polymerase chain reaction thermocycler. Agarose gel electrophoresis unit. Electrophoresis power pack able to operate up to 250 V. UV-visible spectrophotometer. Sonicator to lyse E. coli cells. High speed centrifuge (e.g. Sorvall RC 5C Plus, Thermo Fisher scientific). SDS-PAGE electrophoresis apparatus. Centrifuge tubes of 0.5 mL, 1.5 mL, 15 mL and 30 mL of capacity. 5 ml Polypropylene Columns. Class II, type A2 biosafety cabinets. Gel and Blot Imaging System. Fluorescence microscope. CO2 incubator. 2.2 Cloning of YY1-IN construct Synthetic DNA primers used to amplify genes encoding Npu DnaE IN and DBD-YY1 (20 nmol scale, HPLC purified) (Table 2). Genomic DNA from Nostoc punctiforme strain ATCC 29133/PCC 73102 (obtained from ATCC). DNA encoding human YY1 proteins (cDNA clone IMAGE: 5261384) (can be obtained from many sources, eg. Genscript, USA). TE buffer: 10 mM Tris-HCl, 1 mM EDTA, pH 8.0. Vent DNA polymerase, TaqDNA polymerase, dNTPs solution, 10X thermopol PCR buffer, 10X Taq DNA polymerase buffer. Restriction enzymes Not I, Nde I, BamH I, Kpn I and Sal I. PCR Purification Kit (e.g. QIAquick from QIAGEN). Miniprep Kit (e.g. QIAprep from QIAGEN). Gel Extraction Kit (e.g. QIAquick from QIAGEN). Chemical competent DH5α cells. Expression plasmid pET28a (Novagen-EMD Millipore). Mammalian Expression vector pcDNA4/TO/myc-His (Invitrogen). T4 DNA ligase and T4 DNA ligase buffer. Ampicillin stock solution: 100 mg ampicillin/mL in pure H2O, sterilized by filtration. Store in 1 mL aliquots at −20° C. Kanamycin stock solution: 25 mg kanamycin/mL in pure H2O, sterilized by filtration. Store in 1 mL aliquots at −20° C. Chloramphenicol stock solution: 34 mg chloramphenicol/mL in EtOH. Store in 1 mL aliquots at −20° C. LB medium: 25 g of LB broth was dissolved in 1 L of pure H2O and sterilized by autoclaving at 120° C for 30 min. LB medium-agar: 3.3 g of LB agar was suspended in 100 mL of pure H2O and sterilized by autoclaving at 120° C for 30 min. To prepare plates, allow LB medium-agar to cool to ≈50° C, then add 0.1 mL of antibiotic stock solution. SOC Medium: 20 g of tryptone, 5 g yeast extract, 0.5 g NaCl and 0.186 g KCl were suspended into 980 ml of pure water and sterilized by autoclaving at 120° C for 30 min. Dissolve 4.8 g MgSO4, 3.603 g dextrose in 20 mL of pure H2O and filter sterilize over a 45 μm filter and add to the autoclaved medium. 2.3 Bacterial expression YY1-IN construct Chemical competent BL21 (DE3) and Origami2 (DE3) cells (EMD Millipore). Isopropyl-thio-β-D-galactopyranoside (IPTG): Prepare a stock solution of 1 M analytical grade IPTG in H2O and sterilize by filtration over 45 μm filter. Store at −20° C. Lysis buffer: 0.1 mM EDTA, 25 mM sodium phosphate, 150 mM NaCl, 10% (v/v) glycerol, pH 7.4. Phosphate buffer saline (PBS): 25 mM sodium phosphate, 150 mM NaCl, pH 7.4 100 mM phenylmethylsulphonyl fluoride (PMSF) in EtOH (better to prepare fresh before use). 4X SDS-PAGE sample buffer: 1.5 mL of 1 M Tris-HCl buffer at pH 6.8, 3 mL of 1 M DTT (dithiothreitol) in pure H2O, 0.6 g of sodium dodecyl sulfate (SDS), 30 mg of bromophenol blue, 2.4 mL of glycerol, bring final volume to 7.5 mL. SDS-PAGE sample buffer: dilute 4 times 4×SDS-PAGE sample buffer in pure H2O and add 20% (v/v) 2-mercaptoethanol. Prepare fresh. SDS-4–20% PAGE gels, 1X SDS running buffer. Gel stain: Coomassie brilliant blue or Gelcode® Blue (Thermo scientific, USA) or silver stain kit. 2.4 In-vitro trans-splicing Pure labeled DnaE IC polypeptide shown in Table 1. The synthetic peptides used in this study were generated in-house but they could be ordered from any chemical supplier specialized in providing synthetic peptides. Trans-splicing buffer: 0.5 mM EDTA, 1 mM tris-(2-carboxyethyl)phosphine (TCEP), 50 mM NaH2PO4, 250 mM NaCl, pH 7.0 (Note 1). 2.5 In-cell trans-splicing reaction U2OS cells, available from ATCC (ATCC® HTB-96™). HeLa cells, available from ATCC (ATCC® CCL-2™). Dulbecco’s modified eagle medium (DMEM) containing 10% heat inactivated FBS (Fetal Bovine Serum), 1% L-glutamine and 1% penicillin-streptomycin solution. RIPA BUFFER: 50 mM Tris-HCl, 150 mM NaCl buffer, pH 8.0, 1% NP-40, 0.5% sodium deoxycholate, 1% Triton X-100. TBST buffer: 50 mM Tris-HCl, 150 mM NaCl, pH 7.6. Nonpyrogenic sterile 100 × 20 mm polystrene plates. Nonpyrogenic sterile 35 mm glass bottom plates (e.g. MatTek). Transfection reagent for mammalian cells (e.g. Fugene-6 from Promega or equivalent). Complete protease cocktail (e.g. Thermo Scientific). PVDF membrane for western blotting. 5% skim milk in TBST buffer. Anti-His Antibody (e.g. murine IgG). Store at −20° C. Secondary antibody (e.g. horseradish peroxidase-conjugated anti-murine IgG, Vector Lab). ECL kit (e.g. Life Technologies). Chariot protein delivery reagent (Active Motif). 3. Methods 3.1 Cloning of DNA encoding Npu DnaE IN into expression vector pET28a(+) Amplify by PCR the gene containing the Npu DnaE IN (residues 770–876, UniProtKB: B2J066) using a plasmid containing the DnaE gene from Nostoc punctiforme (Strain ATCC 2913/PCC 73102) as template using primers p5-IN and p3-IN (Table 2). The 5′-primer encodes a Sal I restriction site. The 3′-primer introduces a Not I restriction site and stop codon. Carry out the PCR reaction as follows: 40 μL sterile pure H2O, 1 μL of DNA template (≈10 ng/μL), 5 μL of 10× thermopol reaction buffer, 1.0 μL of dNTP solution (10 mM each), 1 μL of p5-IN primer solution (0.2 μM), 1 μL of p3-IN r primer solution (0.2 μM), and 1 μL Vent DNA polymerase (2 units). PCR cycle conditions used: initial denaturation at 94° C for 5 min followed by 30 cycles (94° C denaturation for 30 s, annealing at 52° C for 45 s, and extension at 72° C for 60 s) and final extension at 72° C for 10 min. Purify the PCR amplified fragment encoding Npu Dna IN using a PCR purification kit following the manufacturer instructions and quantify it by UV absorption (for a 1-cm pathlength, an optical density at 260 nm (OD260) of 1.0 equals to a concentration of 50 μg/mL solution of dsDNA). Digest plasmid pET28a(+) (Novagen-EMD Millipore) and PCR-amplified gene encoding the DnaE IN polypeptide with restriction enzymes Sal I and Not I. Use a 0.5 mL centrifuge tube and add 5 μL of 10× restriction buffer (e.g. NEB buffer 2.1 from New England Biolabs), add enough pure sterile water to have a final volume reaction of 50 μL, add ≈10 μg of the corresponding dsDNA to be digested and finally add 1 μL (20 units) of restriction enzyme Not I. Incubate at 37° C for 3 h. Then, add 1 μL (20 units) of restriction enzyme Sal I to the same tube and incubate at 37° C for 1 h. Purify the double digested PCR-product and pET28a plasmid by agarose (0.8% and 2% agarose gels for pET28a(+) and PCR product should be used, respectively) gel electrophoresis. Cut out the bands corresponding to the double digested DNA and purify the DNA using a gel extraction kit. Elute DNA from spin columns with TE buffer and quantify using UV. Ligate double digested pET28a(+) and PCR-product encoding DnaE IN. Use a 0.5 mL centrifuge tube, add ≈ 100 ng of Sal I, Not I-digested pET28a, ≈ 50 ng of Sal I, Not I-digested PCR-amplified DNA encoding DnaE IN, enough pure sterile H2O to make a final reaction volume of 20 μL, 2 μL of 10X T4 DNA ligase buffer, 1 μL of 10 mM ATP and 1 μL (400 units) T4 DNA ligase. Incubate at 16° C overnight. Transform the ligation mixture into DH5α competent cells. ≈100 μL of chemical competent cells are thawed on ice and mixed with the ligation mixture (20 μL) for 30 min. Heat-shock the cells are heat-shocked at 42° C for 45 s and then keep on ice for an extra 10 min. Add 900 μL of SOC medium and incubate at 37° C for 1 h in an orbital shaker. Plate 100 μL on LB agar plate containing kanamycin (25 μg/mL) and incubate the plate at 37° C overnight. Pick up several colonies (most of the times 5 colonies should be enough) and inoculate into 5 ml of LB medium containing kanamycin (25 μg/mL). Incubate tubes at 37° C overnight in an orbital shaker. Pellet down cells and extract DNA using a miniprep kit following the manufacturer protocol and quantify plasmid using UV spectroscopy. Verify the presence of DNA encoding DnaE IN in each colony using PCR and the same conditions described in step 1 of this section. Screen colonies containing DNA encoding DnaE IN for protein expression (Steps 11 through 17). Transform chemical competent BL21 (DE3) cells with plasmids containing the DNA encoding DnaE IN (Note 2). Transformed cells are plated on LB plate containing kanamycin (25 μg/mL) and incubated at 37° C overnight. Resuspend the colonies from 1 plate in 1 mL of LB and inoculate 100 mL of LB containing kanamycin (25 μg/mL) in a 250 mL flask. Grow cells in an orbital shaker incubator at 37° C for 2–3 h to reach mid-log phase (OD at 600 ≈ nm 0.5). Add IPTG to reach a final concentration of 1 mM and incubate cells for 3 h at 37° C. Pellet 1 mL of cells by centrifugation at 6,000 × g for 15 min at 4° C. Discard the supernatant and resuspend pellets in fresh SDS-PAGE sample buffer. Heat samples at 94 °C for 5 min and separate the soluble cell lysate fraction by centrifugation at 15,000 × g for 20 min at 4° C Analyze the soluble fraction by SDS-PAGE analysis to estimate the protein expression of the DnaE IN intein in each clone analyzed. The DnaE IN polypeptide should give a band around 15 kDa. 3.2 Cloning of DNA encoding YY1-IN construct into expression vector pET28a Amplify the gene containing the DNA binding domain of YY1 by PCR using the cDNA for human YY1 (cDNA clone IMAGE: 5261384) as template using primers p5-YY1 and p3-YY1 (Table 2). The 5′-primer and 3′-primer introduce Nde I and BamH I restriction sites, respectively. Carry out the PCR reaction as follows: 40 μL sterile pure H2O, 1 μL of DNA template (≈10 ng/μL), 5 μL of 10× thermopol reaction buffer, 1.0 μL of dNTP solution (10 mM each), 1 μL of p5-YY1 primer solution (0.2 μM), 1 μL of p3-YY1 primer solution (0.2 μM), and 1 μL Vent DNA polymerase (2 units). Use the following PCR cycle conditions: initial denaturation at 94° C for 5 min followed by 30 cycles (94° C denaturation for 30 s, annealing at 52° C for 45 s, and extension at 72° C for 60 s) and final extension at 72° C for 10 min. Purify the PCR amplified fragment encoding YY1 using a PCR purification kit following the manufacturer instructions and quantify by UV spectroscopy. Digest plasmid pET28a encoding DnaE IN (pET-IN) (obtained in section 3.1) and PCR-amplified gene encoding YY1 construct with restriction enzymes Nde I and BamH I. Use a 0.5 mL centrifuge tube and add 5 μL of 10× restriction buffer (e.g. NEB buffer 3.1 from New England Biolabs), add enough pure sterile water to have a final volume reaction of 50 μL, add ≈10 μg of the corresponding dsDNA to be digested and finally add 1 μL (20 units) of restriction enzyme Nde I. Incubate at 37° C for 3 h. Then, add 1 μL (20 units) of restriction enzyme BamH I to the same tube and incubate at 37° C for 1 h. Purify the double digested PCR-product and pET-IN plasmid by agarose (0.8% and 2% agarose gels for pET-IN and PCR product should be used, respectively) gel electrophoresis. Cut out the bands corresponding to the double digested DNA and purify them using a gel extraction kit. Elute DNA from the spin columns with TE buffer and quantify using UV spectroscopy. Ligate double digested pET-IN and PCR-product encoding the DBD of YY1. Use a 0.5 mL centrifuge tube, add ≈100 ng of Nde I, BamH I-digested pET-IN, ≈50 ng of Nde I, BamH I-digested PCR-amplified DNA encoding the DBD of YY1, enough pure sterile H2O to make a final reaction volume of 20 μL, 2 μL of 10X T4 DNA ligase buffer, 1 μL of 10 mM ATP and 1 μL (400 units) T4 DNA ligase. Incubate at 16° C overnight. Transform the ligation mixture into DH5α competent cells, plate them on a LB agar plate containing kanamycin (25 μg/mL), and incubate the plate at 37° C overnight as described previously (Section 3.1, steps 5 and 6). Pick up several colonies (most of the times 5 colonies should be enough) and screen for the presence of DNA inserts encoding the DBD of YY1 by PCR as described previously (Section 3.1, steps 7 through 9). Screen colonies containing DNA encoding the YY1-IN construct for protein expression (Steps 9 through 12). Transform chemical competent Rosetta (DE3) cells with plasmids containing the DNA encoding YY1-IN (Note 3). Transformed cells are plated on LB plate containing kanamycin (25 μg/mL) and and chloramphenicol (34 μg/mL) incubated at 37° C. Resuspend the colonies from 1 plate in 1 mL of LB and inoculate 100 mL of LB containing kanamycin (25 μg/mL) and chloramphenicol (34 μg/mL) in a 250 mL flask. Grow cells in an orbital shaker incubator at 37° C for 2–3 h to reach mid-log phase (OD at 600 nm ≈ 0.5). Add IPTG to reach a final concentration of 1 mM and incubate cells for 3 h at 37° C. Pellet, lyse cells and analyze protein expression level of YY1-IN construct by SDS-PAGE as described in sections 3.1.14 through 3.1.17. The YY1-IN construct should give a band around 30 kDa. 3.3 Bacterial expression of the YY1-IN construct Transform chemical competent Rosetta (DE3) cells with plasmid containing the DNA encoding YY1-IN (plasmid pET-YY1-IN). Transformed cells are plated on LB plate containing kanamycin (25 μg/mL) and chloramphenicol (34 μg/mL) and incubated at 37° C overnight (Note 4). Resuspend the colonies from 2 plates in 2 mL of LB and inoculate 1 L of LB containing kanamycin (25 μg/mL) and chloramphenicol (34 μg/mL) in a 2.5 L flask. Grow cells in an orbital shaker incubator at 37° C for 2–3 h to reach mid-log phase (OD at 600 nm ≈ 0.5). Add IPTG to reach a final concentration of 1 mM and incubate cells for 3 h at 37° C. Pellet cells by centrifugation at 6,000 × g for 15 min at 4° C. Discard the supernatant and process the pellet immediately (Note 5). 3.4 Purification of YY1-IN Protein YY1-IN was purified from inclusion bodies. Resuspend cell pellet with 30 mL of lysis buffer containing 1 mM PMSF. Lyse cells by sonication on ice using 25 s bursts spaced 30 s each (Note 6). Repeat the cycle six times (Note 7). Separate the soluble cell lysate fraction by centrifugation at 15,000 × g for 20 min at 4° C. Store the pellets at −80° C in case they need to be re-processed. Resuspend the insoluble fraction with lysis buffer (30 mL) containing detergent (0.1% Triton X-100). Separate soluble fraction by centrifugation at 15,000 × g for 20 min at 4° C. Repeat this process 2 more times. Wash the insoluble fraction with lysis buffer without detergent as described in step 3. Extract YY1-IN with 10 mL of PBS containing 4 M urea (Note 8). Characterize protein purity by SDS-PAGE analysis and determine concentration by UV spectroscopy (ε280 = 17,180 M−1 cm−1). Use immediately for in vitro trans-splicing experiments. 3.5 In vitro labeling of YY1 using protein trans-splicing Pure polypeptide IC (Table 1) and YY1-IN fusion protein are combined in freshly prepared and degassed trans-splicing buffer to a concentration of ≈ 1 μM and 0.1 μM, respectively. Dissolve IC peptide in pure 30% acetonitrile in pure H2O containing 0.1% trifluoroacetic acid (TFA) to obtain a stock solution of ≈ 1 mg/mL (≈160 μM). Add 3.1 μL of IC stock solution and the appropriate amount of YY1-IN dissolved in lysis buffer containing 4 M urea (Section 3.4, step 6) to 500 μL of fresh trans-splicing buffer. Keep the reaction at room temperature with occasional gently shaking. Take aliquots (50 μL) at different times (1, 3, 7, 10, 15, 30, and 45 min) and rapidly quench them by adding 15 μL of 4X SDS-PAGE sample buffer and heating them at 94° C for 2 min. Monitor the progression of the trans-splicing reaction by using SDS-PAGE (Fig. 3). Load 25 μL of each time point onto an SDS-4–20% PAGE gel. Run the samples at 125 V for about 1 h and 30 min in 1X SDS running buffer. Remove SDS with pure water. Visualize trans-splice reaction by epifluorescence (e.g. Storm 860 Molecular Imager), and silver stain kit (e.g. Thermo scientific, USA) following manufacturer protocol (Fig. 3) (Note 9). 3.6 Cloning of DNA encoding YY1-IN construct into mammalian expression vector pcDNA4/TO/myc-His 1 Amplify the DNA encoding the DBD of YY1 fused to the N-terminus of the DnaE IN (YY1-IN) by PCR using plasmid pET-YY1-IN (Section 3.2) as template using primers p5-YY1-IN and p3-YY1-IN (Table 2). The 5′-primer and 3′-primer introduce Kpn I and Not I restriction sites, respectively. Carry out the PCR reaction as follows: 40 μL sterile pure H2O, 1 μL of DNA template (≈10 ng/μL), 5 μL of 10X thermopol reaction buffer, 1.0 μL of dNTP solution (10 mM each), 1 μL of p5-YY1-IN primer solution (0.2 μM), 1 μL of p3-YY1-IN primer solution (0.2 μM), and 1 μL Vent DNA polymerase (2 units). Use the following PCR cycle conditions: initial denaturation at 94° C for 5 min followed by 30 cycles (94° C denaturation for 30 s, annealing at 52° C for 45 s, and extension at 72° C for 60 s) and final extension at 72° C for 10 min. 2 Purify the PCR amplified fragment encoding the YY1-IN fusion protein using the a PCR purification kit following the manufacturer instructions and quantify it by UV-visible spectroscopy. 3 Digest plasmid pcDNA4/TO/myc-His (Invitrogen) and PCR-amplified gene encoding YY1-IN with restriction enzymes Not I and Kpn I. Use a 0.5 mL centrifuge tube and add 5 μL of 10× restriction buffer (e.g. NEB buffer 2.1 from New England Biolabs), add enough pure sterile water to have a final volume reaction of 50 μL, add ≈10 μg of the corresponding dsDNA to be digested, add 1 μL (20 units) of restriction enzyme Not I and 1 μL (20 units) of restriction enzyme Kpn I. Incubate at 37° C for 16 h. 4 Purify the double digested PCR-product and pcDNA4 plasmid by agarose (0.8% and 2% agarose gels for pCDNA4 and PCR product should be used, respectively) gel electrophoresis. Cut out the bands corresponding to the double digested DNA and purify using a gel extraction kit. Elute DNA from the spin columns with TE buffer and quantify using UV-visible spectroscopy. 5 Ligate double digested pcDNA4 and PCR-product encoding YY1-IN. Use a 0.5 mL centrifuge tube, add ≈ 100 ng of Kpn I, Not I-digested pcDNA4, ≈ 50 ng of Kpn I, Not I-digested PCR-amplified DNA encoding YY1-IN, enough pure sterile H2O to make a final reaction volume of 20 μL, 2 μL of 10X T4 DNA ligase buffer, 1 μL of 10 mM ATP and 1 μL (400 units) T4 DNA ligase. Incubate at 16° C overnight. 6 Transform the ligation mixture into DH5α competent cells. Thaw ≈ 100 μL of chemical competent cells on ice and mix with the ligation mixture (20 μL) for 30 min. Heat-shock the cells at 42° C for 45 s and then keep on ice for an extra 10 min. Add 900 μL of SOC medium and incubate at 37° C for 1 h in an orbital shaker. Plate 100 μL on LB agar plate containing ampicillin (100 μg/mL) and incubate the plate at 37° C overnight. 7 Pick up several colonies (most of the times 5 colonies should be enough) and inoculate into 5 mL of LB medium containing ampicillin (100 μg/mL). Incubate tubes at 37° C overnight in an orbital shaker. 6 Pellet down cells and extract DNA using a miniprep kit following the manufacturer protocol and quantify plasmid using UV-visible spectroscopy. 7 Verify the presence of DNA encoding YY1-IN construct in each colony using PCR. Carry out the PCR reaction out as follows: 40 μL sterile pure H2O, 1 μL of plasmid DNA (≈50 ng/μL), 5 μL of 10× TaqDNA polymerase buffer, 1.0 μL of dNTP solution (10 mM each), 1 μL primer p5-YY1-IN solution (0.2 μM), 1 μL of primer p3-YY1-IN (0.2 μM), and 1 μL Taq DNA polymerase (5 units). 8 Use the following PCR cycle conditions: initial denaturation at 94° C for 5 min followed by 30 cycles (94° C denaturation for 30 s, annealing at 56° C for 45 s, and extension at 72° C for 60 s) and final extension at 72° C for 10 min. 3.7 Expression of YY1-IN fusion protein in U2OS and HeLa cells Grow U2OS and HeLa cells in DMEM containing 10% heat inactivated FBS, 1% L-glutamine and 1% penicillin-Streptomycin solution in a humidified incubator under 5% CO2 atmosphere at 37 °C. The transfection conditions and expression conditions should be optimized for each protein and cell line. Grow U2OS and HeLa cells in 100 mm nonpyrogenic sterile polystrene plates to ≈60% confluency. Transiently transfect the cells with plasmid pcDNA4-YY1-IN (1 μg DNA) using transfection reagent (e.g. Fugene-6) following the manual instructions. The transfected cells are then incubated for 24 h at 37° C. Expression of YY1-IN protein is estimated by western-blot analysis (steps 5–14). Wash cells twice with PBS. Discard PBS wash. Resuspend cells in 0.2 mL RIPA buffer in the presence of 0.1% SDS, 1 mM PMSF and complete protease cocktail. Use a cell scrapper to dislodge cells. Transfer the cell suspension to 1.5 mL centrifuge tubes and incubate on ice for 10 min. Separate the insoluble and soluble fractions by centrifugation at 14,000 rpm in a microcentrifuge at 4° C for 30 min. Incubate 100 μL of cell lysate with 20 μL of 4X SDS-PAGE sample buffer containing 20% mercaptoethanol. Heat the sample at 94°C for 2 min. Analyze the cell lysate samples using SDS-PAGE. Load samples (25 μL) onto an SDS-4–20% PAGE gel. Run the samples at 125 V for about 1 h and 30 min in 1X SDS running buffer. Transfer proteins from PAGE gel to a PVDF membrane using standard protocol. Block membrane with 5% skim milk in TBST buffer for 1 h at room temperature. Add primary anti-His antibody (e.g. murine anti-His IgG, Invitrogen) (Note 10) in TBST buffer containing 5% skim milk and incubate overnight at 4°C on an orbital shaker. Wash the membrane 3 times with TBST buffer for 5 min. Add secondary antibody (e.g. horseradish peroxidase-cojugated anti-murine IgG, Vector Lab) in TBST buffer containing 5% skim, and incubate at room temperature for 1 h. Wash membrane 3 times with TBST buffer for 5 min. Develop membrane with ECL mix (Life Technologies) following manufacturer instructions. Visualize bands using a molecular imager system (e.g. Storm 860, Amersham Biosciences) or using an X-ray film (Fig. 4) 3.8 In-cell labeling using protein trans-splicing 1 Grow U2OS and HeLa cells in either 35 mm glass bottom plates (when fluorescence microscopy is required) or 100 mm nonpyrogenic sterile polystrene plates to 40–50% confluency (see 3.7.1–3.7.3) and then transiently transfected with plasmid pcDNA4-YY1-IN (1 μg DNA) using a transfection reagent (e.g. Fugene-6) following the manual instructions. Incubate transfected cells for 24 h at 37° C. 2 Wash cells with PBS. Transfect peptide IC (50 nM) using the Chariot protein delivery reagent (Active Motif) for 1 h as described in the manufacturer manual (Note 11). 3 Wash transfected cells with full media 2 times and incubate for 1 h at 37° C. 4 Wash cells three times with serum free media and observe fluorescence-labeled proteins in live cells using a fluorescence microscope (Fig. 5A and 5B) (Note 12). 4 Lyse cells using RIPA buffer and analyze soluble cell lysate by SDS-PAGE as described previously (Section 3.7, Steps 5 through 9). 5 Visualize fluoresceine-labeled protein on PAGE-gel using a molecular imaging system (Fig. 5C). 5 Use the same PAGE-gel to analyze the efficiency of the labeling reaction by Western blotting as described previously (Section 3.7, steps 10 through 14) (Fig. 5C). Figure 1 A. Site-specific labeling and fluorescence activation of a protein of interest (POI) by FRET-quenched protein trans-splicing. Key to this approach is the introduction of fluorescence quencher (Q) into the IC polypeptide, which blocks the fluorescence signal of the fluorophore (F) located at the C-terminus of the IC polypeptide before protein trans-splicing happens. When protein trans-splicing occurs the fluorophore is covalently attached to the C-terminus of the POI triggering its fluorescence. The use of this approach for in-cell modification and fluorescence tagging of proteins minimizes the fluorescence background from the unreacted IC polypeptide thus facilitating the optical tracking of the labeled protein inside the cell. B. Scheme showing the approach used for in-cell labeling of a POI with a fluorophore inside a live cell using protein trans-splicing (figure modified from [13]). Figure 2 Design of FRET-quenched DnaE split inteins. A. Multiple sequence alignment of the DnaE IC for different species indicating the positions used for the introduction of the quencher group in the IC polypeptide. Multiple sequence alignment was performed using T-Coffee and visualized using Jalview [26]. Molecular representations of the DnaE inteins were generated using the PyMol software package. B. Crystal structure of the Npu DnaE intein in the pre-spliced state (PDB code: 2KEQ) [27]. DnaE IC and IN are shown in red and blue respectively. The structural secondary elements are also shown. The position used to place the quencher and fluorophore groups at the IC and C-extein, as well as the distances are indicated (figure modified from [13]). Figure 3 SDS-PAGE analysis of the in vitro protein trans-splicing/labeling reaction between FRET-quenched DnaE IC (Table 1) and YY1-IN. Protein detection was performed by silver staining (top) and epifluorescence (bottom). TS = trans-splicing. Figure 4 Expression of protein YY1-IN in U2OS (and HeLa) cells. Cells were collected after 24 h post-transfection, lysed and the soluble fraction analyzed and quantified by western blot using an anti-His antibody. Figure 5 In-cell site-specific labeling of YY1 DBD with a nuclear localization signal and concomitant fluorescence activation using protein trans-splicing. A. U2OS (and HeLa) cells were first transiently transfected with a plasmid encoding YY1-IN and with DnaE IC polypeptides as described. Cells were then extensively washed and examined by fluorescence microscopy. B. Magnification of cells after 18 h of incubation showing the migration of the labeled-YY1 to the nuclear compartment. C. Quantification of labeling yield for in-cell trans-splicing reaction. Identification of labeled YY1 DBD protein and quantification of in-cell trans-splicing yield was performed by western blot (right panel) and epifluorescence (left panel), respectively. Bar represents 25 μm in panels A and B. TS = trans-splicing. Table 1 Sequence of the DnaE C-intein (IC) polypeptide used in this protocol. Standard single code letters are used for the peptides sequences. Single letter codes B and X stand for norleucine and 6-amino hexanoic acid, respectively. 5-(Iodoacetamide)-fluoresceine (IAF) is used to introduce fluorescein into specific Lys or Cys residues, respectively. Dabcyl, Cam, Cam-Fl stand for 4-dimethylaminoazobenzene-4’-carboxyl, carboxamidomethyl, and fluorescein-carboxamidomethyl, respectively. Residues in blue, magenta and yellow represent the Npu DnaE IC intein, Npu DnaE C-extein and NLS peptide, respectively. The quencher and fluorophore groups are in red and green respectively. Peptide Name Compound Sequence Molecular Weight Found (Expecteda)/Da Npu QM-IC-NLS-Fl IC 6449.0±0.1 (6447.6) a Average molecular weight Table 2 DNA oligonucleotides used to generate the dsDNA encoding the YY1-IN and DnaE IN. Primer Name Nucleotide Sequence p5-YY1 5′-AAA GAA GAT CAT ATG CCA AGA ACA ATA GCT TGC CCT C-3′ p3-YY1 5′-C TTC GGA TCC CTG GTT GTT TTT GGC CTTA GC-3′ p5-IN 5′-CTA GTC GAC AAG CTT TTA AGT TTG CGG AAT ATT GTT TAA G CTA TG -3′ p3-IC 5′-TTT GCG GCC GCT TAA TTC GGC AAA TTA TCA ACC CGC AT -3′ p5-YY1-IN 5′-AGG GGT ACC ACC ATG GGC AGC AGC -3′ p3-YY1-IN 5′-GT GGT GCT CGA GTG CGG CCG CA -3′ 1 The buffer should be degassed before adding the TCEP. The buffer should be prepared fresh every time and not stored as TCEP has a limited stability at pH 7.0 in the presence of air. 2 We recommend to screen at least 3–4 different colonies for protein expression in BL21(DE3) cells to make sure that the polypeptide encoding DnaE IN is expressed efficiently. 3 We recommend to screen at least 3–4 different colonies for protein expression in Rosetta(DE3) cells to make sure that the clone with highest expression yield is selected. Expression in Rosetta cells is recommended when the protein of interest (POI) contains rare codons in E. coli. Select the clone with the highest expression yield and proceed to the next section. 4 When plating the transformed cells with pET-YY1-IN, it is better to aim for plates containing 200–300 colonies. 5 Cell pellets can be stored at −80° C for no more than 2–3 weeks before being processed. 6 During sonication, be sure the temperature of the sample does not overheat. 7 A french press can be also used to lyse cells, depending on the availability. 8 Use the highest quality urea available. PBS containing urea should be prepared fresh and not stored for more than 1 day, as urea is not stable in buffers at pH above 7. 9 It is extremely import to check that the labeling/trans-splicing reaction works efficiently in vitro before trying it in live cells. 10 In this particular work the N-terminal of the DBD of YY1 contained a His-tag and therefore an anti-His IgG was used to detect the protein by Western blotting. 11 Peptide transfection should be optimized for any particular case. In our hands, the best peptide transfection efficiency was obtained for 50 nM IC using a molecular ratio IC: Pep-1 of 1:20. Pep-1 is the name of the peptide used in the Chariot transfection system and its sequence is Ac-KETWWETWWTEWSQPKKKRKV-cysteamine. This peptide can be also obtained from different commercial sources (e.g. Anaspec). 12 The use of a FRET-quenched IC polypeptide allows concomitant labeling of the protein with the corresponding fluorophore and activation of its fluorescence. 1 Xie XS Yu J Yang WY 2006 Living cells as test tubes Science 312 228 230 16614211 2 Chen I Ting AY 2005 Site-specific labeling of proteins with small molecules in live cells Curr Opin Biotechnol 16 35 40 15722013 3 Miller LW Cornish VW 2005 Selective chemical labeling of proteins in living cells Curr Opin Chem Biol 9 56 61 15701454 4 Perler FB 2005 Protein splicing mechanisms and applications IUBMB Life 57 469 476 16081367 5 Saleh L Perler FB 2006 Protein splicing in cis and in trans Chem Rec 6 183 193 16900466 6 Xu MQ Evans TC Jr 2005 Recent advances in protein splicing: manipulating proteins in vitro and in vivo Curr Opin Biotechnol 16 440 446 16026977 7 Muir TW 2003 Semisynthesis of proteins by expressed protein ligation Annu Rev Biochem 72 249 289 12626339 8 Shi J Muir TW 2005 Development of a tandem protein trans-splicing system based on native and engineered split inteins J Am Chem Soc 127 6198 6206 15853324 9 Kwon Y Coleman MA Camarero JA 2006 Selective Immobilization of Proteins onto Solid Supports through Split-Intein-Mediated Protein Trans-Splicing Angew Chem Int Ed 45 1726 1729 10 Kurpiers T Mootz HD 2008 Site-specific chemical modification of proteins with a prelabelled cysteine tag using the artificially split Mxe GyrA intein Chembiochem 9 2317 2325 18756552 11 Giriat I Muir TW 2003 Protein semi-synthesis in living cells J Am Chem Soc 125 7180 7181 12797783 12 Woo YH Stubbs L Camarero JA 2008 In vivo protein labeling via protein trans-splicing The 21st Annual Symposium of the Protein Society 32 San Diego 13 Borra R Dong D Elnagar AY Woldemariam GA Camarero JA 2012 In-cell fluorescence activation and labeling of proteins mediated by FRET-quenched split inteins J Am Chem Soc 134 6344 6353 22404648 14 Dassa B Amitai G Caspi J Schueler-Furman O Pietrokovski S 2007 Trans protein splicing of cyanobacterial split inteins in endogenous and exogenous combinations Biochemistry 46 322 330 17198403 15 Zettler J Schutz V Mootz HD 2009 The naturally split Npu DnaE intein exhibits an extraordinarily high rate in the protein trans-splicing reaction FEBS Lett 583 909 914 19302791 16 Mootz HD Blum ES Tyszkiewicz AB Muir TW 2003 Conditional protein splicing: a new tool to control protein structure and function in vitro and in vivo J Am Chem Soc 125 10561 10569 12940738 17 Vila-Perello M Hori Y Ribo M Muir TW 2008 Activation of protein splicing by protease- or light-triggered O to N acyl migration Angew Chem Int Ed 47 7764 7767 18 Berrade L Kwon Y Camarero JA 2010 Photomodulation of protein trans-splicing through backbone photocaging of the DnaE split intein Chembiochem 11 1368 1372 20512791 19 Binschik J Zettler J Mootz HD 2011 Photocontrol of protein activity mediated by the cleavage reaction of a split intein Angew Chem Int Ed Engl 50 3249 3252 21384476 20 Wu H Hu Z Liu XQ 1998 Protein trans-splicing by a split intein encoded in a split DnaE gene of Synechocystis sp. PCC6803 Proc Natl Acad Sci USA 95 9226 9231 9689062 21 Wei XY Sakr S Li JH Wang L Chen WL Zhang CC 2006 Expression of split dnaE genes and trans-splicing of DnaE intein in the developmental cyanobacterium Anabaena sp. PCC 7120 Res Microbiol 157 227 234 16256311 22 Iwai H Zuger S Jin J Tam PH 2006 Highly efficient protein trans-splicing by a naturally split DnaE intein from Nostoc punctiforme FEBS Lett 580 1853 1858 16516207 23 Wang CC Chen JJ Yang PC 2006 Multifunctional transcription factor YY1: a therapeutic target in human cancer? Expert Opin Ther Targets 10 253 266 16548774 24 Gronroos E Terentiev AA Punga T Ericsson J 2004 YY1 inhibits the activation of the p53 tumor suppressor in response to genotoxic stress Proc Natl Acad Sci U S A 101 12165 12170 15295102 25 Sui G Affar el B Shi Y Brignone C Wall NR Yin P Donohoe M Luke MP Calvo D Grossman SR 2004 Yin Yang 1 is a negative regulator of p53 Cell 117 859 872 15210108 26 Taki M Sisido M 2007 Leucyl/phenylalanyl(L/F)-tRNA-protein transferase-mediated aminoacyl transfer of a nonnatural amino acid to the N-terminus of peptides and proteins and subsequent functionalization by bioorthogonal reactions Biopolymers 88 263 271 17216634 27 Oeemig JS Aranko AS Djupsjobacka J Heinamaki K Iwai H 2009 Solution structure of DnaE intein from Nostoc punctiforme: structural basis for the design of a new split intein suitable for site-specific chemical modification FEBS Lett 583 1451 1456 19344715

PMC005xxxxxx/PMC5117463.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0370522 482 Am Surg Am Surg The American surgeon 0003-1348 1555-9823 27657564 5117463 NIHMS813549 Article Overwhelming Recurrent Clostridium difficile Infection after Reversal of Diverting Loop Ileostomy Created for Prior Fulminant C. difficile Colitis Fashandi Anna Z. M.D. Ellis Scott R. Smith Philip W. M.D. Hallowell Peter T. M.D. Department of Surgery, University of Virginia School of Medicine, Charlottesville, Virginia, USA Address correspondence to: Peter T. Hallowell, University of Virginia, Department of Surgery, PO Box 800679, Charlottesville, Virginia 22908-1394, USA, Phone: (434) 243-4811, Fax: (434) 243-7272, PTH2F@hscmail.mcc.virginia.edu 31 8 2016 8 2016 01 8 2017 82 8 194195 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Clostridium difficile infection (CDI) is the leading cause of infectious diarrhea in the developed world and the incidence and severity of CDI began increasing at the turn of the 21st century. “Fulminant” CDI refers to illness that is life-threatening and, importantly, 36–75% of those who develop fulminant CDI have had recent surgery.1,2 Initial treatment of CDI depends on severity of infection and involves oral metronidazole, vancomycin or fidaxomicin. For CDI which is not responding to medical management, early surgical consultation is recommended with the gold standard surgical intervention being total abdominal colectomy and end ileostomy.1 In 2011, however, Neal and colleagues published the novel technique of performing diverting loop ileostomy with colonic lavage and postoperative antegrade vancomycin enemas in patients with severe CDI. They reported preservation of the colon in 93% of patients in addition to reduced mortality (19% vs 50%, odds ratio 0.24) when compared to matched historical patients treated with total abdominal colectomy.3 Although their outcomes interested many in the surgical community, critics have cited concerns over study design, small sample size, and results that have not been reliably replicated.1 We present the case of a 77 year old male who developed severe CDI after taking ciprofloxacin for prostatitis. He underwent emergent diverting loop ileostomy with colonic lavage and antegrade vancomycin enemas resulting in control of his severe colitis. Colonoscopy five months later showed nonspecific colitis and inflammation. He was referred to our facility three months later regarding ileostomy reversal, after his health had returned to baseline. One year after his index operation, he underwent technically uneventful ileostomy reversal. He had early return of bowel function on postoperative day one with loose bowel movements and, by postoperative day two, he felt well and was tolerating a full liquid diet. On the morning of postoperative day three, he was febrile, tachycardic and began having watery diarrhea. Several hours later, he was hypotensive with altered mental status, a markedly distended abdomen, a white blood cell count of 4.5 k/μl, a lactic acid of 5.5 mmol/L and rising serum creatinine. Abdominal plain film showed colonic distension and no pneumoperitoneum. He then required vasopressors for persistent hypotension and broad-spectrum antibiotics were initiated for concern for anastomotic leak. He developed respiratory distress requiring emergent intubation along with central and arterial line placement. His C. difficile test came back positive and a CT scan was consistent with recurrent C. difficile colitis (Fig. 1A & B). That evening, he was taken emergently for exploratory laparotomy and total abdominal colectomy. A short segment enterectomy also was performed due to a single enterotomy. His bowels were left in discontinuity and an Abthera was placed with plans for re-exploration. The following day, he remained acidotic on a bicarbonate drip with increasing pressor requirements and multisystem organ failure requiring high flow oscillating ventilator use as well as continuous dialysis. Based on the severity of his illness, his family elected to withdraw vasopressor and ventilator support. He was extubated and expired shortly thereafter, five days after his ileostomy reversal. Final pathology showed pseudomembranous enteritis and pseudomembranous colitis. This case highlights several questions including appropriate timing of ileostomy reversal and testing for persistent C. difficile. Neal and colleagues did not discuss their criteria for determining when and on whom to perform ileostomy reversal in their series. At the time of initial publication, fifteen of their nineteen surviving patients (79%) had undergone ileostomy reversal. Because diverting loop ileostomy and colonic lavage is a new approach to managing fulminant CDI, there is not yet any long-term follow up data regarding recurrence of CDI after ileostomy reversal. Neal and colleagues did not report subsequent testing for C. difficile prior to ileostomy reversal, nor did they document any immediate recurrences after reversal. Abe and colleagues, however, published a case report in Japan, in which a 69 year old man developed fulminant and fatal CDI several days after ileostomy closure in an ileostomy formed for reasons unrelated to CDI.2 There have been other reports of CDI after ileostomy closure and one series found a statistically significant association of postoperative CDI with delayed ileostomy closure (>6 months from index procedure, p=0.003).4 After this alarming case of fatal recurrence of CDI, we recommend counseling patients about the risk of C. difficile recurrence prior to offering ileostomy closure to those who had their stomas initially created for treatment of CDI. A high index of suspicion as well as a multidisciplinary approach including surgeons and gastroenterologists is necessary for swift diagnosis and treatment of this postoperative complication. Additionally, we are looking into the possibility of perioperative fecal transplantation to both prevent and treat recurrent CDI after ileostomy reversal. Fig. 1 Abdominal computed tomography scan of patient. Representative (A) axial and (B) coronal images showing inflammation and fat stranding around the cecum and ascending colon, consistent with recurrent C. difficile colitis. 1 Steele SR McCormick J Melton GB Practice parameters for the management of clostridium difficile infection Dis Colon Rectum 2015 58 1 10 24 25489690 2 Abe I Kawamura YJ Sasaki J Konishi F Acute fulminant pseudomembranous colitis which developed after ileostomy closure and required emergent total colectomy: A case report J Med Case Rep 2012 6 130-1947-6-130 3 Neal MD Alverdy JC Hall DE Diverting loop ileostomy and colonic lavage: An alternative to total abdominal colectomy for the treatment of severe, complicated clostridium difficile associated disease Ann Surg 2011 254 3 423 7 discussion 427–9 21865943 4 Rubio-Perez I Leon M Pastor D Increased postoperative complications after protective ileostomy closure delay: An institutional study World J Gastrointest Surg 2014 6 9 169 174 25276286

PMC005xxxxxx/PMC5117630.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8500909 5316 J Subst Abuse Treat J Subst Abuse Treat Journal of substance abuse treatment 0740-5472 1873-6483 27776677 5117630 10.1016/j.jsat.2016.08.015 NIHMS812095 Article From long-term injecting to long-term non-injecting heroin and cocaine use: the persistence of changed drug habits Des Jarlais Don C. 1 Arasteh Kamyar 1 Feelemyer Jonathan 1 McKnight Courtney 1 Barnes David M. 1 Tross Susan 2 Perlman David C. 1 Campbell Aimee N. C. 2 Cooper Hannah LF 3 Hagan Holly 4 1 Icahn School of Medicine at Mount Sinai, New York, NY 2 New York State Psychiatric Institute, Department of Psychiatry, Columbia University Medical Center, New York, NY 3 Department of Behavioral Sciences and Health Education, Rollins School of Public Health, Emory University, Atlanta Georgia 4 College of Nursing, New York University, New York NY Contact Author Information: Don C. Des Jarlais, PhD, Professor of Psychiatry, The Baron Edmond de Rothschild Chemical Dependency Institute, Icahn School of Medicine at Mount Sinai, 39 Broadway 5th Floor Suite 530, New York, NY 10006, 212.256.2548, ddesjarlais@chpnet.org 24 8 2016 21 8 2016 12 2016 01 12 2017 71 4853 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objectives Transitioning from injecting to non-injecting routes of drug administration can provide important individual and community health benefits. We assessed characteristics of persons who had ceased injecting while continuing to use heroin and/or cocaine in New York City. Methods We recruited subjects entering Mount Sinai Beth Israel detoxification and methadone maintenance programs between 2011 and 2015. Demographic information, drug use histories, sexual behaviors, and “reverse transitions” from injecting to non-injecting drug use were assessed in structured face-to-face interviews. There were 303 “former injectors,” operationally defined as persons who had injected at some time in their lives, but had not injected in at least the previous 6 months. Serum samples were collected for HIV and HCV testing. Results Former injectors were 81% male, 19% female, 17% White, 43% African-American, 38% Latino/a, with a mean age of 50 (SD=9.2), and were currently using heroin and/or cocaine. They had injected drugs for a mean of 14 (SD=12.2) years before ceasing injection, and a mean of 13 (SD=12) years had elapsed since their last injection. HIV prevalence among the sample was 13% and HCV prevalence was 66%. The former injectors reported a wide variety of reasons for ceasing injecting. Half of the group appeared to have reached a point where relapse back to injecting was no longer problematic: they had not injected for three or more years, were not deliberately using specific techniques to avoid relapse to injecting, and were not worried about relapsing to injecting. Conclusions Former injectors report very-long term behavior change towards reduced individual and societal harm while continuing to use heroin and cocaine. The behavior change appears to be self-sustaining, with full replacement of an injecting route of drug administration by a non-injecting route of administration. Additional research on the process of long-term cessation of injecting should be conducted within a “combined prevention and care” approach to HIV and HCV infection among persons who use drugs. Former-injectors Relapse Non-injection drugs use HIV Hepatitis C 1. Introduction The transition from using illicit drugs through non-injecting routes of administration to injecting greatly increases the risk of harmful consequences (Fuller et al., 2002). Injecting greatly increases the likelihood of infection with blood-borne viruses such as HIV and hepatitis C (HCV), the likelihood of other infections such as skin abscesses and endocarditis, the likelihood of developing a substance use disorder, and the likelihood of an overdose (Degenhardt & Hall, 2012; Garfein, Vlahov, Galai, Doherty, & Nelson, 1996; Kerr et al., 2007; Spijkerman, van Ameijden, Mientjes, Coutinho, & van den Hoek, 1996). As injecting drug use is heavily stigmatized, injecting also increases the likelihood that the user will experience severe social discrimination (Ahern, Stuber, & Galea, 2007). Compared to intranasal administration, injecting typically produces more intense drug effects and is a relatively cost-efficient method of drug administration. The transition from non-injecting to injecting drug use is often considered a permanent change in the preferred route of drug administration. While the person who injects may also use non-injectable drugs (nicotine, alcohol, marijuana), continued use of injectable drugs is likely to be primarily through injecting. “Reverse transitions” from injecting to non-injecting drug use have been observed in a variety of locations, including New York City (Des Jarlais et al., 2007), New Haven (Schottenfeld, O'Malley, Abdul-Salaam, & O'Connor, 1993), Baltimore (Genberg et al., 2011), Amsterdam (Buster et al., 2009), Brazil (Inciardi et al., 2006), China (Li et al., 2011) and Malaysia (Tejani, Chawarski, Mazlan, & Schottenfeld, 2011). These “former injectors” can maintain heroin/cocaine use for long periods of time—often decades—without relapsing back to injecting (Des Jarlais et al., 2007). They do as well in methadone treatment as persons who inject (Schwartz et al., 2015) and they are less likely to acquire HCV than persons who continue to inject (Des Jarlais et al., 2013). There is still much that is not known about reverse transitions from injecting to non-injecting drug use, in particular, how some former injectors avoid relapse to injecting over long periods of time. Relapse is considered one of the defining characteristics of substance use disorders. One would expect that persons who have reverse transitioned from injecting to non-injecting drugs would have great difficulties in avoiding relapse back to a route of administration that produces an intense drug effect at a relatively low monetary cost. For example, in a Baltimore study of 936 persons who ceased injecting, three-quarters of them relapsed back to injecting, the median time before relapse was only 1 year, and both non-injecting use of heroin and of cocaine after ceasing injecting were significantly associated with shorter time to relapse (Shah, Galai, Celentano, Vlahov, & Strathdee, 2006). In this report we examine various characteristics of a group of 303 former injectors, most of whom have maintained their non-injecting drug use over long periods of time. We specifically examine their reasons for stopping injecting, the length of time since their last injection, whether they are worried about relapsing to injecting, their use of specific mechanisms for avoiding relapse, and their attitudes towards the different highs produced by injecting versus non-injecting drug use. Finally, we consider implications of the findings for reducing drug injecting and for changing drug use behavior in general. 2. Materials and Methods The findings reported here are derived from data collected from patients/subjects entering the Mount Sinai Beth Israel drug detoxification and methadone maintenance programs in New York City. The methods for this “Risk Factors” study have been previously described in detail (Des Jarlais et al., 1989; Des Jarlais et al., 2009) so only a summary will be presented here. The Mount Sinai Beth Israel detoxification program serves New York City as a whole, with approximately half of the patients residing in Manhattan, one quarter in Brooklyn, one-fifth in the Bronx, and the remainder (about 5%) in other areas. Patients enter the program voluntarily. Research staff visited the general admission wards of the program in a preset order and examined all intake records of a specific ward to construct lists of all patients admitted within the prior 3 days. All of the patients on the list for the specific ward were then asked to participate in the study. Among patients approached by our interviewers, willingness to participate was more than 95%. After all of the patients admitted to a specific ward in the 3-day period had been asked to participate and interviews had been conducted among those who agreed to participate, the interviewer moved to the next ward. As there was no relationship between assigning patients to wards and the order that the staff rotated through the wards, these procedures would yield an unbiased sample of persons entering the detoxification program. A structured questionnaire covering demographic characteristics, HIV risk behavior, and drug use history was administered by a trained interviewer to each patient. Most drug use and HIV risk behavior questions referred to the 6 month period prior to the interview. Subjects were asked if they had ever injected illicit drugs, and if yes, their age at first injection and how long it had been since their last injection. “Former injectors” were operationally defined as subjects who 1) reported that they had injected drugs at some point in their lives, 2) reported that they had not injected within the last 6 months, and 3) that they had continued to use injectable drugs (heroin, cocaine or amphetamines) through non-injecting routes of administration (intranasal use and/or smoking and/or oral use). Former injectors were also asked about reasons for stopping injecting, whether they were using specific strategies to avoid relapse, and if yes, to describe those strategies. Multiple responses were permitted for both reasons for ceasing injecting and strategies. Subjects were also asked to compare the “highs” of cocaine and heroin when used through different routes of administration. They were asked “How do you compare the high of injecting heroin (and/or cocaine) to the high of snorting heroin (and/or cocaine)?” After completing the interview, each participant was seen by an HIV counselor for pretest counseling for HIV and HCV, along with specimen collection. (It was not possible to collect serum samples from all participants due to problems with collapsed veins.) HIV testing was conducted at the New York City Department of Health laboratory by using a commercial, enzyme-linked, immunosorbent assays (EIA) test with Western blot confirmation (BioRad Genetic Systems HIV-1-2+0 EIA and HIV-1 Western Blot, BioRad Laboratories, Hercules, CA). Samples were tested for HCV antibody with the Ortho HCV enzyme immunoassay (EIA) 3.0 (Ortho-Clinical Diagnostics, Inc., Raritan, NJ). Samples with optical density values of > 8.0 were considered positive, samples with values of 1.0 to 8.0 were confirmed positive with radio-immune blotting assay (RIBA) (Chiron RIBA HCV 3.0 Strip Immunoblot Assay, Novartis Vaccines & Diagnostics, Inc. Emeryville, CA) and samples with values < 1.0 were considered negative. Serial cross-sectional data have been collected for the project since 1990. We did permit individuals to participate in the study in different years. For the analyses reported here we used only the last interview from the 9 former injectors who participated more than once. STATA 13 (StataCorp, College Station, TX), was used for analyses and for generating graphs. We utilized the Epanechnikov kernel density estimates for smoothing curves in the graphs. The study was approved by the Mount Sinai Beth Israel Institutional Review Board. 3. Results A total of 303 participants were included in the analysis; 63% from the detoxification program and 37% from the methadone maintenance program. Below we describe the participant sample along with the results of the qualitative analysis of injecting patterns, strategies to avoid relapse back to injecting, and how different routes of administration produce different highs. 3.1 Demographics and Drug Use Behavior Table 1 presents demographic characteristics, current drug use behaviors, HIV and hepatitis C (HCV) serostatus. The subjects were predominantly male and predominantly African-American and Latino/a. They had a mean age of 50 (SD 9.2), and had injected drugs for considerable amounts of time, with a mean of 14 and median of 12 years of injecting. Intranasal heroin was used by a substantial majority (71%) of subjects in the 6 months prior to the interview, though a majority (53%) also reported smoking crack cocaine and a substantial minority (34%) reported using intranasal cocaine. All subjects reported previous episodes of substance use treatment, with 98% having previously received methadone and/or detoxification programs, and 61% having previously received both detoxification and methadone maintenance. Figure 1 shows the distribution of times since last injection. We represented the distribution of time since last injection in a histogram. The height of each column in the histogram depicts the number of participants corresponding to the years-since-last-injection at the base of that column. In order to also present a continuous depiction of this distribution, we used the Epanechnikov kernel density estimate to connect the column data points. In this techniques the data point being evaluated is placed at the center of the kernel and other data points are weighted according to their distance from the point being evaluated, thereby giving greater influence to the data points closer to the center of the kernel. While the modal time since last injection was within the previous year (15%), substantial proportions reported very long time periods since their last injection: 56% of these former injectors reported a decade or longer since their last injection, 35% reported 20 years or longer since their last injection, and 15% reported 30 years or longer since their last injection. With allowances for recall rounding to the nearest decade by some subjects in their reported time since last injection, the curve is similar to an exponential decay curve. Table 2 shows a cross-tabulation of last drug injected (only a single drug could be named in response to this question) and the non-injecting drug use reported for the last 6 months by the former injectors (multiple drugs could be mentioned). There is considerable continuity in the use of heroin and cocaine. Of the 227 persons who reported heroin as their last drug injected, 75% of them reported intranasal use of heroin in the 6 months prior to the interview, and of the 26 who reported cocaine as their last drug injected 85% of them reported smoking crack cocaine in the 6 months prior to the interview. Thus, even within a context of poly-drug use, these subjects maintained considerable continuity in the drugs they were using while having made a major change in their route of drug administration. 3.2 Reasons for ceasing injection Table 3 presents the reasons cited by former injectors for why they stopping injecting. Subjects were permitted to endorse multiple reasons (the percentages thus add to greater than 100%). A considerable variety of reasons were endorsed, and no single reason was endorsed by as many as 20% of the former injectors. Protecting health (avoiding HIV/AIDS endorsed by 12%, afraid of overdose by 15%, other health concerns by 17%) and problems with injecting as a route of administration (don't like needles by 16%, got tired of injecting by 14%, loss of veins by 9%, prefer mellow high of snorting by 5%, difficulties in obtaining injection equipment by 4%) were the two most frequently endorsed major categories of reasons for ceasing to inject. A substantial percentage also cited social and self-image concerns (avoiding stigmatization by 12%, avoiding track marks by 12%, maintaining self-image as a non-injector by 4%) as reasons for ceasing to inject. 3.3 Worry about relapse Many former drug users report avoiding relapse back to injecting drug use as being a “one day at a time” continuous struggle, and with great concern that they will relapse (Binswanger et al., 2012; Evans, Hahn, Lum, Stein, & Page, 2009; McKay, 1999). We asked these former injectors whether they were “worried that they would relapse back to injecting?” A very large majority (86%) reported that they were not worried about relapsing back to injecting. Length of time since last injection was negatively associated with being worried about relapsing to injecting in univariate logistic regression (see Table 4a). Among the 85 former injectors whose last injection was ≤ 36 months prior to the interview, 21 (25%) reported being worried about relapsing, while among the 218 former injectors whose last injection was > 36 months prior to the interview, only 20 (9%) were worried about relapsing back to injecting (chi squared = 12.57, df=1, p < 0.001; OR=0.31, 95% CI= 0.15 – 0.61). Worrying about relapse was not significantly associated with any other demographic and drug use variables in Table 1. 3.4 Strategies to avoid relapse Relapse prevention programs typically teach former drug users a variety of techniques to prevent relapse, including avoiding situations that would cue drug craving, seeking social support, and realizing that a single “lapse” does not have to mean a full relapse back to continued drug use (Dejong, 1994; Marlatt & Donovan, 2005; Witkiewitz & Marlatt, 2004). We asked the former injectors if they were using any specific strategy to avoid relapse, and, if they were, to briefly describe the strategy. A large majority (235/303, 78%) reported that they were not consciously using any specific strategy. Among the 67 former injectors who reported that they were using a specific strategy, 43% reported using self-talk/thinking negative thoughts about injecting, 18% reported avoiding places and people that might lead them back to injecting, 13% reported that they attended meetings, 9% reported that they were using alternative methods of drug administration, and 8% reported aversion to needles. There was a positive relationship between worrying about relapse and using specific strategies to avoid relapse—former injectors who reported worrying about relapse were more likely to report that they were using specific strategies than former injectors who were not worried about relapse (see Table 4b). Having a specific strategy was not significantly related to any of the demographic or current drug use variables in Table 1 (full data available from the first author). 3.5 Comparisons of “highs” between injecting and intranasal drug use As noted above, it is generally accepted that injecting a psychoactive drug produces a more intense drug effect than intranasal administration, though there may be considerable individual variation in the differences. We asked these former injectors if injecting heroin gave them a more intense high compared to intranasal use and if injecting cocaine gave them a more intense high compared to intranasal use. (These questions were asked only of the persons who were currently using heroin or cocaine intranasally and had injected that drug in the past.) Modest majorities of the current heroin and cocaine users reported that injecting gave them a more intense high—63% endorsing that injecting heroin gave a more intense high and 52% endorsing that injecting cocaine gave a more intense high. We assessed whether endorsing the greater intensity statements was associated with being worried about relapse (results presented in Tables 5a and 5b). The association between endorsing greater intensity of injecting and being worried about relapse approached statistical significance for heroin (p =0.066) and was significant for cocaine (p = 0.03). 4. Discussion Addiction is often defined as a chronic, relapsing condition in which it is very difficult—though not impossible—for individuals to make long-lasting changes in their patterns of drug use. The common image of a former drug user is a person who is attempting to avoid relapse, a person who must be vigilant to avoid people, places, and things that can trigger craving, and is deliberately using multiple evidence-based techniques to avoid relapse. As shown in Table 3, the former injectors interviewed in this study reported a great number of reasons for ceasing to inject drugs. Ceasing to inject usually was not a goal in itself, but rather was a means to achieve a wide variety of individual goals. From the content of the reasons in Table 3, it would appear that ceasing to inject would probably have been very effective in achieving these goals. For example, a third of the former injectors ceased injecting over 20 years ago, before large-scale syringe exchange programs were implemented in New York City. It is likely that ceasing to inject at that time provided considerable protection against HIV and HCV infection, as well as protection from transmitting to others if they were seropositive. And, as the stigmatization of injecting drug use has clearly continued in New York City, it is likely that ceasing to inject also reduced stigmatization. A recent study from Finland would also suggest that transitioning to non-injecting drug use would reduce the risk of fatal drug overdose. In that study, fatal overdoses among people who injected drugs were three times more likely than among people using the same drugs though other routes of administration (Onyeka et al., 2016). As a group, the former injectors in this study reported relatively long periods of injecting drug use—a mean of 12 years—followed by long periods since their last injection, with over half reporting 10 years or more since their last injection and over a third reporting 20 years or more since their last injection. There was, however, considerable variation among the former injectors. A subgroup of former injectors appear to be still “in the process” of actively avoiding relapse to injection drug use. They were worried about relapsing, had shorter times since last injection, deliberately used specific strategies to avoid relapse and were likely to recall injecting as providing a more intense high than intranasal drug use. However, this description fits only a modest percentage of these former injectors—only 6/303 (2%) had all four of these characteristics, and 39 (13%) had three of these characteristics. A second, much larger proportion of these former injectors appear to have reached stability where avoiding relapse to injecting was no longer problematic for them. It had been a long time since their last injection (at least 3 years and with a mean of 19 years), they were not worried about relapsing, they were not consciously using specific strategies to avoid relapse, and only about half of them recalled injecting as providing a more intense high than intranasal use. This description fits 145/303 (48%) of the former injectors in our sample. It would seem that they had fully replaced the habit of injecting drug use with a habit of non-injecting drug use. Changing drug use behavior in a direction of reduced individual and societal harm and in a way that becomes self-sustaining is a major goal of harm reduction and substance use treatment programs. It is worth considering several aspects of the sustained behavior change exhibited by these former injectors. In addition to individual goals, former injectors also ceased injecting and avoided relapse in an environment that provided facilitators for doing so. The quality of street heroin and cocaine in New York City has been relatively high for a long time, so that it is economically feasible to maintain a drug habit without having to inject. There are also low threshold, harm reduction-oriented substance use treatment programs in New York City, so that it is possible to receive treatment that provides a respite from the difficulties of street drug use, reduce one's level of drug tolerance, and reduce (or cease) drug use. 98% of the former injectors in this study reported that they had previously received either detoxification or methadone maintenance treatment and 61% reported that they had previously received both detoxification and methadone maintenance treatment. This episodic treatment for substance use problems clearly did not lead to permanent abstinence, but it is likely to have reduced the chances of relapsing to injecting. These former injectors did not, however, receive any formal treatment services for avoiding relapse to injecting while continuing non-injecting use of heroin and cocaine. Currently, neither drug treatment programs nor syringe exchange programs in New York City support ceasing to inject while continuing to use heroin or cocaine through non-injecting routes of administration as a formal “treatment goal.” About a sixth of the former injectors gave their reason for injecting as “They don't like needles.” Fear of/dislike for needles persisted during their injecting careers. Further research into fear of/dislike for needles may provide valuable insights for the treatment of persons with substance use disorders who are currently injecting and for programs to prevent initiation into injecting drug use. 4.1 Memories of injecting Almost half of these former injectors did not recall injecting as producing a “more intense” high than intranasal heroin or cocaine use is a particularly interesting finding deserving of additional research. It may be that experiencing injecting as not producing a more intense high has a strong biological component and is a selective factor for ceasing to inject and for avoiding relapse to injecting. It is also possible that their memories of injecting were evoked during the long time periods since last injection, and that interference in reconsolidation of these memories led to qualitative changes in the memories towards injecting being a less intense experience. Understanding how memories of injecting might change after ceasing to inject—or ceasing to use specific drugs—could provide important insights into avoiding relapse. 4.2 Limitations The major limitation is that the data come from cross-sectional surveys rather than a longitudinal study. Thus, we do not know how many former injectors relapsed back to injecting, or the factors that led them to relapse. We also do not know how many ceased injecting and then later ceased using heroin and cocaine. We also do not know when changes from “being in the process” of avoiding relapse changed to reaching a point where “relapse is no longer problematic.” In the associations between the reasons for ceasing to inject and current (past 6 month) non-injecting drug use, the reason for ceasing to inject would necessarily have preceded current drug use, but considerable caution is needed with respect to drawing causal inferences because of the many other factors that might have operated over the very long time periods involved. This study also does not permit estimating the number of former injectors in New York City. In a respondent driven sampling (RDS) study conducted in 2004 (Abdul-Quader et al., 2006), current injectors were permitted to recruit current non-injecting heroin and cocaine users and current non-injecting users were permitted to recruit current injectors. The two groups (current injectors and current non-injecting users) were equally likely to recruit members of the other group as they were to recruit members of their own group (RDS homophilies near 1.0). Half of the subjects recruited in that study were non-injecting drug users and half of the non-injecting users reported that they had previously injected heroin and/or cocaine. Thus, it would appear that about a quarter of street heroin and cocaine users in the city are former injectors, though this may be changing due to the new increases in prescription opioid users beginning to inject heroin. We did ask the former injectors why they had ceased injecting. From the content of the reasons for ceasing to inject (see Table 3), we would expect that the persons who had ceased injecting and had avoided relapse to injecting are likely to have accomplished these goals. We did not, however specifically determine if they had accomplished these goals. These limitations would need to be addressed in a cohort study, but note that this may require a very long-term study to capture the full experience of avoiding relapse to injecting over decades during which these former injectors avoided relapse back to injecting. 4.3 Implications for promoting long-term reductions in very harmful drug use behaviors These former injectors did not cease using heroin and cocaine but they did make a long-term change in their drug use behavior that almost certainly reduced important harmful effects of drug use for both themselves and for the community. This sustained change process appears to involve at least three factors: The majority set individual, achievable behavior change goals, with ceasing injection a means to these goals. Ceasing injection was not an externally imposed goal. Abstinence from heroin and cocaine use was not required, though it was not precluded as part of the long-term drug behavior change process. That important behavior change occurred while the former injectors continued to use psychoactive drugs supports the logic of using agonist drug therapies and not requiring abstinence as a precondition for other important behavior changes. They had a safety net of substance use treatment programs that could be utilized when they could no longer manage their non-injecting drug use. Note that all of these former injectors were interviewed on entry into substance use treatment programs, and all reported episodes of previous drug treatment. 4.4 Combined prevention and care for HIV, HCV, and overdose among PWID Implementing programs that support “reverse transitions” from injecting to non-injecting drug use might be an important part of combined prevention and care for HCV. HCV is much more efficiently transmitted than HIV through sharing of injection equipment and high rates of HCV infection have persisted among injection drug users in many places where HIV has been brought under control. PWID who reverse transition to non-injecting drug use should greatly reduce their chances of acquiring HIV and/or HCV. PWID who reverse transition after being infected with HIV or HCV should greatly reduce their chances of transmitting HIV or HCV to others. Additionally, HCV infected PWID who reverse transitioned and were successfully treated for HCV would greatly reduce their chances of re-infection with HCV. Programs to support “reverse transitions” back to non-injecting drug use might be particularly helpful in areas experiencing transitions from non-medical use of opioid analgesics to heroin injecting. 5. Conclusions The former injectors in this study have made an important change in their drug use behavior that would reduce adverse consequences at both the individual and societal level. For the majority, this change appears to have reached a point where relapsing back to injecting is no longer problematic for them. Understanding how long-term, self-sustaining changes occur in drug use behaviors is a fundamental question in research on problematic drug use. Additional research on former injectors may provide valuable insights into such change processes. We would suggest that research on cessation of injection among persons not ready or able to completely cease drug use be conducted with a framework of combined prevention and care for HIV/HCV among persons who use drugs. Acknowledgments We would like to thank the staff at Mount Sinai Beth Israel who collected the data and performed serologic testing for the participants included in the analysis. We would also like to thank all co-authors for their careful revisions and recommendations of the manuscript. Role of Funding Source This work was supported through grants R01DA003574, R01DA035707, and P30DA011041 from the US National Institute on Drug Abuse The funding agency had no role in the design, conduct, data analysis or report preparation for the study. Figure 1 Distribution of time since last injection (with Epanechnikov kernel density smoothing) Table 1 Demographics, current drug use characteristics, and HIV and HCV seroprevalence Mean SD Age 50 9.2 Years injected 12 12.2 N % Total 303 100 Gender     Female 56 18     Male 246 81 Race/ethnicity     White 51 17     African American 131 43     Latino/a 115 38 Current (past 6 month) drug use     Speedball (nasal) 34 11     Heroin (nasal) 214 71     Cocaine (nasal) 103 34     Crack cocaine 162 53 Previous drug treatment experience 303 100 Total with valid test results 258     HIV+ 34 13     HCV+ 170 66 Table 2 Current drug use by last drug injected among former injectors Last drug injected Current drug use Speedball (nasal) N (%) Heroin (nasal) N (%) Heroin (smoked) N (%) Cocaine N (%) Crack cocaine N (%) Speedball (N=47) 9 (19) 33 (70) 1 (2) 20 (43) 23 (49) Heroin (N=227) 20 (9) 170 (75) 13 (6) 68 (30) 114 (50) Cocaine (N=26) 4 (15) 10 (38) 2 (8) 13 (50) 22 (85) Table 3 Reasons reported for stopping injecting among former injectors Reason reported N (%) Total 303 (100) Health concerns (avoid AIDS, avoid overdose, other health concerns) 131 (44) Problems with injecting (don't like needles, tired of injecting, loss of veins, problems obtaining injecting equipment) 134 (45) Social reactions/self-image (stigmatization by others, wanting to preserve self-image as non-injector) 85 (28) Other reasons 66 (22) Table 4a Being “worried about relapsing to injecting” among former injectors Worried Yes N (%) No N (%) OR 95% CI Time since last injection ≤ 36 months 21 (25) 64 (75) 1.00 > 36 months 20 (9) 198 (91) 0.31 0.15 - 0.61 Table 4b Being “worried about relapsing to injecting” among former injectors Worried Yes N (%) No N (%) OR 95% CI Specific strategy to avoid relapse Yes 15 (37) 52 (20) 1.00 No 26 (63) 209 (80) 0.43 0.21 - 0.88 Table 5a Intensity of the high from heroin injection and “worried about relapsing to injecting” among former injectors Worried Yes N (%) No N (%) OR 95% CI More intense high from injection No 22 (16) 112 (84) 1.00 Yes 6 (8) 73 (92) 2.39 0.93 – 6.18 Table 5b Intensity of the high from cocaine injection and “worried about relapsing to injecting” among former injectors Worried Yes N (%) No N (%) OR 95% CI More intense high from injection No 4 (8) 45 (92) 1.00 Yes 13 (24) 41 (76) 3.57 1.08 – 11.82 Highlights Former PWID reported long term behavior changes, reducing individual and societal harms Many former PWID permanently transitioned back to non-injection drug use Over 50% of the sample was not worried about relapsing back to injecting drugs Multiple reasons for ceasing injection drug use were cited by former PWID Ceasing to inject provides considerable protection against HIV and HCV infection This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. References Abdul-Quader AS Heckathorn DD McKnight C Bramson H Nemeth C Sabin K Des Jarlais DC Effectiveness of respondent-driven sampling for recruiting drug users in New York City: findings from a pilot study. Journal of Urban Health 2006 83 3 459 476 16739048 Ahern J Stuber J Galea S Stigma, discrimination and the health of illicit drug users. Drug and Alcohol Dependence 2007 88 2 188 196 17118578 Binswanger IA Nowels C Corsi KF Glanz J Long J Booth RE Steiner JF Return to drug use and overdose after release from prison: a qualitative study of risk and protective factors. Addiction Science & Clinical Practice 2012 7 1 1 22966407 Buster MC Witteveen E Prins M van Ameijden EJ Schippers G Krol A Transitions in drug use in a new generation of problem drug users in Amsterdam: a 6-year follow-up study. European Addiction Research 2009 15 4 179 187 19622884 Degenhardt L Hall W Extent of illicit drug use and dependence, and their contribution to the global burden of disease. The Lancet 2012 379 9810 55 70 Dejong W Relapse prevention: an emerging technology for promoting long-term drug abstinence. International Journal of the Addictions 1994 29 6 681 705 8034380 Des Jarlais DC Arasteh K Perlis T Hagan H Heckathorn DD McKnight C Friedman SR The transition from injection to non-injection drug use: long-term outcomes among heroin and cocaine users in New York City. Addiction 2007 102 5 778 785 17506155 Des Jarlais DC Friedman SR Novick DM Sotheran JL Thomas P Yancovitz SR Maslansky R HIV-1 infection among intravenous drug users in Manhattan, New York City, from 1977 through 1987. JAMA 1989 261 7 1008 1012 2915408 Des Jarlais DC McKnight C Arasteh K Feelemyer J Perlman DC Hagan H Cooper HL Transitions from injecting to non-injecting drug use: Potential protection against HCV infection. Journal of Substance Abuse Treatment 2013 46 3 325 3 24161262 Des Jarlais DC Arasteh K Hagan H McKnight CM Perlman D Friedman SR Persistence and change in disparities in hiv infection among injection drug users in New York City after large-scale syringe exchange programs. American Journal of Public Health 2009 99 S2 S445 451 19797757 Evans JL Hahn JA Lum PJ Stein ES Page K Predictors of injection drug use cessation and relapse in a prospective cohort of young injection drug users in San Francisco, CA (UFO Study). Drug and Alcohol Dependence 2009 101 3 152 157 19181458 Fuller CM Vlahov D Ompad DC Shah N Arria A Strathdee SA High-risk behaviors associated with transition from illicit non-injection to injection drug use among adolescent and young adult drug users: a case-control study. Drug and Alcohol Dependence 2002 66 2 189 198 11906806 Garfein RS Vlahov D Galai N Doherty MC Nelson KE Viral infections in short-term injection drug users: the prevalence of the hepatitis C, hepatitis B, human immunodeficiency, and human T-lymphotropic viruses. American Journal of Public Health 1996 86 5 655 661 8629715 Genberg BL Gange SJ Go VF Celentano DD Kirk GD Mehta SH Trajectories of injection drug use over 20 years (1988-2008) in Baltimore, Maryland. American Journal of Epidemiology 2011 173 7 829 836 21320867 Inciardi JA Surratt HL Pechansky F Kessler F von Diemen L da Silva EM Martin SS Changing patterns of cocaine use and HIV risks in the south of Brazil. Journal of Psychoactive Drugs 2006 38 3 305 310 17165373 Kerr T Fairbairn N Tyndall M Marsh D Li K Montaner J Wood E Predictors of non-fatal overdose among a cohort of polysubstance-using injection drug users. Drug and Alcohol Dependence 2007 87 1 39 45 16959438 Li J Luo J Des Jaralis D Koram N Li P Liu H Role of sexual transmission of HIV among young non-injection and injection opiate users: a respondent driven sampling study. 2011 Paper presented at the American Public Health Association Washington, DC. Marlatt GA Donovan DM Relapse prevention: maintenance strategies in the treatment of addictive behaviors 2005 Guilford Press New York McKay JR Studies of factors in relapse to alcohol, drug and nicotine use: a critical review of methodologies and findings. Journal of Studies on Alcohol 1999 60 4 566 576 10463814 Onyeka IN Basnet S Beynon CM Tiihonen J Föhr J Kauhanen J Association between routes of drug administration and all-cause mortality among drug users. Journal of Substance Use 2016 00 00 1 7 Schottenfeld RS O'Malley S Abdul-Salaam K O'Connor PG Decline in intravenous drug use among treatment-seeking opiate users. Journal of Substance Abuse Treatment 1993 10 1 5 10 8450573 Schwartz RP Kelly SM Gryczynski J Mitchell SG O'Grady KE Jaffe JH Heroin use, HIV-Risk, and criminal behavior in Baltimore: findings from clinical research. Journal of Addictive Diseases 2015 34 2-3 151 161 26079104 Shah NG Galai N Celentano DD Vlahov D Strathdee SA Longitudinal predictors of injection cessation and subsequent relapse among a cohort of injection drug users in Baltimore, MD, 1988–2000. Drug and Alcohol Dependence 2006 83 2 147 156 16364568 Spijkerman IJ van Ameijden EJ Mientjes GH Coutinho RA van den Hoek A Human immunodeficiency virus infection and other risk factors for skin abscesses and endocarditis among injection drug users. Journal of Clinical Epidemiology 1996 49 10 1149 1154 8826995 Tejani E Chawarski M Mazlan M Schottenfeld RS Temporal changes in initiation of injection use in heroin users in Malaysia 1968 to 2009. 18-23 6 2011 Paper presented at the College on Problems of Drug Dependence Hollywood, FL. Witkiewitz K Marlatt GA Relapse prevention for alcohol and drug problems: that was Zen, this is Tao. American Psychologist 2004 59 4 224 15149263

PMC005xxxxxx/PMC5117632.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101231976 32624 Nat Chem Biol Nat. Chem. Biol. Nature chemical biology 1552-4450 1552-4469 27748750 5117632 10.1038/nchembio.2207 NIHMS810261 Article Discovery of MRSA active antibiotics using primary sequence from the human microbiome Chu John 1‡ Vila-Farres Xavier 1‡ Inoyama Daigo 3 Ternei Melinda 1 Cohen Louis J. 1 Gordon Emma A. 1 Reddy Boojala Vijay B. 1 Charlop-Powers Zachary 1 Zebroski Henry A. 2 Gallardo-Macias Ricardo 3 Jaskowski Mark 3 Satish Shruthi 3 Park Steven 4 Perlin David S. 4 Freundlich Joel S. 3 Brady Sean F. 1* 1 Laboratory of Genetically Encoded Small Molecules, The Rockefeller University, New York, New York, USA 2 Proteomics Resource Center, The Rockefeller University, New York, New York, USA 3 Department of Pharmacology, Physiology, and Neuroscience, Rutgers University – New Jersey Medical School, Newark, New Jersey, USA 4 Public Health Research Institute, Rutgers University – New Jersey Medical School, Newark, New Jersey, USA * Corresponding author: sbrady@rockefeller.edu ‡ These authors contributed equally to this work. 23 8 2016 17 10 2016 12 2016 17 4 2017 12 12 10041006 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Here, we present a natural product discovery approach whereby structures are bioinformatically predicted from primary sequence and produced by chemical synthesis (synthetic-bioinformatic natural products, syn-BNPs), circumventing the need for bacterial culture and gene expression. When applied to nonribosomal peptide synthetase gene clusters from human-associated bacteria we identified the humimycins. These antibiotics inhibit lipid II flippase and potentiate β-lactam activity against methicillin-resistant Staphylococcus aureus in mice, potentially providing a new treatment regimen. The characterization of small molecules produced by bacteria in laboratory culture has been a key step to understanding bacterial physiology and developing small molecule therapeutics.1 As successful as this approach has been for identifying novel bioactive small molecules, extensive sequencing of bacterial genomes and metagenomes has revealed that the bacterial biosynthetic diversity traditionally accessed in the laboratory represents only a small fraction of what is predicted to exist in nature.2,3 This shortcoming arises from our inability to culture most bacteria in the laboratory and from the fact that most biosynthetic gene clusters remain silent under laboratory fermentation conditions.4 Here, we present a bioactive small molecule discovery pipeline that circumvents the requirement for either bacterial culture or gene cluster expression. In our approach, natural product structures are bioinformatically predicted from primary sequence data and produced by chemical synthesis. Since natural products often appear in nature as families of related structures with the same biological activity, we reasoned that even if our structural predictions were not perfect, many syn-BNPs would be sufficiently accurate representations of nature to elicit the intended bioactivities. We have called these bioinformatically inspired compounds syn-BNPs for Synthetic Bioinformatic Natural Products (Fig. 1a). The human microbiome is an exemplary test case for a syn-BNP discovery approach. Tremendous resources have been allocated to the sequencing and bioinformatic analysis of the human microbiome.5,6 Nevertheless, the functional characterization of this data, including commensal bacteria-encoded natural product biosynthetic gene clusters, remains rare. Our interest in exploring syn-BNPs encoded by the human microbiota stems from the potential use of these metabolites as therapeutics and as tools for improving our understanding of human microbiome functions. Because antibiotics can serve as medicines and as modulators of the composition of the human microbiome, we chose to screen syn-BNPs predicted from the human microbiome for antibacterial activity against human associated commensal and pathogenic bacteria. Systematic bioinformatic analysis of sequenced bacterial genomes indicate that nonribosomal peptides (NRPs) are one of the most common and diverse families of complex secondary metabolites produced by bacteria.7,8 Over the past two decades, a number of models have been developed for predicting the identity, order, and modification of the amino acids comprising an NRP, based solely on the primary sequence of NRP megasynthetases.9–12 Concurrently, solid phase peptide synthesis (SPPS) of structurally diverse peptides has become rapid and economical, making NRP gene clusters an ideal test case for a syn-BNP approach (Fig. 1b). Genomic sequence data from human (commensal and pathogenic) associated bacteria were bioinformatically queried for gene clusters predicted to encode large NRPs (≥5 residues), as short NRPs are often highly modified and are therefore not easily accessible using SPPS alone. This analysis led to the identification of 57 unique nonribosomal peptide synthetase (NRPS) gene clusters, from which we removed those that appeared to be incomplete in the existing sequence data and those containing more than one PKS module, a thioreductase domain, or any heterocyclization domains. The chemical outputs of the remaining 25 gene clusters, which we believed to be amenable to SPPS, were predicted using three published NRPS prediction algorithms (Stachelhaus, Minowa, and NRPSPredictor2) to produce syn-BNP targets.9–12 In instances where bioinformatic predictions for human microbiome associated gene clusters diverged strongly between algorithms (e.g., side-chains were predicted to carry opposite charges), multiple syn-BNP peptides were designed and synthesized. In all cases where NRPS gene clusters were bioinformatically predicted to encode an N-terminally acylated peptide, we elected to design the syn-BNP to be N-acylated with β-hydroxymyristic acid (HMA), a fatty acid commonly observed in NRPs.13 In total, 30 syn-BNPs targets were designed based on the gene clusters found in human commensal bacterial sequence data. After two rounds of SPPS using standard Fmoc chemistry, we obtained pure samples for 25 of the 30 targeted syn-BNPs (Supplementary Results, Supplementary Table 1). To identify novel antibiotic scaffolds with potential in vivo roles in shaping the ecology of the human microbiome, we assayed this collection of syn-BNPs for antibacterial activity against a panel of common human commensal and pathogenic bacteria. This led to the identification of two antibiotics we have trivially named humimycin A (1) and B (2) (human microbiome mycin, Fig. 2a). The humimycins were predicted from closely related NRPS gene clusters found in the genomes of Rhodococcus equi and Rhodococcus erythropolis, respectively. Bioinformatic analyses of these two NRPS gene clusters indicated that they encoded hepta-peptides that differed at only the fourth and sixth residues (F/Y and V/I, respectively, Fig. 2b). Both syn-BNPs were synthesized with N-terminal HMA modifications, due to the presence of starter condensation domains (Cs) in the gene clusters, which are associated with acylation of the first amino acid of an NRP.13 Rhodococcus species have been extensively studied for natural product production using traditional fermentation-based discovery methods. None of these studies report the identification of a metabolite resembling the humimycins.14 Likewise, our extensive analysis of Rhodococcus species culture broth extracts by both LCUV and LCMS analysis did not reveal any metabolites related to the humimycins, suggesting that the humimycin gene cluster is silent under laboratory fermentation conditions. The humimycins were found to be broadly active against Firmicutes and to show some activity against Actinobacteria, when screened for antibiosis against commensal and pathogenic bacteria (Fig. 2c). The humimycins are particularly active against Staphylococcus and Streptococcus species, including common members of the normal human flora such as S. aureus (minimum inhibitory concentration (MIC) 8 μg/mL) and S. pneumoniae (MIC 4 μg/mL). This spectrum of activity is interesting in light of the fact that Firmicutes and Actinobacteria dominate the human microbiota of the gut (Fig. 2d).15 In a structure-activity relationship study, we found that no residue in 1 could be replaced with alanine without dramatically impacting the potency of the antibiotic (Supplementary Table 3). Humimycin A exhibits MICs ranging from 8–128 μg/mL against methicillin-resistant S. aureus (MRSA) clinical isolates (Supplementary Table 4). To study the antibacterial mode of action of the humimycins, we selected S. aureus USA300 mutants that could survive on 2.5 times the MIC (20 μg/mL) and sequenced the genomes of 23 resistant mutants. Upon comparison to the parent strain, we found that all 23 mutants contained one non-synonymous mutation in SAV1754, an essential gene in S. aureus (Supplementary Figure 3 and Table 5). Fifteen of these strains contained no other detectable mutations. Overexpression of SAV1754 in S. aureus confers resistance to humimycin A (MIC >128 μg/mL), further supporting inhibition of SAV1754 as a likely mode of action of the humimycins (Supplementary Table 6). The gene product of SAV1754 is believed to be a homolog of MurJ, a flippase responsible for the translocation of peptidoglycan precursors from the inside to the outside of the cell.16 While MurJ is essential in many bacteria, including many important pathogens, it remains an underexplored antibacterial target.17 The ability of SAV1754 inhibitors to potentiate β-lactam antibiosis is thought to arise from the fact that both antibiotics target the same essential pathway, peptidoglycan biosynthesis (Fig. 3a). In a high throughput screen for molecules that could potentiate β-lactam antibiosis against otherwise resistant strains Merck & Co. identified synthetic small molecule inhibitors of SAV1754.18,19 Humimycin A exhibits a similar ability to restore β-lactam sensitivity to β-lactam resistant bacteria. For example, the MIC of carbenicillin (carboxypenicillin) was reduced from 32 to 1 μg/mL in the presence of 2 μg/mL humimycin A (0.25× MIC) against MRSA USA300 (Fig. 3b), one of the most predominant community-associated MRSA strains in the U.S. Humimycin A’s ability to potentiate β-lactam activity is also seen with strains where it alone shows no detectable antibacterial activity. For example, the MRSA COL strain, while not susceptible to humimycin A (MIC >512 μg/ml) and exhibiting a very high MIC for the β-lactam dicloxacillin (MIC 256 μg/mL), is sensitive to dicloxacillin at 8 μg/mL in the presence of as little as 4 μg/mL of humimycin A (Fig. 3c). The ability of humimycin A to potentiate β-lactam activity in vitro led us to explore the possibility that it might do the same in vivo. In murine tolerability studies humimycin A is tolerated at concentrations (>50 mg/kg), far exceeding those expected to be necessary for β-lactam potentiation. In a murine peritonitis-sepsis model treatment of a MRSA COL infection with dicloxacillin and humimycin together dramatically increases survival compared to treatment with either humimycin or dicloxacillin alone (Fig. 3d), potentially providing a novel MRSA treatment regimen. While R. equi has historically been regarded as an opportunistic pathogen seen in animals and immune-compromised patients,20 R. erythropolis is found as a part of the normal human nasal, mouth and eye microbiota.21,22 Interestingly, the occurrence of Rhodococcus species in the gut increases dramatically to a median of 30% in some patients diagnosed with ulcerative colitis (UC).23 The production of an antibiotic with activity against Firmicutes and Actinobacteria could play a role in establishing the overpopulation of R. erythropolis in the UC gut as Firmicutes and Actinobacteria normally represent nearly half of the gut microbiota.15 In addition to the potential to provide new small molecule therapeutics, characterization of molecules inspired by commensal bacteria biosynthetic gene clusters can provide a means for developing hypotheses about how commensal bacteria affect human physiology. For example, the discovery of the humimycins provides a testable mechanistic hypothesis for how dysbiosis of gut microbiota might evolve in UC.24 Our identification of the humimycins using a syn-BNP approach validates this as a strategy for identifying bioactive metabolites and highlights the unique state of the field of natural product chemistry today. Extensive biosynthetic studies have culminated in our emerging ability to predict the structures of many natural products from primary sequence alone. While in this study we focused on the synthesis of linear peptides because of the ease with which they can be generated by SPPS, there are many ways to expand this approach to more topologically and functional complex NRPs. For example, the construction of cyclic peptides using purified thioesterase domains is compatible with SPPS.25 Based on our analysis of high-quality sequenced bacterial genomes in GenBank not associated with the human microbiome, there are currently more than 1,500 unique large NRPS gene clusters (encoding ≥5 amino acids) amenable to a SPPS-based syn-BNP approach. As the sequencing of microbial genomes is still in an early exponential growth phase, this number should only continue to grow for the foreseeable future (Supplementary Figure 2). With the development of improved bioinformatic prediction algorithms for biosynthetic gene cluster families beyond NRPSs and the incorporation of more sophisticated chemical and chemo-enzymatic synthesis steps into the production of syn-BNPs, we believe this approach will enable broad and rapid access to diverse bioactive compounds that are inspired by gene clusters found within the ever-growing assemblage of microbial sequence data. Online Methods Bioinformatic prediction of NRPs Genome sequences of the human microbiota were downloaded from the NIH Human Microbiome Project (HMP, (ftp://ftp.ncbi.nlm.nih.gov/genomes/HUMAN_MICROBIOM/Bacteria)26 and the Human Oral Microbiome Database (HOMD, (ftp://ftp.homd.org/HOMD_annotated_genomes).27 The software package Antibiotics and Secondary Metabolite Analysis Shell (antiSMASH) v2.0 was used for the identification and prediction of NRP biosynthetic gene clusters encoded by these genomes.28 Syn-NRPs originating from the HMP and HOMD databases were named serially as [Human.N] and [Oral.N]. All syn-NRPs discussed in this manuscript are listed in Supplementary Table 1. AntiSMASH consults three prediction algorithms to call the amino acid substrate specificity of an adenylation domain (NRPSPredictor2, Stachelhaus code, and Minowa). A consensus prediction refers to the situation wherein two (or all three) algorithms make consistent substrate predictions for a given adenylation domain. In this case the predicted amino acid was used in the synthesis of the syn-BNP. In case of a minor conflict between prediction algorithms we opted for the amino acid with the smaller side-chain, e.g., Val/Leu/Ile and Ser/Thr. In case of major conflicts (e.g., where side-chains were predicted to carry opposite charges), both peptides were synthesized. Tyrosine and phenylalanine prediction made by NRPSPredictor2 or Stachelhaus code were chosen over tryptophan (Trp) predictions made Minowa, as we noticed that Trp is overrepresented in Minowa predictions. Lastly, tyrosine (Tyr) was used at the first residue in place of p-hydroxyphenylglycine (Hpg) in Human.8v1 and v2. To check the robustness of these NRPS prediction algorithms we carried out a similar analysis of NRPS gene clusters deposited in the MiBIG database and found that the core peptide encoded by the vast majority NRPS gene clusters was predicted correctly (Supplementary Figure 2). Peptide synthesis Resins for peptide synthesis were purchased from AnaSpec. Coupling reagents (PyBOP) and Nα-Fmoc/side-chain protected amino acids were purchased from P3BioSystems. 3-Hydroxymyristic acids were purchased from TCI America (racemic mixture) and Santa Cruz Biotechnology (pure enantiomers). All other chemical reagents and solvents were purchased from Sigma Aldrich. Reaction vessels were custom made by the Scientific Glassblowing Laboratory at the Department of Chemistry of Yale University. Pure samples were obtained for 25 of the 30 syn-BNP peptides targeted for chemical synthesis (Supplementary Table 1). 20 of these peptides were purchased through the custom peptide synthesis service of GenScript Biotech Corporation and five were synthesized in-house. Peptides from GenScript were delivered as lyophilized materials that had been HPLC-purified and MS-verified (MALDI). All pure peptides were dissolved in DMSO at 12.8 mg/mL as stock solutions and stored at −20 °C. In-house peptide syntheses, including humimycin A and B, were built on Wang resin29 following standard Fmoc/tBu SPPS methods. The first amino acid (6 equiv.) was activated using DIC (3 equiv.) in 10% DMF/DCM (0°C), added to the resins in the presence of DMAP as a catalyst (0.1 equiv.) and shaken under nitrogen (4 h at 0°C). Unreacted resins were capped using acetic anhydride in pyridine (1 h). Fmoc removal was accomplished using three rounds of treatment with 20% piperidine in DMF (15, 10, and 5 min. each). All ensuing amino acids were coupled twice. In each coupling an Nα-Fmoc and side-chain protected amino acid was activated using a mixture of PyBOP (4 equiv.) and DIEA (8 equiv.), followed by reaction with the peptide on-resin (1 h). Peptides were cleaved by 95% TFA supplemented with TIS and H2O (2.5% of each, v/v) for 2 h, concentrated to approximately 10% of the original volume, diluted with aqueous MeCN (75%, v/v), passed through a 0.45 μm filter and HPLC-purified. All purified peptides were examined by LC/MS (ESI). Characterization of the humimycins A racemic mixture of 3-hydroxymyristic acid was used for N-terminal modification in our initial syntheses of all syn-BNPs. Humimycin A diastereomers showed different MIC values when tested against MRSA USA300 (Supplementary Table 3). The absolute stereochemistry of the most active diastereomer was determined by comparing HPLC-purified peptides from the bulk synthesis to independent batches of small-scale syntheses using enantiopure (R) and (S)-3-hydroxymyristic acid (Supplementary Figure 1). The more potent (S)-isomer is referred to as humimycin A (1). In the case of humimycin B the analogous (S)-isomer was purified and is referred to as compound 2. Humimycin A (1) HRMS: m/z calculated for [M − H]− (C58H84N7O14): 1102.6076, found: 1102.6075. Humimycin B (2) HRMS: m/z calculated for [M − H]− (C59H86N7O15): 1132.6182, found: 1132.6194. Syn-BNP screening Syn-BNPs were screened against a panel of commensal and pathogenic bacteria covering the four major phyla associated with the human microbiome. This included five Actinobacteria, four Bacteroidetes, six Firmicutes and three Proteobacteria species. All peptides were tested in duplicate for antibiosis activity. Assays were performed in microtiter plates, wherein each well contained growth media (see Supplementary Table 2 for a list of growth media) (100 μL), syn-BNP (32 μg/mL) and bacteria diluted 1,000-fold from a stationary phase culture. Binary antibiosis results for most bacteria were determined by visual inspection after static incubation at 37 °C for 18 h. P. melaninogenica and Eubacterium sp. 3_1_31 were grown for 36 h, and C. amycolactum was grown for 60 h. Specific MICs were determined for syn-BNPs that inhibited bacterial growth in this initial screen (see Susceptibility assays, part a). Bacteria species associated with the human flora were obtained from BEI Resources. Susceptibility assays a) Standard assays MIC assays were performed in duplicate in 96-well microtiter plates based on the protocol recommended by Clinical and Laboratory Standards Institute.30 DMSO stock solutions of syn-BNPs (12.8 mg/mL) were added to the first well in a row and serially diluted (2 fold per transfer) across the microtiter plate. The last well was reserved for a peptide-free control. Overnight cultures of bacteria were diluted 5,000-fold and 50 μL was used as an inoculum in each well. MIC values were determined by visual inspection after 18 h incubation (37 °C, static growth). b) Synergy assays Synergistic β-lactam-humimycin activities were assessed through a two-dimensional (2D) susceptibility assay. Two fold serial dilutions were carried out as described above. Carbenicillin (a β-lactam antibiotic) was diluted serially from left to right, and humimycin was diluted serially from top to bottom. The highest concentration tested for both antibiotics was 32 μg/mL. Fractional inhibitory concentration (FIC) is defined as the ratio of the apparent synergistic MIC divided by the MIC of the antibiotic measured alone.29 Selection of humimycin A resistant mutants A single S. aureus USA300 colony (the parent) from a freshly struck plate was inoculated into LB medium and grown overnight at 37 °C. Part of the overnight culture (4 mL) was spun down and kept frozen at −20 °C. The rest of the overnight culture was diluted 100-fold, supplemented with humimycin A at 20 μg/mL (2.5X MIC) and 100 μL aliquots was distributed into 200 unique microtiter plate wells. Growth was observed in 50 wells after overnight incubation, indicating the presence of bacteria with mutation(s) conferring humimycin A resistance. Approximately 2 μL of culture from each of these wells was used to inoculate freshly prepared 100 μL aliquots of LB media supplemented with humimycin A (20 μg/mL). The resulting cultures after overnight incubation were struck out for single colonies on LB/agar plates supplemented with humimycin A (20 μg/mL) for single colonies. Genome sequencing Single colonies of 23 humimycin A resistant mutants as well as the USA300 parent were individually inoculated into 4 mL of LB media free of any antibiotics. After overnight incubation cells were collected by centrifugation. DNA extractions were performed using a MasterPure Purification Kit (EpiCentre Biotechnologies). Multiplex sequencing libraries were prepared from the resulting genomic DNA using a Nextera XT DNA Sample Preparation Kit (FC-131-1024) with Nextera XT Index kit (FC-131-1001) based on protocols provided by the manufacturer (Illumina). Briefly, the genomic DNA was treated with RNase and quantified using the Qubit dsDNA HS Assay System (Q32854, ThermoFisher Scientific). Tagmentation and PCR amplification proceeded according to the manufacturer’s protocol, after which the quality and size of the libraries were verified using HS D1000 ScreenTape (TapeStation 2200, Agilent Technologies). Libraries were pooled at equimolar concentrations and column purified by NucleoSpin Gel and PCR Cleanup (MN-750609-250, Macherey-Nagel). The resulting tagged DNA library was size-selected by E-Gel (Life Technologies) and the 450 bp band was excised. The final library pool was checked for molarity on TapeStation and sequenced using MiSeq Reagent Kit v3 (MS-102-3003, Illumina). Mutation (SNP) identification De-barcoded MiSeq reads were assessed for mutations by comparing each read against the reference genome of Staphylococcus aureus USA300_FPR3757 (RefSeq assembly accession: GCF_000013465.1). All reads were mapped to the reference genome using SNIPPY (https://github.com/tseemann/snippy) for the identification of variants. SNIPPY is a wrapper of several programs including freebayes (https://github.com/ekg/freebayes).32 Single-nucleotide polymorphisms (SNP) observed in the parent strain were then subtracted from those observed in the humimycin A resistant strains, resulting in a final list of SNPs (Supplementary Table 5). Cloning and overexpression of SAV1754 The SAV1754 gene was PCR amplified from wild type S. aureus USA300 and a mutant resistant to humimycin A (mutant no. 8, Supplementary Table 5). PCR products and the pRMC233 vector were digested (SacI/KpnI) and ligated, followed by transformation into S. aureus RN4220 and selection on BHI agar plates containing chloramphenicol (10 μg/mL). The recombinant plasmids were verified by DNA sequencing. Overnight cultures of the resulting S. aureus strains were used to inoculate LB containing chloramphenicol (10 μg/mL). Late log-phase cultures (OD600 ~0.8) were induced by anhydrotetracycline (50 ng/mL) for 5 h and then tested in the presence of anhydrotetracycline (5 ng/mL) for susceptibilities against humimycin A. Primer sequences, PCR conditions, and susceptibility data are listed in Supplementary Table 6. Murine peritonitis-sepsis model Female outbred Swiss Webster mice were used for this study. MRSA COL was grown in Mueller-Hinton broth at 37°C overnight and diluted with 5% hog mucin and 0.9% NaCl to provide challenge inoculum of approximately 5 × 108 CFU per mouse in a volume of 0.5 mL via intraperitoneal injection. Forty mice were randomly grouped into 10 per cohort, and each group was given single doses of vehicle (20% DMA, 40% PEG, 40% D5W), humimycin (HM) at 50 mg/kg, dicloxacillin (DCX) at 125 mg/kg, and HM:DCX combination at 12.5 mg/kg HM:125 mg/kg DCX 1 h post-infection via IV injection. Mice were maintained in accordance with American Association for Accreditation of Laboratory Care criteria. The Rutgers University Institutional Animal Care and Use Committee approved all animal procedures. Supplementary Material 1 We thank the Fischetti (MRSA), Tomasz (MRSA) and Marraffini (S. aureus, S. delphini, S. intermedius, and S. pseudo-intermedius) laboratories at the Rockefeller University for providing strains. This work was support by the Rainin Foundation, NIH grants U19AI109713 (D.S.P) and F32 29 AI110029 (Z.C.P.). Figure 1 Overview of the Syn-BNP approach a) Advances in our understanding of natural product biosynthesis have enabled the prediction of natural product structures from primary sequence data alone. In a syn-BNP approach these structures are accessed through chemical synthesis instead of biosynthesis. b) Here we apply a syn-BNP approach to NRPs predicted from human microbiome sequence data and assay these new molecules for antibiosis activities. Figure 2 Discovery and screening of the humimycins a) The humimycins were predicted from closely related gene clusters found in two Rhodococcus spp. cultured from human subjects. b) Chemical structures of humimycin A (1) and B (2). The two antibiotics differ only at the fourth (F/Y) and sixth (V/I) residues. c) MIC values for the humimycins against a panel of human commensal and pathogenic bacteria were determined (n = 2). The right panel shows that the humimycins are particularly active against bacteria in the Staphylococcus and Streptococcus genus (n = 3). Figure 3 Humimycin A and β-lactam act in synergy a) SAV1754 is the S. aureus homolog of MurJ, which is a flippase responsible for the transportation of peptidoglycan precursors across the cytoplasmic membrane. b) Carbenicillin (C) and humimycin A (HM) act synergistically to inhibit the growth of MRSA USA300 (n = 2). Fraction inhibitory concentration (FIC) values ≤0.5 defines synergy between two agents (shaded in light gray); [C:HM] denotes the respective inhibitory concentrations at each data point (μg/mL). c) The minimum inhibitory concentration (MIC) of humimycin A with dicloxacillin (DCX) alone and at various humimycin A concentrations against MRSA COL (n = 2) are shown in red and purple, respectively. Humimycin A alone does not inhibit MRSA COL growth (MIC >512 μg/mL, blue). d) Survival data for mice treated with humimycin A or dicloxacillin either alone or together using a MRSA COL peritonitis model (n = 10 mice per cohort) are shown. In this model humimycin potentiates β-lactam activity in vivo. Author Contributions SFB conceived of the project. JC and XVF carried out antibiosis assays, spectrum of activity screening, and resistant mutant selection. DI, HAZ, RGM, MJ, SS, and JSF carried out peptide synthesis on large scale. MAT carried out genome sequencing. LJC and EAG screened anaerobic bacteria. BVBR and ZCP carried out bioinformatic analysis. SP and DSP carried out mice studies. Financial Interest Statement The authors declare no competing financial interests. 1 Newman DJ Cragg GM J Nat Prod 79 629 661 2016 26852623 2 Charlop-Powers Z Milshteyn A Brady SF Curr Opin Microbiol 19 70 75 2014 25000402 3 Piel J Annu Rev Microbiol 65 431 453 2011 21682647 4 Rutledge PJ Challis GL Nat Rev Microbiol 13 509 523 2015 26119570 5 Qin J Nature 464 59 65 2010 20203603 6 Turnbaugh PJ Nature 457 480 484 2009 19043404 7 Charlop-Powers Z eLife 4 2015 8 Doroghazi JR Nat Chem Biol 10 963 968 2014 25262415 9 Stachelhaus T Mootz HD Marahiel MA Chem Biol 6 493 505 1999 10421756 10 Minowa Y Araki M Kanehisa M J Mol Biol 368 1500 1517 2007 17400247 11 Rottig M Nucleic Acids Res 39 W362 367 2011 21558170 12 Weber T Nucleic Acids Res 43 W237 243 2015 25948579 13 Rausch C Hoof I Weber T Wohlleben W Huson DH BMC Evol Biol 7 78 2007 17506888 14 Kitagawa W Tamura T Microbes Environ 23 167 171 2008 21558704 15 D’Argenio V Salvatore F Clin Chim Acta 451 97 102 2015 25584460 16 Sham LT Science 345 220 222 2014 25013077 17 Sewell EW Brown ED J Antibiot 67 43 51 2014 24169797 18 Lee SH Sci Transl Med 8 329ra332 2016 19 Huber J Chem Biol 16 837 848 2009 19716474 20 Kraal L Abubucker S Kota K Fischbach MA Mitreva M PLoS One 9 e97279 2014 24827833 21 Rasmussen TT Kirkeby LP Poulsen K Reinholdt J Kilian M APMIS 108 663 675 2000 11200821 22 Graham JE Invest Ophthalmol Vis Sci 48 5616 5623 2007 18055811 23 Lepage P Gastroenterology 141 227 236 2011 21621540 24 Jostins L Nature 491 119 124 2012 23128233 25 Kohli RM Walsh CT Burkart MD Nature 418 658 661 2002 12167866 26 Human Microbiome Project Consortium Nature 2012 486 215 22699610 27 Chen T Database (Oxford) 2010 baq013 20624719 28 Blin K Nucleic Acids Res 2013 41 W204 23737449 29 Wang SS J Am Chem Soc 1973 95 1328 4687686 30 Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard 9 Clinical and Laboratory Standards Institute Wayne, PA 2012 31 Hall MJ Middleton RF Westmacott D J Antimicrob Chemother 1983 11 427 6874629 32 Garrison E Marth G arXiv 2012 33 Corrigan RM Foster TJ Plasmid 2009 61 126 18996145

PMC005xxxxxx/PMC5117677.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 1300217 533 Anesthesiology Anesthesiology Anesthesiology 0003-3022 1528-1175 27753644 5117677 10.1097/ALN.0000000000001390 NIHMS817098 Article Tryptophan and Cysteine Mutations in M1 Helices of α1β3γ2L γ-Aminobutyric Acid Type A Receptors Indicate Distinct Inter-subunit Sites for Four Intravenous Anesthetics and One Orphan Site Nourmahnad Anahita B.S. 1* Stern Alex T B.S. 1*a Hotta Mayo B.S. 1a Stewart Deirdre S. Ph.D. 1 Ziemba Alexis M. B.A. 1b Szabo Andrea B.S. 1 Forman Stuart A. M.D., Ph.D. 1 1 Dept. of Anesthesia Critical Care & Pain Medicine, Massachusetts General Hospital, Boston, MA, USA. Corresponding author: Stuart A. Forman, MD-PhD, Dept. of Anesthesia Critical Care & Pain Medicine, Jackson 444, Massachusetts General Hospital, 55 Fruit Street, Boston, MA, USA. saforman@partners.org; Phone: 617-724-5156; Fax: 617-724-8644 * These authors contributed equally to this research. a Current affiliation: University of Southern California, Keck School of Medicine, Los Angeles, CA, USA. b Current affiliation: Dept. of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA. 17 9 2016 12 2016 01 12 2017 125 6 11441158 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background GABAA receptors mediate important effects of intravenous general anesthetics. Photolabel derivatives of etomidate, propofol, barbiturates, and a neurosteroid incorporate in GABAA receptor transmembrane helices M1 and M3 adjacent to inter-subunit pockets. However, photolabels have not been consistently targeted at heteromeric αβγ receptors and don’t form adducts with all contact residues. Complementary approaches may further define anesthetic sites in typical GABAA receptors. Methods Two mutation-based strategies, substituted tryptophan sensitivity and substituted cysteine modification-protection, combined with voltage-clamp electrophysiology in Xenopus oocytes, were used to evaluate interactions between four intravenous anesthetics and six amino acids in M1 helices of α1, β3, and γ2L GABAA receptor subunits: two photolabeled residues, α1M236 and β3M227, and their homologs. Results Tryptophan substitutions at α1M236 and positional homologs β3L231 and γ2L246 all caused spontaneous channel gating and reduced GABA EC50. Substituted cysteine modification experiments indicated etomidate protection at α1L232C and α1M236C, mTFD-MPAB protection at β3M227C and β3L231C, and propofol protection at α1M236C and β3M227C. No alphaxalone protection was evident at the residues we explored and none of the tested anesthetics protected γ2I242C or γ2L246C. Conclusions All five inter-subunit transmembrane pockets of GABAA receptors display similar allosteric linkage to ion channel gating. Substituted cysteine modification and protection results were fully concordant with anesthetic photolabeling at α1M236 and β3M227, and reveal overlapping non-congruent sites for etomidate and propofol in β+ – α− interfaces and mTFD-MPAB and propofol in α+ – β− and γ+ – β− interfaces. Our results identify the α+ – γ− transmembrane interface as a potentially unique “orphan” modulator site. Ligand-gated ion channel glycine receptor GluCl alcohol general anesthetic cysteine modification photolabel electrophysiology allosteric modulator allosteric agonist Introduction Etomidate, propofol, alphaxalone, and barbiturates are intravenous general anesthetics that enhance the activity of γ-aminobutyric acid type A (GABAA) receptors, the dominant inhibitory neurotransmitter receptors in mammalian brain and members of the pentameric ligand-gated ion channel (pLGIC) superfamily 1–4. GABAA receptor subunits contain structural elements common to all pLGICs, including an N-terminal extracellular domain and a transmembrane domain with four alpha helices (M1 to M4). Most GABAA receptors consist of two a, two b, and one g subunit arranged βαβαγ counterclockwise 5, creating four distinct types of subunit interfaces: α+ – β−, α+ – γ−, β+ – γ−, and two β+ – α− (Fig 1). Anesthetic photolabels form adducts with GABAA receptor residues that are homologs of ivermectin contacts in transmembrane inter-subunit pockets of GluCl pLGICs, imaged with x-ray crystallography 6. Azi-etomidate labeled αM236 in α-M1 and βM286 in β-M3 in GABAA receptors from bovine brain or cells expressing α1β3γ2L 7,8. The barbiturate R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid (mTFD-MPAB), photolabeled residues in b3-M1 (β3M227), α1-M3, and γ2-M3 8. m-Azi-propofol (azi-Pm) photolabeled both α1M236 and β3M227 in α1β3 receptors 9. 6-Azi-pregnanalone (6-aziP), a neurosteroid derivative, photolabeled b3F301 in β3 homomeric receptors 10. Propofol, but not alphaxalone, displaced both azi-etomidate and mTFD-MPAB labeling in α1β3γ2L receptors 8,11,12. These results suggest that α1β3γ2L receptors form overlapping etomidate and propofol sites in β3+ – α1− interfaces adjacent to α1-M1s, and overlapping mTFD-MPAB and propofol sites in homologous α1+ – β3− and γ2+ – β3− pockets abutting β3-M1s (Fig 1). However, we don’t know how propofol sites overlap with vs. differ from etomidate and mTFD-MPAB sites, because photolabels don’t adduct all residues that contact parent drugs 13,14. Additionally, no anesthetics have photolabeled γ2-M1, and because receptors photolabeled by aziPm and 6-aziP lacked γ2 subunits, propofol or neurosteroid interactions with γ2 remain untested. To both complement and supplement photolabeling data, two approaches based on voltage-clamp electrophysiological studies of GABAA receptors with mutations at putative anesthetic contact residues have been widely applied. Tryptophan substitutions in putative anesthetic sites have been reported to both mimic the channel-gating effects of anesthetic binding and impair anesthetic modulation 15–19. Cysteines substituted at putative contact residues have been covalently modified by probes such as p-chloromercuricbenzenesulfonate (pCMBS) and protected from modification by anesthetics 14,20–24. However, neither of these approaches has been rigorously compared to photolabeling. In the current study, we created tryptophan and cysteine mutations at two photolabeled residues in M1 helices, α1M236 and β3M227, and their homologs based on sequence alignment: α1L232, β3L231, γ2I242 and γ2L246 (Table 1). At each locus, we assessed tryptophan mutant functions and performed substituted cysteine modification-protection (SCAMP) experiments to probe interactions with etomidate, propofol, alphaxalone, and mTFD-MPAB. Our analysis addressed three key questions: 1) Do SCAMP results and/or tryptophan mutant sensitivity results for the study drugs agree with photolabeling at α1M236 and β3M227? 2) Do the α1-M1 contacts for etomidate and propofol or the β3-M1 contacts for mTFD-MPAB and propofol differ? 3) Do alphaxalone, propofol, or other anesthetics bind in the α1+ – γ2− interface? Materials and Methods Animals Female Xenopus frogs were used as a source of oocytes in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Approval for animal use in this study was obtained from the Massachusetts General Hospital Institutional Animal Care and Use Committee (protocol #2005N000051). Frogs were housed and maintained in a veterinarian-supervised facility and anesthetized in tricaine prior to oocyte collection. All efforts were made to minimize animal suffering. Materials R-mTFD-MPAB was a gift from Prof. Karol Bruzik, PhD (Dept of Medicinal Chemistry and Pharmacognosy, University of Illinois, Chicago, USA). It was stored as a 100 mM solution in DMSO and diluted in electrophysiology buffer to 4, 8, or 16 µM for experiments. Alphaxalone was purchased from Tocris Bioscience (Bristol, UK), stored as a 10 mM solution in DMSO, and diluted to 1.25, 2.5, or 5 µM in electrophysiology buffer for experiments. Propofol (2,6-diisopropylphenol) was purchased from Sigma-Aldrich (St. Louis, MO, USA), stored as a 10 mM solution in DMSO and diluted to 2.5, 5, or 10 µM in electrophysiology buffer for experiments. R-Etomidate was purchased from Hospira, Inc (Lake Forest, IL, USA) as a 2 mg/ml (~8.2 mM) solution in 35% propylene glycol:water and diluted to 1.6, 3.2, or 6.4 µM in electrophysiology buffer for experiments. We have previously shown that DMSO and propylene glycol at the dilutions used here produce no effects on GABAA receptor function 23. Para-cloromercuribenzenesulfonic acid sodium salt (pCMBS) was purchased from Toronto Chemical Research (Toronto, ON, Canada) and fresh stock solutions in electrophysiology buffer were prepared on the day of use, and kept on ice until final dilution. γ-Aminobutyric acid (GABA), picrotoxin, salts, and buffers were purchased from Sigma-Aldrich. GABAA Receptor Expression in Xenopus Oocytes Oocytes were prepared for use as previously described 15. Complementary DNAs encoding human α1, β3, and γ2L GABAA receptor subunits in pCDNA3.1 expression vectors (Thermo Fisher Scientific, Waltham, MA, USA) were used as mutagenesis templates. Tryptophan and cysteine mutations were introduced into cDNAs by site-directed mutagenesis using QuikChange kits (Agilent Technologies, Santa Clara, CA, USA). Selected clones were sequenced through the entire coding region, and a single clone for each mutant was chosen for subsequent use. Messenger RNAs were synthesized on linearized DNA templates using mMessage Machine kits (Ambion Thermo Fisher), purified, and mixed in a ratio of 1α:1β:3γ, then diluted in RNAase-free water to 1 ng/nl. Oocytes were injected with ~12 ng total mRNA mix and incubated in ND96 buffer (in mM: 96 NaCl, 2 KCl, 1 CaCl2, 0.8 MgCl2, 1 EGTA, 10 HEPES, pH 7.5) supplemented with ciprofloxacin (2 mg/ml) and amikacin (100 µg/ml) at 17 °C for 48 to 72 hours before electrophysiological studies. Two Electrode Voltage-Clamp Electrophysiology Electrophysiological experiments were performed in ND96 buffer at room temperature (21 to 23 °C). Oocytes were positioned in a low volume (30 µl) custom-built flow-cell and impaled with two borosilicate glass micro-electrodes filled with 3 M KCl (resistance < 1 MΩ). Oocytes were voltage-clamped at −50 mV (model OC-725C, Warner Instruments, Hamden CT, USA). Superfusion solutions based on ND96 were selected and delivered at a rate of 2–3 ml/min from glass reservoir syringes via PTFE tubing and a PTFE micro-manifold (MP-8, Warner Instruments). Solutions were selected by activating electrical valves (VC-8, Warner Instruments). Specialized software and a digital input/output interface (pClamp 8.0 and Digidata 1322, both from Molecular Devices, Sunnyvale, CA) were used to coordinate the delivery of different superfusion solutions and recordings of voltage and current signals. Currents were filtered at 1 kHz, digitized at 100 Hz, and stored on a computer disk for offline analysis. GABA concentration-responses Currents in voltage-clamped oocytes expressing GABAA receptors were exposed to solutions containing GABA (range 0.1 µM to 10 mM) with or without anesthetics for 10 to 20 s, followed by 5 minute ND96 wash. Normalization sweeps at 1 to 10 mM GABA alone, (i.e. maximum GABA for the specific receptor) were recorded every other or every third experiment. GABA concentration-responses in the presence of etomidate or mTFD-MPAB were studied using the same method with solutions of variable GABA combined with 3.2 µM etomidate or 8 µM mTFD-MPAB. Baseline responses were measured using anesthetic without GABA. Oocytes were not pre-exposed to anesthetics when co-applied with GABA, and maximal normalization sweeps were in either high GABA alone or high GABA plus anesthetic. At least 3 oocytes from two different frogs were used for each concentration response. Spontaneous receptor activity and maximal GABA efficacy Spontaneous activation of GABAA receptors (in the absence of GABA or anesthetics) was assessed by applying 2 mM picrotoxin to voltage-clamped oocytes. Outward currents that reversed during picrotoxin washout were assumed to represent spontaneously active channels. Spontaneous activity, if detected, was normalized to maximally activated inward GABA-elicited current in the same cell (n ≥ 3 cells). Maximal GABA efficacy for each receptor was determined by comparing peak currents elicited with high GABA to currents elicited with high GABA supplemented with 2.5 µM alphaxalone, which positively modulated all receptors in this study. GABA efficacy was calculated by normalizing GABA responses to GABA + alphaxalone responses, which we assumed represents 100% activation, in the same cell (n ≥ 3 cells). GABA EC5 enhancement Voltage-clamped oocytes expressing wild-type or mutant GABAA receptors were repetitively exposed for 20 s to GABA EC5 (eliciting ~ 5% of maximal GABA response) separated by 5 min wash until three stable peak responses (varying by less than 5%) were sequentially recorded. The oocyte was then exposed for 30 s to anesthetic, followed by 20 s exposure to a solution containing GABA EC5 combined with anesthetic at 2 × EC50 for loss-of-righting-reflexes (LoRR) in tadpoles: 2.5 µM alphaxalone 25, 3.2 µM etomidate 26, 5 µM propofol 27, or 8 µM mTFD-MPAB 28. For each receptor type (wild-type and 12 mutants) and four anesthetics, triplicate measurements of current response to GABA EC5 and GABA EC5 + anesthetic were obtained in at least four oocytes from two different frogs. Because receptors with the γ2L246W mutation displayed substantial spontaneous activity, 2 × EC50 anesthetic enhancement of GABA EC5 responses approached maximal GABA responses for all drugs. This receptors was also studied using 1 × EC50 for tadpole LoRR of each anesthetic. Substituted Cysteine Modification and Protection (SCAMP) Cysteine modification and protection were performed in the presence of maximally activating GABA, as previously described 23. Maximal GABA serves to 1) enhance anesthetic binding site occupancy in protection experiments, 2) assure in most cases that the mix of receptor states is similar in both control modification and protection studies, and 3) increase the rate of sulfhydryl modification. Each substituted cysteine mutant receptor was expressed in Xenopus oocytes, voltage clamped, and exposed to both GABA EC5 (low) and a maximally activating GABA concentration (high). After 5 minute wash, oocytes were then exposed for 30 s to ND96 containing high GABA plus pCMBS, a water-soluble cysteine modifying reagent, ranging in concentration from 1 µM to 1 mM, followed by 5 minute wash in ND96. Responses to low and high GABA were then repeated. We thus confirmed that pCMBS produced irreversible changes in receptor function (in most cases an increase in the low:high GABA response ratio), and identified pCMBS concentrations suitable for comparing initial rates of modification. For control modification rate experiments, voltage-clamped oocytes were repeatedly tested for responses to both EC5 and maximal GABA, then washed for 5 min in ND96, to assure that the response ratio was stable (< 5% change) before pCMBS exposure. Oocytes were then exposed for 5 to 10 s to solutions of maximal GABA plus pCMBS at concentrations estimated to produce about 10% of the maximal modification effect, followed by 5 minute wash, and re-testing for low and high GABA responses. At least three cycles of pCMBS exposure/wash/low:high GABA response testing were performed on each oocyte. A final modification cycle was performed using 10 × pCMBS concentration for 20 s to complete modification of receptors, and electrophysiological response was assessed as the maximal modification effect. For anesthetic protection experiments, a similar experimental protocol was used, except that oocytes were exposed to anesthetic for 30 s just before exposure to a solution of pCMBS + GABA + anesthetic. Post-modification wash and response tests were identical to control conditions (no anesthetic present). Anesthetic concentrations used in protection studies were 2 to 4 × LoRR EC50. For each cysteine mutant, at least 5 oocytes were studied in control modification experiments and at least 4 oocytes were studied in protection experiments. Group sample sizes were based both on prior experience and a power analysis performed with G*Power 3.08 software (Franz Faul, Universitat Kiel, Germany). Based on our experience with SCAMP, SD/mean ratios for modification rates range from 0.2 to 0.5 and in protection experiments we aimed to detect drops of more than 50% from control modification rates. We estimated sample sizes for two groups with respective means of 2 and 1, both with SD = 0.4 (an effect size of 2.5), in a one-tail t-test with α = 0.013 (adjusted for four drug comparisons to control) and β = 0.85, indicating a need for 5 samples per group. Data analysis and statistics Results in text and figures are mean ± sem unless otherwise indicated. GABA concentration-responses Digitized GABA concentration-response data was corrected for baseline leak currents and digitally filtered (10 Hz low-pass, Bessel function) using Clampfit 9.0 software (Molecular Devices). Peak currents were normalized to control (maximal currents), and combined GABA data from multiple cells (n ≥ 3) was fitted with logistic equations using Prism 5.02 (GraphPad Software Inc, La Jolla, CA): Eq. 1 Inorm=(1max−1min)/(1+10(LogEC50−Log[GABA]×nH))+1min where EC50 is the half-maximal activating GABA concentration, and nH is the Hill slope. Mean GABA EC50 and 95% confidence interval are reported. GABA EC50 shifts in the presence of anesthetics were calculated from the difference in log(EC50) values (control – anesthetic) with propagation of log(SD) errors 29. Mean GABA EC50 shift ratios and 95% confidence intervals were calculated. Statistical comparison of GABA EC50s and anesthetic shifts between mutants and wild-type was based on the calculated confidence intervals of the differences of means. Functional characteristics of mutant receptors Comparison of both spontaneous activity and GABA efficacy between mutants and wild-type was based on one-way ANOVA with post-hoc Dunnett’s tests (in Prism 5.02). EC5 enhancement data for the four equi-potent anesthetic concentrations in wild-type and all six tryptophan mutants was tabulated and analyzed with two-way ANOVA and Bonferroni posttests for wild-type vs. mutation for each anesthetic (Prism 5.02). SCAMP Apparent pCMBS modification rates were calculated from data for individual oocytes expressing cysteine mutants. Either normalized maximal GABA response (for α1β3M227Cγ2L) or normalized low:high GABA response ratio (for the other five cysteine mutants) was plotted against cumulative pCMBS exposure (M×s) and linear least squares analyses with y-axis intercepts fixed at 1.0 were applied. For each mutant, the resulting slopes from control and anesthetic protection studies were tabulated and compared using one-way ANOVA in Prism 5.02. Comparison of photolabeling with mutation-based approaches at α1M236 and β3M227 We treated photolabeling results as “gold standard” evidence of anesthetic contact with a residue. For each anesthetic at each residue studied, we categorized substituted tryptophan sensitivity as positive if EC5 enhancement was significantly reduced compared to wild-type. SCAMP results were categorized as positive if addition of anesthetic significantly reduced the initial rate of modification by pCMBS. We constructed 2 × 2 contingency tables and calculated the percentage agreement between each mutation-based method and photolabeling (i.e. concordance of positive and negative results), as well as Cohen’s kappa, a measure of agreement between methods that corrects for chance. Contingency analysis with Fisher’s exact test was used to calculate a conservative p-value, with the implicit null hypothesis being that mutation-based results are uncorrelated with photolabeling. Statistical significance was inferred for p < 0.05. Results Modulation of wild-type α1β3γ2L receptors by four intravenous general anesthetics We expressed GABAA receptors in Xenopus oocytes and characterized their function using two-electrode voltage clamp, assessing GABA concentration-response, maximal GABA efficacy, spontaneous activation, and modulation by intravenous general anesthetics (Table 2). To quantitatively compare the effects of multiple general anesthetics in GABAA receptors, we used equipotent drug concentrations based on loss-of-righting-reflexes (LoRR) tests in tadpoles. The EC50s for LoRR are 1.6 µM etomidate, 2.5 µM propofol, 4 µM mTFD-MPAB, and 1.25 µM alphaxalone 25–28. Wild-type α1β3γ2L GABA responses were similarly enhanced by equipotent 2 × EC50 anesthetic concentrations, measured by either the shift in GABA EC50s (Fig 2 shows results for etomidate and mTFD-MPAB) or enhancement of receptor activation at EC5 GABA (Fig 3 shows results for all four anesthetics). Functional characteristics and anesthetic sensitivity of α1-M1 Trp substitutions: α1M236Wβ3γ2L and α1L232Wβ3γ2L receptors Photolabel analogs of both etomidate and propofol form adducts with α1M236 7,9. We have previously described some of the functional characteristics of α1M236Wβ2γ2L GABAA receptors 15, which mimic the effects of bound anesthetic, including increased sensitivity to GABA, spontaneous channel activity, as well as reduced modulation by etomidate, quantified as the ratio of GABA EC50s in the absence vs. presence of anesthetic. We hypothesized that tryptophan substitutions within anesthetic binding sites may, as a general rule, mimic the effects of anesthetics and reduce modulation by drugs that occupy those sites. The functional characteristics of α1M236Wβ3γ2L receptors were very similar to those of α1M236Wβ2γ2L (Table 2), including very weak modulation by etomidate (Figs 2 and 3). mTFD-MPAB potently activated α1M236Wβ3γ2L receptors, while inducing a smaller GABA EC50 shift than in wild-type receptors (Fig 2, Table 2). This may be due to this receptor’s spontaneous activity. However, EC5 enhancement by mTFD-MPAB was similar in wild-type and α1M236Wβ3γ2L receptors (Fig 3). Propofol, and unexpectedly alphaxalone, also produced significantly less EC5 enhancement in this mutant than in wild-type (Fig 3). The α1 homolog of β3M227 is α1L232 (Table 1). A tryptophan mutation at this site was described in an earlier study of volatile anesthetic modulation 30. Oocyte-expressed α1L232Wβ3γ2L receptors were characterized by a low GABA EC50, low GABA efficacy, and no spontaneous activation (Table 2). Based on GABA EC50 shifts or EC5 enhancement metrics, modulation of α1L232Wβ3γ2L receptors by etomidate and propofol was significantly reduced, while modulation by mTFD-MPAB and alphaxalone was similar to wild-type (Table 2, Figs 2 and 3). Functional characteristics and anesthetic sensitivity of β3-M1 Trp substitutions: α1β3M227Wγ2L and α1β3L231Wγ2L receptors Both mTFD-MPAB and aziPm form photo-adducts with β3M227 8,9. The effects of mutations at β3M227 and β3L231 have not been reported previously. Oocyte-expressed α1β3M227Wγ2L receptors displayed no detectable spontaneous activity, GABA EC50 about twice that of wild-type, and high GABA efficacy (Table 2). Anesthetic modulation of α1β3M227Wγ2L receptors was similar to that in wild-type for etomidate, propofol, and alphaxalone, but significantly reduced for mTFD-MPAB (Table 2, Figs 2 and 3). The β3 homolog of α1M236 is β3L231 (Table 1). The functional characteristics of α1β3L231Wγ2L receptors included spontaneous channel gating similar to that in α1M236Wβ3γ2L, a low GABA EC50, and high GABA efficacy (Table 2). Unexpectedly, etomidate modulation of α1β3L231Wγ2L receptors was significantly less than in wild-type, while modulation by mTFD-MPAB, propofol and alphaxalone was comparable to effects in wild-type (Table 2, Figs 2 and 3). SCAMP in α1-M1: α1M236Cβ3γ2L and α1L232Cβ3γ2L receptors Basal leak and both low and high GABA responses in voltage clamped oocytes expressing wild-type α1β3γ2L GABAA receptors were unaffected by exposure to 1 mM pCMBS for up to 60 s (n = 4 oocytes; data not shown). We have previously described the functional effects of α1M236C mutations in GABAA receptors (see Table 3), the effects of pCMBS modification at α1M236C, and evidence that both etomidate and propofol protect this substituted sidechain from modification, while alphaxalone does not 21,31. For our current experiments, we first tested α1M236Cβ3γ2L receptors for sensitivity to anesthetics, confirming that GABA EC5 responses are enhanced similarly by all four study drugs (Table 3). For modification and protection experiments, we assessed low:high GABA response ratios, which increased by 8.9 ± 0.38-fold (n = 6) after maximal pCMBS modification. Compared to previous studies, we used lower concentrations of etomidate and propofol in protection experiments, and also tested protection with 8 µM mTFD-MPAB. In agreement with our prior studies, we found that addition of 2.5 µM alphaxalone dramatically increased the pCMBS modification rate relative to that in the presence of GABA alone (Fig 4). This is explained by observations that GABA alone activates only about 24% of α1M236Cβ3γ2L receptors, while pCMBS modification of this receptor is much faster in GABA-activated versus inactive receptors 21. Therefore, we used pCMBS + GABA + alphaxalone as the control condition for protection experiments with pCMBS + GABA + other anesthetics. These indicated that both etomidate and propofol protect α1M236C, while mTFD-MPAB does not (Fig 4). We and others have also previously described the functional effects of α1L232C mutations (see Table 3) and evidence that etomidate protects this substituted sidechain from pCMBS modification 21,32,33. All four anesthetics similarly enhanced GABA EC5 responses in α1L232Cβ3γ2L receptors (Table 3). By systematically testing various pCMBS concentrations, we found that very low concentrations (1 µM) of pCMBS produced irreversible enhancement of gating in α1L232Cβ3γ2L receptors (Fig 5A and B), in contrast to the inhibitory effects of 200 to 500 µM pCMBS that were previously reported 21,33. Maximal pCMBS modification increased the low:high GABA response ratio by 7.1 ± 0.47-fold (n =7). In comparison to pCMBS + GABA, addition of etomidate significantly slowed modification of α1L232Cβ3γ2L receptors (Fig 5B and C). Alphaxalone, propofol, and mTFD-MPAB did not significantly affect the rate of pCMBS modification in this mutant. SCAMP in β3-M1: α1β3M227Cγ2L and α1β3L231Cγ2L receptors Studies of cysteine substitutions at β3M227 and β3L231 have not been reported previously. The functional properties of α1β3M227Cγ2L receptors are summarized in Table 3. Modulation of EC5 responses in α1β3M227Cγ2L receptors by etomidate and other study drugs was similar to that in wild-type. The function of oocyte-expressed α1β3M227Cγ2L receptors was irreversibly modified by pCMBS (100 µM; Fig 6A and B), but unlike other cysteine mutants in this study, modification reduced receptor activation without altering GABA sensitivity (the ratio of responses to EC5 vs. high GABA remained constant). Maximal pCMBS modification reduced peak currents by 71 ± 5.4 % (n = 6). Protection experiments showed that both mTFD-MPAB and propofol significantly slowed modification of α1β3M227Cγ2L receptors, while alphaxalone and etomidate did not (Fig 6C). The functional properties of α1β3L231Cγ2L receptors are summarized in Table 3. This mutant was modulated similarly by all four anesthetics at equipotent concentrations. Modification by pCMBS resulted in enhanced GABA sensitivity based on the ratio of electrophysiological currents elicited with low vs. high GABA (Fig 6D and E). Maximal pCMBS modification increased this ratio 5.7 ± 0.44-fold (n = 6). Anesthetic protection experiments showed strong protection by mTFD-MPAB, but not by etomidate, propofol, or alphaxalone (Fig 6F). Comparison of tryptophan anesthetic sensitivity and SCAMP results with photolabeling at α1M236 and β3M227 Concordance between photolabeling, substituted tryptophan anesthetic sensitivity and SCAMP was assessed for the four anesthetics at the two photolabeled residues that we studied: α1M236 and β3M227. In this set of eight anesthetic-residue pairs, there were four positive photolabeling pairs and four negative pairs. Substituted tryptophan sensitivity was categorized as positive if EC5 enhancement by anesthetic in the mutant was significantly lower than in wild-type (Fig 3). SCAMP was categorized as positive if the anesthetic significantly slowed the initial rate of pCMBS modification at the substituted cysteine (Fig 4 and Fig 6C). The categorized results for all three methods are summarized in Table 4. The agreement between photolabeling and substituted tryptophan sensitivity was 75% (Cohen’s kappa = 0.5; p = 0.49 by Fisher’s exact test), with mismatches for propofol-β3M227 (a false negative) and alphaxalone-α1M236 (a false positive) interactions. The concordance between photolabeling and SCAMP results was 100%, and statistically significant (Cohen’s kappa = 1.0; p = 0.029 by Fisher’s exact test). Functional characteristics and anesthetic sensitivity of γ2-M1 Trp substitutions: α1β3γ2I242W and α1β3γ2L246W receptors Both substituted tryptophan sensitivity and SCAMP were used to determine if any of the study anesthetics bind near the γ2-M1 helix. Effects of mutations at γ2I242 and γ2L246 have not been reported previously. The functional properties of α1β3γ2I242W and α1β3γ2L246W receptors are summarized in Table 2. The γ2I242W mutation did not produce spontaneous receptor activation or enhance GABA sensitivity. Instead we observed a high GABA EC50 and reduced GABA efficacy in this mutant (Fig 7A). Oocyte-expressed α1β3γ2L246W receptors were characterized by ~11% spontaneous channel activation, low GABA EC50, and very high GABA efficacy (Table 2; Fig 7B). Modulation of α1β3γ2I242W by all four intravenous general anesthetics was similar to that observed in wild-type receptors (Fig 7C). In α1β3γ2L246W receptors, all the anesthetics produced a large amount of direct channel activation (e.g. Fig 7B) and GABA EC5 enhancement with pre-exposure to the anesthetic concentrations used in other mutants produced near-maximal activation, reducing our ability to discern differences between drugs. We therefore studied EC5 enhancement with half the usual anesthetic concentrations (i.e. 1 × EC50 for tadpole LoRR; Fig 7C). Under these altered conditions, the overall amount of enhancement was lower, and results indicated that none of the four anesthetics produced significantly more or less enhancement than the others. SCAMP in γ2-M1: α1β3γ2I242C and α1β3γ2L246C receptors The functional properties of oocyte-expressed α1β3γ2I242C and α1β3γ2L246C receptors (Table 3) were similar to wild-type (Table 2). Neither of the γ2-M1 cysteine mutations altered sensitivity to modulation by any of the four study anesthetics. Application of 10 µM pCMBS to α1β3γ2I242C receptors irreversibly enhanced GABA sensitivity (Fig. 8A and B). Maximal pCMBS modification increased the low:high GABA response ratio by 5.8 ± 0.37-fold (n = 8). None of the four anesthetic drugs significantly reduced the rate of pCMBS modification, even at twice the concentration used for most other cysteine mutants in this study (Fig 8C). Modification of α1β3γ2L246C receptors required high pCMBS concentrations (500 µM), and also enhanced GABA sensitivity (Fig 8D and E). Maximal pCMBS modification increased the low:high GABA response ratio by 5.0 ± 0.38-fold (n = 5). None of the study anesthetics significantly reduced the rate of pCMBS modification in α1β3γ2L246C receptors. Given the negative SCAMP results at both γ2I242 and γ2L246 with the four intravenous anesthetics, we also tested whether a flexible linear alcohol might bind in the α+ - γ− interface. Additional SCAMP experiments showed no protection by 120 µM n-octanol in either α1β3γ2I242C or α1β3γ2L246C receptors (n = 4 each, data not shown). Discussion Anesthetic photolabeling data in α1β3γ2L GABAA receptors remains limited. We aimed to extend the map of drug contacts and test the functional roles of residues using two mutant-based approaches: substituted tryptophan sensitivity and SCAMP. For two residues, α1M236 and β3M227, and four intravenous anesthetics, published photolabeling results were 100% concordant with SCAMP, but only 75% concordant with substituted tryptophan sensitivity. SCAMP experiments also indicated etomidate contact at α1L232 and mTFD-MPAB contact at β3L231, but no propofol or alphaxalone contact at these loci. At γ2I242 and γ2L246, both approaches consistently indicated no anesthetic interactions. SCAMP fully agrees with photolabeling evidence of selective anesthetic binding to inter-subunit GABAA receptor sites SCAMP is based on formation of a covalent bond between a sulfhydryl engineered into the target receptor and a chemical probe. SCAMP is thus analogous to photolabeling in requiring proximity between probe and site. Both photolabeling and functional studies indicate that anesthetics bind preferentially to GABA-bound open and desensitized states 14,34. Thus, our experiments were designed to promote formation of GABA-bound receptors with high anesthetic affinity, and were powered to detect large protection effects indicating site occupation. We extended our previous SCAMP studies in α1M236C and α1L232C 21 to include additional drugs, and investigated four new cysteine mutants in β3-M1 and γ2-M1. All six substituted cysteines were accessible to pCMBS, forming covalent adducts that irreversibly altered receptor function. In all but one case, cysteine modification enhanced receptor sensitivity to GABA (increased low:high GABA response ratio; Table 3), echoing the effects of tryptophan substitution at most of these positions (Table 2). The exception was β3M227C, where pCMBS modification reduced GABA responses. Notably, the β3M227W mutation reduced GABA sensitivity (Table 2). SCAMP results for α1M236C and β3M227C were in perfect accord with both direct and indirect (drug competition) anesthetic photolabeling studies (Table 4). Contingency analysis with Fisher’s exact test indicated that SCAMP is a sensitive and specific predictor of photolabeling results for these two residues and four anesthetics (four true positives, four true negatives, no false positives or negatives; p = 0.029). We can extend this analysis to include βM286, another residue photolabeled by both azietomidate and aziPm and where SCAMP evidence confirms propofol and etomidate contact, but no interaction with alphaxalone 7,9,12,20,22. Adding these two true positives and one true negative to our Fisher’s exact test analysis results in p = 0.0022. These results support SCAMP as a reliable technique to test putative anesthetic contact points beyond those identified with photolabels. SCAMP also has limitations as an approach to mapping ligand-receptor contact residues 14. Cysteine modification is frequently dependent on the receptor state. Establishing conditions where the mix of functional receptor states is similar in both control modification and protection experiments can be challenging 21. Protection studies may not be feasible or interpretable when cysteine-substitutions induce insensitivity to the ligand (possibly impaired binding) or when exposure to reactive probes causes little or no functional change 21,31. To infer steric interference from SCAMP results, allosteric inhibition of cysteine modification must be ruled out. We addressed this issue by studying multiple drugs that produce similar functional effects. SCAMP identifies additional anesthetic contact residues in α1-M1 and β3-M1 helices While α1L232 has not been photolabeled by anesthetic derivatives, β3L231 was photolabeled by a convulsant barbiturate 35. Our current SCAMP results indicate that α1L232C is protected by etomidate, confirming an earlier study using higher drug concentrations 21. Neither alphaxalone nor mTFD-MPAB protected α1L232C from modification. Despite evidence that propofol binds adjacent to α1-M1 and displaces azietomidate, α1L232C was not protected by propofol. Thus, etomidate and propofol sites in the β+ – α− interfaces overlap, but not at α1L232. Of note, aziPm also photolabels α1I239 9, but pCMBS does not affect the function of α1I239Cβ2γ2L receptors 21, making SCAMP studies unfeasible. Our data indicate that β3L231C is protected by mTFD-MPAB, extending the list of this drug’s likely contacts. However, while propofol displaces mTFD-MPAB photolabeling it does not protect β3L231C, again suggesting overlapping but non-congruent sites adjacent to β3-M1. Consistent with photolabeling results, neither etomidate nor alphaxalone protected β3L231C. Considered together, the above SCAMP results indicate that etomidate binds exclusively in the two β+ – α− interfaces near α1L232 and α1M236, while mTFD-MPAB binds exclusively in homologous α+ – β− and γ+ – β− sites abutting βM227 and βL231. Propofol binds to all four of these sites, adjacent to either α1M236 or βM227. Do M1 helix tryptophan substitutions mimic anesthetic binding in GABAA receptors? This study extended our earlier analysis of α1M236W and assessed five more tryptophan mutants in M1 helices. Both α1M236W and β2M286W mutations mimic the effects of anesthetic binding on GABAA receptors, including decreased GABA EC50, increased GABA efficacy, and spontaneous channel activation, all of which indicate enhanced channel gating 15. We hypothesized that these effects were produced when tryptophan sidechains partially filled the β+ – α− interfacial pockets where etomidate binds, and this approach has been applied to other transmembrane residues in putative anesthetic sites 17–19. Both β3L231W and γ2L246W mutations produced effects similar to α1M236W (Table 2). Thus, tryptophan substitution at the 10th residue after the conserved GYF sequence in any of the M1 helices of α1β3γ2L receptors (Table 1) produced gating effects that mimic anesthetic binding. Receptors with α1L232W mutations also displayed increased apparent GABA sensitivity, but not spontaneous channel activation or GABA efficacy higher than wild-type. On the other hand, β3M227W and γ2I242W produced decreases in GABA sensitivity and none of the W mutations at the 6th residue after the conserved GYF caused spontaneous receptor gating. Overall, these results indicate that the middle segments of all M1 helices of α1β3γ2 receptors are similarly linked to ion channel gating and suggest that all five adjacent inter-subunit pockets, including α+ – γ−, are potential allosteric modulator-agonist sites. Substituted tryptophan sensitivity does not consistently agree with photolabeling or SCAMP evidence of anesthetic binding interactions If Trp substitutions in M1 helices occlude adjacent inter-subunit anesthetic sites, they should reduce modulation by anesthetics occupying these pockets. In agreement with photolabeling, α1M236W reduced receptor sensitivity to both etomidate and propofol, while β3M227W reduced sensitivity to mTFD-MPAB. However, anesthetic sensitivities of α1M236Wβ3γ2L and α1β3M227Wγ2L receptors were not in full accord with photolabeling and SCAMP results at these sites and others (Table 4). Unexpectedly, α1M236W reduced sensitivity to alphaxalone (Fig 3). Interestingly, Henin et al 36 predicted cholesterol binding sites near all six of the M1 residues we studied, and Hosie et al 37 implicated αT237, immediately adjacent to αM236, as a contact for neurosteroids, based on altered tetrahydrodeoxycorticosterone sensitivity in mutants. However, SCAMP indicated no alphaxalone protection at αM236, and alphaxalone enhances azietomidate photolabeling at αM236, implying allosteric effects 12. Indeed, SCAMP provided no evidence of alphaxalone binding near any of the six residues we studied. Correlation with photolabeling is limited, because photolabeling with 6-aziP used β3 homomeric receptors (thus, we lack data for α or γ subunits), and mass spectroscopic analysis missed a sequence including β3M227 and β3L23110. The combined negative SCAMP and photolabeling results together suggest that mutant effects at αM236 and αT237 are indirectly (allosterically) coupled to neurosteroid sites. AziPm photolabeled β3M227 9 and SCAMP indicates propofol protection at β3M227C (Fig 6), but β3M227W did not reduce sensitivity to propofol (Figs 2 and 3). To explain this result, we suggest that propofol exerts most of its gating effects through the two sites that it shares with etomidate, and relatively weak effects through sites adjacent to βM227 that it shares with mTFD-MPAB. Thus, the β3M227W mutation could occlude propofol binding to α+ – β− and γ+ – β− sites while preserving propofol modulation mediated by β+ – α− sites. This hypothesis is supported by evidence that βN265 mutations reduce propofol binding and reduce most of its modulatory effects 31,38,39. For anesthetic interactions at α1L232 and β3L231, SCAMP and substituted tryptophan sensitivity agree in five of eight drug-residue pairs (Table 4). Discordant results were obtained at α1L232 for propofol and at β3L231 for both etomidate and mTFD-MPAB. No anesthetics occupy the α+ – γ− transmembrane interface The extracellular α+ – γ− interface of GABAA receptors forms the high-affinity benzodiazepine site 40,41, but no anesthetic photolabels incorporate in γ2-M1 in the transmembrane portion of this interface. In our current study, neither γ2I242W nor γ2L246W mutations selectively altered EC5 enhancement by the four intravenous drugs we studied. Moreover, SCAMP studies at both γ2I242C and γ2L246C indicated no protection by these anesthetics, even at twice the concentrations that protected residues in other inter-subunit pockets. Thus, we found no evidence of intravenous anesthetic binding in the α+ – γ− transmembrane interfacial pocket. We tested whether the α+ – γ− site was accessible to a small flexible drug, n-octanol, but observed no protection. Others have suggested that ivermectin, a macrocyclic lactone that modulates GABAA receptors, may bind in this interface 42. Conclusions Our SCAMP results are in remarkable agreement with anesthetic photolabeling studies in GABAA receptors, while identifying new drug interactions that indicate overlapping but non-congruent sites for etomidate/propofol in β+ – α− interfaces and mTFD-MPAB/propofol in α+ – β− and γ+ – β− interfaces. The functional characteristics of substituted tryptophan mutants appear useful for probing allosteric linkages to the channel gate, but unreliable for identifying drug binding contacts. SCAMP provided no evidence that alphaxalone binds near any of the six residues we studied. Finally, our studies in γ2-M1 indicate coupling to channel gating similar to that in the four inter-subunit anesthetic sites, but no anesthetic contacts. Thus, the α+ – γ− transmembrane interface in α1β3γ2L GABAA receptors is a potentially unique modulator site with no established ligands; in other words, an orphan site. We thank Youssef Jounaidi, Ph.D. (Instructor, Dept. of Anesthesia Critical Care & Pain Medicine, Massachusetts General Hospital, Boston, MA, USA) for his help with molecular biology, and Timothy Houle, Ph.D. (Faculty Member, Dept. of Anesthesia Critical Care & Pain Medicine, Massachusetts General Hospital, Boston, MA, USA) for expert advice on statistical analysis. Keith W. Miller, D.Phil. (Professor, Dept. of Anesthesia Critical Care & Pain Medicine, Massachusetts General Hospital, Boston, MA, USA), Douglas Raines, M.D. (Professor, Dept. of Anesthesia Critical Care & Pain Medicine, Massachusetts General Hospital, Boston, MA, USA) and Jonathan B. Cohen, Ph.D. (Professor of Neurobiology, Harvard Medical School, Boston, MA, USA) provided helpful comments on the manuscript. Funding: This work was supported by grants (GM089745 and GM058448) from the National Institutes of Health, Bethesda, MD, USA. Figure 1 GABAA receptor transmembrane inter-subunit anesthetic sites The diagram depicts the arrangements of α1 (yellow), β3 (blue), and γ2L (green) subunits and each subunit’s transmembrane four-helix bundle (M1 to M4). The ‘+’ and ‘–’ interfacial surfaces of each subunit, corresponding respectively to M3 and M1 aspects, are identified. The approximate position of photolabeled residues, α1M236 and β3M227 are labeled in magenta, while their homologs on other subunits (see Table 1) are labeled in black. We also depict the hypothesized inter-subunit sites for etomidate (red rhombi), mTFD-MPAB (R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; green rectangles), and propofol (white hexagons), based on photolabeling data. Figure 2 GABA concentration-responses and anesthetic-shifts in α1-M1 and β3-M1 tryptophan substituted GABAA receptor mutants The top panel shows wild-type peak current data (mean ± sem) for GABA alone (black circles), combined with 3.2 µM etomidate (red triangles), or combined with 8 µM mTFD-MPAB (R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; green diamonds). Normalization for control responses was to maximal GABA and to GABA plus anesthetic for anesthetic shift studies. The four other panels show similar data for two tryptophan-substituted α1-M1 mutants and two β3-M1 mutants (labeled in each panel). Fitted GABA EC50s and EC50 shifts in the presence of anesthetics are summarized in Table 2. Figure 3 Anesthetic EC5 enhancement in wild-type vs α1-M1 and β3-M1 substituted tryptophan GABAA receptor mutants The bar-graph depicts GABA EC5 enhancement ratios (mean ± sem) for five receptor types (x-axis labels) and four equi-potent anesthetic solutions: 3.2 µM etomidate (ETO; red), 8 µM mTFD-MPAB (R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; green), 5 µM propofol (PRO; white), and 2.5 µM alphaxalone (ALFAX; purple). The number of oocytes studied for each interaction is indicated by the numbers in each bar. Each drug’s effect in mutants was compared to the same drug effect in wild-type (by two-way ANOVA). Differs from wild-type at ** p < 0.01, *** p < 0.001. Figure 4 Substituted cysteine modification and anesthetic protection in α1M236Cβ3γ2L receptors The bar graph summarizes pCMBS modification rate data in the presence of GABA alone and in the presence of four anesthetics (x-axis labels: ALX is alphaxalone; ETO is etomidate; MPAB is R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; PRO is propofol). We have previously published data showing the functional effect of pCMBS modification in this mutant and protection by etomidate and propofol 21,31. Color coding by drug is the same as in Figure 3 and the number of cells studied for each condition is indicated by the numbers in each bar. This mutant is characterized by low GABA efficacy (Table 2) and pCMBS modification in the presence of alphaxalone represents a control modification condition with high channel open probability matching that in the presence of other anesthetics 21. ** Differs from the rate with pCMBS + GABA + alphaxalone at p < 0.01 (one way ANOVA). Figure 5 Substituted cysteine modification and anesthetic protection in α1L232Cβ3γ2L receptors Panel A shows a series of current sweeps recorded from a single oocyte expressing α1L232Cβ3γ2L receptors before and after a series of 10 s exposures to p-chloromercuribenzenesulfonate (pCMBS) + GABA (arrows). Red traces show current elicited by 3 µM GABA (~EC5) and the black traces show current elicited by 1 mM GABA (black bars above traces indicate GABA application). The final sweeps depict the effect of full modification by 10 µM pCMBS + GABA for 20 s (asterix and arrow). Panel B shows normalized low:high GABA response ratios and linear least squares fits for individual oocytes from all control modification studies (black symbols and lines) and in the presence of 3.2 µM etomidate (red symbols and lines). Panel C is a bar graph summarizing α1L232Cβ3γ2L modification rate data in control studies and in the presence of four anesthetics (x-axis labels: ALX is alphaxalone; ETO is etomidate; MPAB is R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; PRO is propofol). The number of cells studied for each condition is indicated by the numbers in each bar. *Differs from pCMBS + GABA at p < 0.05. Figure 6 Substituted cysteine modification and anesthetic protection in α1β3M227Cγ2L and α1β3L231Cγ2L receptors Panel A shows a series of traces recorded from a single oocyte expressing α1β3M227Cγ2L receptors. Currents were elicited with 2 mM GABA (black bars above traces indicate GABA applications) before and after a series of 10 s exposures to GABA + 100 µM p-chloromercuribenzenesulfonate (pCMBS; arrows). The final trace was recorded after modification with GABA + 1 mM pCMBS for 20 s. Panel B shows normalized peak current data and linear rate analyses for cells modified with pCMBS + GABA (black symbols and lines) and cells modified with pCMBS + GABA + 8 µM mTFD-MPAB (green symbols and lines). Panel C is a bar graph summarizing α1β3M227Cγ2L modification rate data in the absence and presence of four anesthetics (x-axis labels: ALX is alphaxalone; ETO is etomidate; MPAB is R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; PRO is propofol). Panel D shows a series of traces recorded from a single oocyte expressing α1β3L231Cγ2L receptors. Currents were elicited both 2 µM GABA (red traces) and 1 mM GABA (black traces) before and after a series of 10 s exposures to GABA + 25 µM pCMBS (arrows). Black bars above traces indicate GABA applications. The final trace was recorded after modification with GABA + 250 µM pCMBS for 20 s. Panel E shows normalized low:high GABA response ratios and linear least squares rate analyses for cells modified with pCMBS + GABA (black symbols and lines) and cells modified with pCMBS + GABA + 8 µM mTFD-MPAB (green symbols and lines). Panel F is a bar graph summarizing α1β3L231Cγ2L modification rate data in the absence and presence of four anesthetics (x-axis labels). In panels C and F, results statistically differing from pCMBS + GABA control are * p < 0.05 and ** p < 0.01. Figure 7 GABA concentration-responses and anesthetic modulation of α1β3γ2I242W and α1β3γ2L246W receptors Panel A shows peak current results (mean ± sem) for GABA-dependent activation of α1β3γ2I242W receptors with GABA alone (black circles) and in the presence of 3.2 µM etomidate (red triangles). Both data sets are normalized to 3 mM GABA responses, illustrating the low efficacy of GABA alone. Lines through data represent logistic fits (Eq. 1, methods). Fitted EC50 and shift results are reported in Table 2. Panel B shows peak current results (mean ± sem) for GABA-dependent activation of α1β3γ2L246W receptors with GABA alone (black circles) and in the presence of 3.2 µM etomidate (red triangles). Both data sets are normalized to 1 mM GABA responses. Lines through data represent logistic fits. Fitted results are reported in Table 2. Panel C is a bar graph summarizing GABA EC5 enhancement results for both α1β3γ2I242W and α1β3γ2L246W receptors. At equipotent concentrations, none of the anesthetics (x-axis labels: ALX is alphaxalone; ETO is etomidate; MPAB is R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; PRO is propofol) differ in their modulation of these two mutant receptors. Figure 8 Substituted cysteine modification and anesthetic protection in α1β3γ2I242C and α1β3γ2L246C receptors Panel A shows a series of traces recorded from a single oocyte expressing α1β3γ2I242C receptors. Currents were elicited with 3.5 µM GABA (red traces) and 1 mM GABA (black traces) before and after a series of exposures to GABA + 10 µM p-chloromercuribenzenesulfonate (pCMBS; arrows). Black bars above traces indicate GABA applications. The final trace was recorded after modification with GABA + 100 µM pCMBS for 20 s. Panel B shows normalized low:high GABA response ratios and linear least squares rate analyses for cells modified with pCMBS + GABA. Panel C is a bar graph summarizing α1β3γ2I242C modification rate data in the absence and presence of four anesthetics (x-axis labels: ALX is alphaxalone; ETO is etomidate; MPAB is R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid; PRO is propofol). Panel D shows a series of traces recorded from a single oocyte expressing α1β3γ2L246C receptors. Currents were elicited both 1.5 µM GABA (red traces) and 1 mM GABA (black traces) before and after a series of exposures to GABA + 500 µM pCMBS (arrows). Black bars above traces indicate GABA applications. The final trace was recorded after modification with GABA + 1 mM pCMBS for 60 s. Panel E shows normalized low:high GABA response ratios and linear least squares rate analyses for cells modified with pCMBS + GABA. Panel F is a bar graph summarizing α1β3γ2L246C modification rate data in the absence and presence of four anesthetics (x-axis labels). None of the rates with anesthetics differed significantly from control. Table 1 GABAA Receptor M1 Helix Amino Acid Sequence Alignments Position after Conserved GYF 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 α1 G224 Y F V I Q T Y L232 P C I M236 T V I L S Q V S F W L N R E β3 G219 Y F I L Q T Y M227 P S I L231 I T I L S W V S F W I N Y D γ2 G234 Y F T I Q T Y I242 P C T L246 I V V L S W V S F W I N K D The aligned amino acid single-letter code sequences are from human genomic data for GABAA receptor α1 (UniProtKB P14867), β3 (UniProtKB P28472) and γ2 (UniProtKB P18507). The conserved pre-M1 GYF sequence is highlighted in pink. The two highlighted residues α1M236 and β3M227 have been photolabeled by derivatives of intravenous anesthetics. Table 2 Functional and Pharmacological Characteristics of α1β3γ2L GABAA Receptors with M1 Tryptophan Substitutions Receptor Type GABA EC50(µM) [95% CI] (n) GABA Efficacy mean ± se (n) Spont. Activation mean ± se (n) Etomidate-induced GABA EC50 Shift [95% CI] (n) MPAB-induced GABA EC50 Shift [95% CI] (n) α1β3γ2L 18 [16 to 20] (14) 0.88 ± 0.025 (5) < 0.005 (5) 16 [13 to 20] (5) 12 [9.7 to 15] (4) α1L232Wβ3γ2L 4.8 [3.6 to 6.4] * (3) 0.69 ± 0.036 ** (3) <0.005 (4) 2.2 [1.6 to 3.1] * (3) 7.4 [4.6 to 12.0] (3) α1M236Wβ3γ2L 1.5 [1.1 to 2.3] * (3) 0.97 ± 0.011** (6) 0.04 ± 0.014 * (4) 1.6 [1.0 to 2.6] * (3) 2.4 [1.4 to 4.4]* (3) α1β3M227Wγ2L 38 [30 to 47] * (3) 0.99 ± 0.013** (3) <0.005 (4) 12 [9.2 to 15] (3) 4.8 [3.8 to 6.2]* (3) α1β3L231Wγ2L 4.2 [3.7 to 4.8] * 0.91 ± 0.023 (3) 0.035 ± 0.0045 * (4) 4.4 [3.1 to 6.3] * (3) 7.8 [5.0 to 12] (3) α1β3γ2I242W 105 [91 to 121] * (3) 0.75 ± 0.042 ** (4) <0.005 (4) 15 [12 to 19] (3) – α1β3γ2L246W 5.3 [3.6 to 7.7] * (5) 0.97 ± 0.011* (3) 0.11 ± 0.017 ** (5) 4.9 [2.7 to 8.8] * (7) – MPAB is R-5-allyl-1-methyl-5-(m-trifluoromethyl-diazirinylphenyl) barbituric acid. Differs from wild-type at: * p < 0.05, ** p < 0.01, *** p < 0.001. Table 3 Functional and Pharmacological Characteristics of α1β3γ2L GABAA Receptors with M1 Cysteine Substitutions Receptor Type GABA EC50(µM) [95% CI] (n) GABA Efficacy mean ± se (n) Spont. Activation mean ± se (n) GABA EC5 Fold- Enhancement a mean [range] (n) Effect of pCMBS Modification α1L232Cβ3γ2L b 77 * [56 to 107] (5) 0.95 ± 0.03 ** (5) <0.005 (4) 9.8 [8.1 to 11.4] (12) Increase low:high GABA response ratio α1M236Cβ3γ2L b 360 * [260 to 510]* (6) 0.24 ± 0.06*** (6) 0.04 ± 0.014 ** (4) 10.6 [8.6 to 12] (15) Increase low:high GABA response ratio α1β3M227Cγ2L 29 * [25 to 33] (6) 0.65 ± 0.050 * (4) <0.005 (4) 10.1 [7.4 to 14.3] (12) Reduce high GABA response α1β3L231Cγ2L 53 * [39 to 71] (3) 0.66 ± 0.031 ** (3) <0.005 (4) 9.3 [5.8 to 12] (12) Increase low:high GABA response ratio α1β3γ2I242C 16 [13 to 19] (3) 0.94 ± 0.033 * (4) <0.005 (4) 9.3 [5.6 to 11.4] (12) Increase low:high GABA response ratio α1β3γ2L246C 14 [12 to 17] (6) 0.86 ± 0.034 (4) <0.005 (4) 12.0 [8.7 to 15] (12) Increase low:high GABA response ratio pCMBS is p-chloromercuricbenzenesulfonate. Differs from wild-type (see Table 2) at: * p < 0.05, ** p < 0.01, *** p < 0.001. a GABA EC5 Enhancement results are for all four study drugs at 2 × EC50 for tadpole LoRR. Each drug was tested in at least 3 oocytes. b GABA EC50, efficacy and spontaneous activation results are from Stewart et al 21. Table 4 Photolabelinga vs. Substituted Tryptophan Sensitivityb and SCAMPc in α1β3γ2L GABAA Receptor M1 Domains Drug Etomidate mTFD-MPAB Propofol Alphaxalone Residue W PL C W PL C W PL C W PL C α1L232 + – + – – – + – – – ND – α1M236 + + + – – – + + + + – – β3M227 – – – + + + – + + – – – β3L231 + – – – – + – – – – – – γ2I242 – – – – – – – ND – – ND – γ2L246 – – – – – – – ND – – ND – a Photolabeling at α1M236 and β3M227 (highlighted rows) is indicated by a bolded ‘+’ in the middle column (PL) under each anesthetic if the specified residue was an incorporation site for anesthetic analogs: R-(+)-azi-etomidate, R(−)-mTFD-MPAB, m-azi-propofol, or 6-azi-pregnanalone. The negative result for α1M236-alphaxalone was inferred from photolabeling competition using azietomidate. ‘ND’ indicates an absence of direct or indirect photolabeling data, due to missing subunits in the photolabeled receptors. b In the left column (W) under each drug, ‘+’ indicates that a Trp substitution significantly reduces enhancement of GABA EC5 responses by the anesthetic (see Figs 3 and 7). c SCAMP is substituted cysteine modification-protection. In the right column (C) under each anesthetic, ‘+’ indicates that the drug significantly inhibited the rate of covalent modification at the cysteine-substituted sidechain (see Figs 4–6 and 8). Summary Statement Anesthetic contacts in GABAA receptors are incompletely mapped. We showed that substituted cysteine modification-protection agrees with anesthetic photolabeling, and used this approach to test contacts between four anesthetics and all five inter-subunit transmembrane pockets. Conflict of interest: The authors have no conflicts of interest related to this work. References 1 Jurd R Arras M Lambert S Drexler B Siegwart R Crestani F Zaugg M Vogt KE Ledermann B Antkowiak B Rudolph U General anesthetic actions in vivo strongly attenuated by a point mutation in the GABA(A) receptor beta3 subunit FASEB J 2003 17 250 252 12475885 2 Zeller A Arras M Jurd R Rudolph U Identification of a molecular target mediating the general anesthetic actions of pentobarbital Mol Pharmacol 2007 71 852 859 17164405 3 Mihalek RM Banerjee PK Korpi ER Quinlan JJ Firestone LL Mi ZP Lagenaur C Tretter V Sieghart W Anagnostaras SG Sage JR Fanselow MS Guidotti A Spigelman I Li Z DeLorey TM Olsen RW Homanics GE Attenuated sensitivity to neuroactive steroids in gamma-aminobutyrate type A receptor delta subunit knockout mice Proc Natl Acad Sci U S A 1999 96 12905 12910 10536021 4 Franks NP General anaesthesia: from molecular targets to neuronal pathways of sleep and arousal Nat Rev Neurosci 2008 9 370 386 18425091 5 Olsen RW Sieghart W GABA(A) receptors: Subtypes provide diversity of function and pharmacology Neuropharmacology 2009 56 141 148 18760291 6 Hibbs RE Gouaux E Principles of activation and permeation in an anion-selective Cys-loop receptor Nature 2011 474 54 60 21572436 7 Li GD Chiara DC Sawyer GW Husain SS Olsen RW Cohen JB Identification of a GABAA receptor anesthetic binding site at subunit interfaces by photolabeling with an etomidate analog J Neurosci 2006 26 11599 11605 17093081 8 Chiara DC Jayakar SS Zhou X Zhang X Savechenkov PY Bruzik KS Miller KW Cohen JB Specificity of intersubunit general anesthetic binding sites in the transmembrane domain of the human alpha1beta3gamma2 GABAA receptor J Biol Chem 2013 288 19343 19357 23677991 9 Jayakar SS Zhou X Chiara DC Dostalova Z Savechenkov PY Bruzik KS Dailey WP Miller KW Eckenhoff RG Cohen JB Multiple Propofol Binding Sites in a gamma-Aminobutyric Acid Type A Receptor (GABAAR) Identified Using a Photoreactive Propofol Analog J Biol Chem 2014 289 456 468 10 Chen ZW Manion B Townsend RR Reichert DE Covey DF Steinbach JH Sieghart W Fuchs K Evers AS Neurosteroid analog photolabeling of a site in the third transmembrane domain of the beta3 subunit of the GABA(A) receptor Mol Pharmacol 2012 82 408 419 22648971 11 Li GD Chiara DC Cohen JB Olsen RW Numerous classes of general anesthetics inhibit etomidate binding to gamma-aminobutyric acid type A (GABAA) receptors J Biol Chem 2010 285 8615 8620 20083606 12 Li GD Chiara DC Cohen JB Olsen RW Neurosteroids allosterically modulate binding of the anesthetic etomidate to gamma-aminobutyric acid type A receptors J Biol Chem 2009 284 11771 11775 19282280 13 Weiser BP Woll KA Dailey WP Eckenhoff RG Mechanisms revealed through general anesthetic photolabeling Curr Anesthesiol Rep 2014 4 57 66 24563623 14 Forman SA Miller KW Mapping General Anesthetic Sites in Heteromeric Gamma-Aminobutyric Acid Type A Receptors Reveals Potential For Targeting Receptor Subtypes Anesth Analg 2016 Epub May 10 2016 15 Stewart DS Desai R Cheng Q Liu A Forman SA Tryptophan mutations at azi-etomidate photo-incorporation sites on α1 or β2 subunits enhance GABAA receptor gating and reduce etomidate modulation Mol Pharmacol 2008 74 1687 1695 18805938 16 Krasowski MD Nishikawa K Nikolaeva N Lin A Harrison NL Methionine 286 in transmembrane domain 3 of the GABAA receptor beta subunit controls a binding cavity for propofol and other alkylphenol general anesthetics Neuropharmacology 2001 41 952 964 11747900 17 Akk G Li P Bracamontes J Reichert DE Covey DF Steinbach JH Mutations of the GABA-A receptor alpha-1 subunit M1 domain reveal unexpected complexity for modulation by neuroactive steroids Mol Pharmacol 2008 74 614 627 18544665 18 Eaton MM Cao LQ Chen ZW Franks NP Evers AS Akk G Mutational analysis of the putative high-affinity propofol binding site in human β3 homomeric GABAA receptors. Mol Pharm 2015 88 736 745 19 Chang CS Olcese R Olsen RW A single M1 residue in the beta-2 subunit alters channel gating of GABAA receptor in anesthetic modulation and direct activation Journal of Biological Chemistry 2003 278 42821 42828 12939268 20 Bali M Akabas MH Defining the propofol binding site location on the GABAA receptor Mol Pharmacol 2004 65 68 76 14722238 21 Stewart DS Hotta M Li GD Desai R Chiara DC Olsen RW Forman SA Cysteine Substitutions Define Etomidate Binding and Gating Linkages in the alpha-M1 Domain of gamma-Aminobutyric Acid Type A (GABAA) Receptors J Biol Chem 2013 288 30373 30386 24009076 22 Stewart DS Hotta M Desai R Forman SA State-Dependent Etomidate Occupancy of Its Allosteric Agonist Sites Measured in a Cysteine-Substituted GABAA Receptor Mol Pharmacol 2013 83 1200 1208 23525330 23 Stern AT Forman SA A Cysteine Substitution Probes beta3H267 Interactions with Propofol and Other Potent Anesthetics in alpha1beta3gamma2L gamma-Aminobutyric Acid Type A Receptors Anesthesiology 2016 124 89 100 26569173 24 McCracken ML Borghese CM Trudell JR Harris RA A transmembrane amino acid in the GABAA receptor beta2 subunit critical for the actions of alcohols and anesthetics J Pharmacol Exp Ther 2010 335 600 606 20826568 25 Bandyopadhyaya AK Manion BD Benz A Taylor A Rath NP Evers AS Zorumski CF Mennerick S Covey DF Neurosteroid analogues. 15. A comparative study of the anesthetic and GABAergic actions of alphaxalone, Delta16-alphaxalone and their corresponding 17-carbonitrile analogues Bioorg Med Chem Lett 2010 20 6680 6684 20875742 26 Husain SS Stewart D Desai R Hamouda AK Li SG Kelly E Dostalova Z Zhou X Cotten JF Raines DE Olsen RW Cohen JB Forman SA Miller KW p-Trifluoromethyldiazirinyl-etomidate: a potent photoreactive general anesthetic derivative of etomidate that is selective for ligand-gated cationic ion channels J Med Chem 2010 53 6432 6444 20704351 27 Tonner PH Poppers DM Miller KW The general anesthetic potency of propofol and its dependence on hydrostatic pressure Anesthesiology 1992 77 926 931 1443748 28 Savechenkov PY Zhang X Chiara DC Stewart DS Ge R Zhou X Raines DE Cohen JB Forman SA Miller KW Bruzik KS Allyl m-Trifluoromethyldiazirine Mephobarbital: An Unusually Potent Enantioselective and Photoreactive Barbiturate General Anesthetic J Med Chem 2012 55 6554 6565 22734650 29 Bevington PR Robinson DK Data Reduction and Error Analysis for the Physical Sciences 2002 3rd New York, NY McGraw-Hill 30 Jenkins A Greenblatt EP Faulkner HJ Bertaccini E Light A Lin A Andreasen A Viner A Trudell JR Harrison NL Evidence for a common binding cavity for three general anesthetics within the GABAA receptor J Neurosci 2001 21 RC136 11245705 31 Stewart DS Pierce DW Hotta M Stern AT Forman SA Beta N265 in Gamma-Aminobutyric Acid Type A Receptors is Both a Binding and Efficacy Determinant for Etomidate and Propofol PLoS One 2014 10 27 9 10 e111470 25347186 32 Borghese CM Hicks JA Lapid DJ Trudell JR Harris RA GABA(A) receptor transmembrane amino acids are critical for alcohol action: disulfide cross-linking and alkyl methanethiosulfonate labeling reveal relative location of binding sites J Neurochem 2014 128 363 375 24117469 33 Bali M Jansen M Akabas MH GABA-induced intersubunit conformational movement in the GABAA receptor alpha1M1-beta2M3 transmembrane subunit interface: Experimental basis for homology modeling of an intravenous anesthetic binding site J Neurosci 2009 29 3083 3092 19279245 34 Rüsch D Zhong H Forman SA Gating allosterism at a single class of etomidate sites on alpha1beta2gamma2L GABA-A receptors accounts for both direct activation and agonist modulation J Biol Chem 2004 279 20982 20992 15016806 35 Jayakar SS Zhou X Savechenkov PY Chiara DC Desai R Bruzik KS Miller KW Cohen JB Positive and Negative Allosteric Modulation of an alpha1beta3gamma2 gamma-Aminobutyric Acid Type A (GABAA) Receptor by Binding to a Site in the Transmembrane Domain at the gamma+-beta- Interface J Biol Chem 2015 290 23432 23446 26229099 36 Henin J Salari R Murlidaran S Brannigan G A predicted binding site for cholesterol on the GABAA receptor Biophys J 2014 106 1938 1949 24806926 37 Hosie AM Wilkins ME da Silva HM Smart TG Endogenous neurosteroids regulate GABAA receptors through two discrete transmembrane sites Nature 2006 444 486 489 17108970 38 Maldifassi MC Baur R Sigel E Functional sites involved in modulation of the GABA receptor channel by the intravenous anesthetics propofol, etomidate and pentobarbital Neuropharmacology 2016 105 207 214 26767954 39 Krasowski MD Koltchine VV Rick CE Ye Q Finn SE Harrison NL Propofol and other intravenous anesthetics have sites of action on the gamma-aminobutyric acid type A receptor distinct from that for isoflurane Mol Pharmacol 1998 53 530 538 9495821 40 Walters RJ Hadley SH Morris KD Amin J Benzodiazepines act on GABAA receptors via two distinct and separable mechanisms Nature Neuroscience 2000 3 1274 1281 11100148 41 Sigel E Buhr A The benzodiazepine binding site of GABA-A receptors T.I.P.S 1997 18 425 429 42 Estrada-Mondragon A Lynch JW Functional characterization of ivermectin binding sites in alpha1beta2gamma2L GABA(A) receptors Front Mol Neurosci 2015 8 55 26441518

PMC005xxxxxx/PMC5117680.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9710002 22013 Clin Liver Dis Clin Liver Dis Clinics in liver disease 1089-3261 1557-8224 27842768 5117680 10.1016/j.cld.2016.08.010 NIHMS812524 Article Herbal and Dietary Supplement Induced Liver Injury de Boer Ynto S. 12 Sherker Averell H. 3 1 Liver Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 2 Department of Gastroenterology and Hepatology, VU University Medical Center, Amsterdam, The Netherlands 3 Liver Diseases Research Branch, Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD Correspondence to: Averell H. Sherker, MD, FRCPC, FAASLD, DDDN, NIDDK, NIH, 6011 Executive Blvd, Room 6003, Bethesda, MD 20852, Telephone: + 1 301 451 6207, Fax: +1 301 480 8300, averell.sherker@nih.gov; Additional Author Contact: Ynto S. de Boer, MD, De Boelelaan 1117, 1081 HV Amsterdam, Netherlands, Telephone: +31 20 444 4444, ynto.deboer@nih.gov, y.deboer@vumc.nl 24 8 2016 14 10 2016 2 2017 01 2 2018 21 1 135149 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary The increase in the use of herbal and dietary supplements (HDS) over the last decades has been accompanied with an increase in the reports of HDS associated hepatotoxicity. The spectrum of HDS induced liver injury is diverse and the outcome may vary from transient liver test elevations to fulminant hepatic failure resulting in death or requiring liver transplantation. There are no validated standardized tools to establish the diagnosis, but some HDS products do have a typical clinical signature that may help to identify HDS induced liver injury. Herbals Dietary supplements Liver Toxicity Drug-induced liver injury Introduction Epidemiology Herbs and botanicals, as well as their metabolites, constituents and extracts, are included in the definition of “dietary supplements” in United States Federal law.(1) The term “Herbal and Dietary Supplements” (HDS) is redundant but commonly used to categorize these products. Although regulated by the Food and Drug Administration, dietary supplements are not subject to the safety monitoring and approval process of pharmaceutical drugs. Despite the facts that these agents generally lack proof of efficacy and that their manufacturers are not permitted to make medical claims, these products have gained extremely wide acceptance and their use has increased over recent decades. During this time, the estimated number of supplements marketed in the United States has increased over ten-fold -- from ~4000 in 1993 to ~55000 in 2012.(2, 3) About half of the adult population in the United States reports having used at least one dietary supplement in the past month.(4, 5) These products are more commonly used by non-Hispanic whites, at older age and with higher levels of education.(6-9) The majority of alternative medicine users feel that the use of HDS products is consistent with their attitudes towards health and life, and that these agents contribute to their wellbeing.(10) The use of HDS is associated with considerable expense. In 2007, $14.8 billion was spent out of pocket on herbal or complementary nutritional products, equivalent to one-third of the out-of-pocket expenditures associated with prescription drug use in the United States.(11) Nationally, it is estimated that 23,000 emergency department visits each year can be attributed to adverse effects associated with the use of HDS.(12) While there have been well-documented outbreaks of acute liver injury associated with specific dietary supplements, the true incidence of HDS-induced liver injury (HILI) is difficult to estimate. In Spain, 2% of investigated cases of drug-induced liver injury have been attributed to HDS(13), while in Iceland the number is approximately 16%.(14) The NIH-funded Drug-Induced Liver Injury Network (DILIN) has recently reported that, of total DILI cases adjudicated between 2004 and 2013, attribution to HDS has increased from 7% to 20% (Figure 1).(15) Among patients presenting with acute liver failure, those whose disease was attributed to HDS use are more likely to undergo liver transplantation than those associated with prescription medicines (56.1 vs. 31.9%, P <0.005).(16) Regulation and quality control In the United States, manufacturers of dietary supplements containing ingredients that were introduced after October 15, 1994, are required to notify the FDA before marketing and to provide a rationale for the safety of the ingredients, such as historical use.(1) Safety testing or FDA approval of dietary supplements is not required before marketing. Only in case of serious adverse events (hospitalization or death) is post marketing notification of the FDA required.(17) Recent examples of HDS products that were withdrawn from the market include OxyElite Pro in 2013 (caused acute liver failure)(18) and Hydroxycut (hepatocellular injury with jaundice).(19) In the European Union (EU), herbal and dietary supplements are regulated under the Traditional Herbals Medicine Products Directive 2004/24/EC.(20, 21) This directive stipulates that if a product has been shown to be safely used over an acceptable long period (over 30 years with 15 years use within the EU), it may be registered through a simplified procedure if the product is not administered parenterally and does not require a medical prescription. In contrast to U.S. regulations, in Europe food supplements such as vitamin and mineral substances are regulated by the European Food Safety Authority (AFSA) according to Directive 2002/46/EC,(22) whereas herbal medical products are overseen by the Committee on Herbal Medicinal Products (HMPC) of the European Medicines Agency (EMA).(23) To further complicate the regulatory landscape, many HDS products are acquired online through the internet, where vendors and manufacturers may not be easily identifiable and enforcement is extremely difficult (Tables 1 and 2). Clinical presentation and Diagnosis HDS induced liver injury may manifest virtually the entire spectrum of acute and chronic liver disease. In epidemic outbreaks (e.g., Oxyelite Pro) affected individuals may present with a relatively consistent phenotype.(24) A small number of agents (e.g., anabolic steroids) have an idiosyncratic clinical presentation which may trigger a high index of suspicion, even in the absence of a disclosed history of exposure. More typically, sporadic cases present with hepatocellular, cholestatic or mixed pattern of liver injury with varying degrees of severity and hepatic dysfunction. Patients may present with asymptomatic liver enzyme elevations, nonspecific constitutional symptoms, symptoms typical of acute hepatitis (icterus, nausea, fatigue, right upper quadrant abdominal pain) or acute liver failure with hepatic encephalopathy. These cases may have an autoimmune phenotype, as the presence of autoantibodies was reported to be 29% in one series of patients with HDS induced liver injury.(25) Other causes for liver injury such as biliary obstruction (cholelithiasis and malignancy), viral hepatitis (hepatitis A, B, C, E, CMV and EBV), alcoholic and nonalcoholic steatohepatitis, autoimmune liver disease (autoimmune hepatitis [AIH], primary sclerosing cholangitis [PSC] and primary biliary cholangitis [PBC]), hemochromatosis and Wilson disease should be considered and excluded. Unlike prescription drugs, HDS are often perceived as ‘natural’ (and, by extension, harmless) products by patients and may not be considered relevant to disclose. Individuals who have been using a product for an extended period of time may legitimately discount its role in their acute illness, not recognizing that formulations may change without notice, sourcing of ingredients may vary, and that unregulated quality control processes may lead to significant lot-to-lot variations. Patients may be reluctant or embarrassed to share their use of alternative therapeutics with conventional medical practitioners and in some cases (e.g., anabolic steroids), consumers will deny use, knowing that the practice is illegal. Patients with liver disease should be questioned directly about their use of prescription medications, over-the-counter products, and HDS. If the diagnosis remains uncertain or the index of suspicion is high, the patient should be questioned about HDS use again. It may be helpful to ask the patient or a family member to bring all of their medications and supplements to the clinic or hospital. Figure 2 is a rolling suitcase full of HDS products brought to clinic (and being consumed) by a patient with marked jaundice and advanced subacute liver disease who repeatedly denied HDS use until told of her physician’s suspicion after she underwent liver biopsy. Figure 3 shows the pharmacopeia of HDS being used by a patient with liver injury, illustrating the challenges of ascribing causality to a specific agent. Even when there is a high degree of suspicion for HILI, it may be difficult to establish a diagnosis with a high degree of certainty. To address this issue, Naranjo et al. (1981) developed an Adverse Drug Reaction Probability Scale (ADRPS) to establish the probability of an adverse drug reaction, primarily in controlled-trials and studies.(26) The score is derived from 10 simple questions that can add up to a total score that ranges from −4 to 14.(26) Although widely used, this system was shown to have a limited applicability in estimating liver injury due to drugs.(27) Instead, they found that the Roussel Uclaf Causality Assessment Method (RUCAM), performed better (See Al Sehu, Xiaochao Ma, and R Venkataramanan’s article “Mechanisms of drug induced hepatotoxicity,” in this issue).(27-30) Pattern of injury As with ‘classical’ DILI, patients with liver injury due to HDS can be classified into hepatocellular, mixed or cholestatic liver injury. This pattern is defined by the R value ([ALT/ULN] ÷ [Alk P/ULN]), in which a value >5 is interpreted as hepatocellular, <2 as cholestatic and 2-5 as mixed hepatic injury. Across the world, HILI appears to be more commonly associated with an hepatocellular pattern of injury than prescription DILI.(14, 25, 31-35) In DILI, hepatocellular injury with jaundice has been described to have a more severe outcome than is seen in mixed or cholestatic patterns of injury (Hy’s Law). Unique aspects of HDS induced liver injury The mechanisms through which HDS products cause hepatoxicity are variable and specific to the substance consumed. In HILI, it is important to note that substances may be safe in their ‘natural’ form but highly concentrated preparations and synthesized chemicals, although marketed as natural, may be associated with toxicities (e.g., catechins found in green tea preparations and synthetic aegeline in OxyELITE Pro(36)). A major challenge in evaluating liver injury due to HDS products is the inaccuracies with respect to product labeling. Contrary to regulations, some products do not display a label listing ingredients. In the DILIN experience, it was found that 29 of 73 HDS products (40%) taken for various purposes (body building, weight loss, immune support and others) and causing liver injury, did not identify green tea extract (GTE) or any of its component catechins on the label despite containing catechins by analytic chemical methods.(37) Interestingly, 3 of 18 (17%) investigated products that did list catechins or GTE on the label did not contain these substances in detectable concentrations. In general, label-reported concentrations of GTE did not accurately reflect the actual contents. Adulteration of HDS products has been described. Tablets of the Chinese herbal product Jin Bu Huan Anodyne listed Polygala chinensis as its single effective ingredient, but were found to contain levo-tetrahydropalmatine, which is found in the plant genera Stephania and Corydalis but not in the genus Polygala.(38) This product was responsible for an outbreak of severe hepatotoxicity before it was removed from the market. Given the lack of regulatory oversight on production and manufacturing, there is a potential for contamination of HDS product. A report on the hepatotoxicity associated with Herbalife products identified bacterial contamination with Bacillus subtilis as a potential cause for the products’ hepatotoxicity profile(39). Hepatotoxicity associated with specific HDS Anabolics Marketed anabolic steroids are generally synthetic chemicals and are not HDS as strictly defined.(1) However, they are typically included in the discussion of HILI. Liver injury due to ingestion of anabolic steroids/bodybuilding compounds has a very typical clinical presentation. It mostly involves young men involved in bodybuilding, weight training, or athletics who, despite modest liver enzyme elevations, present with marked jaundice and pruritus.(15, 40) It typically has a relatively mild course and completely resolves, albeit often slowly, after the cessation of the product. Pruritus may be debilitating. The use of anabolic steroids or enhancing products is often emphatically denied by the patient, yet the diagnosis can be made confidently based on the presentation and clinical course. Frequently, the patient does not return for a scheduled follow-up visit when feeling better (“Jay’s Law”a). Patients should be warned that the use of these agents may be illegal. Black cohosh Black cohosh (Cimicifuga/Actaea racemosa) is an herbal extract that was traditionally used by Native Americans to treat a wide variety of symptoms, including joint aches, myalgia and gynecologic symptoms. Today it is primarily used for the treatment of post-menopausal symptoms. The mechanism of action is unknown, but there have been reports on hepatotoxicity with and without autoimmune features,(41-43) which has led to the publication of a cautionary statement by the US Dietary Supplement Information Expert Committee.(44) However, a more recent meta-analysis of five randomized, double-blind, controlled clinical trials found no evidence that isopropanolic extracts of black cohosh have any adverse effect on liver function.(45) Black cohosh has been known to be adulterated with other species of Actaea (Actaea pachypoda Ell. (white cohosh) and Actaea podocarpa DC. (yellow cohosh) from China which may be responsible for the hepatotoxicity reported.(46) Germander The blossoms of wall germander (Teucrium chamaedrys) have long been used in folk medicine in the Middle East and Mediterranean region as treatment for dyspepsia, obesity, diabetes and abdominal colic. Despite its wide use, it was found in the early 1990s that herbal preparations, in the form of tea or capsules, could cause significant liver injury. The injury is characterized by an hepatocellular pattern associated with marked jaundice, in the absence of immunoallergic or autoimmune features.(47) The latency to onset of injury is relatively short, usually within 30 days of starting the preparation. Although fatal cases and liver transplantation have been reported, the injury generally resolves after the cessation of the agent.(48) Re-exposure to germander leads to rapid recurrence of the injury.(47) The toxicity is thought to arise due to CYP3A4 activation of the component furan ring Teucrin A, which can then alkylate intracellular epoxide hydrolase, leading to formation of anti-microsomal epoxide hydrolase autoantibodies(49). It has been hypothesized that the anorexogenic properties of germander may actually relate to a mild hepatitis. Green tea Green tea (Camellia sinensis) contains polyphenols known as catechins (+-catechin, gallocatechin, epicatechin, epigallocatechin, epicatechin gallate, and epigallocatechin gallate). An intake of 2-3 cups of green tea per day will not generally lead to hepatotoxicity. However, HDS products such as SlimQuick, generally intended for weight loss may contain higher doses of GTE, which can induce hepatotoxicity.(50-53) Epigallocatechin gallate (EGCG) is the most abundant green tea polyphenol, and is believed to be the most active and potent hepatotoxic component.(50) Genomic investigation in outbred mice identified genes that were associated with EGCG toxicity. There is a suggestion that analogous human genetic variants may be associated with susceptibility to GTE hepatotoxicity.(54) Pyrrolizidine alkaloids Pyrrolizidine alkaloids are found in a large number of plants, including several used as HDS. Among these are Senecio, and Symphytum (Comfrey) species. Sinusoidal obstruction syndrome (SOS, previously known as hepatic veno-occlusive disease) was first described in 1954 among Jamaicans drinking “bush teas” brewed from Senecio.(55) Reports from South Africa(56) (Senecio-contaminated bread), India (57) (Crotalaria-contaminated cereal), Afghanistan(58) (Heliotropium-contaminated wheat) and the southwestern United States(59-61) (Comfrey used as HDS) have implicated pyrrolizidine alkaloids in SOS. Many reports describe the disease in children suggesting either an increased susceptibility or a dose effect.(62) Interestingly, the pulmonary vascular bed is also sensitive to the effects of pyrrolizidine alkaloids.(63) Sinosoidal obstruction syndrome may present as an acute, subacute or chronic liver injury characterized by weight gain, ascites and tender hepatomegaly. Hepatic sinusoidal cells appear to be the primary target of pyrrolizidine alkaloids. These cells are damaged and swell, impeding sinusoidal blood flow, inducing hemorrhage, and ultimately resulting in sinusoidal obstruction.(64) Kava kava Kava kava (Piper methysticum) is used to treat anxiety and depressive disorders. However, numerous worldwide reports of fulminant hepatotoxicity, both hepatocellular and cholestatic, have led to the withdrawal of distribution licenses in the US, Europe and Australia.(65-67) Both immunoallergic and idiosyncratic factors (including CYP2D6 deficiency), have been implicated. (65, 66) Traditional Chinese Medicine In the art of Traditional Chinese Medicine (TCM), specific herbs are selected in different preparations for their supposed properties to treat disease within the human body. TCMs have been used to treat conditions such as viral hepatitis for centuries. In China, currently, approximately 40% of cases of DILI are attributed to the use of TCMs, and have been responsible for cases of acute liver failure with associated coagulopathy.(68, 69) Proprietary mixes Herbalife In 2004, a report by Elinav et al. implicated ingestion of Herbalife products in in 12 patients who developed DILI, manifest as acute fulminant hepatitis.(70) Herbalife products consist of a wide range of different mixtures, usually being taken for the purpose of weight loss or general well-being. Identified ingredients include Solidago gigantea, Ilex paraguariensis, Petroselinum crispum, Garcinia cambogia, Spiraea, Matricaria chamomilla, Liquiritia, Foeniculum amare, Humulus lupulus, Chromium and numerous others. Additionally, the proprietary formula of these products, contain a wide range of listed and unlisted ingredients, which makes it challenging to identify a single responsible component with any degree of certainty.(71) In the initial cohort of cases implicated, the injury resolved spontaneously in 11 of 12 (92%) patients; one patient with preexisting chronic hepatitis B died after undergoing liver transplantation. Three patients developed recurrent liver test abnormalities after resuming ingestion of Herbalife products. Since then, several reports have shown similar associations of HILI with Herbalife products, also suggesting contamination with Bacillus subtilis as a potential cause for its hepatotoxicity profile.(39, 72) Employees of Herbalife have aggressively criticized reports of Herbalife-associated hepatotoxicity (73, 74), but their criticisms have been effectively rebutted.(75) OxyELITE Pro Between February 2012 and February 2014 the FDA received 55 reports of liver disease in consumers of OxyELITE Pro. The typical clinical course consisted of a severe acute hepatitis pattern of injury with a median time to onset of 60 days. Hospitalization was required in 33 (60%) cases and liver transplantation in 3 (5%).(76) In early 2013 the formula of OxyELITE Pro had been changed, substituting 1,3-Dimethylamylamine, which had been associated with cardiovascular toxicity, with aegeline.(76) Early reports of liver injury were from Hawaii, where an initial cluster of 7 patients was reported to develop liver injury in the period between May and September 2013.(24, 77) Following this report, other cases were identified in an outbreak investigation performed by the Hawaii Department of Health, Centers for Disease Control and Prevention (CDC) and FDA.(77) The product was recalled and the manufacturer was required to discontinue the distribution of OxyELITE Pro.(18) Aegeline, derived from the bark of the Bael tree in India, has long been used as a traditional remedy but the component implicated in the OxyELITE Pro outbreak was synthetic.(36) Hydroxycut Hydroxycut products are generally marketed and used as a weight loss supplements. Two published case series implicated the use of some Hydroxycut products to the occurrence of liver injury, presenting predominantly with an hepatocellular pattern of injury and symptoms of jaundice, fatigue, nausea, vomiting, and abdominal pain.(78, 79) Several Hydroxycut products were voluntarily recalled in 2009, following a published FDA warning related to the use of Hydroxycut.(19) Move Free Advanced Move Free Advanced is a widely distributed dietary supplement, sold over the counter in the United States for treatment of sore joints and to improve flexibility and mobility. The product contains glucosamine, chondroitin, hyaluronic acid, and proprietary Uniflex consisting of Chinese skullcap (Scutellaria baicalensis) and black catechu. In a 2010 report, the ingestion of Move Free was identified as a probable cause for the development of cholestatic hepatitis which resolved after discontinuation of the supplement.(80) In one patient, Move Free was not initially recognized as the agent responsible for the injury and the patient restarted the supplement, after which liver injury recurred. A liver biopsy performed at that time was consistent with acute drug induced liver injury.(81) In one patient, pulmonary infiltrates developed simultaneous with the hepatotoxicity and resolved completely with cessation of the supplement.(82) Diterpenoid compounds in Scutellaria baicalensis, have previously been shown to cause apoptosis in isolated rat hepatocytes, through reactive metabolites formed by CYP3A.(83) Conclusion/summary The increase in the use of herbal and dietary supplements (HDS) and a growing awareness of the potential for these agent to cause liver injury has been associated with an increase in reports of HDS associated hepatotoxicity. Limited regulatory oversight, inaccurate product labeling, adulterants and inconsistent sourcing of constituent ingredients may all contribute to the potential for toxicity. The spectrum of HDS induced liver injury is diverse and the outcome may vary from transient liver test abnormalities to acute hepatic failure requiring liver transplantation, or resulting in death. The most commonly implicated products include bodybuilding and weight loss products. There are no validated standardized tools to establish the diagnosis, but some HDS products do have a clear clinical signature that can make diagnosis almost certain. The keys to diagnosis are a high level of suspicion and a comprehensive workup to eliminate competing etiologies. Management is generally supportive and nonspecific. Figure 1 Increase of the proportion of enrolled DILI patients due to HDS products in the DILIN prospective study. Light gray bar represents medications, medium gray bar represents nonbodybuilding HDS, and dark gray bar represents bodybuilding HDS. From Navarro VJ, Barnhart H, Bonkovsky HL, et al. Liver injury from herbals and dietary supplements in the U.S. Drug-Induced Liver Injury Network. Hepatology. 2014;60(4):1399-408, with permission. Figure 2 This patient presented with jaundice and moderately severe subacute hepatitis. She denied any drug or HDS ingestion on repeated questioning over several visits. A liver biopsy was performed as liver tests were slow to improve, and was suspicious for hepatotoxicity. The patient was asked again about ingestions and she admitted that she was taking “one or two” HDS products. She wheeled this suitcase in to her next visit and admitted that she was regularly using all the products in the bag. Figure 3 A patient was referred with abnormal liver tests. He readily admitted to using HDS and brought in all of the products seen here. This illustrates the challenges of ascribing causality to a specific agent. (Courtesy of Dr. Victor Navarro) Table 1 Use and mechanism of specific HDS products Herbals Common use Mechanism and comments References Anabolic steroids bodybuilding, weight training, or athletics unknown (40) Black cohosh (Cimicifuga/Actaea racemosa) joint aches, myalgia and menopausal symptoms unknown possibly adultrated with Actaea pachypoda Ell. (white cohosh) and Actaea podocarpa DC. (yellow cohosh) (46) Chaparral (Larrea tridentata) antioxidant properties, anti- inflammatory, liver disease, skin disorders unknown possibly, interference with cyclo- oxygenase or CYP450, estrogen-like activity (84) Green tea extracts (Camellia sinensis) weight loss epigallocatechin gallate (EGCG) toxicity is possibly heightened in individuals with a genetic predisposition (54) Pyrrolizidine alkaloids/Comfrey (Senecio, Symphytum) natural home remedy hepatic sinusoidal cells are damaged, ultimately resulting in sinusoidal obstruction syndrome (55, 64) Germander (Teucrium chamaedrys) dyspepsia, obesity, diabetes and abdominal colic CYP3A4 dependent alkylation of microsomal protein leading to autoantibody formation (47-49) Greater celandine (Chelidonium majus) dyspepsia unknown (85) Kava Kava (Piper methysticum) gall bladder disease, biliary colic, cholelithiasis, and jaundice immunoallergic and idiosyncratic factors, including CYP2D6 deficiency (65, 66) Mistletoe (Viscum Album) asthma, infertility, hypertension mistletoe lectins have immunostimulating properties and a strong dose-dependent cytotoxic activity (86) Pennyroyal (Mentha pulegium, Hedeoma pulegoides) abortifacient oxidation of pulegone by cytochrome P450 into menthofuran, depletion of glutathione (87) Skullcap (Scutellaria baicalensis) arthritic symptoms CYP3A dependent apoptosis demonstrated in isolated rat hepatocytes (80-82) Jin Bu Huan sedation, analgesic adulteration with other plant genera (38) Ma huang stimulant, weight loss idiosyncratic ephidrine alkaloid toxicity (88) Table 2 Use and mechanism of specific HDS proprietary mixes Proprietary mixes Common use Mechanism and comments Reference s Herbalife weight-loss or improvement of well-being wide range of different products with listed and unlisted ingredients hepatotoxicity potentially due to contamination with Bacillus subtilis in some cases (39) OxyELITE Pro weight loss hepatotoxicity emerged after reformulation with synthetic aegeline product was recalled (18, 24) Hydroxycut weight loss different products, changing formulations voluntarily recalled in 2009 (79) Move Free Advanced See skullcap, table 1 (80-82) SlimQuick weight loss, see green tea extracts , table 1 (54) Key Points The increase in the use of herbal and dietary supplements (HDS) and a growing awareness of the potential for these agent to cause liver injury has been associated with an increase in reports of HDS associated hepatotoxicity. Limited regulatory oversight, inaccurate product labeling, adulterants and inconsistent sourcing of constituent ingredients may all contribute to the potential for toxicity. The spectrum of HDS induced liver injury is diverse and the outcome may vary from transient liver test abnormalities to acute hepatic failure requiring liver transplantation, or resulting in death. The most commonly implicated products include bodybuilding and weight loss products. There are no validated standardized tools to establish the diagnosis, but some HDS products do have a clear clinical signature that can make diagnosis almost certain. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. The authors have nothing to disclose. a Observation made by Dr. Jay H. Hoofnagle in DILIN Causality Assessment, 2013. References 1 Dietary Supplement Health and Education Act of 1994 1994 Public Law 103-417, 108 Stat 4325 2 Report of the Commission on Dietary Supplement Labels 1997 Department of Health and Human Services, Office of Disease Prevention and Health Promotion Washington, DC 3 Dietary supplements: FDA may have opportunities to expand its use of reported health problems to oversee products (GAO-13-244) 13 3 2013 Government Accountability Office Washington, DC Report No 4 Bailey RL Gahche JJ Lentino CV Dwyer JT Engel JS Thomas PR Dietary supplement use in the United States, 2003-2006 J Nutr 2011 141 2 261 6 Epub 2010/12/24 21178089 5 Picciano MF Dwyer JT Radimer KL Wilson DH Fisher KD Thomas PR Dietary supplement use among infants, children, and adolescents in the United States, 1999-2002 Arch Pediatr Adolesc Med 2007 161 10 978 85 Epub 2007/10/03 17909142 6 Radimer K Bindewald B Hughes J Ervin B Swanson C Picciano MF Dietary supplement use by US adults: data from the National Health and Nutrition Examination Survey, 1999-2000 Am J Epidemiol 2004 160 4 339 49 Epub 2004/08/03 15286019 7 Bailey RL Gahche JJ Miller PE Thomas PR Dwyer JT Why US adults use dietary supplements JAMA Intern Med 2013 173 5 355 61 Epub 2013/02/06 23381623 8 Foote JA Murphy SP Wilkens LR Hankin JH Henderson BE Kolonel LN Factors associated with dietary supplement use among healthy adults of five ethnicities: the Multiethnic Cohort Study Am J Epidemiol 2003 157 10 888 97 Epub 2003/05/15 12746241 9 Block G Jensen CD Norkus EP Dalvi TB Wong LG McManus JF Usage patterns, health, and nutritional status of long-term multiple dietary supplement users: a cross-sectional study Nutr J 2007 6 30 Epub 2007/10/26 17958896 10 Astin JA Why patients use alternative medicine: results of a national study JAMA 1998 279 19 1548 53 Epub 1998/05/30 9605899 11 Nahin RL Barnes PM Stussman BJ Bloom B Costs of complementary and alternative medicine (CAM) and frequency of visits to CAM practitioners: United States, 2007 Natl Health Stat Report 2009 18 1 14 Epub 2009/09/24 12 Geller AI Shehab N Weidle NJ Lovegrove MC Wolpert BJ Timbo BB Emergency Department Visits for Adverse Events Related to Dietary Supplements N Engl J Med 2015 373 16 1531 40 Epub 2015/10/16 26465986 13 Andrade RJ Lucena MI Fernandez MC Pelaez G Pachkoria K Garcia-Ruiz E Drug-induced liver injury: an analysis of 461 incidences submitted to the Spanish registry over a 10-year period Gastroenterology 2005 129 2 512 21 Epub 2005/08/09 16083708 14 Bjornsson ES Bergmann OM Bjornsson HK Kvaran RB Olafsson S Incidence, presentation, and outcomes in patients with drug-induced liver injury in the general population of Iceland Gastroenterology 2013 144 7 1419 25 25 e1 3 quiz e19-20. Epub 2013/02/20 23419359 15 Navarro VJ Barnhart H Bonkovsky HL Davern T Fontana RJ Grant L Liver injury from herbals and dietary supplements in the U.S. Drug-Induced Liver Injury Network Hepatology 2014 60 4 1399 408 Epub 2014/07/22 25043597 16 Hillman L Gottfried M Whitsett M Rakela J Schilsky M Lee WM Clinical Features and Outcomes of Complementary and Alternative Medicine Induced Acute Liver Failure and Injury Am J Gastroenterol 2016 Epub 2016/04/06 17 Dietary Supplement and Nonprescription Drug Consumer Protection Act 2006 Public Law 109-462, 120 Stat 4500, Pub. L. No. 109-462 Stat. 4500 18 U.S. Food and Drug Administration FDA investigates acute hepatitis illnesses potentially linked to products labeled OxyElite Pro 30 7 2014 Report No 19 U.S. Food and Drug Administration Recall - Firm Press Release Iovate Health Sciences U.S.A., Inc. Voluntarily Recalls Hydroxycut-Branded Products 1 5 2009 Report No 20 European Commission Herbal Medicinal Products 31 3 2004 Report No 21 Directive 2004/24/EC Of The European Parliament And Of The Council 2004 22 Directive 2002/46/EC of The European Parliament And Of The Council 2002 23 European Medicines Agency Committee on Herbal Medicinal Products (HMPC) 2004 24 Roytman MM Porzgen P Lee CL Huddleston L Kuo TT Bryant-Greenwood P Outbreak of severe hepatitis linked to weight-loss supplement OxyELITE Pro Am J Gastroenterol 2014 109 8 1296 8 Epub 2014/08/06 25091255 25 Bessone F Hernandez N Sanchez A di Pace M Arrese M Brahm JR The Spanish-Latin American DILI Network: preliminary results from a collaborative strategie initiative J Hepatol 2013 58 Suppl 1 212 3 23022497 26 Naranjo CA Busto U Sellers EM Sandor P Ruiz I Roberts EA A method for estimating the probability of adverse drug reactions Clin Pharmacol Ther 1981 30 2 239 45 Epub 1981/08/01 7249508 27 Garcia-Cortes M Lucena MI Pachkoria K Borraz Y Hidalgo R Andrade RJ Evaluation of naranjo adverse drug reactions probability scale in causality assessment of drug-induced liver injury Aliment Pharmacol Ther 2008 27 9 780 9 Epub 2008/02/21 18284654 28 Benichou C Criteria of drug-induced liver disorders. Report of an international consensus meeting J Hepatol 1990 11 2 272 6 Epub 1990/09/01 2254635 29 Danan G Benichou C Causality assessment of adverse reactions to drugs--I. A novel method based on the conclusions of international consensus meetings: application to drug-induced liver injuries J Clin Epidemiol 1993 46 11 1323 30 Epub 1993/11/01 8229110 30 Benichou C Danan G Flahault A Causality assessment of adverse reactions to drugs--II. An original model for validation of drug causality assessment methods: case reports with positive rechallenge J Clin Epidemiol 1993 46 11 1331 6 Epub 1993/11/01 8229111 31 Chalasani N Fontana RJ Bonkovsky HL Watkins PB Davern T Serrano J Causes, clinical features, and outcomes from a prospective study of drug-induced liver injury in the United States Gastroenterology 2008 135 6 1924 34 34 e1 4 Epub 2008/10/29 18955056 32 Reuben A Koch DG Lee WM Drug-induced acute liver failure: results of a U.S. multicenter, prospective study Hepatology 2010 52 6 2065 76 Epub 2010/10/16 20949552 33 Wai CT Tan BH Chan CL Sutedja DS Lee YM Khor C Drug-induced liver injury at an Asian center: a prospective study Liver Int 2007 27 4 465 74 Epub 2007/04/04 17403186 34 Suk KT Kim DJ Kim CH Park SH Yoon JH Kim YS A prospective nationwide study of drug-induced liver injury in Korea Am J Gastroenterol 2012 107 9 1380 7 Epub 2012/06/27 22733303 35 Medina-Cáliz I González-Jiménez A Bessone F Hernández N Sánchez A Di Pace M Variations in drug-induced liver injury (DILI) between different prospective dili registries Clinical Therapeutics 2013 35 8 e24 36 FDA Uses New Authorities To Get OxyElite Pro Off the Market 2013 37 Navarro VJ Bonkovsky HL Hwang SI Vega M Barnhart H Serrano J Catechins in dietary supplements and hepatotoxicity Dig Dis Sci 2013 58 9 2682 90 Epub 2013/04/30 23625293 38 Woolf GM Petrovic LM Rojter SE Wainwright S Villamil FG Katkov WN Acute hepatitis associated with the Chinese herbal product jin bu huan Ann Intern Med 1994 121 10 729 35 Epub 1994/11/15 7944049 39 Stickel F Droz S Patsenker E Bogli-Stuber K Aebi B Leib SL Severe hepatotoxicity following ingestion of Herbalife nutritional supplements contaminated with Bacillus subtilis J Hepatol 2009 50 1 111 7 Epub 2008/11/18 19010564 40 Robles-Diaz M Gonzalez-Jimenez A Medina-Caliz I Stephens C Garcia-Cortes M Garcia-Munoz B Distinct phenotype of hepatotoxicity associated with illicit use of anabolic androgenic steroids Aliment Pharmacol Ther 2015 41 1 116 25 Epub 2014/11/15 25394890 41 Lynch CR Folkers ME Hutson WR Fulminant hepatic failure associated with the use of black cohosh: a case report Liver Transpl 2006 12 6 989 92 Epub 2006/05/25 16721764 42 van de Meerendonk HW van Hunsel FP van der Wiel HE Autoimmune hepatitis induced by Actaea racemosa. Side affects of an herb extract Ned Tijdschr Geneeskd 2009 153 6 246 9 Epub 2009/03/11. Auto-immuunhepatitis door zilverkaars. Bijwerking van een kruidenextract 19271447 43 Cohen SM O’Connor AM Hart J Merel NH Te HS Autoimmune hepatitis associated with the use of black cohosh: a case study Menopause 2004 11 5 575 7 Epub 2004/09/10 15356412 44 Mahady GB Low Dog T Barrett ML Chavez ML Gardiner P Ko R United States Pharmacopeia review of the black cohosh case reports of hepatotoxicity Menopause 2008 15 4 Pt 1 628 38 Epub 2008/03/15 18340277 45 Naser B Schnitker J Minkin MJ de Arriba SG Nolte KU Osmers R Suspected black cohosh hepatotoxicity: no evidence by meta-analysis of randomized controlled clinical trials for isopropanolic black cohosh extract Menopause 2011 18 4 366 75 Epub 2011/01/14 21228727 46 Verbitski SM Gourdin GT Ikenouye LM McChesney JD Hildreth J Detection of Actaea racemosa adulteration by thin-layer chromatography and combined thin-layer chromatography-bioluminescence J AOAC Int 2008 91 2 268 75 Epub 2008/05/15 18476337 47 Larrey D Vial T Pauwels A Castot A Biour M David M Hepatitis after germander (Teucrium chamaedrys) administration: another instance of herbal medicine hepatotoxicity Ann Intern Med 1992 117 2 129 32 Epub 1992/07/15 1605427 48 Dag MS Aydinli M Ozturk ZA Turkbeyler IH Koruk I Savas MC Drug- and herb-induced liver injury: a case series from a single center Turk J Gastroenterol 2014 25 1 41 5 Epub 2014/06/12 24918129 49 De Berardinis V Moulis C Maurice M Beaune P Pessayre D Pompon D Human microsomal epoxide hydrolase is the target of germander-induced autoantibodies on the surface of human hepatocytes Mol Pharmacol 2000 58 3 542 51 Epub 2000/08/23 10953047 50 Lambert JD Kennett MJ Sang S Reuhl KR Ju J Yang CS Hepatotoxicity of high oral dose (−)-epigallocatechin-3-gallate in mice Food Chem Toxicol 2010 48 1 409 16 Epub 2009/11/04 19883714 51 Bonkovsky HL Hepatotoxicity associated with supplements containing Chinese green tea (Camellia sinensis) Ann Intern Med 2006 144 1 68 71 Epub 2006/01/04 16389263 52 Mazzanti G Menniti-Ippolito F Moro PA Cassetti F Raschetti R Santuccio C Hepatotoxicity from green tea: a review of the literature and two unpublished cases Eur J Clin Pharmacol 2009 65 4 331 41 Epub 2009/02/10 19198822 53 Weinstein DH Twaddell WS Raufman JP Philosophe B Mindikoglu AL SlimQuick - associated hepatotoxicity in a woman with alpha-1 antitrypsin heterozygosity World J Hepatol 2012 4 4 154 7 Epub 2012/05/09 22567188 54 Church RJ Gatti DM Urban TJ Long N Yang X Shi Q Sensitivity to hepatotoxicity due to epigallocatechin gallate is affected by genetic background in diversity outbred mice Food Chem Toxicol 2015 76 19 26 Epub 2014/12/03 25446466 55 Bras G Jelliffe DB Stuart KL Veno-occlusive disease of liver with nonportal type of cirrhosis, occurring in Jamaica AMA Arch Pathol 1954 57 4 285 300 Epub 1954/04/01 13147641 56 Wilmot FC Robertson GW Senecio disease or cirrhosis of the liver due to Senecio poisoning Lancet 1920 2 848 9 57 Tandon RK Tandon BN Tandon HD Bhatia ML Bhargava S Lal P Study of an epidemic of venoocclusive disease in India Gut 1976 17 11 849 55 Epub 1976/11/01 1001974 58 Mohabbat O Younos MS Merzad AA Srivastava RN Sediq GG Aram GN An outbreak of hepatic veno-occlusive disease in north-western Afghanistan Lancet 1976 2 7980 269 71 Epub 1976/08/07 59848 59 Stillman AS Huxtable R Consroe P Kohnen P Smith S Hepatic veno-occlusive disease due to pyrrolizidine (Senecio) poisoning in Arizona Gastroenterology 1977 73 2 349 52 Epub 1977/08/01 873137 60 Bach N Thung SN Schaffner F Comfrey herb tea-induced hepatic veno-occlusive disease Am J Med 1989 87 1 97 9 Epub 1989/07/01 2741990 61 Ridker PM Ohkuma S McDermott WV Trey C Huxtable RJ Hepatic venocclusive disease associated with the consumption of pyrrolizidine-containing dietary supplements Gastroenterology 1985 88 4 1050 4 Epub 1985/04/01 3972224 62 Steenkamp V Stewart MJ Zuckerman M Clinical and analytical aspects of pyrrolizidine poisoning caused by South African traditional medicines Ther Drug Monit 2000 22 3 302 6 Epub 2000/06/13 10850397 63 Shubat PJ Banner W Jr. Huxtable RJ Pulmonary vascular responses induced by the pyrrolizidine alkaloid, monocrotaline, in rats Toxicon 1987 25 9 995 1002 Epub 1987/01/01 3124302 64 DeLeve LD Shulman HM McDonald GB Toxic injury to hepatic sinusoids: sinusoidal obstruction syndrome (veno-occlusive disease) Semin Liver Dis 2002 22 1 27 42 Epub 2002/04/03 11928077 65 Russmann S Lauterburg BH Helbling A Kava hepatotoxicity. Ann Intern Med 2001 135 1 68 9 Epub 2001/07/04 66 Stickel F Baumuller HM Seitz K Vasilakis D Seitz G Seitz HK Hepatitis induced by Kava (Piper methysticum rhizoma) J Hepatol 2003 39 1 62 7 Epub 2003/06/25 12821045 67 From the Centers for Disease Control and Prevention Hepatic toxicity possibly associated with kava-containing products--United States, Germany, and Switzerland, 1999-2002 JAMA 2003 289 1 36 7 Epub 2003/01/08 12515265 68 Zhao P Wang C Liu W Chen G Liu X Wang X Causes and outcomes of acute liver failure in China PLoS One 2013 8 11 e80991 Epub 2013/11/28 24278360 69 Zhao P Wang C Liu W Wang F Acute liver failure associated with traditional Chinese medicine: report of 30 cases from seven tertiary hospitals in China* Crit Care Med 2014 42 4 e296 9 Epub 2013/12/18 24335449 70 Elinav E Pinsker G Safadi R Pappo O Bromberg M Anis E Association between consumption of Herbalife nutritional supplements and acute hepatotoxicity J Hepatol 2007 47 4 514 20 Epub 2007/08/19 17692424 71 Teschke R Wolff A Frenzel C Schulze J Eickhoff A Herbal hepatotoxicity: a tabular compilation of reported cases Liver Int 2012 32 10 1543 56 Epub 2012/08/30 22928722 72 Johannsson M Ormarsdottir S Olafsson S Hepatotoxicity associated with the use of Herbalife Laeknabladid 2010 96 3 167 72 Epub 2010/03/04. Lifrarskadi tengdur notkun a Herbalife 20197595 73 Appelhans K Frankos V Shao A Misconceptions regarding the association between Herbalife products and liver-related case reports in Spain Pharmacoepidemiol Drug Saf 2012 21 3 333 4 author reply 5. Epub 2012/03/13 22407600 74 Appelhans K Najeeullah R Frankos V Letter: retrospective reviews of liver-related case reports allegedly associated with Herbalife present insufficient and inaccurate data Aliment Pharmacol Ther 2013 37 7 753 4 Epub 2013/03/06 23458533 75 Reddy KR Bunchorntavakul C Letter: retrospective reviews of liver-related case reports allegedly associated with Herbalife present insufficient and inaccurate data--authors’ reply Aliment Pharmacol Ther 2013 37 7 754 5 Epub 2013/03/06 23458534 76 Klontz KC DeBeck HJ LeBlanc P Mogen KM Wolpert BJ Sabo JL The Role of Adverse Event Reporting in the FDA Response to a Multistate Outbreak of Liver Disease Associated with a Dietary Supplement Public Health Rep 2015 130 5 526 32 Epub 2015/09/04 26327730 77 Johnston DI Chang A Viray M Chatham-Stephens K He H Taylor E Hepatotoxicity associated with the dietary supplement OxyELITE Pro - Hawaii, 2013 Drug Test Anal 2016 8 3-4 319 27 Epub 2015/11/06 26538199 78 Dara L Hewett J Lim JK Hydroxycut hepatotoxicity: a case series and review of liver toxicity from herbal weight loss supplements World J Gastroenterol 2008 14 45 6999 7004 Epub 2008/12/06 19058338 79 Fong TL Klontz KC Canas-Coto A Casper SJ Durazo FA Davern TJ 2nd Hepatotoxicity due to hydroxycut: a case series Am J Gastroenterol 2010 105 7 1561 6 Epub 2010/01/28 20104221 80 Linnebur SA Rapacchietta OC Vejar M Hepatotoxicity associated with chinese skullcap contained in Move Free Advanced dietary supplement: two case reports and review of the literature Pharmacotherapy 2010 30 7 750 258e 62e Epub 2010/07/01 20586134 81 Yang L Aronsohn A Hart J Jensen D Herbal hepatoxicity from Chinese skullcap: A case report World J Hepatol 2012 4 7 231 3 Epub 2012/08/03 22855699 82 Dhanasekaran R Owens V Sanchez W Chinese skullcap in move free arthritis supplement causes drug induced liver injury and pulmonary infiltrates Case Reports Hepatol 2013 2013 965092 Epub 2013/01/01 25431706 83 Haouzi D Lekehal M Moreau A Moulis C Feldmann G Robin MA Cytochrome P450-generated reactive metabolites cause mitochondrial permeability transition, caspase activation, and apoptosis in rat hepatocytes Hepatology 2000 32 2 303 11 Epub 2000/08/01 10915737 84 Sheikh NM Philen RM Love LA Chaparral-associated hepatotoxicity Arch Intern Med 1997 157 8 913 9 Epub 1997/04/28 9129552 85 Benninger J Schneider HT Schuppan D Kirchner T Hahn EG Acute hepatitis induced by greater celandine (Chelidonium majus) Gastroenterology 1999 117 5 1234 7 Epub 1999/10/27 10535888 86 Kienle GS Grugel R Kiene H Safety of higher dosages of Viscum album L. in animals and humans--systematic review of immune changes and safety parameters BMC Complement Altern Med 2011 11 72 Epub 2011/08/30 21871125 87 Thomassen D Slattery JT Nelson SD Menthofuran-dependent and independent aspects of pulegone hepatotoxicity: roles of glutathione J Pharmacol Exp Ther 1990 253 2 567 72 Epub 1990/05/01 2338648 88 Neff GW Reddy KR Durazo FA Meyer D Marrero R Kaplowitz N Severe hepatotoxicity associated with the use of weight loss diet supplements containing ma huang or usnic acid J Hepatol 2004 41 6 1062 4 Epub 2004/12/08 15582145

PMC005xxxxxx/PMC5117809.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9508162 20933 Inflamm Bowel Dis Inflamm. Bowel Dis. Inflammatory bowel diseases 1078-0998 1536-4844 27749455 5117809 10.1097/MIB.0000000000000922 NIHMS811343 Article A Low Neutrophil CD64 Index is Associated with Sustained Remission during Infliximab Maintenance Therapy Minar Phillip MD 1* Jackson Kimberly BA 1 Tsai Yi-Ting MS 1 Rosen Michael J. MD 1 Northcutt Michael MD 2 Khodoun Marat PhD 23 Finkelman Fred D. MD 3 Denson Lee A. MD 1 1 Department of Pediatrics, Division of Gastroenterology, Hepatology and Nutrition, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio 2 Department of Internal Medicine, University of Cincinnati, Cincinnati, Ohio 3 Division of Cellular and Immunobiology, Cincinnati Children’s Hospital Medical Center, Cincinnati Ohio * to whom correspondence should be addressed: Phillip Minar, MD, Division of Pediatric Gastroenterology, Hepatology & Nutrition, Cincinnati Children’s Hospital Medical Center, MLC 2010, 3333 Burnet Avenue, Cincinnati, OH 45229, Tel: 513-803-4688, Fax: 513-636-5581, phillip.minar@cchmc.org 24 8 2016 11 2016 01 11 2017 22 11 26412647 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background We have previously shown that CD64 surface expression on circulating neutrophils is significantly elevated in children with newly diagnosed Crohn’s disease (CD). Our primary aim was to investigate whether elevations in neutrophil CD64 in asymptomatic patients could be utilized to predict treatment failure during maintenance infliximab. Methods Pediatric CD subjects receiving maintenance infliximab in clinical remission (short pediatric CD activity index [shPCDAI] <15) were enrolled. We measured neutrophil CD64 expression (CD64 index, Trillium Diagnostics, LLC) and infliximab trough concentrations. Infliximab failure was defined as a shPCDAI>15 on two consecutive infusions, discontinuation of infliximab, hospitalization, endoscopic ulcerations or surgery during the following year of maintenance infliximab. Results We enrolled 36 subjects, 22/36 were male and 29/36 were white. Mean(SD) age at study entry was 15(4) years with a median of 14 (5–20) infusions prior to study entry. 4/36 were receiving a concurrent immunomodulator. Over one year, 15/36 were classified as infliximab failures. Asymptomatic subjects with a neutrophil CD64 index >1 at study entry had a higher probability of treatment failure compared to asymptomatic subjects with a CD64 index <1 (log-rank =.002). We found only neutrophil CD64 index >1 and non-white race were risk factors for treatment failure by univariate regression analysis. We found no difference in the mean infliximab trough concentration at study entry between treatment failures (2.8 μg/ml, SD 1.2) and subjects remaining in remission on infliximab (4.2 μg/ml, SD 3.4; p=.17). Conclusion Neutrophil CD64 index >1 is a significant risk factor for treatment failure during infliximab maintenance therapy. therapeutic drug monitoring biomarkers pediatrics Crohn’s disease Introduction The therapeutic strategy for Crohn’s disease (CD) is to rapidly induce clinical remission and maintain sustained, steroid-free remission by targeting mucosal healing (MH).1 Yet, despite effective induction treatments, the clinical course during the maintenance phase for many patients with CD waxes and wanes with incomplete intestinal healing likely contributing to the >60% lifetime risk of surgery in CD.2, 3 With the advent of the anti-TNFα agents, the likelihood of achieving MH was markedly improved in comparison to thiopurine monotherapy.4 MH, the absence of intestinal ulcerations viewed by endoscopy, has been associated with reduced incidence of CD related complications.5 Monitoring for MH by endoscopy, however, requires sedation, is costly and is associated with a small risk of complications. In contrast, multiple studies have shown that patient reported symptom indices are unreliable surrogates of MH as both the pediatric CD activity index (PCDAI) and short pediatric CD activity index (shPCDAI) poorly correlate with endoscopic scores of intestinal injury.6, 7 As enthusiasm spreads to adopt the treat-to-target therapeutic strategy, it is vital to further validate surrogate markers of MH as frequent use of endoscopy or abdominal imaging are impractical and costly. Infliximab has been shown to have excellent efficacy in moderate-severe pediatric CD, either used alone (monotherapy) or in combination with an immunomodulator such as 6-mercaptopurine or methotrexate.8–10 Although response rates during induction exceed 80%, a recent review found the probability of children with CD remaining on infliximab monotherapy at 5 years was 0.48 (±0.08).8, 10 Similar studies have shown that over 50% of patients receiving infliximab will require dose intensification (increased dose or frequency) to maintain response.11, 12 A recent retrospective investigation found that utilization of proactive therapeutic drug monitoring (TDM, drug monitoring in asymptomatic patients) prolonged infliximab durability as the authors showed ≥90% probability of remaining on infliximab in inflammatory bowel disease patients who achieved a trough concentration of ≥5 μg/ml.10 While studies have shown direct correlations between detectable infliximab trough concentrations, absence of drug antibodies and long-term efficacy,10, 11, 13 there is a paucity of compelling studies to recommend routine use of blood inflammatory biomarkers to improve infliximab durability. Only recently, Click et al. found that over two years, C-reactive protein (CRP) elevation in asymptomatic patients (which they designated as “silent CD”) was significantly associated with risk of hospitalization (adjusted hazard ratio 2.12, 95% confidence interval (CI) 1.13–3.98, p = .02).14 In our initial investigation, we found that the neutrophil Fcγ Receptor I (CD64) surface expression was a promising candidate biomarker to detect CD-related intestinal injury as we found a significant correlation between neutrophil CD64 index (a marker of CD64 expression) and endoscopic severity scores in CD.7 CD64 is a high affinity cell surface receptor for immunoglobulin (Ig)G1, IgG3 and is expressed by innate immune cells linking antibody specificity with effector cell functions such as clearance of immune complexes and antibody-dependent cell-mediated cytotoxicity.15 While inflammation or infection results in upregulation by interferon-γ, CD64 is an attractive biomarker for CD-related gut injury as neutrophil CD64 is minimally expressed during health.16 In a small cohort of CD patients receiving a variety of maintenance medications, we found asymptomatic patients with increased neutrophil CD64 expression (index >1) were more likely to flare over the following year compared to the asymptomatic group with a neutrophil CD64 index <1 (44% vs. 5%, p<.01).7 The primary aim of our prospective, observational study was to investigate the rate of treatment failures in CD children receiving maintenance infliximab. We hypothesized that asymptomatic CD patients with an elevated neutrophil CD64 at study entry were at a higher risk of treatment failure (clinical relapse, hospitalization, presence of endoscopic ulcerations, surgery, or the discontinuation of infliximab) compared to asymptomatic patients with a neutrophil CD64 index <1. Materials and Methods Patient Population We enrolled pediatric CD subjects receiving infliximab at Cincinnati Children’s Hospital Medical Center from 2013–2014. All subjects recruited for this longitudinal study were receiving a stable maintenance dose of infliximab (≥4 infusions) and were in steroid-free clinical remission. We utilized the shPCDAI (<15) to define clinical remission. Although subjects were recruited at varying time points during their maintenance regimen, all were prospectively followed for clinical relapse, hospitalization, surgery or discontinuation of infliximab for one year. We recorded clinical and demographic characteristics, including the CD phenotype according to the Paris Classification,17 and logged the shPCDAI at each infusion. The primary outcome was time to treatment failure one year from study enrollment. We defined treatment failure as a clinical relapse (shPCDAI ≥15 on two consecutive infusions), experiencing a CD complication (such as an intra-abdominal abscess or fistula development), having abdominal surgery for a CD-related complication, presence of endoscopic ulcerations (if performed during study period), a CD-related hospitalization or discontinuing infliximab therapy. The shPCDAI is a validated assessment to determine clinical activity in pediatric CD and is calculated by combining the adjusted scores from 6/9 measures (abdominal pain, number of stools, general well-being, extraintestinal manifestations, weight, and abdominal exam) from the PCDAI.18, 19 We calculated the shPCDAI at each infusion by recording the subject’s weight and providing the family or subject with a questionnaire to evaluate abdominal pain, number of daily stools (+/−blood), well-being and extraintestinal CD manifestations (rash, fever, mouth sores, and joint or eye complaints). As it was not feasible to perform an abdominal exam at each infusion for each subject, we included the abdominal exam documented during the previous clinic visit if the infusion did not coincide with a physician visit. We recorded all clinician initiated changes in the infliximab regimen; however, an infliximab intensification was not considered a treatment failure unless the subject had a shPCDAI ≥15 on two consecutive infusions. As this was an observational study, additional laboratory markers such as CRP, erythrocyte sedimentation rate (ESR), albumin, hemoglobin and platelet count were recorded if performed in conjunction with the infusion but were not mandated during the study. We also recorded the results of clinician-driven TDM and recorded the clinical decisions that followed. All clinician-driven TDM was conducted by Esoterix Laboratory Services, INC (Austin, TX). For primary hypothesis testing, we defined deep remission as asymptomatic (shPCDAI <15) and a neutrophil CD64 index <1 while silent CD was defined as asymptomatic with a neutrophil CD64 index >1. Biologic Assays Neutrophil CD64 index Blood samples were collected at study entry and prior to each infliximab infusion for one year. Measurement of CD64 surface expression on granulocytes was performed by quantitative flow cytometry on a FACSCalibur (BD Biosciences, San Jose, CA) using the Leuko64™ assay kit (Trillium Diagnostics, Brewer, ME20). The kit includes fluorescent beads and antibodies to CD64 and CD163. The lymphocyte, monocyte and granulocyte populations are defined by their forward and side scatter characteristics along with surface CD163 staining to further define the monocyte population. The CD64 index is calculated by the ratio of the mean fluorescent intensity (MFI) of the granulocytes to that of the calibration beads. The granulocyte specific index is referred as the Neutrophil CD64 index. Infliximab concentration We determined free infliximab concentration from blood drawn prior to the infliximab infusion at study entry and at each subsequent infusion. The infliximab concentration was measured by a quantitative enzyme-linked immunosorbent assay (ELISA) with a detection limit of 0.7 μg/ml (Immundiagnostik IDKmonitor®, Germany) following the manufacturer’s instructions with an inter-assay precision of 6% at 2.7 μg/ml, 4% at 5 μg/ml and 4% at 9.6 μg/ml. Concentrations below the cutoff value were calculated as 0.69 μg/ml and reported as undetectable. Antibodies to infliximab were not measured with this assay, however, the presence of antibodies to infliximab were logged if the subject had TDM with the commercial assay as part of their clinical care. Briefly, detection limit for antibodies to infliximab using the commercial assay (Esoterix Laboratory, LLC) is 22 ng/ml with the electrochemiluminescence immunoassay (ECLIA). Statistical Analysis Continuous variables are presented as mean (standard deviation, SD) or median (25–75% interquartile range) depending on the data distribution. We used the Student’s t test for normally distributed data and the Mann-Whitney U test for non-normally distributed data to detect differences between two continuous variables. The Fisher’s exact test was used for comparison of categorical data. We used Kaplan-Meier survival analysis to determine the one year risk of treatment failure between subjects with deep remission and silent CD with the log-rank test used to determine a difference between the two groups. The Cox proportional hazard regression analysis was used to evaluate the relationship between treatment failure and various subject specific covariates to produce hazard ratios and 95% confidence intervals. Both pre-selected (CD64, albumin, CRP, ESR) and post-hoc subject characteristics were included in the univariate model with covariates removed from our multivariable Cox regression analysis if they were nonsignificant by the Wald test (p<0.1). We tested the validity of assumption of the proportional hazards using statistical and graphing methods. Receiving operating characteristic analysis was performed to determine sensitivity, specificity, positive predictive value, and negative predictive value for a neutrophil CD64 index <1 and one-year treatment success. Statistical analysis for this study were performed using R (Core Team 2012, Vienna, Austria) with the ‘survival’ package used to perform Kaplan-Meier and Cox regression analyses. P value of <.05 was considered statistically significant. Ethical Considerations All subjects and/or their parents/guardians provided informed consent, with subjects over the age of 11 providing assent prior to collection of study data. The study was approved by the Institutional Review Board at Cincinnati Children’s Hospital Medical Center. Results Patient Demographics We enrolled thirty-six CD subjects receiving maintenance infliximab. This cohort included 22 males (61%) with a mean age of 15 (SD 4) years at study entry. The cohort had been receiving infliximab for a median of 14 (range 5–20 17) months with 28 (78%) subjects having an inflammatory (B1) phenotype and only 4 subjects on concurrent immunomodulator (three on methotrexate, one on 6-mercaptopurine) therapy. During the 52 weeks of follow up, 15 subjects were classified as treatment failures and 21 remained in steroid-free, clinical remission. Table 1 details the subject’s demographics and disease phenotypes by study outcome. The two most common reasons for treatment failure included an elevated shPCDAI on two consecutive visits (n=8) and infliximab discontinuation (n=3) with all three subjects having elevated antibodies to infliximab. The remaining treatment failures included one subject with an intra-abdominal abscess, one subject with mucosal ulcerations visualized by endoscopy, one subject undergoing ileal resection while one subject was hospitalized to receive intravenous corticosteroids for a severe CD flare. Although the two groups were very similar in terms of disease phenotype and length of time on infliximab, we found that the 20 of the 21 subjects (95.2%) who remained in remission for the follow up period were white in comparison to only 9/15 (60%) in the treatment failure group (Fischer Exact, p=.013). The six non-white included three African American and three biracial children (two were Asian/white and one subject was African American/white). There was no difference in the number of infusions, time (months) on infliximab prior to study entry or mean infliximab dose (p=0.22) between the two groups. Similarly, there was no difference in concurrent medication exposures (immunemodulators) or previous history of abdominal surgery (14.3% vs. 33.3%, p=.24) between the two groups. Serum biomarkers and treatment failures We found that the neutrophil CD64 index at study entry was significantly higher (median 1.00, min-max range 0.5–4.9) in the 15 subjects who subsequently flared compared to the median CD64 index from the 21 subjects who maintained remission (median 0.66, range 0.38–1.7, p =.03). We also found that the mean albumin at study entry was significantly lower 3.6 gm/dL (SD 0.3) in the treatment failure group compared to the subjects who remained in remission (4 mg/dL, SD 0.4, p <.01). We found there was no difference in the CRP or the infliximab trough between these two groups (Table 2, baseline measurements of CRP and ESR were available in 68% and 73% of the enrolled subjects at study entry). Time to Treatment Failure The primary aim of the study was to determine the rate of treatment failure in pediatric CD patients receiving maintenance infliximab who had a normal neutrophil CD64 index. Kaplan-Meier survival analysis was performed to evaluate the probability of treatment failure by separating asymptomatic subjects into CD64high (silent CD) and CD64low (deep remission). Subjects with a neutrophil CD64 index >1 were classified as CD64high (n=12) while those with a neutrophil CD64 index <1 were classified as CD64low (n=24). We found that CD64high subjects were significantly more likely to relapse over 52 weeks (log-rank =.002, Figure 1a) as compared to CD64low. The median time to treatment failure for subjects with a CD64 index >1 was 90 (57–253) days with 9/15 of the relapse events occurring in the first 4 months from study entry. The median survival time for the 24 subjects with a CD64<1 was 246 (174–335) days. With a true hazard ratio (relative risk) for CD64<1 (control) subjects to CD64>1 (experimental) subjects of 0.236, we post-hoc calculated a probability (power) of .906 with an alpha set at .05. We also found there was a significant negative correlation between the neutrophil CD64 index and days until treatment failure (Spearman r = −0.68, p=.007). In order to examine the strength of our findings, we conducted a sensitivity analysis by performing the same survival analysis for treatment failure utilizing the traditional markers of inflammation (including ESR, CRP and albumin). We found that there was no difference in the probability of treatment failure in our cohort when albumin <3.7 gm/dL (log-rank =.062, Figure 1b), CRP or ESR were tested. Finally, we evaluated the probability of infliximab failure after excluding subjects (n=4) with baseline elevations in CRP in order to control for confounding factors associated with a higher likelihood of failure. After controlling for CRP, we found the CD64high subjects continued to have a higher probability of infliximab failure compared to CD64low subjects (log-rank =.038) We performed a univariate cox regression analysis to further evaluate additional pre-selected and post-hoc patient specific characteristics that could have been associated with infliximab treatment failure. In addition to the neutrophil CD64, we found non-white race and albumin <3.7 gm/dL were associated with a higher probability of treatment failure (p value <.1, Table 3). We then performed a multivariable Cox regression analysis with these three variables and found that neutrophil CD64 index >1 and non-white race were the only significant risk factors (p<.05) for treatment failure in our cohort (Table 4). Neutrophil CD64 testing characteristics We found that 8/12 CD64high (pre-study median CD64 index 1.7, range 1.1–42.2) subjects relapsed compared to 7/24 CD64low subjects (median CD64 index 0.62, range 0.54–0.76). Using receiver operator characteristic curve analysis, a neutrophil CD64 index >1 had a sensitivity of 53.3%, specificity of 81%, positive predictive value of 66.7%, and negative predictive value of 70.8% for treatment failure over the following year (area under the curve 0.72, 95% CI 0.54–0.89, p=.029). The median neutrophil CD64 index at the time of treatment failure was 1.9 (range 0.83–3.2) with a mean infliximab trough of 1.9 (SD 1.2) μg/ml at the time the subject was classified as a treatment failure. In comparison, the mean infliximab trough for treatment responders at week 52 (or at censorship) was 3.4 (SD 2.8) μg/ml with a CD64 index of 0.88 (0.64–1.3). In figure 2, we have graphed the longitudinal assessments of neutrophil CD64 index and infliximab trough concentration at each infusion up until treatment outcome (treatment failure or continued remission) for one year with the dotted line representing a CD64 index =1. Neutrophil CD64 and TDM Although CD64high subjects were much more likely to lose response to infliximab, we found there was no difference in infliximab dose [CD64low, mean 6.5 (SD 2.7) vs. CD64high, 7.5 (SD 2.8) mg/kg, p=.3)], time on infliximab (months or number of infusions), or the infliximab trough concentration at study entry for these subjects compared to CD64low subjects. CD64high subjects had a mean infliximab trough at study entry of 2.5 (SD 1.6) μg/ml compared to a CD64low trough of 4.2 μg/ml (SD 3.2, p=.11). However, CD64low subjects had a higher mean albumin (4 gm/dl, SD 0.32) compared to a mean albumin of 3.5 gm/dL (SD 0.25, p<.001) in the CD64high group. As noted previously, we did not routinely measure antibodies to infliximab in all 36 subjects, however, 25/36 subjects had clinician-driven TDM (which included checking for drug antibodies) during the one year study. Overall, there were 16 events (14 subjects) of antibodies to infliximab during the study; all 16 events had concurrent detectable drug levels. There was no difference in the number of antibodies to infliximab events between the CD64high (41.7%) and CD64low (37.5%, p = 1.0) subjects. The median antibody to infliximab concentration was 69 (min-max 26–1478) ng/ml with only 4/16 of the antibodies to infliximab >100 ng/ml. In the four subjects with antibodies >100 ng/ml, the corresponding neutrophil CD64 index was 3.74, 2.79, 1.11 and 0.61 which directly correlated to the anti-infliximab antibody concentration (1478, 754, 292, 121 ng/ml, respectively). Likewise, there was a significant negative correlation between infliximab trough level and the neutrophil CD64 index (Spearman r = −0.36, p<.05) measured at study entry using the ELISA with a similar correlation between CD64 and infliximab troughs measured at subsequent infusions during the one year observational study (Spearman r = −0.35, p<.001, n=133). Finally, when we evaluated all infliximab trough measurements (repeated measures) for the 36 subjects, we found the 21 treatment responders had a median infliximab trough of 3.6 (1.6–7.3) μg/ml compared to 2.6 (1.5–3.6) μg/ml for the 15 treatment failures (p=.008). Discussion Patients with silent CD, defined by Click et al. as asymptomatic (treated) patients with an elevated CRP, were found to have a nearly 2-fold risk for hospitalization over the subsequent two years compared with asymptomatic CD patients with a normal CRP.14 In our longitudinal, observational study designed to evaluate infliximab efficacy in pediatric CD, we found a neutrophil CD64 index >1 in asymptomatic patients (“silent CD”) was not only associated with infliximab failure over one year but also correlated with lower infliximab trough concentrations during maintenance. In addition, in a secondary analysis, we found our survival testing (risk of treatment failure) remained statistically significant (log-rank =.038) after removing subjects in our cohort who had baseline elevations in CRP at study entry (as a method to control for possible clinically-silent but ongoing intestinal inflammation). While clinical remission remains a goal of CD therapy, the intestinal healing achieved with anti-TNF therapy has compelled clinicians to consider treating patients to “biologic remission.” Biologic remission remains a moving target in terms of its fundamental definition, however, many clinicians would agree that complete absence of intestinal ulcerations by endoscopy would suffice.5 However, it is vital to explore surrogate biomarkers (fecal and blood) of biologic remission as repeat endoscopy is a costly endeavor for CD patients. While we did not measure MH at the beginning of our study with endoscopy, we have previously found that the neutrophil CD64 index correlates significantly with the endoscopic severity score for CD (Spearman r = 0.66, p<.001) and have demonstrated that neutrophil CD64 measured during the course of therapy can be utilized to assess for disease progression.7 Although larger studies are needed, our results indicate that patients with a repeated CD64 index >1 require close follow-up as they were found to have a higher probability of earlier clinical relapse which is likely secondary to ongoing intestinal ulcerations. Our data also suggests neutrophil CD64 is a reliable surrogate biomarker of intestinal healing as we found subjects with sustained remission rarely had consecutive CD64 index measurements >1 (Figure 2a) and enjoyed longer infliximab durability. TDM has and will continue to improve the care of CD patients receiving infliximab, yet over-zealous TDM is likely impractical considering the financial cost and the prolonged testing turnaround time.21 With the significant negative correlation we found between the neutrophil CD64 index and the infliximab trough (Spearman r = −0.35, p<.001), we propose that frequent monitoring for silent CD (biologically active) with the neutrophil CD64 biomarker in CD patients who are receiving maintenance therapy will lead to an individualized, data-driven strategy to optimize medical therapy by targeting TDM to patients with persistent elevations in the CD64 index. From this study, we cannot conclude whether there is a direct relationship between CD64 elevations and infliximab immunogenicity given the small numbers of patients with antibodies to infliximab, however, we found it interesting that subjects with antibodies to infliximab >100 ng/ml had concurrent elevations in neutrophil CD64 (in 3 of 4 subjects). We hypothesize the cause for acute neutrophil CD64 elevation is the result of cumulative intestinal inflammation initiated by the presence of anti-TNF neutralizing antibodies. This is the first study to evaluate longitudinal assessments of the neutrophil CD64 surface expression in predicting infliximab failures. Our study, however, is not the first to propose an association between CD64 and infliximab treatment nonresponse. Wojtal et al., retrospectively, found that colonic CD64 (Fcγ Receptor I) mRNA expression was significantly increased in the inflamed intestinal segment of infliximab non-responders.22 The strength of our study derives from our access to a well-defined, closely monitored longitudinal pediatric cohort receiving infliximab infusions at the same hospital. In addition, our statistical analysis was robust and controlled for confounders such as infliximab trough values and additional biomarkers at study entry. Because neutrophil CD64 surface expression has been shown to be elevated during acute bacterial infections (e.g. Streptococcus spp. and sepsis)23 we used strict criteria to define treatment failure as a shPCDAI >15 on two consecutive infusions which could control for possible skewing of a symptom index by infections or functional gastrointestinal complaints. Stool cultures for enteric infections as well as testing for Clostridium difficile were clinician-driven, however, there was no evidence of active infection (by history, exam or testing) at study entry or during treatment in any of the subjects. We had to rely on the shPCDAI to define clinical remission at study entry as it is uncommon to perform endoscopy (to detect MH) in clinically quiescent pediatric patients and only six patients had a fecal calprotectin at study entry (another potential surrogate marker to assess baseline intestinal inflammation). Although the shPCDAI has been validated as a surrogate for clinical activity in pediatric CD, Turner et al. recently showed the shPCDAI and other pediatric activity indices demonstrated only modest capacity to distinguish between endoscopic healing and ongoing disease activity.24 We also found that it was unfeasible to score the abdominal exam component of the shPCDAI at each infusion as not every infusion corresponded to a clinic visit and therefore we had to rely on the previous documented physician exam to calculate the shPCDAI at each infusion which increased our risk of misclassification. While non-white race was found to be a risk factor for anti-TNF failure in our univariate regression model, we have not found the intensity of the neutrophil CD64 index to vary by race or gender in this or in any other pediatric CD cohorts. The anti-TNF agents are currently the most effective medical treatment for CD, yet, fewer than 50% will remain on infliximab monotherapy after five years.10 Recently, proactive TDM was shown to be highly effective in improving long-term efficacy with infliximab monotherapy.11 Our results show that repeated, longitudinal assessments of the neutrophil CD64 index can be utilized in clinical practice to closely monitor biologic activity and infliximab response in pediatric CD. In addition, our findings support that a persistently elevated neutrophil CD64 index in an asymptomatic patient should trigger closer follow-up for an opportunity to intervene with TDM, abdominal imaging, or repeat endoscopy before anti-TNF failure. This work was supported by NIH K23DK105229 (PM), the NASPGHAN Foundation/Crohn’s and Colitis Foundation of America Young Investigator Development Award (PM), NIH K23DK094832 (MR), and NIH R21AI103816 (FDF, LAD) Abbreviations CI Confidence interval CRP c-reactive protein CD Crohn’s disease ESR erythrocyte sedimentation rate MH mucosal healing shPCDAI short Pediatric Crohn’s disease activity index SD standard deviation TDM therapeutic drug monitoring TNF tumor necrosis factor Figure 1 Kaplan-Meier curves showing the proportion of Crohn’s subjects with treatment failure over one year. (A) Survivor analysis by the circulating neutrophil CD64 expression (index) at study entry and (B) survivor analysis by serum albumin at study entry. Figure 2 Neutrophil CD64 index scores and infliximab concentrations of individual study subjects at baseline and throughout the one year follow up by those who remained in remission (A, C) during the study and infliximab failures (B, D). The dotted line represents the cut off (≥1) for an abnormal neutrophil CD64 index in this study. Infliximab concentrations were measured serially prior to each infusion (trough) by an ELISA for (C) subjects remaining in remission and (D) infliximab failures. Table 1 Baseline characteristics by treatment outcome Remission (n=21) Treatment Failure (n=15) Female 7 (33%) 7(47%) White race 20 (95%)a 9 (60%) Age at diagnosis, years (SD) 11 (4) 12 (5) Age at study entry, years (SD) 16 (4) 14 (5) No. of infusions prior to entry, (SD) 14 (8) 13 (9) Months on infliximab prior, (SD) 26 (18) 18 (15) Infliximab dose at entry, mg/kg (SD) 6.3 (2.6) 7.5 (2.9) Crohn’s location  Ileum (L1) 4 1  Colonic (L2) 1 1  Ileocolonic (L3) 16 13  Perianal location (p) 6 2 Crohn’s behavior  Inflammatory (B1) 18 10  Stricturing (B2) 2 3  Penetrating (B3) 1 1  Both penetrating/stricturing (B2B3) 0 1 Concomitant medications  6-mercaptopurine 1 0  Methotrexate 1 2  5-ASA 4 2 a Fisher exact test <.05 Table 2 Blood biomarkers at study entry for the 36 clinically quiescent Crohn’s disease subjects receiving infliximab maintenance infusions. Biomarker Remission (n=21) Treatment Failure (n=15) p Neutrophil CD64, (range) 0.66 (0.38–1.7) 1.00 (0.5–4.9) .03 ESR mm/hr., (SD, n=27) 9 (4) 26 (27) .02 CRP mg/dL, (range, n=25) 0.29 (0.29–0.33) 0.29 (0.29–0.62) .24 Albumin gm/dL, (SD) 4 (0.4) 3.6 (0.3) <.01 Infliximab trough μg/ml, (SD) 4.2 (3.4) 2.8 (1.2) .17 Table 3 Univariate Cox regression to evaluate risk factors (at study entry) for treatment failure in asymptomatic subjects receiving maintenance infliximab. Covariate Hazard Ratio 95% CI p value Neutrophil CD64 index >1 4.43 1.6–12.4 .005 Female 1.5 0.53–4 .47 Non-white race 4.5 1.6–12.9 .005 <10 years old at diagnosis 0.68 0.23–2 .48 BMI z-score <0 1.4 0.47–3.9 .57 Penetrating or stricturing phenotype 1.8 0.62–5.3 .28 Perianal disease 0.52 0.12–2.3 .39 Previous surgery 2 0.7–6 .2 Received infliximab <1 year 1.6 0.57–4.7 .36 Study entry infliximab trough  <1 μg/ml 1.1 0.25–5.1 .88  <3 μg/ml 1.2 0.38–3.8 .75 Infliximab dose >5 mg/kg 1.8 0.6–5.2 .3 CRP >0.5 mg/dL 2.6 0.69–9.5 .16 ESR >10 mm/hr. 1.7 0.42–6.4 .47 Albumin <3.7 gm/dL 2.7 0.92–7.9 .071 Table 4 Multivariable Cox regression analysis to predict infliximab treatment failure. Covariate Hazard Ratio 95% CI p value Neutrophil CD64 index >1 9.2 2.7–32 <.001 Non-white race 8.5 1.9–39 .006 Albumin <3.7 gm/dL 0.76 0.19–3 .69 Financial disclosures: The authors have no financial, professional, or personal arrangement(s) with a company whose product figures prominently in the submitted manuscript or with a company making a competing product. 1 Baert F Moortgat L Van Assche G Mucosal healing predicts sustained clinical remission in patients with early-stage Crohn’s disease Gastroenterology 2010 138 2 463 8 quiz e10–1. Epub 2009/10/13 19818785 2 Hanauer SB Inflammatory bowel disease: epidemiology, pathogenesis, and therapeutic opportunities Inflammatory bowel diseases 2006 12 Suppl 1 S3 9 Epub 2005/12/27 16378007 3 Thia KT Sandborn WJ Harmsen WS Risk factors associated with progression to intestinal complications of Crohn’s disease in a population-based cohort Gastroenterology 2010 139 4 1147 55 Epub 2010/07/20 20637205 4 Colombel JF Sandborn WJ Reinisch W Infliximab, azathioprine, or combination therapy for Crohn’s disease The New England journal of medicine 2010 362 15 1383 95 Epub 2010/04/16 20393175 5 Shah SC Colombel JF Sands BE Systematic review with meta-analysis: mucosal healing is associated with improved long-term outcomes in Crohn’s disease Alimentary pharmacology & therapeutics 2016 43 3 317 33 26607562 6 Zubin G Peter L Predicting Endoscopic Crohn’s Disease Activity Before and After Induction Therapy in Children: A Comprehensive Assessment of PCDAI, CRP, and Fecal Calprotectin Inflammatory bowel diseases 2015 7 Minar P Haberman Y Jurickova I Utility of neutrophil Fcgamma receptor I (CD64) index as a biomarker for mucosal inflammation in pediatric Crohn’s disease Inflammatory bowel diseases 2014 20 6 1037 48 24788216 8 Hyams J Crandall W Kugathasan S Induction and maintenance infliximab therapy for the treatment of moderate-to-severe Crohn’s disease in children Gastroenterology 2007 132 3 863 73 quiz 1165–6. Epub 2007/02/28 17324398 9 Walters TD Kim MO Denson LA Increased effectiveness of early therapy with anti-tumor necrosis factor-alpha vs an immunomodulator in children with Crohn’s disease Gastroenterology 2014 146 2 383 91 Epub 2013/10/29 24162032 10 Grossi V Lerer T Griffiths A Concomitant Use of Immunomodulators Affects the Durability of Infliximab Therapy in Children With Crohn’s Disease Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association 2015 13 10 1748 56 25911120 11 Vaughn BP Martinez-Vazquez M Patwardhan VR Proactive therapeutic concentration monitoring of infliximab may improve outcomes for patients with inflammatory bowel disease: results from a pilot observational study Inflammatory bowel diseases 2014 20 11 1996 2003 25192499 12 De Bie CI Hummel TZ Kindermann A The duration of effect of infliximab maintenance treatment in paediatric Crohn’s disease is limited Alimentary pharmacology & therapeutics 2011 33 2 243 50 21083595 13 Afif W Loftus EV Jr Faubion WA Clinical utility of measuring infliximab and human anti-chimeric antibody concentrations in patients with inflammatory bowel disease The American journal of gastroenterology 2010 105 5 1133 9 Epub 2010/02/11 20145610 14 Click B Vargas EJ Anderson AM Silent Crohn’s Disease: Asymptomatic Patients with Elevated C-reactive Protein Are at Risk for Subsequent Hospitalization Inflammatory bowel diseases 2015 21 10 2254 61 26197446 15 Hoffmeyer F Witte K Schmidt RE The high-affinity Fc gamma RI on PMN: regulation of expression and signal transduction Immunology 1997 92 4 544 52 Epub 1998/03/14 9497497 16 Buckle AM Hogg N The effect of IFN-gamma and colony-stimulating factors on the expression of neutrophil cell membrane receptors J Immunol 1989 143 7 2295 301 Epub 1989/10/01 2476506 17 Levine A Griffiths A Markowitz J Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification Inflammatory bowel diseases 2011 17 6 1314 21 Epub 2011/05/12 21560194 18 Kappelman MD Crandall WV Colletti RB Short pediatric Crohn’s disease activity index for quality improvement and observational research Inflammatory bowel diseases 2011 17 1 112 7 20812330 19 Hyams JS Ferry GD Mandel FS Development and validation of a pediatric Crohn’s disease activity index Journal of pediatric gastroenterology and nutrition 1991 12 4 439 47 Epub 1991/05/01 1678008 20 http://www.trilliumdx.com/24/Leuko64. 21 Katz L Gisbert JP Manoogian B Doubling the infliximab dose versus halving the infusion intervals in Crohn’s disease patients with loss of response Inflammatory bowel diseases 2012 18 11 2026 33 Epub 2012/02/02 22294554 22 Wojtal KA Rogler G Scharl M Fc gamma receptor CD64 modulates the inhibitory activity of infliximab PloS one 2012 7 8 e43361 Epub 2012/09/01 22937039 23 Davis BH Olsen SH Ahmad E Neutrophil CD64 is an improved indicator of infection or sepsis in emergency department patients Arch Pathol Lab Med 2006 130 5 654 61 16683883 24 Turner D Levine A Walters TD Which PCDAI Version Best Reflects Intestinal Inflammation in Pediatric Crohn’s Disease? Journal of pediatric gastroenterology and nutrition 2016

PMC005xxxxxx/PMC5117824.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7508686 6347 Pain Pain Pain 0304-3959 1872-6623 27842048 5117824 10.1097/j.pain.0000000000000704 NIHMS813027 Article CaMKIIα underlies spontaneous and evoked pain behaviors in Berkeley sickle cell transgenic mice He Ying 13 Chen Yan 1 Tian Xuebi 1 Yang Cheng 1 Lu Jian 1 Xiao Chun 1 DeSimone Joseph 23 Wilkie Diana J. 4 Molokie Robert E. 1235 Wang Zaijie Jim 13* 1 Department of Biopharmaceutical sciences, University of Florida, Gainesville, FL 2 Division of Hematology/Oncology, Department of Medicine, University of Florida, Gainesville, FL 3 University of Illinois Sickle Cell Center, University of Florida, Gainesville, FL 4 Department of Biobehavioral Nursing Science, University of Florida, Gainesville, FL 5 Jesse Brown Veteran’s Administration Medical Center, Chicago, IL 60612 * Corresponding author: Zaijie Jim Wang, Ph.D., University of Illinois, Chicago, IL, 60607 (MC865), (312)996-0888 (voice); (312)996-0098 (fax); zjwang@uic.edu 28 8 2016 12 2016 01 12 2017 157 12 27982806 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Pain is one of the most challenging and stressful conditions to patients with sickle cell disease (SCD) and their clinicians. Patients with SCD start experiencing pain as early as three months old and continue having it throughout their lives. Although many aspects of the disease are well understood, little progress has been made in understanding and treating pain in SCD. This study aimed to investigate the functional involvement of Ca2+/calmodulin-dependent protein kinase II (CaMKIIα) in the persistent and refractory pain associated with SCD. We found non-evoked ongoing pain as well as evoked hypersensitivity to mechanical and thermal stimuli were present in Berkeley sickle cell transgenic mice (BERK mice), but not non-sickle control littermates. Prominent activation of CaMKIIα was observed in the dorsal root ganglia and spinal cord dorsal horn region of BERK mice. Intrathecal administration of KN93, a selective inhibitor of CaMKII, significantly attenuated mechanical allodynia and heat hyperalgesia in BERK mice. Meanwhile, spinal inhibition of CaMKII elicited conditioned place preference in the BERK mice, indicating the contribution of CaMKII in the ongoing spontaneous pain of SCD. We further targeted CaMKIIα by siRNA knockdown. Both evoked pain and ongoing spontaneous pain were effectively attenuated in BERK mice. These findings elucidated, for the first time, an essential role of CaMKIIα as a cellular mechanism in the development and maintenance of spontaneous and evoked pain in SCD, which can potentially offer new targets for pharmacological intervention of pain in SCD. Sickle cell disease pain spontaneous pain phosphorylation Ca2+/calmodulin-dependent protein kinase II 1. Introduction Sickle cell disease (SCD) is a group of autosomal recessive genetic disorders that are caused by mutations in the hemoglobin genes and result in polymerization of deoxyhemoglobin and red cell sickling. Pain is a life-long companion of people living with SCD and is a predictor of disease severity and mortality.39 A nationwide epidemiological study reported that over 60% of SCD patients have at least one pain crisis episode annually.39 In addition to severe acute pain that is associated with vaso-occlusive crises, chronic pain is also prevalent in SCD. A longitudinal diary survey found over half of patients with SCD reported the presence of chronic pain on more than 50% of the days.42 In our self-reported, computerized McGill Pain Questionnaire study conducted in the clinic for non-urgent visits (i.e., not on crisis days), over 60% of subjects reported continuous or constant pain.48 Strikingly, 90% of these subjects also chose pain quality descriptors that are consistent with the presence of neuropathic pain. On average, patients reported using 4.9 analgesics, yet at least one third of patients were not satisfied with their pain control.48 As better treatment and care are extending the life expectancy of patients with SCD, pain is increasingly becoming a big problem that negatively impacts the health and quality of life of these patients. Limited progress, however, has been made in understanding the basic neurobiological mechanisms underlying chronic pain in SCD. In this study, we employed Berkeley sickle cell transgenic mice (BERK mice),37 a model of severe SCD, to identify and characterize neurobiological mechanisms of chronic pain in SCD. BERK mice express exclusively human sickle hemoglobin and have a phenotype that closely mimics many features of severe SCD in humans, including severe hemolytic anemia, irreversibly sickled red cells, increased rigidity of erythrocytes, extensive multiple organ damage, vascular ectasia, intravascular hemolysis, exuberant hematopoiesis, cardiomegaly, glomerulosclerosis, visceral congestion, hemorrhages, multiorgan infarcts, pyknotic neurons, progressive siderosis, gallstones, and priapism.25,33,37 Ongoing spontaneous pain is frequently reported by patients with SCD; however, it is rarely studied in preclinical research.28 In our ongoing work with other, but not all, chronic pain conditions, we found the Ca2+/calmodulin-dependent protein kinase IIα (CaMKIIα) to be a critical molecular mechanism in experimental models of chronic inflammatory and nerve injury neuropathic pain.6,7,32 In this study, we examined the role of CaMKIIα in chronic pain behaviors including ongoing spontaneous pain in BERK mice. 2. Materials and methods 2.1. Materials 2-[N-(2-hydroxyethyl)]-N-(4-methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine) (KN93) and 2-[N-(4-Methoxybenzenesulfonyl)]amino-N-(4-chlorocinnamyl)-N-methylbenzylamine (KN92) were purchased from Tocris Bioscience (Ellisville, MO). Lidocaine HCl (2%) was from Hospira (Lake Forest, IL). Other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). CaMKIIα siRNA (sense, 5′-CACCACCAUUGAGGACGAAdTdT-3′, antisense, 5′-UUCGUCCUCAAUGGUGdTdT-3′) and scrambled RNA duplex control (sense, 5′-AUACGCGUAUUAUACGCGAUUACGAC-3′; antisense, 5′-CGUUAAUCGCGUAUAAUACGCGUAT-3′) were synthesized by Integrated DNA Technologies (Coralville, IW). Lidocaine, KN93, KN92 and RNA duplexes were administered intrathecally (i.t.) in a volume of 5 μL by percutaneous puncture through the L5–L6 intervertebral space.26,32 The RNAs were mixed with a transfection reagent i-Fect (Neuromics, Minneapolis, MN) at a ratio of 1:5 (w/v).7 2.2. Animals BERK mouse38 breeding colony was established from breeders that were obtained from the Jackson Laboratories (Bar Harbor, MA). These mice have been crossed once with C57Bl/6J mice in the Jackson Laboratories, so they have > 50% C57Bl/6J background. We used the following breeding schedule that has been found to be most efficient: female breeders: heterozygous for Hbb (non-sickle); male breeders: homozygous for Hbb (sickle). Genotyping was performed as previously published38 to obtain the BERK mice (sickle) and non-sickle wild-type control littermate mice. The latter serves as control for BERK mice. Age- and sex- matched adult mice (3–5 months old; 20–30g) were used in the study. Prior to actual experimental procedures, mice were provided with food and water ad libitum. Experiments were carried out in accordance with the International Association for the Study of Pain (IASP) recommendations and the NIH Guide for the Care and Use of Laboratory Animals after securing approval from the University of Illinois Institutional Animal Care and Use Committee. The researchers who performed tests were blinded to genotype information before and during the experiments. 2.3. Assessment of ongoing spontaneous pain The conditioned place preference (CPP) method was employed to unmask the ongoing spontaneous pain as we have previously described.21 The CPP apparatus (San Diego Instruments, San Diego, CA) consists of 3 Plexiglas chambers separated by manual doors. A center chamber (6 1/4″ W × 8 1/8″ D × 13 1/8″ H) connects the two end-chambers that are identical in size (10 3/8″ W × 8 1/8″ D × 13 1/8″ H), but can be distinguished by the texture of the floor (rough vs. smooth) and wall pattern (vertical vs. horizontal stripes). Movement of mice and time spent in each chamber were monitored by 4 × 16 photobeam arrays and automatically recorded in SDI CPP software. Preconditioning was performed across 3 days for 30 min each day when mice were exposed to the environment with full access to all chambers. On day 3, a pre-conditioning bias test was performed to determine whether a preexisting chamber bias exists. In this test, mice were placed into the middle chamber and allowed to explore open field with access to all chambers for 15 min. Data were collected and analyzed for duration spent in each chamber. Animals spending more than 80% or less than 20% of the total time in an end-chamber were eliminated (~ 10% of total animals) from further testing. A single trial conditioning protocol was used in the experiments. On conditioning day (day 4), mice first received vehicle control (saline, i.t.) paired with a randomly chosen chamber in the morning and, 4 h later, either KN93 (45 nmol, i.t.) or lidocaine (0.04 %, i.t.) was paired with the other chamber in the afternoon for 15 min (lidocaine) or 30 min (KN93). Whereas the pairing time with lidocaine was based on previous experience,21 pairing condition with KN93 was determined in pilot experiments. On the test day, 20 h after the afternoon pairing, mice were placed in the middle chamber of the CPP box with all doors open so the animals had free access to all chambers. The movement and duration that each mouse spent in each chamber were recorded for 15 min for analysis of chamber preference. Difference scores were calculated as (test time-preconditioning time) spent in the drug chamber. 2.4. Assessment of cold sensitivity Sensitivity to cold stimulus was examined as previously described.8 Mice were placed in individual Plexiglas containers to adapt to the environment for 30 min. A cold stimulus was applied by a brief application (1 s) of tetrafluoroethane to the ventral surface of left hindpaw. Mice were observed for 5 min and the number of licks and the duration of lifting of the sprayed paw were recorded. 2.5. Assessment of heat sensitivity Sensitivity to heat stimulus was determined by the following methods: 1) paw withdrawal latency to radiant heat using a plantar tester (UGO BASILE Model 7372, Stoelting, Wood Dale, IL).7,20 Mice were allowed to acclimate within Plexiglass enclosures on a clear glass plate maintained at 30°C. Radiant heat stimulation was applied to the center of the planter surface of the hindpaw and the latency to paw withdrawal was recorded. A cut-off time of 20s was applied to avoid tissue damage, which was appropriate as the mean control latency is about 12 sec, although the control baseline can be adjusted by varying the intensity of the stimulus. The presence and degree of hyperalgesia were determined by comparing the withdrawal latencies of the test animals and those of controls; 2) The hotplate test was conducted by placing the mice in a glass cylinder on a cold/hot plate analgesia meter (Stoelting) with floor temperature controlled to 50, 52 or 55°C and determining the latency to hindpaw licking or escaping. A 60, 45, and 30 s cutoff time was set for 50°C, 52°C, and 55°C plate temperature, respectively, to prevent injury; 3) The tail immersion test was performed by dipping the distal half of the tail into a water bath maintained at 48, 52, or 55 °C and recording the latency to a rapid tail flick response. A 15, 12, or 10 s cutoff was applied to 48, 52, or 55 °C test, respectively. 2.6. Assessment of mechanical sensitivity Non-noxious mechanical sensitivity was assessed by probing with von Frey filaments.7 Mice were placed in individual Plexiglas containers on a wire mesh platform and tested after 30 min acclimation to the environment. Calibrated von Frey filaments (Stoelting, Wood Dale, IL) were used to press upward to the midplantar surface of the hindpaw for 5 s or until a withdrawal response happened. Using the “up-down″ algorithm,11 50% probability of paw withdrawal threshold was determined. 2.7. Formalin-induced inflammatory pain model Tonic inflammatory pain was induced in BERK mice and non-sickle littermate controls (~ 3 months) by a subcutaneous injection of formalin (2% in saline, 20 μL/mouse) into the dorsal surface of the left hindpaw, as described previously.44,46 The hindpaw was observed for 60 min for the number and duration of paw flinching and the results tabulated for successive 5 min intervals. This test produces a distinct biphasic response. The total number of flinches during the early phase (0–10 min) and late phase (10.01–60 min) were summed, respectively. 2.8. Immunohistochemistry Immunostaining of spinal activated CaMKIIα (pCaMKIIα) was performed as described previously.7 After deep anesthesia with ketamine (100 mg/kg) and xylazine (5 mg/kg, i.p.), mice were initially perfused with phosphate buffer (0.1 M, pH 7.4) for 5 min, followed by 4% paraformaldehyde in phosphate buffer for 20 min (~10 mL/min). Spinal cord lumbar regions were dissected out and sectioned at 20 μm thickness with a cryostat. The floating sections were incubated with the primary antibody for pCaMKIIαThr286 (1:1000, Santa Cruz Biotechnology, Santa Cruz, CA) overnight at room temperature, followed by another incubation with biotinylated goat anti-rabbit IgG secondary antibody (1:200, Vector Laboratories, Burlingame, CA) at room temperature for 2 h. The sections were developed using Elite Vectastain ABC kit (Vector Laboratories). Diaminobenzidine (DAB) stained sections were imaged using Olympus IX71 inverted fluorescence microscope (Olympus Corp., Lake Success, NY) and quantified by the MetaMorph Imaging Software (Molecular Devises, Sunnyvale, CA). For each animal, 5–10 consecutive sections (depending on the size of the region) were imaged. For each group, 5 sections and 6 areas from each section were analyzed and averaged. 2.9. Immunoblotting The lumbar spinal cord and dorsal root ganglia (DRG) tissues were harvested for western blot analysis as previously described.23 Briefly, tissue samples were homogenized in a modified RIPA buffer (1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 5 mM EDTA, 1.0 mM NaF, 10mM sodium pyrophosphate, and 2 mM sodium vanadate in PBS, pH 7.4) in the presence of protease inhibitors and centrifuged (11,000g, 60 min). Supernatants were separated by 12% SDS-PAGE and electro-transferred onto PVDF membrane. The membrane was probed with a rabbit antibody against pCaMKIIα (1:1000, Santa Cruz Biotechnology) at room temperature overnight, followed by incubation with HRP-conjugated donkey anti-rabbit secondary antibody (1:1000, GE Healthcare, Marlborough, MA)at room temperature for 1 h. An enhanced chemiluminescence detection system (ECL, GE Healthcare) was applied for development. The membrane was then stripped and reprobed with a mouse anti-CaMKIIα antibody (1:1000, Santa Cruz Biotechnology) followed by a HRP-conjugated anti-mouse secondary antibody (1:1000, GE Healthcare) and developed as above. Finally, the membrane was stripped again and probed with a mouse anti-β-actin antibody (1:10,000, Sigma) followed by a HRP-conjugated anti-mouse secondary antibody (1:10,000, GE Healthcare). ECL signals were detected using a ChemiDoc system and analyzed using the Quantity One program (Bio-Rad). CaMKII immunoreactivity was expressed as the ratio of the optical densities of pCaMKIIα or CaMKIIα to those of β-actin. 2.10. Statistical Analysis All data are presented as Mean ± SEM. For evoked pain behavior data, differences between groups were analyzed using a one-way ANOVA (treatment) followed by Tukey post hoc test (multiple groups) or Student’s t test (two groups). To analyze the CPP data, two-way ANOVA (pairing vs. treatment) was applied followed by Bonferroni post hoc test. Difference scores were analyzed using Student’s paired t test by computing the differences between test time and preconditioning time for each mouse. Statistical significance was established at the 95% confidence limit. 3. Results 3.1. Presence of ongoing spontaneous pain in BERK mice Patients with SCD experience ongoing pain that is different from the sickle cell crisis pain. We established a conditioned place preference (CPP) paradigm that has been validated in mice to detect non-evoked pain in SCD mice.21 The chronic pain in mice with SCD may present as an aversive state, and the suppression of which may produce CPP, as we have observed in mice with tissue or nerve injury.10,22 First, we performed a preconditioning test to ensure that there was no existing chamber bias for any of the mice with different genotypes. After a single trial conditioning with saline and lidocaine, BERK mice spent significantly more time in the chambers that were paired with lidocaine (420 ± 32 s) than in saline-paired chamber (256 ±37 s, P < 0.01), indicating that lidocaine induced CPP in the sickle mice (Fig. 1A). On the contrary, non-sickle littermate mice spent equal amount of time in the saline (330 ± 25 s) or lidocaine (294 ± 34 s) paired chambers, suggesting the absence of lidocaine-CPP in the non-sickle mice (Fig. 1A). This observation was in agreement with what we reported for naïve ICR and C57Bl/6 mice in the absence of spontaneous pain.10,21,22. We also analyzed the different score for each mouse in each chamber. In comparison to the time spent in the chambers during pre-conditioning, the sickle mice displayed significant difference scores in lidocaine-paired chamber (P < 0.05). None of the other mice groups, including the sickle mice paired with saline treatment, exhibited significant difference scores after conditioning (Fig. 1B). These scores further confirmed the presence of ongoing spontaneous pain in BERK sickle cell mice. 3.2. Hypersensitivity to noxious cold stimulus in BERK mice Patients with SCD have enhanced pain sensitivity to cold environments.2,35,41,47 To assess the noxious cold hyperalgesia in BERK mice, the plantar surface of the left hindpaw was exposed to tetrafluoroethane for 1 s.8 Under this condition, it produced reflex responses as measured by the paw licking and lifting behaviors. We counted the number of licks occurred in 5 min triggered by the cold stimulus. As compared with the non-sickle littermates, BERK mice displayed significantly enhanced licking behavior, with an increased number of licks by more than 2.6 fold (P < 0.01, Fig. 2A). The total time of hindpaw lifting and licking in BERK mice (139.9 ± 5.0 s) was increased by around 3.5 fold of that in the non-sickle mice (40.5 ± 10.0 s) (P < 0.001, Fig. 2B). Consistent with the findings from other cold pain measurements from other investigators,24,34 these observations demonstrated the existence of cold-induced hypersensitivity in BERK mice. 3.3. Evoked mechanical and heat pain behaviors in BERK mice We also determined the baseline nociceptive response of BERK mice towards mechanical and heat stimuli. As compared with the age/sex-matched non-sickle control mice (1.52 ± 0.14 g), BERK mice displayed significantly reduced pain threshold (0.10 ± 0.05 g) to normally innocuous mechanical stimulus by von Frey filament probing (P < 0.001, Fig. 2C), indicative of the presence of tactile allodynia in BERK mice. In response to the radiant heat challenge, BERK mice exhibited decreased latencies to noxious heat stimuli applied to the hindpaw (6.32 ± 0.78 s in BERK mice vs. 11.97 ± 0.78 s in non-sickle mice, P < 0.001, Fig. 2D), indicating the presence of heat hyperalgesia. The enhanced thermal sensitivity was further confirmed in the tail-flick assay, as BERK mice consistently displayed shortened tail-withdrawal latencies to the water bath maintained at 48°C (6.28 ± 0.56 s in BERK mice vs. 8.35 ± 0.63 s in non-sickle mice, P < 0.05), 52°C (1.84 ± 0.11 s in BERK mice vs. 2.31 ± 0.15 s in non-sickle mice, P < 0.05), and 55°C (1.02 ± 0.07 s in BERK mice vs. 1.33 ± 0.09 s in non-sickle mice, P < 0.05) (Fig. 2E). In addition, BERK mice showed a significant decrease in the latency to noxious input in the hot plate test at 52 °C (6.61 ± 0.90 s in BERK mice vs. 8.90 ± 0.55 s in non-sickle mice, P < 0.05) (Fig. 2F). Therefore, these data indicated that BERK mice have pronounced evoked hypersensitivity to mechanical and heat stimuli. 3.4. Inflammatory pain sensitivity in BERK mice As seen in SCD patients, BERK mice have been reported to have increased inflammation and inflammatory mediators.1,17,33 A logical question to ask is whether BERK mice responded differently to commonly used inflammatory pain stimuli such as formalin. Intraplantar formalin injection induced typical biphasic spontaneous pain in both BERK mice and non-sickle littermate controls (Fig. 3A). There was no difference between the BERK and non-sickle mice in Phase I (1–10 min) response (P > 0.05, Fig. 3B). For Phase II (10.01–60 min), there was a significantly increased response in BERK mice (P < 0.05, Fig. 3A). 3.5. Activation of CaMKIIα in BERK mice We hypothesized that CaMKIIα, a key cellular protein kinase and regulator of neuronal activity, may play an important role in promoting chronic pain in SCD. As a first step to address this question, we determined CaMKIIα activity in the spinal cord and dorsal root ganglia (DRG) of BERK mice and non-sickle control mice. Western blot analysis showed the expression of activated or phosphorylated CaMKIIα (pCaMKIIα) was significantly elevated in the spinal cord in BERK mice (Fig. 4A). Compared with the non-sickle mice, the level of spinal pCaMKIIα in BERK mice increased by 1.7-fold (P < 0.01). On the other hand, the expression of total CaMKIIα in BERK and control mice were not different (P > 0.05). Similar observations were found in DRG region (Fig. 4B). There was a prominent activation of CaMKIIα in DRG of the sickle mice, with a 2.3-fold increase of pCaMKIIα as compared with that in the non-sickle mice. We further conducted immunohistochemical analysis in BERK mice. The pCaMKIIα immunoreactivity was found mostly in the superficial dorsal horn of the spinal cord (Fig. 4C). Quantitative analysis of pCaMKIIα immunoreactivity was performed by counting the number of positively stained cells using the MetaMorph Imaging Software. The percentage of pCaMKIIα-positive cells was drastically enhanced in the sickle mice (48%, 105/220,Fig. 4C, right panel) compared with that of non-sickle mice (29%, 45/155, Fig. 4C left panel). The latter was similar to the percentage of pCaMKIIα-positive spinal neurons in the Sprague-Dawley rat.4 These data clearly demonstrated that spinal CaMKIIα activity is enhanced in BERK sickle cell mice. 3.6. Pharmacologic intervention of CaMKIIα in BERK mice To assess a possible functional role of CaMKIIα in sickle cell pain, we examined the effect of KN93, a potent and cell permeable inhibitor of CaMKIIα in BERK and control mice. CPP test was performed first using KN93 as the pairing drug. If CaMKIIα participates in the ongoing spontaneous pain in BERK mice, its inhibition is expected to suppress ongoing pain and produce CPP. In these experiments, BERK mice demonstrated a strong preference for KN93-paired chamber (415 ± 14 s) over the saline chamber (311 ± 17 s, P < 0.05, Fig. 5A), after pairing with KN93 (45nmol, i.t.) for 30 min. In contrast, the non-sickle mice spent similar amount of time in the saline chamber (336 ± 20 s) and the KN93-paired chamber (361 ± 22 s) (P > 0.05). As compared before and after conditioning, BERK mice preferred to stay in the KN93-paired chamber as illustrated by the significant difference score generated in the KN93-paired chamber (P < 0.05, Fig. 5B). In addition, KN92, a kinase-inactive derivative of KN93 failed to produce CPP in either BERK or control mice (Fig. 5C–D). Since KN93 selectively elicited CPP in BERK mice, these data suggested a functional role for CaMKIIα in mediating ongoing spontaneous pain in SCD. We next examined the evoked pain behaviors in BERK and control mice after receiving a bolus injection of KN93 (45 nmol, i.t.).The mechanical hypersensitivity in BERK mice (0.06 ± 0.01 g) was transiently, but effectively, reversed by KN93 (Fig. 5E). With an initial onset at around 30 min, the peak anti-allodynic effect of KN93 was observed at 2 h, when the paw withdrawal threshold in BERK mice (1.55 ± 0.26 g, P < 0.001) was restored to a level that was indistinguishable from that in the non-sickle mice (1.59 ± 0.27g). In response to the radiant heat stimuli, spinal administration of KN93 was capable of reversing thermal hyperalgesia in BERK mice (Fig. 5F). Significant attenuation was achieved at 2 h (11.17 ± 1.15 s in BERK mice treated with KN93 vs. 6.33 ± 0.79 s in BERK mice treated with KN92, P < 0.01). Moreover, KN93 did not change mechanical and thermal sensitivity in the non-sickle mice (Fig. 5E–F). Neither did KN92 affect baseline nociception in BERK or the non-sickle mice. Therefore, inhibiting spinal CaMKIIα is effective in transiently reversing evoked hypersensitivity in BERK sickle cell mice. 3.7. CaMKIIα silencing in BERK mice To further investigate the essential role of CaMKIIα in sickle cell pain, siRNA targeting CaMKIIα (i.t.) was applied to knock down the expression of CaMKIIα. This approach is known to produce a highly selective repression of the target protein. BERK and non-sickle littermate mice received CaMKIIα siRNA or scrambled RNA duplex (2 μg/injection, twice per day for 3 consecutive days, i.t.). Sensitivities to mechanical and thermal stimuli were measured daily. Treatment with CaMKIIα siRNA gradually attenuated mechanical allodynia (Fig. 6A) and thermal hyperalgesia (Fig. 6B) in BERK mice. When tested 24 h after first siRNA treatment, thermal hyperalgesia was significantly reversed in BERK mice (6.03 ± 0.33 s in siRNA group vs. 2.90 ± 0.37 g in scrambled RNA group, P < 0.01, Fig. 6B). After 3 days’ treatment, CaMKIIα knockdown was confirmed in the spinal cord by western blot analysis (Fig. 6A insert). In BERK mice, mechanical hypersensitivity was completely reversed (0.96 ± 0.11 g in siRNA group vs. 0.10 ± 0.03 g in scrambled RNA group, P < 0.001, Fig. 6A). Meanwhile, thermal hyperalgesia in BERK mice was significantly suppressed by CaMKIIα siRNA and lasted for 5 d (Fig. 6B). On the other hand, CaMKIIα siRNA did not alter mechanical or thermal sensitivity in the non-sickle mice (Fig. 6A–B). Furthermore, BERK and non-sickle mice were subjected to the CCP paradigm to investigate ongoing spontaneous pain on Day 4. BERK mice treated with scrambled RNA spent significantly more time in the lidocaine-paired chamber (412 ± 44 s) than in the saline chamber (231 ± 38 s) (P < 0.05, Fig. 6C), similar to BERK mice without any treatment (Fig. 1). BERK mice received 3d CaMKIIα siRNA failed to show preference for saline- (344 ± 36 s) or lidocaine-paired (340 ± 41 s) chambers (P > 0.05), which was in stark contrast to BERK mice received scrambled RNA or BERK mice received no treatment (Fig. 1). Lidocaine didn’t produce CPP in the non-sickle mice treated with CaMKIIα siRNA or scrambled RNA. Analysis of difference scores confirmed that CaMKIIα siRNA disrupted lidocaine-CPP in BERK mice (Fig. 6D). Since persistent ongoing spontaneous pain and evoked hypersensitivities were no longer present in BERK mice after knocking down spinal CaMKIIα, these data strongly indicated that CaMKIIα is required for chronic pain behaviors in SCD. 4. Discussion Pain in SCD is characterized by the presence of chronic pain with episodes of acute pain crises. The neurobiology of chronic pain in SCD is poorly understood. Studies employing animal models of SCD become an indispensable approach in vivo to unravel the complicated molecular mechanisms in the central and peripheral nervous systems. Early murine models of SCD, such as HbSAD mice and HbS/HbS-Antilles mice expressed a “supersickling” variant of HbS and developed organ and tissue characteristics of SCD. However, these mice retain the production of mouse globins, which interfere significantly with the sickling of erythrocytes. To overcome this limitation, the recent creation of transgenic mice, such as BERK mice completely replace mouse hemoglobins with human hemoglobins, producing a phenotype that closely mimics many features of severe SCD in humans.25,33,37 In the present study, we profiled two distinctive pain components (ongoing spontaneous pain and evoked pain) in BERK mouse. When asked to characterize their pain, patients described daily, chronic pain without any known stimulus.42,48 As a dominant clinical manifestation, it is imperative to investigate the presence of ongoing spontaneous pain in animal models of SCD. Our study is the first to demonstrate the presence of ongoing spontaneous pain in BERK mice using the CPP paradigm that we have validated in mice with chronic tissue inflammation or nerve injury.21 Indeed, a single trial conditioning with lidocaine was sufficient to produce CPP selectively in BERK mice, but not non-sickle mice (Fig 1). This is not only in agreement with the clinic characteristics of pain in patients, but also the first study to demonstrate the feasibility of applying CPP paradigm to detect ongoing spontaneous pain in a transgenic mouse model. BERK mice showed heightened responses to formalin in the second phase, but not the first phase, suggesting a central mechanism. Unlike the Phase I that is mostly due to direct activation of primary afferent nociceptors, the second phase is considered as an index for inflammatory response involving central mechanisms.44 To identify heat hypersensitivity in BERK mice, we used three different types of heat stimuli with varying intensities. These results corroborated with each other to demonstrate that BERK mice displayed prominent heat hyperalgesia as well as mechanical allodynia, as have been reported previously by two other groups,24,30 reflecting the presence of fully developed evoked pain behaviors in mice with SCD. The enhanced responses and/or lowered thresholds to heat and mechanical stimuli were also found in patients with SCD by thermal and mechanical quantitative sensory testing.2,12 Allodynia to noxious cold was also found in the study. In an operant assay using two plates with different temperature settings, BERK mice spent less time on the plate set at 23 °C than that at 30 °C.49 Another group observed changes in mouse facial expression, body length and back curvature in response to noxious cold (4 °C).34 Together with the findings of the current study, it is clear that BERK mice exhibited hypersensitivity to both mild and extremely cold stimuli, which was also reported by patients with SCD.2,12,35 Our findings provided the first evidence that CaMKIIα is a critical mediator for chronic pain in SCD. CaMKIIα is a multifunctional, Ca2+/calumodulin (CaM) activated serine/threonine protein kinase that is a key component of intracellular Ca2+ signaling pathways.9,16,31 It is involved in a variety of Ca2+-mediated cellular processes including the biosynthesis of neurotransmitters, hormone secretion, neurotransmitter release, gene expression and neuronal plasticity.18,31 Autophosphorylation at threonine 286/287 of CaMKIIα renders the kinase fully active even in the absence of Ca2+. Robust activation of CaMKIIα was identified in the spinal cord of BERK mice (Fig. 4), suggesting a correlation between increased CaMKIIα activity and persistent pain state in SCD. Moreover, spinal inhibition or knock-down of CaMKIIα was able to abolish both ongoing spontaneous pain and evoked hypersensitivities in BERK mice, further illustrating a functional role of spinal CaMKIIα in promoting chronic pain in SCD. CaMKIIα is specifically expressed in the superficial laminae in the spinal dorsal horn and in the small to medium diameter primary sensory neurons in the primary afferent, where nociceptive signals are transmitted and processed.3,4 CaMKIIα activity was significantly increased in the spinal cord within minutes after an intradermal injection of capsaicin.13 Spinally administered KN93 inhibited the enhanced response in the spinal nociceptive neurons and changes in exploratory behavior evoked by capsaicin.13 KN93 selectively and directly binds to the CaM-binding site of CaMKII, preventing the activation of CaMKII.43 We have previously reported that CaMKIIα is required for the initiation and/or maintenance of chronic inflammatory pain, L5/L6 nerve ligation-induced neuropathic pain, and opioid-induced hyperalgesia.6,7,32 Furthermore, CaMKIIα has been reported to contribute to hyperalgesia priming for transition from acute to chronic pain.14,15 Establishing CaMKIIα mechanisms will open a new scientific arena to explore in future studies the upstream and downstream signaling components of CaMKIIα-mediated pain transmission in SCD. Both the release of Ca2+ from intracellular storages and Ca2+ influx from extracellular sources as a result of receptor or ion channel activation may activate CaMKIIα. Activation of the transient receptor potential vanilloid 1 (TRPV1) or the N-methyl-D-aspartate (NMDA) receptors can lead to Ca2+ influx and CaMKII activation.27,47 The latter, in turn, phosphorylates and activates the TRPV1 and the NMDA receptors.19,29 Therefore, there exist potential feed-forward loops to keep CaMKIIα-mechanisms active after the original Ca2+ signaling has subsided or disappeared, which is crucial to maintain chronicity in pain conditions such as that in SCD. Hargreaves and colleagues found that capsaicin increased CaMKIIα activity in TRPV1-positive neurons in rat trigeminal ganglion neurons.40 In these trigeminal ganglion neurons, capsaicin- or n-arachidonoyl-dopamine (NADA)-evoked calcitonin gene-related peptide (CGRP) release was inhibited by KN93.40 Indeed, TRPV1has been identified as a contributor to the mechanical hypersensitivity in SCD. TRPV1 antagonist A-425619 attenuated mechanical hypersensitivity in BERK mice 30–60 minutes after administration (i.p.), implicating a unique TRPV1-mediated mechanism for mechanical hypersensitivity in SCD.24 Several other mechanisms have also been proposed for SCD pain. BERK mice were found to have decreased expression of the mu opioid receptor and increased immunoreactivity for CGRP, substance P, and several activators of neuropathic and inflammatory pain, which may contribute to hyperalgesia.5,30 These investigations found that morphine, the cannabinoid receptor agonist CP 5594015, or inhibitors of mast cell activation attenuated chronic pain behaviors in BERK mice.30,45 As an intracellular protein kinase, CaMKIIα may be activated by some of these inflammatory or pronociceptive mediators, although future studies are needed to investigate potential interactions. In summary, we characterized pain behaviors in BERK mice using age- and sex-controlled non-sickle littermate mice as controls. Moreover, a CaMKIIα-mediated cellular mechanism for chronic pain in SCD was identified by two complementary approaches employing pharmacological inhibition and genetic silencing. These findings may lead to translational research targeting CaMKIIα for alleviating chronic pain in SCD. The latter is poorly managed in the clinic because it is frequently refractory to the currently available pain medications, underscoring an urgent need for novel medications for treating chronic pain in SCD. Although specific inhibitors of CaMKIIα have not made to the clinical testing, we have previously identified a FDA-approved antipsychotic drug trifluoperazine as a potent inhibitor of CaMKIIα.6,32 In a small mechanism-based translational study, trifluoperazine was not only found to be safe, but also showed some promise in alleviating chronic pain after a single dose in patients with SCD.36 Additional studies with trifluoperazine and other CaMKIIα inhibitors may ultimately lead to rational designing, identifying, and optimizing clinically useful treatment for the chronic pain in SCD. This work was supported by a grant (R01HL098141) from the National Heart Lung and Blood Institute (NHLBI), National Institutes of Health (NIH). Y.H. is a Sickle Cell Scholar supported by U01HL117658 from NHLBI. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NHLBI or NIH. Figure 1 Lidocaine induced conditioned place preference (CPP) in BERK sickle cell mice. (A) BERK mice spent significantly more time in lidocaine-paired chamber, whereas the non-sickle control littermates showed no chamber preference to lidocaine. ** P < 0.01, saline paired vs. lidocaine paired group, two-way ANOVA followed by Bonferroni post hoc test; n = 6. (B) Difference score analysis confirmed that the sickle mice, but not non-sickle mice produced CPP to lidocaine. * P < 0.05, test time vs. preconditioning time in the lidocaine paired chamber, paired t test; n = 6. Figure 2 Evoked pain behaviors displayed in BERK sickle cell mice. A cold stimulus was applied by a brief (1 s) spray of tetrafluoroethane to the ventral surface of hindpaw. (A) The number of licks and (B) the duration of lifting of the sprayed paw were recorded. *** P < 0.001, “sickle” group vs. “non-sickle”, n = 9. (C) When compared with the non-sickle control littermates, BERK mice exhibited significantly reduced threshold of response to probing by von Frey filaments, P < 0.001, n = 12. (D) BERK mice showed significantly reduced withdrawal latency to radiant heat, indicating the presence of thermal hyperalgesia, ***P < 0.001 vs. “non-sickle” mice group, n = 12. (E) In the tail immersion test performed at 48, 52, or 55 °C. BERK mice showed reduced latencies to a rapid tail flick response as compared with non-sickle control mice. *P < 0.05, n = 9. (F) BERK mice showed significantly reduced withdrawal latency to the hot plate maintained at 52 °C, * P < 0.05 vs. “non-sickle” group, n = 9. Figure 3 BERK sickle cell mice responded to the inflammatory pain stimulus induced by a subcutaneous injection of formalin (2% in saline, 20 μL/mouse) into the dorsal surface of the left hindpaw. (A) Mice paw flinching was observed for 60 min, the number of flinches was recorded for successive 5 min intervals. (B) The total number of flinches during the early phase (0–10 min) and late phase (10.01–60 min) were summed, respectively.* P < 0.05 “sickle” vs. “non-sickle” group, n = 10. Figure 4 Activation of CaMKIIα was observed in BERK sickle cell mice. Western blot analysis showed significant increase of phosphorylated CaMKIIα (pCaMKIIα) in the spinal cord (A) and DRG (B) of BERK mice as compared with the non-sickle control mice, ** P < 0.01, n =3. (C) Compared with the non-sickle mice (left panel), elevated immunohistochemical staining of pCaMKIIα was found in the superficial lamina region of the dorsal spinal cord in the sickle cell mice (right panel). Quantitative analysis of pCaMKIIα immunoreactivity was performed by counting the number of positively stained cells using the MetaMorph Imaging Software. Figure 5 KN93 suppressed chronic pain in BERK sickle cell mice. (A) KN93 (45nmol in 5 μL saline, i.t.) produced CPP in the sickle mice. The sickle mice spent significantly more time in KN93-paired chamber, whereas the non-sickle control mice showed no chamber preference, spending similar amount of time in saline- and KN93-paired chambers. (B) Difference scores confirmed the presence of KN93-CPP in the sickle mice, but not the non-sickle control mice. * P < 0.05; n = 6.(C) KN92, an inactive analog of KN93 (45 nmol in 5 μL saline, i.t.), did not produce CPP in the sickle mice. Neither the sickle mice nor the non-sickle mice exhibited chamber preference. (D) Difference scores confirmed the absence of KN92-CPP in the sickle mice. n = 6. Mice paw withdrawal threshold to von Frey filament probing (E) and withdrawal latency to radiant heat (F) were measured before (0) and 0.5, 1, 2, 4 and 24 h after the injection of KN93 (45 nmol in 5 μL of saline, i.t.). * P < 0.05, ** P < 0.01, ***P < 0.001, compared with the “non-sickle w/KN93” group; # P < 0.05, ## P < 0.01, ### P < 0.001, compared with the “sickle w/KN92” group; n = 6. Figure 6 CaMKIIα knockdown by siRNA reversed mechanical allodynia (A), thermal hyperalgesia (B) and ongoing spontaneous pain (C–D) in BERK sickle cell mice. Mice were treated with CaMKIIα siRNA or scrambled (scr) RNA duplex (2 μg, i.t. twice per day for 3 days). Mechanical and thermal sensitivities were tested daily. * P < 0.05, ** P < 0.01, ***P < 0.001, compared with the “non-sickle w/siRNA” group; # P < 0.05, ## P < 0.01, ### P < 0.001, compared with the “sickle w/scr” group; scr: scrambled RNA duplex; n = 6. Arrows indicated siRNA/scr injections. The expression of CaMKIIα was reduced by 3d-treatment with CaMKIIα siRNA, compared with the same treatment with scrambled siRNA (A-insert). (C) When tested on Day 4 by lidocaine-CPP paradigm, spontaneous ongoing pain was present in “sickle w/scr” mice, but not “sickle w/siRNA” mice. (D) Difference score confirmed the absence of chamber preference in BERK mice after CaMKIIα siRNA treatment, * P < 0.05, n = 6. Conflict of Interest Statement The authors declare no conflict of interest. 1 Belcher JD Bryant CJ Nguyen J Transgenic sickle mice have vascular inflammation Blood 2003 101 3953 3959 12543857 2 Brandow AM Stucky CL Hillery CA Hoffmann RG Panepinto JA Patients with sickle cell disease have increased sensitivity to cold and heat Am J Hematol 2013 88 37 43 23115062 3 Bruggemann I Schulz S Wiborny D Hollt V Colocalization of the mu-opioid receptor and calcium/calmodulin-dependent kinase II in distinct pain-processing brain regions Brain Res Mol Brain Res 2000 85 239 250 11146127 4 Carlton SM Localization of CaMKIIalpha in rat primary sensory neurons: increase in inflammation Brain Res 2002 947 252 259 12176168 5 Cataldo G Rajput S Gupta K Simone DA Sensitization of nociceptive spinal neurons contributes to pain in a transgenic model of sickle cell disease Pain 2015 156 722 730 25630029 6 Chen Y Luo F Yang C Kirkmire CM Wang ZJ Acute inhibition of Ca2+/calmodulin-dependent protein kinase II reverses experimental neuropathic pain in mice J Pharmacol Exp Ther 2009 330 650 659 19478130 7 Chen Y Yang C Wang ZJ Ca2+/calmodulin-dependent protein kinase II alpha is required for the initiation and maintenance of opioid-induced hyperalgesia J Neurosci 2010 30 38 46 20053885 8 Chen Y Yang C Wang ZJ Proteinase-activated receptor 2 sensitizes transient receptor potential vanilloid 1, transient receptor potential vanilloid 4, and transient receptor potential ankyrin 1 in paclitaxel-induced neuropathic pain Neuroscience 2011 9 Colbran RJ Targeting of calcium/calmodulin-dependent protein kinase II Biochem J 2004 378 1 16 14653781 10 Corder G Doolen S Donahue RR Constitutive mu-opioid receptor activity leads to long-term endogenous analgesia and dependence Science 2013 341 1394 1399 24052307 11 Dixon WJ Efficient analysis of experimental observations Annu Rev Pharmacol Toxicol 1980 20 441 462 7387124 12 Ezenwa MO Molokie RE Wang ZJ Safety and Utility of Quantitative Sensory Testing among Adults with Sickle Cell Disease: Indicators of Neuropathic Pain? Pain Pract 2016 16 282 293 25581383 13 Fang L Wu J Lin Q Willis WD Calcium-calmodulin-dependent protein kinase II contributes to spinal cord central sensitization J Neurosci 2002 22 4196 4204 12019337 14 Ferrari LF Araldi D Levine JD Distinct terminal and cell body mechanisms in the nociceptor mediate hyperalgesic priming J Neurosci 2015 35 6107 6116 25878283 15 Ferrari LF Bogen O Levine JD Role of nociceptor alphaCaMKII in transition from acute to chronic pain (hyperalgesic priming) in male and female rats J Neurosci 2013 33 11002 11011 23825405 16 Fink CC Meyer T Molecular mechanisms of CaMKII activation in neuronal plasticity Curr Opin Neurobiol 2002 12 293 299 12049936 17 Frenette PS Atweh GF Sickle cell disease: old discoveries, new concepts, and future promise J Clin Invest 2007 117 850 858 17404610 18 Fukunaga K Miyamoto E A working model of CaM kinase II activity in hippocampal long-term potentiation and memory Neurosci Res 2000 38 3 17 10997573 19 Fukunaga K Soderling TR Miyamoto E Activation of Ca2+/calmodulin-dependent protein kinase II and protein kinase C by glutamate in cultured rat hippocampal neurons J Biol Chem 1992 267 22527 22533 1358879 20 Hargreaves K Dubner R Brown F Flores C Joris J A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia Pain 1988 32 77 88 3340425 21 He Y Tian X Hu X Porreca F Wang ZJ Negative reinforcement reveals non-evoked ongoing pain in mice with tissue or nerve injury J Pain 2012 13 598 607 22609247 22 He Y Wang ZJ Nociceptor Beta II, Delta, and Epsilon Isoforms of PKC Differentially Mediate Paclitaxel-Induced Spontaneous and Evoked Pain J Neurosci 2015 35 4614 4625 25788678 23 He Y Yang C Kirkmire CM Wang ZJ Regulation of Opioid Tolerance by let-7 Family MicroRNA Targeting the {micro} Opioid Receptor J Neurosci 2010 30 10251 10258 20668208 24 Hillery CA Kerstein PC Vilceanu D Transient receptor potential vanilloid 1 mediates pain in mice with severe sickle cell disease Blood 2011 118 3376 3383 21708890 25 Hsu L Diwan B Ward JM Noguchi CT Pathology of “Berkeley” sickle-cell mice includes gallstones and priapism Blood 2006 107 3414 3415 16597602 26 Hylden JL Wilcox GL Intrathecal morphine in mice: a new technique Eur J Pharmacol 1980 67 313 316 6893963 27 Jones TL Lustig AC Sorkin LS Secondary hyperalgesia in the postoperative pain model is dependent on spinal calcium/calmodulin-dependent protein kinase II alpha activation Anesth Analg 2007 105 1650 1656 table of contents 18042863 28 King T Vera-Portocarrero L Gutierrez T Unmasking the tonic-aversive state in neuropathic pain Nat Neurosci 2009 12 1364 1366 19783992 29 Kitamura Y Miyazaki A Yamanaka Y Nomura Y Stimulatory effects of protein kinase C and calmodulin kinase II on N- methyl-D-aspartate receptor/channels in the postsynaptic density of rat brain J Neurochem 1993 61 100 109 7685812 30 Kohli DR Li Y Khasabov SG Pain-related behaviors and neurochemical alterations in mice expressing sickle hemoglobin: modulation by cannabinoids Blood 2010 116 456 465 20304807 31 Lisman J Schulman H Cline H The molecular basis of CaMKII function in synaptic and behavioural memory Nat Rev Neurosci 2002 3 175 190 11994750 32 Luo F Yang C Chen Y Reversal of chronic inflammatory pain by acute inhibition of Ca2+/calmodulin-dependent protein kinase II J Pharmacol Exp Ther 2008 325 267 275 18178903 33 Manci EA Hillery CA Bodian CA Zhang ZG Lutty GA Coller BS Pathology of Berkeley sickle cell mice: similarities and differences with human sickle cell disease Blood 2006 107 1651 1658 16166585 34 Mittal A Gupta M Lamarre Y Jahagirdar B Gupta K Quantification of pain in sickle mice using facial expressions and body measurements Blood Cells Mol Dis 2016 57 58 66 26852657 35 Molokie RE Wang ZJ Wilkie DJ Presence of neuropathic pain as an underlying mechanism for pain associated with cold weather in patients with sickle cell disease Med Hypotheses 2011 77 491 493 21763079 36 Molokie RE Wilkie DJ Wittert H Mechanism-driven phase I translational study of trifluoperazine in adults with sickle cell disease Eur J Pharmacol 2014 723 419 424 24211787 37 Paszty C Transgenic and gene knock-out mouse models of sickle cell anemia and the thalassemias Curr Opin Hematol 1997 4 88 93 9107524 38 Paszty C Brion CM Manci E Transgenic knockout mice with exclusively human sickle hemoglobin and sickle cell disease Science 1997 278 876 878 9346488 39 Platt OS Thorington BD Brambilla DJ Pain in sickle cell disease. Rates and risk factors N Engl J Med 1991 325 11 16 1710777 40 Price TJ Jeske NA Flores CM Hargreaves KM Pharmacological interactions between calcium/calmodulin-dependent kinase II alpha and TRPV1 receptors in rat trigeminal sensory neurons Neurosci Lett 2005 389 94 98 16095822 41 Smith WR Bauserman RL Ballas SK Climatic and geographic temporal patterns of pain in the Multicenter Study of Hydroxyurea Pain 2009 146 91 98 19683393 42 Smith WR Penberthy LT Bovbjerg VE Daily assessment of pain in adults with sickle cell disease Ann Intern Med 2008 148 94 101 18195334 43 Sumi M Kiuchi K Ishikawa T The newly synthesized selective Ca2+/calmodulin dependent protein kinase II inhibitor KN-93 reduces dopamine contents in PC12h cells Biochem Biophys Res Commun 1991 181 968 975 1662507 44 Tang L Chen Y Chen Z Blumberg PM Kozikowski AP Wang ZJ Antinociceptive pharmacology of N-(4-chlorobenzyl)-N′-(4-hydroxy-3-iodo-5-methoxybenzyl) thiourea, a high-affinity competitive antagonist of the transient receptor potential vanilloid 1 receptor J Pharmacol Exp Ther 2007 321 791 798 17312187 45 Vang D Paul JA Nguyen J Small-molecule nociceptin receptor agonist ameliorates mast cell activation and pain in sickle mice Haematologica 2015 100 1517 1525 26294734 46 Wang Z Gardell LR Ossipov MH Pronociceptive actions of dynorphin maintain chronic neuropathic pain J Neurosci 2001 21 1779 1786 11222667 47 Wang ZJ Molokie RE Wilkie DJ Does cold hypersensitivity increase with age in sickle cell disease? Pain 2014 155 2439 2440 25172823 48 Wilkie DJ Molokie R Boyd-Seal D Patient-reported outcomes: descriptors of nociceptive and neuropathic pain and barriers to effective pain management in adult outpatients with sickle cell disease J Natl Med Assoc 2010 102 18 27 20158132 49 Zappia KJ Garrison SR Hillery CA Stucky CL Cold hypersensitivity increases with age in mice with sickle cell disease Pain 2014 155 2476 2485 24953902

PMC005xxxxxx/PMC5118062.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101217933 32339 J Neural Eng J Neural Eng Journal of neural engineering 1741-2560 1741-2552 27705958 5118062 10.1088/1741-2560/13/6/066002 NIHMS822278 Article Chronic In Vivo Stability Assessment of Carbon Fiber Microelectrode Arrays Patel Paras R. a Zhang Huanan b Robbins Matthew T. a Nofar Justin B. c Marshall Shaun P. d Kobylarek Michael J. a Kozai Takashi D. Y. efgh Kotov Nicholas A. abij Chestek Cynthia A. aklm a Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, United States b Department of Chemical Engineering, University of Michigan, Ann Arbor, MI, United States c Psychology Department, University of Michigan, Ann Arbor, MI, United States d Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI, United States e Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA, United States f Center for the Neural Basis of Cognition, University of Pittsburgh, Pittsburgh, PA, United States g McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, PA, United States h NeuroTech Center of the University of Pittsburgh Brain Institute, University of Pittsburgh, Pittsburgh, PA, United States i Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, United States j Department of Macromolecular Science and Engineering, University of Michigan, Ann Arbor, MI, United States k Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI, United States l Neurosciences Program, University of Michigan, Ann Arbor, MI, United States m Robotics Program, University of Michigan, Ann Arbor, MI, United States 16 10 2016 05 10 2016 12 2016 01 12 2017 13 6 066002066002 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objective Individual carbon fiber microelectrodes can record unit activity in both acute and semi-chronic (∼1 month) implants. Additionally, new methods have been developed to insert a 16 channel array of carbon fiber microelectrodes. Before assessing the in vivo long-term viability of these arrays, accelerated soak tests were carried out to determine the most stable site coating material. Next, a multi-animal, multi-month, chronic implantation study was carried out with carbon fiber microelectrode arrays and silicon electrodes. Approach Carbon fibers were first functionalized with one of two different formulations of PEDOT and subjected to accelerated aging in a heated water bath. After determining the best PEDOT formula to use, carbon fiber arrays were chronically implanted in rat motor cortex. Some rodents were also implanted with a single silicon electrode, while others received both. At the end of the study a subset of animals were perfused and the brain tissue sliced. Tissue sections were stained for astrocytes, microglia, and neurons. The local reactive responses were assessed using qualitative and quantitative methods. Main results Electrophysiology recordings showed the carbon fibers detecting unit activity for at least 3 months with average amplitudes of ∼200 μV. Histology analysis showed the carbon fiber arrays with a minimal to non-existent glial scarring response with no adverse effects on neuronal density. Silicon electrodes showed large glial scarring that impacted neuronal counts. Significance This study has validated the use of carbon fiber microelectrode arrays as a chronic neural recording technology. These electrodes have demonstrated the ability to detect single units with high amplitude over 3 months, and show the potential to record for even longer periods. In addition, the minimal reactive response should hold stable indefinitely, as any response by the immune system may reach a steady state after 12 weeks. Flexible electrodes Minimal injury High density array Neural electrodes 1. Introduction Recording stable, low-noise, high-amplitude unit activity in the motor cortex is crucial for the long-term stability of any brain machine interface (BMI) system [1–9] and can be equally important in many neuroscience studies [10–13]. To accomplish this goal, a system of electrodes should ideally elicit little to no immune response, have the capacity to concurrently record from a large population of neurons to either access more information content or to better understand local population dynamics, and demonstrate the ability to chronically record neural activity. The initial insertion of any electrode is a traumatic event to the local cellular network and vasculature and is greatly influenced by insertion speed [14,15], location [16], and technique [17,18]. If the electrode is removed soon after insertion, the local area will heal [19,20]. Permanent implantation of the electrode leads to the eventual formation of a localized glial scar comprised chiefly of astrocytes and microglia [20–32]. Accompanying the scar is a varying degree of neuronal cell death within the immediate vicinity of the electrode [20,21,25,30]. The persistence of the scar can be attributed to multiple factors including the continual release of inflammatory factors by the locally activated glial cells [27,28,33] and a breached blood brain barrier that cannot completely heal, therefore allowing the infiltration of pro-inflammatory cells and chemokines [19,34,35]. The impact that these chemokines have on the local environment has been shown through the use of genetic knockouts. The removal of monocyte chemoattractant protein 1 led to improved neuronal density [36] while caspase-1 knockout mice demonstrated significantly better recording quality as compared to wild type mice [37]. One way to mitigate the inflammatory response is through a reduced probe footprint. The reduction in probe size has been shown to reduce the mechanical strain on nearby neurons [37], lessen the long-term glial response, and improve the survival rates of the local neuronal population [38–40]. It should be noted that these studies demonstrating an improved tissue response made use of hard materials, such as silicon [39,40], and softer materials, such as SU-8 and parylene [38]. While an electrode's material properties may play an important role in bridging the mechanical mismatch between an implant and the brain, the previous studies on electrode dimensions point to probe size as being a more critical factor. We have recently proposed a multi-electrode array design using carbon fibers as the basis for the recording electrode [41,42]. Carbon fiber electrodes are small (d = 6.8 μm), and with the addition of a parylene-c insulating coating (t = 800 nm), the overall diameter is only increased to 8.4 μm. In addition, this electrode material is extremely amenable to creating high density arrays and with a site coating of poly(3,4-ethylenedioxythiophene) (PEDOT) has been shown to record high quality unit activity [41,43,44]. This work evaluates the longevity of these arrays, by first testing two different formulations of the site coating material, PEDOT, using an accelerated aging test. Carbon fiber arrays functionalized with parylene-c and PEDOT were further evaluated by chronically implanting them into rat motor cortex. Additionally, some animals were implanted with a commercially available planar silicon electrode in the contralateral hemisphere's motor cortex. Impedance and electrophysiology recordings were taken on a regular basis and analyzed to demonstrate the carbon fiber's viability as a chronic electrode technology. Lastly, a subset of animals were perfused and stained to quantitatively analyze the glial response and neuronal density surrounding both electrode types. 2. Materials & Methods 2.1 Soak Test 2.1.1 Probe Assembly Printed circuit boards (PCBs) for accelerated soak testing were first roughened in the non-trace and non-bond pad areas with a Dremel tool, to allow for better adherence of the final epoxy coating (figure 1(a)). Once roughened, eight individual carbon fibers (T-650/35 3K, Cytec Thornel, Woodland Park, NJ), with length of approximately 1 cm, were placed on the individual bond pads using conductive silver epoxy (H20E, Epoxy Technology, Billerica, MA) (figure 1(b)). The conductive epoxy was then oven cured using the manufacturer's recommended settings. The silver epoxy bond was then covered with insulating epoxy (353NDT, Epoxy Technology, Billerica, MA) and oven cured using the manufacturer's recommended settings (figure 1(c)). Probes were then insulated with a conformal coating of parylene-c (t=800 nm) using a Parylene Deposition System 2010 (SCS Coatings, Indianapolis, IN). After insulation, the tips of each probe were cut to re-expose a bare carbon fiber site. At this site, one of two solutions was electrodeposited. The first was a solution of 0.01 M 3,4-ethylenedioxythiophene (483028, Sigma-Aldrich, St. Louis, MO):0.1 M sodium p-toluenesulfonate (152536, Sigma-Aldrich, St. Louis, MO). The second solution was composed of 0.01 M 3,4-ethylenedioxythiophene (483028, Sigma-Aldrich, St. Louis, MO):0.1 M polystyrene sulfonate (m.w. 70.000, 222271000, Acros, NJ). For each solution the electrodeposition was carried out by applying 100 pA/channel for 600 seconds to form a layer of poly(3,4-ethylenedioxythiophene):sodium p-toluenesulfonate (PEDOT:pTS) or poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS). All channels to be coated with a given solution were shorted together during the electrodeposition step and the current delivered was scaled accordingly. 2.1.2 Accelerated Soak Test Setup Boards with parylene-c and PEDOT:pTS or PEDOT:PSS coated carbon fibers were mounted to the underside of a jar lid (figure 1(d)). The lids were then secured to jars that contained 1× phosphate buffered saline (PBS) solution (BP3994, Fisher, Waltham, MA) (figure 1(e)). The 1× PBS was at a level such that only the fibers were submerged and not the entire printed circuit board. The jars were then lowered into a water bath maintained at 60 °C. At each time point the fibers were removed from the heated 1× PBS and rinsed once with deionized water. Next, the fibers' impedances were recorded. Once recordings were complete the assembly was returned to the heated 1× PBS. According to works by Green et al. [45] and Hukins et al. [46], equation (1) can be used to determine the aging time that the fibers have undergone: (1) t37=tT×Q10(T−37)/10 Where t37 is the simulated aging time at 37 °C, tT is the amount of real time that the samples have been kept at the elevated temperature, T, and Q10 is an aging factor that is equal to 2, according to ASTM guidelines for polymer aging [47]. Calculating the simulated time for tT = 1 and T = 60 °C results in t37 = 4.92. This value of 4.92 is the acceleration factor and all real time measurements are scaled by this amount to obtain the simulated time. 2.1.3 Electrochemical Impedance Spectroscopy (EIS) EIS measurements were taken with a PGSTAT12 Autolab (EcoChemie, Utrecht, Netherlands), controlled by vendor-supplied NOVA software. Measurements were obtained by applying a 10 mVRMS signal from 10 Hz to 31 kHz. Custom Matlab (Mathworks, Natick, MA) scripts were used to determine frequency specific impedance values. All reported values are mean ± standard error of the mean. 2.2 Chronic Electrode Implantation 2.2.1 Carbon Fiber Array Preparation Carbon fiber arrays were fabricated as previously described [42]. Briefly, individual fibers were secured to bare traces (152.4 μm pitch) of a custom made PCB with silver epoxy that was then heat cured. This exposed contact was then coated with a heat cured insulating epoxy to protect the connection between the fiber and trace. Once fully assembled, all carbon fiber arrays for neural recordings were coated with an 800 nm thick insulating layer of parylene-c using a Parylene Deposition System 2010 (SCS Coatings, Indianapolis, IN). Probe tips also received a site coating of PEDOT:pTS with the same formula and deposition parameters used on the fibers that underwent soak testing. The final preparation step was a coating of poly(ethylene glycol) (PEG) as described in [42]. 2.2.2 Surgery for Chronic Implantation of Carbon Fibers and Silicon Probes Chronic implantation of carbon fiber arrays (figure 2(a)) and silicon probes (figure 2(b)) used adult male Long Evans rats (n=3 rats with only carbon fibers, 3 with both electrodes, and 2 with only silicon probes) weighing 300 – 350 g. Rats were first anesthetized using 5% isoflurane (v/v) for induction and then 1 – 3% isoflurane (v/v) for maintenance. The head was then shaved and triple swabbed using alternating applications of betadine and 70% ethanol. Ointment was applied to the eyes to keep them from drying during surgery. Once mounted in the stereotax, the shaved area was swabbed one more time with betadine and 70% ethanol. A subcutaneous injection of lidocaine (4 mg/mL) was given at the incision site at a maximum dosage of 4 mg lidocaine per 1 kg of body weight. After incision, the skin flaps were pulled away using hemostats and the skull surface cleaned. A burr bit (19008-07, Fine Science Tools, Foster City, CA) was used to drill seven holes around the periphery of the skull for seven bone screws (19010-00, Fine Science Tools, Foster City, CA). Next, 2 mm × 3 mm craniotomies were made over the left and right motor cortex using coordinates from a reference atlas [48]. Before resecting the dura, a layer of Kwik-Sil (World Precision Instruments, Sarasota, FL) was applied to the skull at the posterior, anterior, and leftmost sides. Following the resection of the dura on the rightmost craniotomy, the PEG coated carbon fiber array was brought to the surface of the brain. The exposed fibers were implanted according to methods previously described [42]. The silicon probes (A1×16-3mm-50-177-HZ16_21mm, NeuroNexus Technologies, Ann Arbor, MI) were 3 mm in length with sixteen 177 μm2 iridium sites spaced 50μm apart, starting from the tip. To implant the silicon probe, a small metal rod was attached to the stereotactic manipulator and positioned above the rat's skull. A drop of melted PEG was applied to the very tip of the rod. The base of the silicon probe was then positioned to rest in the still liquid PEG, which secured the probe as it solidified. The dura over the leftmost craniotomy was resected and the probe was implanted to the desired depth. The polyimide cable connecting the probe to the PCB was secured to the nearest bone screws using Kwil-Sil. After the Kwil-Sil had cured, the PEG was dissolved away using sterile Lactated Ringer's. Additional Kwik-Sil was then applied to the skull at the lateral side of the rightmost craniotomy, forming a complete barrier around both craniotomies. The Kwik-Sil barrier was flooded with either Kwik-Cast (World Precision Instruments, Sarasota, FL), petroleum jelly, or alginate [49]. Reference and ground wires from both PCBs were attached to the posterior most bone screw. The PCBs were then anchored to all of the skull's bone screws using dental acrylic. The skin flaps were brought up over the dental acrylic headcap on each side and sutured together at the anterior and posterior ends. Triple antibiotic ointment was liberally applied around the headcap. Animals were then removed from the stereotax and allowed to recover on a heated pad placed under their cage. During surgery, animal vitals were monitored using a pulse-oximeter and rectal temperature probe. All procedures and post-operative care complied with the University of Michigan's University Committee on Use and Care of Animals. A detailed breakdown of each animal's implant(s) and implant depth can be found in table 1. 2.2.3 Electrophysiology Recordings and Spike Sorting Electrophysiology recordings using chronic implants of carbon fiber arrays and silicon probes were done while the rats were awake and moving about freely in their cage. All acquisition of electrophysiology recordings were taken using a ZC16 headstage, RA16PA preamplifier, and RX5 Pentusa base station (Tucker-Davis Technologies, Alachua, FL). During data acquisition, the pre-amplifier high pass filtered at 2.2 Hz, anti-aliased filtered at 7.5 kHz, and sampled at a rate of ∼25 kHz. Each recording session lasted 5 or 10 minutes. Recording sessions were imported into Offline Sorter (Plexon, Dallas, TX) and first high-pass filtered (250 Hz corner, 4th order Butterworth). Each channel was manually thresholded and the resultant waveforms sorted by a trained operator. Sorted waveforms belonging to the same neuronal unit were averaged together to obtain a peak-to-peak amplitude for that unit, which was averaged with all other unit peak-to-peak amplitude values to obtain the mean value for each recording day for each probe type. All reported values are mean ± standard error of the mean. 2.2.4 Noise Floor and Signal-to-Noise Ratio Calculations To determine the noise floor for each recording channel, a trained operator picked out five 100 ms snippets of filtered electrophysiology recording data that did not contain sorted units and did not display amplifier saturation indicative of a motion artifact. The snippets of data, best characterized as non-spiking neural activity, were then joined together in a single 500 ms block which was used to calculate VRMS-Channel. The signal-to-noise ratio (SNR) of each sorted unit was calculated by dividing the peak-to-peak voltage of the waveform by 3·VRMS-Channel. All reported noise and SNR values are mean ± standard error of the mean. 2.2.5 Channel Exclusion and Count It was discovered throughout the study that certain datasets were corrupted by either the use of a broken headstage or from fibers themselves that showed signs of breakage. A full explanation of these types of problems and how they were mitigated can be found in the supplementary section. The primary goal in removing corrupted datasets or channels was to avoid skewing the analysis in any one direction. The number of channels used for impedance analysis at each time point can be seen in figure S1. The number of channels used for calculating the percentage of channels with units and the noise levels at each time point can be seen in figure S2. The number of units detected used for amplitude analysis at each time point can be seen in figure S3. 2.3 SEM Imaging A FEI Nova 200 Nanolab Focused Ion Beam Workstation and Scanning Electron Microscope (FEI, Hillsboro, OR) was used for SEM imaging. Prior to imaging, samples were gold sputter coated with a SPI-Module Sputter Coater (SPI Supplies, West Chester, PA). 2.4 Histology 2.4.1 Perfusion and Tissue Staining At day 91, 2 animals were transcardially perfused with 250 – 300 mL of 1× phosphate buffered saline (PBS) (BP3994, Fisher Scientific, Waltham, MA) followed by 250 – 300 mL of 4% (w/v) paraformaldehyde (P6148, Sigma Aldrich, St. Louis, MO) in 1× PBS. Extracted brains were then soaked in 4% (w/v) paraformaldehyde for an additional 24 hours. Once fixed, the tissue was cryoprotected by successive 24 hour long soaks in 10%, 20%, and finally 30% (w/v) sucrose (BP220, Fisher Scientific, Waltham, MA) in 1× PBS. If the tissue sample had not sunk to the bottom of the solution after 24 hours, additional time was given before moving to the next higher concentration of sucrose. Next, tissue was embedded in Optimal Cutting Tissue Compound (4583, Sakura, Netherlands) and frozen to -20 °C. The frozen sample was sectioned into 20 μm slices using a Microm 550 Cryostat (Fisher Scientific, Waltham, MA) and mounted directly onto slides. A hydrophobic barrier using a PAP pen (22312, Ted Pella, Redding, CA) was drawn around each slice and allowed to dry. To stain the slices they were first rinsed with 1× PBS for 10 minutes. Next, slices were blocked with 10% goat serum (S-1000, Vector Labs, Burlingame, CA) in 1× PBS for one hour at room temperature. Slices were then incubated in a primary antibody solution containing one or more of the following: Rabbit anti-Iba1 (1:500 dilution) (019-19741, Wako, Richmond, VA), Rabbit anti-GFAP (1:500 dilution) (Z033429-2, Dako, Carpinteria, CA), or Mouse anti-Neun (1:500 dilution) (MAB377, Millipore, Billercia, MA), mixed with 0.3% Triton X-100 (T8787, Sigma Aldrich, St. Louis, MO) and 3% goat serum in 1× PBS, overnight in a covered chamber. The next day, slices were triple rinsed with 1× PBS, with each wash allowed to sit for 10 minutes. Slices were then incubated in a solution of Alexa 488 Goat anti-Rabbit IgG (1:200 dilution) (A-11034, Invitrogen, Grand Island, NY), Alexa 555 Goat anti-Mouse IgG (1:200 dilution) (A-11031, Invitrogen, Grand Island, NY), 0.2% Triton X-100, and 5% goat serum in 1× PBS at room temperature for two hours. The slices were then rinsed twice with 1× PBS with each rinse lasting 10 minutes. Slides were then cover slipped using Prolong Gold (P36930, Invitrogen, Grand Island, NY) and allowed to dry overnight before imaging. 2.4.2 Confocal Imaging & Processing A LSM 510-META Laser Scanning Confocal Microscope (Zeiss, Oberkochen, Germany) was used to image the stained slices. Pixel-based image intensity analytics was performed using previously published custom MATLAB script I.N.T.E.N.S.I.T.Y. v1.1 [50]. Laser, imaging, and PMT settings were constant between contralateral hemispheres in each slice. Gain and contrast settings were altered during image processing. Briefly, to prevent holes in the tissue (such as major blood vessels and shuttle tracts) from artificially reducing the average activity-dependent fluorescence, background noise intensity threshold was calculated from 5% of the corners of each image. To calculate the background noise intensity threshold, pixels with intensity greater than one standard deviation dimmer than mean pixel intensity were considered “signal” and removed from the threshold calculation. The threshold was then determined by calculating the pixel intensity of one standard deviation below the mean of the remaining pixel intensities. Bins with intensity values dimmer than average intensities of the control images were considered tissue “holes”. Using MATLAB, the center of the silicon or carbon fiber track (15 μm × 123 μm or 8.4 μm diameter, respectively) was identified on each image, after which the script generated masks every 25 μm of rounded rectangles or concentric rings, respectively (see I.N.T.E.N.S.I.T.Y. v1.1 readme). Carbon fiber probe tracks were identified as holes in the tissue surrounded by an increased intensity “cloud” of anti-mouse secondary antibody label over non-cellular features. Weak cross-talk between anti-mouse secondary antibodies and rat primary antibodies likely indicate implant sites where rat IgG entered the brain parenchyma from insertion injury related to blood brain barrier leakage. The tissue reactions to single fibers have been previously characterized [41], and the summation of the tissue response from multiple shanks influence the overall tissue health and recording performance. Therefore, the tissue reaction was quantified as a summation of neighboring shanks similar to electrophysiology performance metrics, instead of disentangling overlapping bins. The average intensity for all pixels above the background noise intensity threshold in each bin was calculated and then normalized against the background to calculate the Signal-to-Noise Intensity Ratio (SNIR) in each bin as follows; (2) SNIR=AvgI>TAvgN where AvgI>T is the mean intensity of all pixels above the noise threshold (>T) in each bin, and AvgN is the mean noise floor intensity. This means, SNIR=1 represents the noise floor. Therefore, it is expected that the SNIR does not asymptote to 1 unless there is no staining signal in the corresponding bin. Data were averaged for each implant type and time point, and then reported as mean and standard error. Neurons were counted on I.N.T.E.N.S.I.T.Y. v1.1 binned images using the built in cell counting function in ImageJ. NeuN density was calculated by counting NeuN positive cells in each bin divided by the total area of the tissue in each bin after the area of the holes were removed. 3. Results 3.1 Accelerated Soak Test Previous work has shown that parylene-c coated carbon fibers with only an exposed carbon tip site are unable to record unit activity due to the high site impedance [41]. To alleviate this issue, PEDOT:PSS was electrodeposited at the tip of each site which greatly reduced the site impedance [41,51]. Recent studies by Green et al. have demonstrated that other formulations of PEDOT are more stable over time when compared to PEDOT:PSS [45,52]. To determine the best site coating for the carbon fiber electrodes, an accelerated soak test was carried out between the original PEDOT:PSS (n=8 fibers) formulation and a different formulation, PEDOT:pTS (n=23 fibers) [45]. In addition to determining the best site coating, the values from this study will establish a baseline that can be compared to later chronic animal implants. Prior to PEDOT deposition the impedance values at 1 kHz were 3809.0 ± 426.7 kΩ and 6781.8 ± 655.3 kΩ, respectively, for the PEDOT:pSS and PEDOT:pTS coated fibers. The large difference in pre-deposition impedances is likely attributable to the uneven surface of the exposed fibers, which results from the manual cutting process used to re-expose the tips after parylene-c coating. At day 0, when the initial PEDOT depositions took place, both sets of impedance values at 1 kHz (PEDOT:PSS 142.8 ± 23.2 kΩ and PEDOT:pTS 117.9 ± 28.4 kΩ) were similar (figure 3), mitigating any differences in the pre-deposition impedances. As time progressed, the average impedance of the PEDOT:PSS coated fibers increased faster than those coated with PEDOT:pTS. On the final day of testing (35 days in real time, 172.2 days in simulated time) the fibers coated with PEDOT:PSS had an average impedance (1921.4 ± 344.5 kΩ) double that of the PEDOT:pTS coated fibers (840.5 ± 117.7 kΩ). During the repeated measurements carried out over the course of 35 days, one PEDOT:PSS coated fiber and three PEDOT:pTS coated fibers were accidentally broken off of the test boards, which resulted in lower sample sizes over the duration of the study. SEM images (figures 4(a) and 4(b)) show good adherence of both PEDOT formulations to the carbon fiber tip's outer edges. A visible void of PEDOT can be seen in the center of both PEDOT formulations, which may help to explain the steady increases in impedance. Based on the impedance results, all chronic implants of carbon fibers received a site coating of PEDOT:pTS. 3.2 Chronic Implant Impedance To assess the longevity and viability of the carbon fiber arrays, 5 Long Evans rats were implanted chronically with carbon fiber arrays (n=75 fibers) in the right motor cortex. Two of those rats were also implanted with silicon electrodes (n=2 electrodes with 16 sites each) in the left motor cortex. In addition, 3 more rats were implanted with only silicon electrodes (n=2 electrodes with 16 sites each and 1 with 15 sites.). A detailed breakdown of each animal's implant type, duration, and depth, can be found in table 1. For all performance metrics, no differences were noted between animals that received one or both probe types. Impedance measurements were taken every day for the first 13 days, every other day from days 13 to 31, and then every third day from days 31 to 91. For the two animals continued out to day 154, measurements were taken once a week after day 91. One animal, ZCR22, was sacrificed at day 73 for histological and surgical technique evaluations. The pre-implant 1 kHz impedances (figure 5(a)) at day 0 for the carbon fibers was 128.1 ± 12.0 kΩ while those of the silicon probes were 1118.5 ± 17.4 kΩ. At day 1, post-implantation impedances for carbon fibers increased to 621.9 ± 14.8 kΩ while the silicon sites also increased to 2018.3 ± 17.6 kΩ. The impedance for both sets of probes continued upward until day 12 for the carbon fibers and day 15 for the silicon probes. At these time points the impedance values were 2044.2 ± 171.1 kΩ for the carbon fibers and 3493.7 ± 90.5 kΩ for the silicon probes. Following the initial increases in impedance, the carbon fiber electrodes saw a leveling off in the impedance values which fluctuated between 1500 – 2500 kΩ from days 11 to 91. The silicon probes saw greater changes after day 9 with average impedance values ranging from approximately 2000 kΩ to just below 4000 kΩ. The large variation in average impedance for the silicon probes before and after day 31 is likely due to the drop in electrode sample size after day 31. On the final days of recording, the carbon fiber electrodes had a mean 1 kHz impedance of 2268.3 ± 253.7 kΩ at day 154 while the silicon probes had a mean 1 kHz impedance of 2742.5 ± 126.5 kΩ at day 91. The differences in recording site sizes between the carbon fibers (36.3 μm2) and silicon probes (177 μm2) can potentially skew the impedance results as a larger site size typically results in lower impedance. When scaled for area (figure 5(b)), carbon fibers at 40 days onwards average approximately 80,000 kΩ μm2, outperforming the silicon probes which averaged 500,000 kΩ μm2. While the silicon probes used here were not functionalized with PEDOT, other studies using the same site size coated with PEDOT found chronic 1 kHz impedance reaching an average of 2210 kΩ (figure 5(a), green) or 391,170 kΩ μm2 (figure 5(b), green) between days 6 & 8 [53]. Two animals, ZCR16 and ZCR17, were not sacrificed at day 91 and were recorded from for an additional two months. Recordings were taken at one week intervals during this extended period. The carbon fiber electrode impedance values rose slightly during this period and fluctuated between 2000 – 3000 kΩ. 3.3 Chronic Unit Activity Electrophysiology recordings followed the same points as those used for the impedance measurements. On day 1 post-implant, 65.3% of the implanted carbon fiber electrodes detected unit activity with a mean peak-to-peak amplitude of 142.1 ± 10.4 μV (figures 6(a) and 6(b)). At the same time point, the silicon electrodes detected unit activity on 3.2% of the electrodes sites with an average peak-to-peak amplitude of 59.8 ± 9.8 μV. By day 6, mean unit amplitude on the 40.3% of carbon fiber electrodes with activity continued to climb to 186.4 ± 17.1 μV while the silicon electrodes showed a small climb in activity with 6.4% of electrode sites detecting units with an average peak-to-peak amplitude of 150.2 ± 20.4 μV. At day 13, the number of carbon fiber electrodes with detectable units had slightly increased to 41.3% with an average detected peak-to-peak amplitude that was maintained at 208.1 ± 22.1 μV. During this same period, the silicon electrode detection rate remained in the single percentage range and at day 13, 3.17% of electrode sites showed an average peak-to-peak amplitude of 103.8 ± 27.0 μV. Following this initial spike during the first two weeks post-implant, the carbon fiber electrodes demonstrated mean peak-to-peak unit activity that stayed within the range of 150 – 250 μV through day 91. During this same period, the silicon electrodes had a very low detection rate of < 5%. When single units were detected, the mean peak-to-peak amplitude was typically between 50 – 150 μV. At day 91, 48.7% of the remaining carbon fiber implanted sites were still able to detect units with a mean peak-to-peak amplitude of 194.7 ± 21.5 μV, while 5.7% of silicon sites detected average peak-to-peak amplitude of 89.0 ± 3.6 μV. Waveforms representative of the unit activity detected for each electrode type can be seen in figure 6(c). The two animals that were carried out to day 154 showed some continued unit activity until day 112, with 27.8% of the sites detecting a mean peak-to-peak unit amplitude of 115.8 ± 23.5 μV. After this time point, no units were detected across the remaining carbon fiber electrodes. The loss of detectable unit activity is likely due to brain tissue swelling into the craniotomy, which was discovered post-mortem. Amplitude values obtained by averaging the largest unit on each carbon fiber channel across time (figure 6(d)) compare favorably to those seen in chronic implants of Utah arrays in primates [4]. 3.4 Baseline Activity and Signal-to-Noise Ratio In addition to unit activity, the baseline activity level (figure 7(a)) was quantified for both implant types. This was in turn was used to calculate the signal-to-noise ratio (SNR) (figure 7(b)) for both electrode styles. Baseline activity levels for both the carbon fiber electrodes and silicon electrodes rose for roughly the first 11 days. The initial levels for the carbon fiber electrodes (8.9 ± 0.4 μVRMS) and silicon electrodes (6.9 ± 0.2 μVRMS) were similar at day 1 and initially peaked at days 11 (15.9 ± 0.8 μVRMS) and 10 (10.8 ± 0.2 μVRMS), respectively. After this time, both sets of baseline activity stayed remarkably consistent with the carbon fibers displaying a noise level around 15 μVRMS and the silicon electrodes around 10 μVRMS, with some slight day-to-day variations. SNR levels for the carbon fiber electrodes initially started at 5.5 ± 0.4 and decreased to 3.3 ± 0.3 at day 6. After this time the average SNR increased and was largely maintained between 3.5 and 5 with some days deviating from this pattern. At day 91, when 4 animals remained in the study, average SNR was still 3.8 ± 0.4. After this time point SNR values rapidly dropped off (figure S4) as the remaining animals (n=2) showed decreased unit activity amplitude as seen in figure 6(b). 3.5 Histological Analysis of Implants The microglial response to the silicon electrode is punctuated by a higher density of cells immediately surrounding the implant site (figure 8(a)). This result is typical of that seen by other groups [20,26,28,54] and agrees with previous results [41]. The global response to the carbon fiber array (figure 8(b)) is markedly different from the silicon implant. While some variability and Iba1 activity can be observed across the array implant region, such as the lower right region of the array showing signs of elevated activity, the standard error on the silicon intensity shows substantially greater variability. Analysis of all microglia images (n=2 images/electrode type) from the silicon electrodes show a consistently elevated, and at times significant (p<0.05), level of activity when compared to the microglial response surrounding the carbon fibers (figure 8(g)). The difference between silicon and carbon fiber is more dramatic when examining GFAP astrocyte activity. A representative image of astrocytic activity shows the formation of a tight scar in the immediate vicinity of the silicon electrode site, with elevated activity that extends outward by as much 1000 μm (figure 8(c)). In contrast, the implant site of the carbon fiber array (figure 8(d)) shows no visible scarring regions and limited fluctuation in GFAP intensity from baseline levels. This is further corroborated by intensity analysis of all astrocyte histology images (n=2 images/electrode type), where the silicon electrodes show a sustained astrocytic response that continues out to 1000 μm from the implant site (figure 8(h)). Closer examination revealed that around the silicon implant, one animal showed a compact microglial sheath surrounded by a large activated astrocyte ring (figures S5(i) & S5(k)), while the other exhibited a more evenly distributed elevated GFAP activity. Despite this large variability in tissue reaction, elevated GFAP activity was commonly observed at the 300-400 μm radius and showed significant difference (p<0.05) compared to carbon fiber response profile, which remained steady around baseline at all distances (figure 8(h)). Neuronal signal intensity shows a marked decrease in the area surrounding the silicon electrode (figure 8(e)). In contrast, the neuronal population surrounding the carbon fibers are well distributed and healthy with no immediately obvious declines in signal intensity (figure 8(f)). Measured normalized neural density (n=4 images/electrode type) confirm these observations with the neural density of the silicon electrodes climbing upwards as distance increases (figure 8(i)). The carbon fibers do start with a high neuronal density. This is in part due to the small tissue area in the inner bin and the relatively large tissue hole from the probe track (relative to the bin area). While the inner bins show large error bars, this data shows that neurons trend closer to the probe track of carbon fibers than the silicon shanks. At further distances, the normalized density levels off to approximately 1 and the standard errors decrease leading to significant differences (p<0.05) (figure 8(i)). All histology images can be found in figure S5. 3.6 SEM Imaging of Explanted Carbon Fiber Electrodes Carbon fiber electrodes from chronic implants were explanted at the end of each animal's time point and imaged (figures 9(a) – 9(d)) to better understand any physical changes the electrodes underwent. Figures 9(a) and 9(b) highlight possible parylene-c delamination along the shank of the carbon fiber electrodes. This delamination, especially near the tip site can affect the probes ability to detect local activity. Figure 9(c) shows what may be a thin coating of biological material that could affect the PEDOT:pTS coating. In addition, the center of the electrode tip demonstrates a void similar to that seen with the soak test fibers (figure 4). This void may initially be caused by an uneven PEDOT:pTS electrodeposition, where more PEDOT:pTS is deposited around the edges [51]. As the PEDOT degrades, the center shows the most pronounced change as it likely has the thinnest coating [51]. This center voiding phenomenon is also seen in figure 9(d). 4. Discussion 4.1 Accelerated Aging Evaluation This work first sought to validate a new site tip coating for the carbon fiber arrays. Accelerated soak tests were implemented to rapidly assess the viability of PEDOT:pTS as compared to PEDOT:PSS. For the first 70 simulated days, the impedances of both sets of fibers remained similar. After this time point the PEDOT:PSS fibers saw a more rapid increase in impedance as compared to the PEDOT:pTS coated fibers. The overall increase in impedance can be attributed to the slow degradation of both PEDOT formulations [45,55]. This is corroborated by SEM images which show a lack of PEDOT in the center of the electrodes. The greater stability of the PEDOT:pTS coating agrees well with results seen by others [45] and led to the decision to switch to a different formulation of PEDOT for the site coating. Using even more stable formulations of PEDOT such as those that use carbon nanotubes [56] will be explored in future studies. In addition, the use of electroplated metals such as platinum [57–60], gold [61], or iridium [62], may also serve as a more stable coating. 4.2 In Vivo Assessment of Carbon Fibers and Silicon Electrodes Impedance levels for both probe types increased dramatically over the course of the first three weeks. These results are typical for chronically implanted electrodes [43,55,63–66]. Historically, this increase in impedance has been largely attributed to the glial scar creating a resistive layer around the probe [28,65,67]. Unfortunately, the lack of a macroscopic scar seen in previous carbon fiber work [41] and confirmed here, makes it difficult to account for the impedance increase seen with the carbon fibers. We propose a more nuanced model, which argues that even the largest of glial scars cannot fully account for the impedance rise seen in implanted Utah arrays [68], which are made of stable materials [69], have similar impedance values to the carbon fiber arrays, and in our own case where the carbon fibers do not create a traditional macroscopic scarring response [41]. Instead, dramatic increases in impedance can best be accounted for by an extremely thin resistive layer (∼0.5 μm) made up of biological material, such as cells or proteins, that is directly adhering to the recording site's surface [68]. Determining a first order approximation of this hypothesized thin layer's resistivity can be accomplished using the following equation: (3) R=ρLA where R = impedance at 1 kHz, ρ = resistivity, L = length or thickness, and A = area of the probe interface. It can be assumed that the probes' own internal resistances are unchanging in the first three weeks, which is a reasonable approximation for implanted metal electrodes and for the carbon fiber site coatings given the results from the soak test. Therefore, any change in resistance can be attributed to the thin adherence layer. The area of the carbon fiber electrodes was scaled by a factor of 10 to conservatively account for the PEDOT:pTS coating's increase on effective surface area [70]. This may be particularly true for porous electrode materials such as PEDOT, where its increased electrochemical surface area can be decreased by biofouling, which clogs the porous matrix [71]. Calculating the resistivity from our own results [68], leads to the values seen in table 2. Across all probe types the resistivity of the adherence layer is within the same order of magnitude and similar in value. The leveling off of all impedance values after approximately the 3rd week is likely due to the adherence layer reaching a steady state in thickness and coverage. The large unit amplitudes seen immediately and well after week 3 on the carbon fiber electrodes indicate that this thin adhesion layer is not severely affecting the ability of the fibers to record activity. Baseline activity levels rose in a similar time course to that of the 1 kHz impedance values for both probe types, though the increase was more pronounced for the carbon fibers, which is counterintuitive given the lower impedance of the carbon fibers over time. This however, can be explained by separating the elements that contribute to overall background activity. Baseline activity is a combination of thermal and biological sources [72,73]. The lower overall impedance of the carbon fiber electrodes should lead to a lower thermal noise level; however, the overall baseline level for the fibers is larger than that of the silicon electrodes. This indicates the presence of a biological component that is larger for the carbon fiber electrodes which also points to a greater survival rate of neurons around the carbon fibers as compared to the silicon electrodes. This is also reflected by the carbon fibers' consistent ability to detect unit activity. The greater number of neurons around the carbon fiber electrodes may not always be detected as individual units, but can still contribute to the overall baseline activity, indicating a healthier local tissue environment. This is also supported by the sparse number of units detected on the silicon probes, which may possibly be suffering from mechanical failures [74]. Additionally, the silicon probes used here had small site sizes of 177 μm2, which other works have also demonstrated as having limited ability to detect unit activity [53]. This is in contrast to previous studies which used 703 μm2 [43] or 1250 μm2 [41] site sizes that were able to chronically detect unit activity. Nevertheless, on average, the amplitude of single units detected on the silicon probes was stable, albeit smaller than those detected on the carbon fibers during the 90 day implantation period. The high unit amplitudes and relatively low baseline activity levels also contribute to a high SNR on the carbon fiber electrodes. This SNR remained stable for the first three months after an initial drop off. This drop off was caused by an increasing baseline activity level (figure 7(a)) and not decreasing unit amplitude, which was rising during the same period (figure 6(b)). Overall, the carbon fiber arrays were able to detect unit activity until day 112, or week 16, after which no more activity was detected. This can be partially attributed to the low number of animals (n=2) at the later time point. In addition, explanted brains from many of the animals showed swelling of the cortex into the craniotomy which likely caused the electrodes to move and not record from their target layer. Unfortunately, this swelling also made it difficult to separate the headcap from the brain without damaging the tissue, which ultimately led to a much lower number of animals that were available for histology. It is important to note however, that the large unit amplitudes detected on the carbon fiber electrodes point to a minimal if not non-existent scar around the electrode. This is further corroborated by histology analysis in figure 8. These images show the formation of a scar around the silicon electrodes coupled with some decreased neuronal density and no evident scar around the carbon fiber arrays coupled with a healthy neuronal population. It is possible that the neuronal density in the bin immediately adjacent to the carbon fiber probe is slightly elevated due to the volumetric displacement of the tissue caused by the carbon fiber probes (figure 8(i)). In this case, large standard deviations in the 0-25 μm bin around the carbon fiber probes suggests that this only occurs in a percentage of probe tracks when the probe displaced the neuron. The decreases in the error bars in subsequent bins suggest that the tissue strain around the carbon fiber probes dramatically decreases by the 25-50 μm bin, and is indistinguishable by a 50 μm radius. Similar results can be observed in previous in vivo neurons around carbon fiber implants [37]. This reduced strain in the tissue and neurons may contribute to the improved recording performance, since it is understood that tissue strain adversely impacts neuronal health [35]. While previous studies show that histology can be a poor predictor of electrophysiological performance [37], these histology results correlate well with the electrophysiology results seen in this study. 5. Conclusions This work has demonstrated the ability of carbon fiber electrodes to chronically record unit activity in the rat motor cortex up to 16 weeks. The units detected were of large amplitude and showed a high SNR. The carbon fibers greatly outperformed silicon electrodes with comparable site sizes and were also shown to detect a larger level of biological noise, indicating a healthier local tissue environment. This is also corroborated by the quality of detected unit activity. It is important to note that the stability of detected unit activity was not tracked across time as this was beyond the scope of the current study, but more analysis will be needed in this area to assess the viability of these electrodes for BMI applications. In addition, while both electrode types were implanted directly in the motor cortex no specific muscle group or region was targeted. This in turn may have led to a lower yield as this study relied on spontaneous awake activity and not activity associated with a specific task, which may have resulted in a higher electrode yield. Further work will seek to improve the performance of the PEDOT and parylene-c coatings. Methods to reduce brain swelling and shield the carbon fibers from mechanical damage are being explored with improved array packaging and fabrication. Improvements in all of these areas could lead to high density recording arrays that cause minimal damage to the surrounding tissue and record high quality unit activity for many years. Supplementary Material JNEaa3f87_Supplementary information.pdf Figure S1. Number of channels used for impedance results. For the silicon electrodes, the jump in channel count from day 23 to 25 was due to eight channels from ZCR22 being incorporated after a headstage had been repaired. The brief dip at day 55 was due to a missed time point. The decline at day 73 was due to another animal being removed from the study with four remaining until day 91. For the carbon fibers all dips and immediate recoveries, except at day 55 which was a missed time point, were due to impedance values that indicated a poor connection. Drops that occur without an immediate recovery are from channels being removed from the study due to breakage. The drop-off at day 91 is from the sacrificing of all remaining animals except for two that continued to day 154. Figure S2. Number of channels used for electrophysiology and noise results. For the silicon electrodes, the jump in channel count from day 23 to 25 was due to eight channels from ZCR22 being incorporated after a headstage had been repaired. The brief dip at day 55 was due to a missed time point. The decline at day 73 was due to another animal being removed from the study with four remaining until day 91. For the carbon fibers, the continual decline in channel count was due to channels being removed as their 10 Hz impedance magnitudes matched those of known broken channels. The two exceptions to this were day 55, which was a missed time point, and after day 91 when only two animals remained. Figure S3. Number of units detected for each probe type. For each probe type, the number of units detected on the valid electrophysiology channels from figure S2, is plotted over time. Figure S4. Chronic baseline activity and SNR to day 154. (a) Recorded baseline activity levels (mean ± standard error of the mean) for both carbon fiber and silicon electrodes for all 154 days. (b) The SNR (mean ± standard error of the mean) for all units detected on the carbon fiber electrodes for all 154 days. Figure S5. Chronic histology from carbon fiber arrays and silicon electrodes. The yellow reactangle depicted in the silicon electrode images show the approximate size and position of the electrode. In the images for the carbon fiber array the outlined yellow profile depicts the footprint of the array. (a), (b), (i), & (j) Microglia staining around implanted electrodes. More elevated responses can be seen around the silicon electrodes. (e), (f), (k), & (l) Astrocyte staining around implanted electrodes. More elevated responses can be seen around the silicon electrodes. (c), (d), (g), & (h) Neuron staining around implanted electrodes. Decreased intensity can be seen around the silicon electrodes and no obvious decreases in the carbon fiber array images. This work was financially supported by the National Institute of Neurological Disorders and Stroke (1RC1NS068396-01, 1U01NS094375-01, & 1R01NS094396) and the McKnight Foundation. The authors would like to thank Daryl Kipke for contributions to the experimental design. The authors would also like to thank Karen Schroeder for paper comments and assistance with statistical analysis & Kip Ludwig for assistance with recording and impedance analysis. Additional thanks to Peter Caintic for assistance with data analysis. Figure 1 Soak test probe assembly and setup (a) Areas between and surrounding the bond pads have been roughened. (b) Silver epoxy on each bond pad for the carbon fibers. (c) Exposed bond pads with carbon fibers are covered with insulating epoxy. (d) Four PCBs with functionalized fibers are secured to the underside of the soak jar's lid. (e) Lids are secured to jars containing 1× PBS. Jars are then placed in a heated water bath. Figure 2 Images of implanted electrodes (a) Carbon fiber array used in implants. (b) Silicon probe, NeuroNexus A1×16-3mm-50-177-HZ16_21mm, with sixteen 177 μm2 iridium sites spaced 50 μm apart, used in implants. Figure 3 PEDOT soak test Impedance values (mean ± standard error of the mean) at 1 kHz for PEDOT:PSS and PEDOT:pTS coated carbon fiber electrodes over the simulated time from the accelerated soak test. Figure 4 SEM images of PEDOT coated and soak tested fibers (a) SEM image of a PEDOT:pTS coated carbon fiber aged to simulated day 172.2 showing PEDOT still at the tip, but with a possible void or loss of PEDOT:pTS in the center. (b) SEM image of a PEDOT:PSS coated carbon fiber aged to simulated day 172.2 showing similar properties to that of the PEDOT:pTS coated fiber. Figure 5 Chronic implant impedances (a) Impedance values (mean ± standard error of the mean) for each probe type across time. Both electrode types saw an approximately 2 MΩ increase in impedance within the first two weeks. Values for the carbon fibers then leveled off while the silicon electrode values dropped before leveling off. Impedance values for 177 um2 silicon sites coated with PEDOT are shown in green [53]. The number of channels used for impedance analysis at each time point can be seen in figure S1. (b) Impedance values scaled by geometric surface area (mean ± standard error of the mean) for each probe type across time. Carbon fibers increased to approximately 80,000 kΩ·μm2 before leveling off, while the silicon electrode values peaked at about 650,000 kΩ·μm2 before steadying at approximately 500,000 kΩ·μm2. Similar to (a), values for 177 um2 silicon sites coated with PEDOT are shown in green [53]. Figure 6 Chronic unit amplitudes and percentage of channels with units (a) On average 20% to 40% of viable carbon fiber electrodes detected unit activity across time, while silicon electrodes did so with a peak of 9.5% at day 10 and most other days detecting no unit activity. After day 91 only two rats remained in the study and the loss of their unit activity is likely explained by brain tissue swelling into the craniotomy which was discovered post mortem. The exact number of channels used for calculating the percentage of channels with units at each time point can be seen in figure S2. (b) Carbon fiber electrodes detected an average unit amplitude of 200 μV across three months. Units detected on silicon electrodes had a mean amplitude of 50 – 100 μV. All values are mean ± standard error of the mean. The exact number of units detected and used for amplitude analysis at each time point can be seen in figure S3. (c) Representative time course of detected unit activity on two different channels, one for each electrode type. (d) The mean of the largest unit detected on each carbon fiber or silicon electrode was calculated for each time point. Figure 7 Chronic baseline activity and SNR (a) Recorded baseline activity levels (mean ± standard error of the mean) for both carbon fiber and silicon electrodes for the first 91 days. Both baseline trends rise during the first two weeks of recording and then level off to steady state values. Carbon fibers demonstrate a higher overall recorded baseline level than silicon, which can be explained by a larger biological background contribution as evidenced by the high amplitude recordings reported in previous sections. (b) The SNR (mean ± standard error of the mean) for all units detected on the carbon fiber electrodes for the first 91 days. After an initial drop-off within the first week, values level off and hold between 3.5 and 5. The exact number of channels used for calculating the noise levels and SNR at each time point can be seen in figure S2. Figure 8 Chronic histology images and analysis (a) & (b) Iba1 (microglia) staining around the implanted carbon fiber array and silicon electrode in ZCR19. Formation of a scar is well defined around the silicon electrode but no so around the carbon fiber array. Yellow rectangles show location and approximate size of implanted electrodes. (c) & (d) GFAP (astrocyte) staining around the implanted carbon fiber array and silicon electrode in ZCR19. Increased glial activity can be observed surrounding the silicon electrode with no obvious uptick in activity around the carbon fiber array. (e) & (f) NeuN (neuron) staining around the implanted carbon fiber array and silicon electrode in ZCR19. Neural density appears much more diminished around the silicon electrode as compared to the carbon fiber array. (g) Signal-to-noise intensity ratio of Iba1 staining around each electrode type (n=2 images/electrode type). Compared to the carbon fiber arrays the silicon electrodes maintain a more elevated level of Iba1 activity for almost all distances. (h) Signal intensity analysis of GFAP staining around each electrode type (n=2 images/electrode type). Similar to (g), the silicon electrodes show more GFAP activity as far out as 1000 μm from the implant site. (i) Normalized neural density around each electrode type (n=2 images/electrode type), illustrating the healthy neuronal population surrounding the carbon fiber arrays and a lack of neurons around the silicon electrodes. *indicates significance at p<0.05. Figure 9 SEM images of chronically implanted carbon fibers (a) & (b) SEM images of chronically implanted carbon fibers that may be experiencing parylene-c delamination. (c) & (d) SEM images of chronically implanted carbon fiber with a loss of surface roughness at the electrode tip indicating a loss of PEDOT in the center, the attachment of a thin adherence layer, or a combination of both. Table 1 Animal implant information Probe implant depth and duration for each animal. Animal Name Carbon Fiber Depth Silicon Probe Depth Days in study ZCR16 1.56 mm No Implant 154 ZCR17 1.505 mm No Implant 154 ZCR18 1.505 mm No Implant 91 ZCR19 1.495 mm 1.45 mm 91 ZCR22 1.45 mm 1.45 mm 73 ZCR28 1.5 mm 1.5 mm 91 ZCR29 No Implant 1.5 mm 91 ZCR30 No Implant 1.5 mm 91 Table 2 Adherence layer resistivity Calculated resistivity values for adherence layers on each probe type are shown to be on the same order of magnitude and similar in value. Probe Type RPre-Implant (MΩ) RWeek 3 (MΩ) ΔR (MΩ) Area (μm2) Length (μm) ρ (Ω-cm) Carbon Fiber 0.136 1.75 1.614 363.1 0.5 23,435 Silicon 1.126 3.061 1.935 177 0.5 13,708 Utah Array 0.3787 0.8441 0.2315 1,100 0.5 20,476 1 Carmena JM Lebedev MA Crist RE O'Doherty JE Santucci DM Dimitrov DF Patil PG Henriquez CS Nicolelis MAL 2003 Learning to control a brain-machine interface for reaching and grasping by primates PLoS Biol 1 193 208 2 Simeral JD Kim SP Black MJ Donoghue JP Hochberg LR 2011 Neural control of cursor trajectory and click by a human with tetraplegia 1000 days after implant of an intracortical microelectrode array J Neural Eng 8 25027 3 Hochberg LR Serruya MD Friehs GM Mukand JA Saleh M Caplan AH Branner A Chen D Penn RD Donoghue JP 2006 Neuronal ensemble control of prosthetic devices by a human with tetraplegia Nature 442 164 71 16838014 4 Chestek CA Gilja V Nuyujukian P Foster JD Fan JM Kaufman MT Churchland MM Rivera-Alvidrez Z Cunningham JP Ryu SI Shenoy K 2011 Long-term stability of neural prosthetic control signals from silicon cortical arrays in rhesus macaque motor cortex J Neural Eng 8 45005 5 Collinger JL Wodlinger B Downey JE Wang W Tyler-Kabara EC Weber DJ McMorland AJC Velliste M Boninger ML Schwartz AB 2013 High-performance neuroprosthetic control by an individual with tetraplegia Lancet 381 557 64 23253623 6 Velliste M Perel S Spalding MC Whitford AS Schwartz AB 2008 Cortical control of a prosthetic arm for self-feeding Nature 453 1098 101 18509337 7 Hochberg LR Bacher D Jarosiewicz B Masse NY Simeral JD Vogel J Haddadin S Liu J Cash SS van der Smagt P Donoghue JP 2012 Reach and grasp by people with tetraplegia using a neurally controlled robotic arm Nature 485 372 5 22596161 8 Taylor DM Tillery SIH Schwartz AB 2003 Information conveyed through brain-control: cursor versus robot IEEE Trans Neural Syst Rehabil Eng 11 195 9 12899273 9 Schwartz AB Cui XT Weber DJ Moran DW 2006 Brain-Controlled Interfaces: Movement Restoration with Neural Prosthetics Neuron 52 205 20 17015237 10 Davidson TJ Kloosterman F Wilson MA 2009 Hippocampal Replay of Extended Experience Neuron 63 497 507 19709631 11 Cohen JY Haesler S Vong L Lowell BB Uchida N 2012 Neuron-type-specific signals for reward and punishment in the ventral tegmental area Nature 482 85 8 22258508 12 Berke JD Breck JT Eichenbaum H 2009 Striatal Versus Hippocampal Representations During Win-Stay Maze Performance J Neurophysiol 101 1575 87 19144741 13 Royer S Sirota A Patel J Buzsaki G 2010 Distinct Representations and Theta Dynamics in Dorsal and Ventral Hippocampus J Neurosci 30 1777 87 20130187 14 Bjornsson CS Oh SJ Al-Kofahi YA Lim YJ Smith KL Turner JN De S Roysam B Shain W Kim SJ 2006 Effects of insertion conditions on tissue strain and vascular damage during neuroprosthetic device insertion J Neural Eng 3 196 207 16921203 15 Johnson MD Kao OE Kipke DR 2007 Spatiotemporal pH dynamics following insertion of neural microelectrode arrays J Neurosci Methods 160 276 87 17084461 16 Kozai TDY Marzullo TC Hooi F Langhals NB Majewska AK Brown EB Kipke DR 2010 Reduction of neurovascular damage resulting from microelectrode insertion into the cerebral cortex using in vivo two-photon mapping J Neural Eng 7 46011 17 Rennaker RL Street S Ruyle AM Sloan AM 2005 A comparison of chronic multi-channel cortical implantation techniques: manual versus mechanical insertion J Neurosci Methods 142 169 76 15698656 18 Rousche P Normann R 1992 A method for pneumatically inserting an array of penetrating electrodes into cortical tissue Ann Biomed Eng 20 413 22 1510293 19 Potter KA Buck AC Self WK Capadona JR 2012 Stab injury and device implantation within the brain results in inversely multiphasic neuroinflammatory and neurodegenerative responses J Neural Eng 9 46020 20 Biran R Martin DC Tresco PA 2005 Neuronal cell loss accompanies the brain tissue response to chronically implanted silicon microelectrode arrays Exp Neurol 195 115 26 16045910 21 Edell DJ Toi VV McNeil VM Clark LD 1992 Factors influencing the biocompatibility of insertable silicon microshafts in cerebral cortex IEEE Trans Biomed Eng 39 635 43 1601445 22 Carter RR Houk JC 1993 Multiple single-unit recordings from the CNS using thin-film electrode arrays IEEE Trans Rehabil Eng 1 175 84 23 Schmidt S Horch K Normann R 1993 Biocompatibility of Silicon-based Electrode Arrays Implanted In Feline Cortical Tissue J Biomed Mater Res 27 1393 9 8263001 24 Turner JN Shain W Szarowski DH Andersen M Martins S Isaacson M Craighead H 1999 Cerebral Astrocyte Response to Micromachined Silicon Implants Exp Neurol 156 33 49 10192775 25 Liu X McCreery DB Carter RR Bullara LA Yuen TGH Agnew WF 1999 Stability of the interface between neural tissue and chronically implanted intracortical microelectrodes IEEE Trans Rehabil Eng 7 315 26 10498377 26 Szarowski DH Andersen MD Retterer S Spence AJ Isaacson M Craighead HG Turner JN Shain W 2003 Brain responses to micro-machined silicon devices Brain Res 983 23 35 12914963 27 Polikov VS Tresco PA Reichert WM 2005 Response of brain tissue to chronically implanted neural electrodes J Neurosci Methods 148 1 18 16198003 28 McConnell GC Rees HD Levey AI Gutekunst CA Gross RE Bellamkonda R 2009 Implanted neural electrodes cause chronic, local inflammation that is correlated with local neurodegeneration J Neural Eng 6 56003 29 Grand L Wittner L Herwik S Göthelid E Ruther P Oscarsson S Neves H Dombovári B Csercsa R Karmos G Ulbert I 2010 Short and long term biocompatibility of NeuroProbes silicon probes J Neurosci Methods 189 216 29 20399227 30 Winslow BD Tresco PA 2010 Quantitative analysis of the tissue response to chronically implanted microwire electrodes in rat cortex Biomaterials 31 1558 67 19963267 31 Kozai TDY Vazquez AL Weaver CL Kim SG Cui XT 2012 In vivo two-photon microscopy reveals immediate microglial reaction to implantation of microelectrode through extension of processes J Neural Eng 9 66001 32 Ersen A Elkabes S Freedman DS Sahin M 2015 Chronic tissue response to untethered microelectrode implants in the rat brain and spinal cord J Neural Eng 12 16019 33 Guilian D Corpuz M Chapman S Mansouri M Robertson C 1993 Reactive Mononuclear Phagocytes Release Neurotoxins After Ischemic and Traumatic Injury To the Central-nervous-system J Neurosci Res 36 681 93 8145296 34 Saxena T Karumbaiah L Gaupp EA Patkar R Patil K Betancur M Stanley GB Bellamkonda R 2013 The impact of chronic blood-brain barrier breach on intracortical electrode function Biomater 34 4703 13 35 Kozai TDY Jaquins-Gerstl AS Vazquez AL Michael AC Cui XT 2015 Brain Tissue Responses to Neural Implants Impact Signal Sensitivity and Intervention Strategies ACS Chem Neurosci 6 48 67 25546652 36 Sawyer AJ Tian W Saucier-Sawyer JK Rizk PJ Saltzman WM Bellamkonda RV Kyriakides TR 2014 The effect of inflammatory cell-derived MCP-1 loss on neuronal survival during chronic neuroinflammation Biomater 35 6698 706 37 Kozai TDY Li X Bodily LM Caparosa EM Zenonos GA Carlisle DL Friedlander RM Cui XT 2014 Effects of caspase-1 knockout on chronic neural recording quality and longevity: Insight into cellular and molecular mechanisms of the reactive tissue response Biomaterials 35 9255 68 25128375 38 Seymour JP Kipke DR 2007 Neural probe design for reduced tissue encapsulation in CNS Biomaterials 28 3594 607 17517431 39 Stice P Gilletti A Panitch A Muthuswamy J 2007 Thin microelectrodes reduce GFAP expression in the implant site in rodent somatosensory cortex J Neural Eng 4 42 17409479 40 Skousen JL Merriam SME Srivannavit O Perlin G Wise KD Tresco PA 2011 Reducing surface area while maintaining implant penetrating profile lowers the brain foreign body response to chronically implanted planar silicon microelectrode arrays Prog Brain Res 194 167 80 21867802 41 Kozai TDY Langhals NB Patel PR Deng X Zhang H Smith KL Lahann J Kotov NA Kipke DR 2012 Ultrasmall implantable composite microelectrodes with bioactive surfaces for chronic neural interfaces Nat Mater 11 1065 73 23142839 42 Patel PR Na K Zhang H Kozai TDY Kotov NA Yoon E Chestek CA 2015 Insertion of linear 8.4 μm diameter 16 channel carbon fiber electrode arrays for single unit recordings J Neural Eng 43 Ludwig KA Uram JD Yang J Martin DC Kipke DR 2006 Chronic neural recordings using silicon microelectrode arrays electrochemically deposited with a poly(3,4-ethylenedioxythiophene) (PEDOT) film J Neural Eng 3 59 70 16510943 44 Venkatraman S Hendricks J King ZA Sereno AJ Richardson-Burns S Martin D Carmena JM 2011 In Vitro and In Vivo Evaluation of PEDOT Microelectrodes for Neural Stimulation and Recording IEEE Trans Neural Syst Rehabil Eng 19 307 16 21292598 45 Green RA Hassarati RT Bouchinet L Lee CS Cheong GLM Yu JF Dodds CW Suaning GJ Poole-Warren LA Lovell NH 2012 Substrate dependent stability of conducting polymer coatings on medical electrodes Biomaterials 33 5875 86 22656446 46 Hukins DWL Mahomed A Kukureka S 2008 Accelerated aging for testing polymeric biomaterials and medical devices Med Eng Phys 30 1270 4 18692425 47 ASTM F-07 2011 Standard Guide for Accelerated Aging of Sterile Barrier Systems for Medical Devices 48 Paxinos G Watson C 2007 The Rat Brain in Stereotaxic Coordinates Academic Press 49 Nunamaker EA Kipke DR 2010 An alginate hydrogel dura mater replacement for use with intracortical electrodes J Biomed Mater Res Part B Appl Biomater 95B 421 9 50 Kozai TDY Gugel Z Li X Gilgunn PJ Khilwani R Ozdoganlar OB Fedder GK Weber DJ Cui XT 2014 Chronic tissue response to carboxymethyl cellulose based dissolvable insertion needle for ultra-small neural probes Biomaterials 35 9255 68 25128375 51 Cui X Martin DC 2003 Electrochemical deposition and characterization of poly(3,4- ethylenedioxythiophene) on neural microelectrode arrays Sensors Actuators B Chem 89 92 102 52 Harris AR Morgan SJ Chen J Kapsa RMI Wallace GG Paolini AG 2013 Conducting polymer coated neural recording electrodes J Neural Eng 10 16004 53 Ludwig KA Langhals NB Joseph MD Richardson-Burns SM Hendricks JL Kipke DR 2011 Poly(3,4-ethylenedioxythiophene) (PEDOT) polymer coatings facilitate smaller neural recording electrodes J Neural Eng 8 14001 54 Biran R Martin DC Tresco PA 2007 The brain tissue response to implanted silicon microelectrode arrays is increased when the device is tethered to the skull J Biomed Mater Res Part A 82 169 78 55 Mandal HS Knaack GL Charkhkar H McHail DG Kastee JS Dumas TC Peixoto N Rubinson JF Pancrazio JJ 2014 Improving the performance of poly(3,4-ethylenedioxythiophene) for brain-machine interface applications Acta Biomater 10 2446 54 24576579 56 Kozai TDY Catt K Du Z Na K Srivannavit O Haque RM Seymour J Wise KD Yoon E Cui XT 2016 Chronic In Vivo Evaluation of PEDOT/CNT for Stable Neural Recordings IEEE Trans Biomed Eng 63 111 9 26087481 57 Arcot Desai S Rolston JD Guo L Potter SM 2010 Improving impedance of implantable microwire multi-electrode arrays by ultrasonic electroplating of durable platinum black Front Neuroeng 3 1 5 20162033 58 Ilic B Czaplewski D Neuzil P Stanczyk T Blough J Maclay GJ 2000 Preparation and characterization of platinum black electrodes J Mater Sci 35 3447 57 59 Whalen JJ III Young J Weiland JD Searson PC 2006 Electrochemical Characterization of Charge Injection at Electrodeposited Platinum Electrodes in Phosphate Buffered Saline J Electrochem Soc 153 C834 9 60 Whalen JJ Weiland JD Searson PC 2005 Electrochemical Deposition of Platinum from Aqueous Ammonium Hexachloroplatinate Solution J Electrochem Soc 152 C738 43 61 Hermans A Wightman RM 2006 Conical Tungsten Tips as Substrates for the Preparation of Ultramicroelectrodes Langmuir 22 10348 53 17129002 62 Meyer RD Cogan SF Nguyen TH Rauh RD 2001 Electrodeposited iridium oxide for neural stimulation and recording electrodes IEEE Trans Neural Syst Rehabil Eng 9 2 11 11482359 63 Vetter RJ Williams JC Hetke JF Nunamaker EA Kipke DR 2004 Chronic neural recording using silicon-substrate microelectrode arrays implanted in cerebral cortex IEEE Trans Biomed Eng 51 896 904 15188856 64 Prasad A Sanchez JC 2012 Quantifying long-term microelectrode array functionality using chronic in vivo impedance testing J Neural Eng 9 26028 65 Williams JC Hippensteel JA Dilgen J Shain W Kipke DR 2007 Complex impedance spectroscopy for monitoring tissue responses to inserted neural implants J Neural Eng 4 410 23 18057508 66 Kozai TDY Du Z Gugel ZV Smith MA Chase SM Bodily LM Caparosa EM Friedlander RM Cui XT 2015 Comprehensive chronic laminar single-unit, multi-unit, and local field potential recording performance with planar single shank electrode arrays J Neurosci Methods 242 15 40 25542351 67 Sankar V Patrick E Dieme R Sanchez JC Prasad A Nishida T 2014 Electrode impedance analysis of chronic tungsten microwire neural implants: understanding abiotic vs. biotic contributions Front Neuroeng 7 1 12 24478695 68 Malaga KA Schroeder KE Patel PR Irwin ZT Thompson DE Bentley JN Lempka SF Chestek CA Patil PG 2016 Data-driven model comparing the effects of glial scarring and interface interactions on chronic neural recordings in non-human primates J Neural Eng 13 16010 69 Negi S Bhandari R Rieth L Solzbacher F 2010 In vitro comparison of sputtered iridium oxide and platinum-coated neural implantable microelectrode arrays Biomed Mater 5 15007 20124668 70 Cui X Hetke JF Wiler JA Anderson DJ Martin DC 2001 Electrochemical deposition and characterization of conducting polymer polypyrrole/PSS on multichannel neural probes Sensors Actuators A Phys 93 8 18 71 Kozai TDY Alba NA Zhang H Kotov NA Gaunt RA Cui XT 2014 Nanostructured Coatings for Improved Charge Delivery to Neurons Nanotechnology and Neuroscience: Nano-electronic, Photonic and Mechanical Neuronal Interfacing De Vittorio M Martiradonna L Assad J Springer New York 71 134 72 Lempka SF Johnson MD Moffitt MA Otto KJ Kipke DR McIntyre CC 2011 Theoretical analysis of intracortical microelectrode recordings J Neural Eng 8 45006 73 Lempka SF Johnson MD Barnett DW Moffitt MA Otto KJ Kipke DR McIntyre CC 2006 Optimization of Microelectrode Design for Cortical Recording Based on Thermal Noise Considerations 2006 Annu Int Conf IEEE Eng Med Biol Soc 3361 4 74 Kozai TDY Catt K Li X Gugel ZV Olafsson VT Vazquez AL Cui XT 2015 Mechanical failure modes of chronically implanted planar silicon-based neural probes for laminar recording Biomaterials 37 25 39 25453935

PMC005xxxxxx/PMC5118073.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9207397 2425 Plant J Plant J. The Plant journal : for cell and molecular biology 0960-7412 1365-313X 27337377 5118073 10.1111/tpj.13249 NIHMS798233 Article Investigating inducible short-chain alcohol dehydrogenases/reductases clarifies rice oryzalexin biosynthesis Kitaoka Naoki 1 Wu Yisheng 2 Zi Jiachen 3 Peters Reuben J. * Roy J. Carver Department of Biochemistry, Biophysics, and Molecular Biology, Iowa State University, Ames, IA 50011, U.S.A * Corresponding author: rjpeters@iastate.edu 1 Current address: Department of Biomolecular Sciences, Graduate School of Life Sciences, Tohoku University, Sendai 980-8577, Japan 2 Current address: Conagen Inc., Bedford, MA 01730, U.S.A. 3 Current address: Biotechnological Institute of Chinese Materia Medica, College of Pharmacy, Jinan University, Guangzhou, Guangdong 510632, China 1 7 2016 1 9 2016 10 2016 01 10 2017 88 2 271279 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Rice (Oryza sativa) produces a variety of labdane-related diterpenoids as phytoalexins and allelochemicals. The production of these important natural products has been partially elucidated. However, the oxidases responsible for production of the keto groups found in many of these diterpenoids have largely remained unknown. Only one short-chain alcohol dehydrogenase/reductases (SDRs), which has been proposed to catalyze the last step in such a pathway, has been characterized to-date. While rice contains >220 SDRs, only the transcription of five has been shown to be induced by the fungal cell wall elicitor chitin. This includes the momilactone A synthase (OsMAS/SDR110C-MS1), with the other four all falling in the same SDR110C family, further suggesting roles in diterpenoid biosynthesis. Here biochemical characterization with simplified substrate analogs was first used to indicate potential functions, which were then supported by further analyses with key biosynthetic intermediates. Kinetic studies were then employed to further clarify these roles. Surprisingly, OsSDR110C-MS2 more efficiently catalyzes the final oxidation to produce momilactone A that was previously assigned to OsMAS/SDR110C-MS1, and we speculate that this latter SDR may have an alternative function instead. On the other hand, two of these SDRs clearly appear to act in oryzalexin biosynthesis, with OsSDR110C-MI3 readily oxidizing the 3α-hydroxyl of oryzalexin D, while OsSDR110C-MS3 can also oxidize the accompanying 7β-hydroxyl. Together, these SDRs then serve to produce oryzalexins A – C from oryzalexin D, essentially completing elucidation of the biosynthesis of this family of rice phytoalexins. diterpenoids oxidases phytoalexins allelochemicals biosynthesis Introduction Upon exposure to microbial pathogens plants produce an arsenal of antibiotic natural products, known as phytoalexins (Hammerschmidt 1999), which are considered to be an important part of the defense response (Bednarek and Osbourn 2009). Thus, the production of these metabolites is a subject of significant interest, as these compounds and/or the genes controlling their biosynthesis might serve as markers for molecular breeding, particularly in important crop plants, such as rice (Oryza sativa). One particularly interesting aspect of such plant natural products biosynthesis has been the observation of gene clusters containing unrelated genes encoding biosynthetic enzymes from a common metabolic pathway in certain cases (Nutzmann and Osbourn 2014). Rice has served as a model cereal crop plant, including for investigation of phytoalexins, which appear to be almost all terpenoids (Schmelz et al. 2014). In particular, labdane-related diterpenoids whose biosynthesis is related to that of the gibberellin phytohormones in the characteristic use of a copalyl diphosphate synthase (CPS) to initiate their production from the general diterpenoid precursor (E,E,E)-geranylgeranyl diphosphate (Zi et al. 2014). These natural products have been suggested to serve as antibiotics against the fungal blast pathogen Magneportha oryzae [i.e., the momilactones, oryzalexins and phytocassanes (Peters 2006)] and bacterial leaf blight pathogen Xanthomonas oryzae [i.e., the oryzalides, oryzadiones and related compounds (Toyomasu 2008)], as well as allelochemicals that inhibit the growth of other plants [i.e., momilactone B (Kato-Noguchi and Peters 2013)]. The rice genome contains two biosynthetic gene clusters associated with labdane-related diterpenoid biosynthesis, one on chromosome 4 that appears to be dedicated to the production of momilactones (Shimura et al. 2007, Wilderman et al. 2004), while the other on chromosome 7 has been associated with the production of phytocassanes (Prisic et al. 2004, Swaminathan et al. 2009), although there is evidence that this cluster has a more general role in such biosynthesis – e.g., production of the oryzalexins and oryzalides as well (Wang et al. 2012a, Wu et al. 2011, Wu et al. 2013). In both cases, there is a gene encoding a CPS and at least one for a kaurene synthase-like (KSL) cyclase specific for the stereoisomer of CPP produced by the co-clustered CPS, along with several encoding cytochrome P450 (CYP) mono-oxygenases that will act on the resulting multicyclic diterpene olefins (e.g., Figure 1). Notably, the chromosome 4 cluster also contains a gene encoding a short-chain alcohol dehydrogenase/reductase (SDR), which has been suggested to catalyze the last step in momilactone A biosynthesis, oxidation of a 3β-hydroxyl to the characteristic keto group (Atawong et al. 2002), leading to its designation as the rice momilactone A synthase, OsMAS (Shimura, et al. 2007). A number of other rice labdane-related diterpenoid phytoalexins also contain keto groups. While the CYPs that catalyze the initial addition of oxygen to form the relevant hydroxyl group have been identified in many cases, the enzyme catalyzing subsequent oxidation to a carbonyl has remained unknown. For example, CYP701A8 and CYP76M8 consecutively oxygenate ent-sandaracopimaradiene at C3α and then C7β, respectively, forming the di-hydroxy oryzalexin D (Wu, et al. 2013), further oxidation of which at C3 and/or C7 forms oryzalexins A – C (Peters 2006). Similarly, CYP76M7 can introduce a C11α hydroxyl group into ent-cassadiene, with the subsequently formed C11-keto found in all the phytocassanes (Swaminathan, et al. 2009), while CYP701A8 can introduce a C3α-hydroxyl, with either this or the derived 3-keto group found in all phytocassanes (Wang et al. 2012b), and CYP71Z7 can introduce a C2α-hydroxyl, which also can be further oxidized to a keto – e.g., as in phytocassane D (Wu, et al. 2011). While there are over 220 SDRs in rice (Moummou et al. 2012), it has already been shown that transcription of the genes encoding the relevant upstream enzymes for labdane-related diterpenoids are inducible (Schmelz, et al. 2014). A particularly broad view was provided by microarray based analysis of the transcriptional response of rice cell culture to induction with the fungal cell wall elicitor chitin (Okada et al. 2007). Notably, this study found significant accumulation of mRNA for five SDRs, including OsMAS. Here biochemical analysis of these SDRs was carried out and used to suggest their roles in rice labdane-related diterpenoid phytoalexin biosynthesis, leading most particularly to clarification of the production of oryzalexins. Results The inducible rice SDRs The SDRs form one of the largest enzymatic super-families and are highly diverse. Hence, the SDRs have been divided into phylogenetically distinct broad classes and more specific families to assist functional annotation (Persson et al. 2009). All five of the chitin-inducible SDRs contain the TGxxxGxG coenzyme binding and YxxxK catalytic motifs that place them in the classical SDR class (Kavanagh et al. 2008), as well as a specific Asp indicating a preference for NAD+ over NADP+ (Kallberg et al. 2002)(Figure S1). More specifically, these all fall within the vascular plant specific SDR110C family, members of which generally oxidize terpenoid or phenolic compounds. Even more particularly, these inducible OsSDR110C family members are divided between two clades therein (Figure 2), one defined by OsMAS and that is termed here the momilactone synthase (MS) clade, and another simply annotated monocot I and termed here the MI clade (Moummou, et al. 2012). According to this nomenclature, OsMAS can be termed OsSDR110C-MS1 (referred to hereafter as MS1), while the other clearly inducible clade member is OsSDR110C-MS3 (MS3). Similarly, those from the MI clade are largely found in a tandem array of 13 such genes on chromosome 7, and the chitin inducible members of this are OsSDR110C-MI2 (MI2), OsSDR110C-MI3 (MI3) and OsSDR110C-MI4 (MI4). Full-length mRNA corresponding to all five of these SDRs have been reported from a previous large-scale cDNA sequence project and were obtained from this source (Kikuchi et al. 2003). In addition to OsMAS, the momilactone biosynthetic gene cluster contains an adjacent and closely related SDR (OsSDR110C-MS2; hereafter MS2), which appears to have arisen by tandem gene duplication and shares >94% sequence identity with MS1 (Miyamoto et al. 2016). While the transcript of this gene was not reported to accumulate in response to chitin, it does not appear to have been separately represented on the microarray. Accordingly, it seemed plausible that this SDR might play a role in momilactone biosynthesis and/or other labdane-related diterpenoid phytoalexin biosynthesis. Thus, MS2 was also investigated here. Due to difficulties in cloning this, presumably in part as a result of the exact identity at both the 5' and 3' ends of the coding regions of this and MS1, a synthetic gene was obtained instead. Initial biochemical analyses with simplified substrate analogs The full length SDRs were recombinantly expressed as N-terminal 6xHis-tagged proteins in Escherichia coli strain C41 and purified via affinity chromatography. Recombinant protein was obtained for all these SDRs except MI4, which was not expressed well and, thus, not further investigated here. Based on previous work, a variety of mono-hydroxylated labdane-related diterpenes are readily available (Kitaoka et al. 2015, Swaminathan, et al. 2009, Wang, et al. 2012a, Wang et al. 2011, Wang, et al. 2012b, Wu, et al. 2011, Wu, et al. 2013). Hypothesizing that activity against these various hydroxyl groups might highlight potential roles for these SDRs in phytoalexin biosynthesis, these diterpene alcohols were tested as potential substrates (Figure 3). For example, 11α-hydroxy-ent-cassadiene provides the alcohol that presumably is oxidized as an early step in phytocassane biosynthesis (Figure 1), while 3β-hydroxy-syn-pimaradiene presents the relevant alcohol that is oxidized to form momilactone A (Figure 3). All assays also included 1 mM NAD+ to provide this necessary enzymatic co-factor, with organic extracts analyzed by GC-MS to detect the presence of oxidized products (i.e., a new peak with an apparent molecular ion of m/z = 286, relative to the m/z = 288 observed with the mono-hydroxylated diterpene precursors). While none of the SDRs reacted with 11α-hydroxy-ent-cassadiene, four of them reacted with 3β-hydroxy-syn-pimaradiene to varying degrees, with only MI2 unable to oxidize this (Table 1 and Figure S2). Surprisingly, MS1 was actually somewhat less active than not only MS2, but also MI3 with this latter substrate, which provides the relevant alcohol for the originally suggested reaction. Indeed, MS1 exhibited greater turnover with the alternative substrate 2α-hydroxy-ent-cassadiene, which was further oxidized by the other SDRs, with the exception of MI2 again (Table 1 and Figure S3). Similarly, these same four SDRs were able to oxidize both 3α-hydroxy-ent-cassadiene and 3α-hydroxy-ent-sandaracopimaradiene to varying degrees as well, although only MS3 was able to oxidize 7β-hydroxy-ent-sandaracopimaradiene (Table 1 and Figures S4 – S6). These initial results suggest that these SDRs are mostly likely to play roles in momilactone and/or oryzalexin biosynthesis, with any role in the production of phytocassanes occurring at later steps for which the relevant intermediates have not yet been identified and, thus, are not yet available. SDRs in momilactone biosynthesis The ability of multiple SDRs to catalyze oxidation of 3β-hydroxy-syn-pimaradiene suggested that the analogous step in momilactone A biosynthesis might be mediated by more than just MS1. This possibility was investigated by directly assaying these with 3β-hydroxy-syn-pimaradien-19,6β-olide, the penultimate intermediate in momilactone A biosynthesis (Atawong, et al. 2002), and originally reported MS1 substrate (Shimura, et al. 2007). Consistent with the results from assays with the simplified substrate analog 3β-hydroxy-syn-pimaradiene, not only MS1, but also MS2, MS3 and MI3 were able to produce momilactone A from 3β-hydroxy-syn-pimaradien-19,6β-olide (Figure 4, Table 1). Nevertheless, kinetic analysis demonstrated that MS1 and MS2 from the momilactone biosynthetic gene cluster exhibited significantly higher catalytic efficiency for this reaction than MS3 and MI3, for which kinetic constants could not be obtained due to poor substrate affinity (see Table 2). Somewhat surprisingly, MS2 exhibited much higher (>20-fold) catalytic efficiency than MS1, with both higher affinity (i.e., lower KM) and rate of catalysis (i.e., kcat). SDRs in oryzalexin biosynthesis The ability of these SDRs to catalyze oxidation of the 3α and/or 7β hydroxylated derivatives of ent-sandaracopimaradiene suggested that these might play a role in oryzalexin biosynthesis. In particular, the oxidation of oryzalexin D to oryzalexins A – C, which was investigated by directly assaying these SDRs with oryzalexin D (Figure 5). Again consistent with the results from the simplified substrate analogs, the same four SDRs were found to oxidize oryzalexin D (Table 1). In each case a new product was observed in which the apparent molecular ion was m/z = 302, representing oxidation of a single hydroxyl of oryzalexin D (m/z = 304). In addition, MS3 yielded two additional products, another with an apparent molecular ion of m/z = 302, and one with an apparent molecular ion of m/z = 300, presumably representing oxidation of both hydroxyl groups (i.e., oryzalexin C). Given that only MS3 was able to oxidize 7β-hydroxy-ent-sandaracopimaradiene, while all four SDRs can oxidize 3α-hydroxy-ent-sandaracopimaradiene, it seemed likely that the common SDR product was selectively oxidized at C3 (i.e., oryzalexin B). This was verified by scaling up the in vitro enzyme reaction to produce sufficient amounts for NMR analysis. Briefly, 0.5 mg of oryzalexin D was obtained by metabolic engineering of E. coli, as previously described (Wu, et al. 2013). This was converted to the common oxidized product with MS1. After extraction and purification, 0.2 mg of this compound was isolated. The resulting 1H-NMR data matched that reported for oryzalexin B rather than oryzalexin A (Kono et al. 1985). Accordingly, the remaining MS3 product is then oryzalexin A, resulting from oxidation of the 7β-hydroxyl group, again consistent with the activity observed with the corresponding simplified substrate analog. To further investigate the relevance of the observed activity to oryzalexin biosynthesis, kinetic analysis was carried out for all four SDR with oryzalexin D. Notably, MS3 and MI3 exhibited much higher catalytic efficiency (>100-fold) with this substrate than MS1 and MS2, the reverse of what is observed with the 3β-hydroxy-syn-pimaradien-19,6β-olide intermediate in momilactone biosynthesis (Table 2). Discussion The studies described here provide some insight into the roles played by the investigated inducible SDRs in rice labdane-related diterpenoid biosynthesis. While all four SDRs that exhibit activity will react with multiple labdane-related diterpenoids (Table 1), such promiscuity is not entirely unexpected (Wu et al. 2007). Moreover, their ability to react with the simplified substrate analogs employed here provided useful suggestions regarding potentially relevant biosynthetic activity, with subsequent kinetic analyses indicating distinct roles for these oxidases (Table 2). Perhaps clearest are the results indicating that MS3 and MI3 catalyze oxidation of the 7β- and/or 3α-hydroxyl groups, respectively, in production of oryzalexins A – C (Figure 6). This is particularly indicated by the strong catalytic efficiency exhibited by these OsSDR110C family members with oryzalexin D (Table 2), and their relatively weak activity with the corresponding mono-hydroxylated derivatives of ent-sandaracompimaradiene argues that oryzalexin D is the relevant precursor to oryzalexins A – C (i.e., rather than proceeding via oxidation and then secondary hydroxylation). Although it remains unclear if there is a preferential route to the fully/dually oxidized oryzalexin C (i.e., via oryzalexin A or B), given that both are observed, and the promiscuity exhibited by MS3, it seems likely that either route can be utilized in planta. Accordingly, these results essentially complete elucidation of the oryzalexin family of phytoalexins. The location of MS1 and MS2 in the momilactone biosynthetic gene cluster imply a role for these in the production of momilactone A and/or B. Somewhat surprisingly, MS2 exhibits higher catalytic efficiency for oxidation of 3β-hydroxy-syn-pimaradien-19,6β-olide to momilactone A than MS1 (Table 2), which was previously suggested to carry out this reaction (Shimura, et al. 2007). Indeed, MS1 actually exhibits higher catalytic efficiency for oxidation of oryzalexin D to oryzalexin B, and both MS1 and MS2 actually seem to have higher affinity for oryzalexin D than 3β-hydroxy-syn-pimaradien-19,6β-olide (Table 2). Thus, while MS2 presumably is responsible for the final step in production of momilactone A, it will be of interest to further probe if MS1 has a different role in momilactone or other rice diterpenoid phytoalexin biosynthesis. Finally, although it seems likely that oxidation of an 11α-hydroxyl group occurs early in phytocassane biosynthesis given the presence of an 11-keto group in all of these phytoalexins, the relevant SDR does not appear to be among those investigated here. On the other hand, the ability of the investigated SDRs to readily oxidize simplified substrate analogs suggests that these SDRs might be relevant to the oxidation of 2α- and 3α-hydroxyl groups that presumably represent latter steps in phytocassane biosynthesis. Nevertheless, it will be necessary to carry out further studies to identify the presumably early acting C11 oxidase. While all three members of the MS clade from rice have already been investigated here, of particular interest in these future studies will be the twelve as of yet uncharacterized members of the MI clade from the SDR110C family found in rice. Material and methods General Unless otherwise noted, chemicals were purchased from Fisher Scientific (Loughborough, Leicestershire, UK), and molecular biology reagents from Invitrogen (Carlsbad, CA, USA). Gas chromatography (GC) was performed with a Varian (Palo Alto, CA) 3900 GC equipped with an HP-5MS column (Agilent, 0.25 μm, 0.25ID, 30 m) and 1.2 L mL/min He flow rate, with detection via a Saturn 2100 ion trap mass spectrometer (MS) in electron ionization mode (70 eV). Samples (1 μL) were injected in splitless mode with the injection port at 250 °C and oven at 50 °C, after holding for 3 min. at 50 °C the oven temperature was raised at 15 °C/min. to 300 °C, where it was held for an additional 3 min. MS data from 90 to 600 mass-to-charge ratio (m/z) was collected from 14 min. after injection until the end of the run. GC with flame ionization detection (GC-FID) was carried out with an Agilent 6890N GC also equipped with an HP-5MS column and using the same temperature program. High pressure liquid chromatography (HPLC) was carried out with an Agilent 1100 HPLC system equipped with auto-sampler, fraction collector and diode-array detector, run in reversed phase at 0.5 mL/min with a ZORBAX Eclipse XDB-C8 column (150 x 4.6 mm, 5 μm), using deionized water (dH2O) and acetonitrile (AcN). NMR spectra were recorded at 25 °C on a Bruker Avance 700 spectrometer (1H 700 MHz; 13C 174 MHz) equipped with a 5-mm HCN cryogenic probe for 1H and 13C, using standard experiments from the Bruker TopSpin v1.3 software. Compounds were dissolved in 0.5 mL deuterated chloroform (CDCl3) and placed in microtubes (Shigemi; Allison Park, PA) for analysis, with chemical shifts referenced using the known chloroform (13C 77.23, 1H 7.24 ppm) signals offset from trimethyl-silane. Substrate preparation Oryzalexin D, 3α-hydroxy-ent-(sandaraco)pimara-8(14),15-diene, 7β-hydroxy-ent-sandaracopimaradiene, 2α-hydroxy-ent-cassa-12,15-diene, 3α-hydroxy-ent-cassadiene, 11α-hydroxy-ent-cassadiene, and 3β-hydroxy-syn-pimara-7,15-diene were produced using a bacterial metabolic engineering system and purified as previously described (Kitaoka, et al. 2015, Swaminathan, et al. 2009, Wang, et al. 2012a, Wang, et al. 2011, Wang, et al. 2012b, Wu, et al. 2011, Wu, et al. 2013). 3β-Hydroxy-syn-pimara-7,15-dien-19,6β-olide was prepared by reduction of momilactone A as previously described (Kato et al. 1977). Briefly, lithium aluminium hydride (0.5 mg) was added to anhydrous ether solution (1 mL) containing 0.5 mg momilactone A. After continuously stirring at room temperature for 10 minutes, water was slowly poured into the reaction mixture. This was then twice extracted with an equal volumne of ether, and the combined organic phase was dried over Na2SO4 and dried in vacuo. The residue was resuspended in 50% AcN/dH2O, and purified by HPLC. After loading the sample, the column was washed for 2 min. with 50% (v/v) AcN/dH2O, with a gradient from 2 – 25 min. raising the AcN to 100%, and the column washed from 25 – 35 min. with 100% AcN. Fractions containing 3β-hydroxy-syn-pimaradien-19,6β-olide were identified by GC-MS analysis, pooled and dried under N2. Structural analysis was carried out using 1D 1H, DQF-COSY, HSQC, HMBC and NOESY spectra acquired at 700 MHz and 13C spectra acquired at 174 MHz. Correlations from the HMBC spectra were used to propose the majority of the structure, while connections between protonated carbons were obtained from DQF-COSY to complete the partial structure and assign proton chemical shifts. NOESY spectra provided Nuclear Overhauser Effect (NOE) cross-peak signal between the proton on C3 and that on C5 to assign the stereochemistry at C3. The 1H-NMR spectrum (700 MHz, CDCl3): δ 5.82 (1H, dd, J = 17.7, 11.7 Hz, H15), 5.64 (1H, d, J = 5.0 Hz, H5), 4.95 (1H, d, J = 17.7 Hz, H16Z), 4.90 (2H, m, H-6, H16E), 3.62 (1H, m, H3), 2.15 (1H, d, J = 11.7 Hz, H14b), 2.04 (1H, m, H2a), 1.99 (1H, dd, J = 11.7, 2.0 Hz, H15), 1.78 (1H, m H2b), 1.77 (1H, d, J = 3.9 Hz, H5), 1.64 (1H, m, H11b), 1.63 (1H, m, H9), 1.52 (1H, m, H12a), 1.52 (3H, s, H18), 1.50 (1H, m, H12b), 1.36 (1H, m, H1a), 1.31 (1H, m, H1b), 1.28 (1H, m, H11a), 1.04 (3H, s, H20), 0.84 (3H, s, H17). The 13C-NMR spectrum (174 MHz, CDCl3): 181.9 (C19), 149.6 (C8), 149.1 (C15), 114.4 (C7), 110.0 (C16), 74.6 (C6), 74.4 (C3), 51.3 (C9), 47.3 (C14), 46.9 (C5), 44.5 (C4), 40.1 (C13), 37.6 (C12), 33.1 (C10), 31.7 (C1), 29.4 (C2), 23.8 (C20), 22.9 (C18), 22.6 (C11), 21.9 (C17). Recombinant construct and protein purification OsMAS/SDR110C-MS1, OsSDR110C-MS3, OsSDR110C-MI2, OsSDR110C-MI3, and OsSDR110C-MI4 were obtained from the KOME rice cDNA databank (GeneBank accessions AK103462, AK110700, AK107157, AK070585, and AK06856, respectively). After several attempts to clone OsSDR110C-MS2 (GeneBank accession AK240900) were unsuccessful, resulting in only amplification of MS1, which has identical sequences at both the 5' and 3' ends of the coding region (corresponding to the primers used here), an alternative version was obtained by gene synthesis (GeneScript), with codon optimization for expression in E. coli (see supporting data for sequence). These were cloned into pENTR/SD/D-TOPO, and then transferred to pDEST17 by LR recombination before verification by complete gene sequencing. The resulting expression vectors were transformed into E. coli C41 OverExpress (Lucigen). The resulting recombinant bacteria were grown in 50 mL NZY medium at 37 °C overnight, and these cultures used to inoculate 1 L NYZ medium, and also shaken at 37 °C until their OD600 reached ~0.6–0.8. The culture was then induced with 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG), shaken at 16 °C for 16 hours. The bacteria were harvested by centrifugation and resuspended in 20 mL lysis buffer (50 mM Bis-Tris, 150 mM KCl, 20 mM Mg2SO4, 10% glycerol, 1 mM DTT, pH6.8). The cells were lysed by brief sonication and the soluble enzyme fraction was obtained by centrifugation (15,000 g, 25 min, 4 °C). The soluble enzyme fraction was incubated (rotating) with a slurry of Ni-NTA resin (2 mL) for 2 hours at 4 °C. The resin was washed by 20 mL wash buffer (50 mM Bis-Tris, 150 mM KCl, 20 mM imidazole, pH 6.8) and then eluted by 5 mL elution buffer (50 mM Bis-Tris, 150 mM KCl, 250 mM imidazole, pH 6.8). The imidazole was removed by dialysis, using 12–15 KDa cut-off membrane (Spectrum Chemical and Laboratory Products, Gardena, CA), twice against 1 L dialysis buffer (20 mM Tris-HCl, 150 mM KCl, 10% glycerol, pH 7.8). The resulting SDRs were utilized for enzymatic assays. Enzymatic assays Initial assays were conducted with 0.5 μM SDR, 50 μM substrate and 1 mM NAD+ in 0.5 mL Tris-HCl buffer (100 mM Tris-HCl, pH 8.0) with 10% (v/v) glycerol. After incubating at 30 °C for 1 hour, the reaction mixture was extracted thrice with 0.5 mL n-hexanes. The organic extracts were combined, dried under N2 gas and resuspended in 80 μL hexanes for analysis by GC-MS. Control assays were conducted without enzyme. Oxidation of the simplified substrate analogs was demonstrated by the reduction in m/z of the molecular ion from 288 to 286 in the mass spectra of the SDR and NAD+ dependent product peaks. Oxidation of 3β-hydroxy-syn-pimara-7,15-dien-19,6β-olide was verified by comparison of the SDR and NAD+ dependent product peak to a previously reported authentic sample of momilactone A (Xu et al. 2012). Kinetic analysis was carried out using 10–40 nM recombinant SDR in 0.5 mL assays run for 5–10 min at 30 °C, as determined by preliminary analyses to lie within the linear response range. The oxygenated diterpene substrates were added in varying concentrations (10–300 μM). To stop the reaction, assay vials were placed on ice and the enzymatic products immediately extracted with n-hexanes as above, and quantified by analysis with GC-FID. Verification of oryzalexin B Oryzalexin D (0.5 mg) was oxidized by MS1 as described above (40 nM SDR and 50 μM substrate), with the assay run overnight at 30 °C. This was then extracted trice with an equal volumne of hexanes, and the pooled extract was dried under N2 gas, and the residue redissolved in 50% AcN/dH2O for HPLC purification. After loading the sample, the column was washed for 2 min. with 50% (v/v) AcN/dH2O, with a gradient from 2 – 7 min. raising the AcN to 100%, and the column washed from 7 – 20 min. with 100% AcN. Fractions containing oryzalexin B were identified by GC-MS analysis, pooled and dried under N2 gas. The structure was confirmed by comparison of the 1H-NMR spectrum (700 MHz, CDCl3) with that previously reported (Kono, et al. 1985): δ 5.70 (1H, dd, J = 17.5, 10.3 Hz), 5.51 (1H, br.s), 4.87 (1H, d, J = 17.5 Hz), 4.86 (1H, d, J = 10.3), 4.18 (1H, br.s), 2.56 (1H, td, J = 14.5, 6.0 Hz), 2.26 (1H, dt, J = 14.5, 3.7 Hz), 2.1 (1H, t, J = 7.5 Hz), 1.98 (1H, dd, J = 11.8, 4.2 Hz), 1.93 (1H, ddd, 13.3, 5.7, 3.3 Hz), 1.20 ~ 1.69 (7H), 1.03 (3H, s), 0.992 (3H, s), 0.987 (3H, s), 0.90 (3H, s). Supplementary Material Supp FigS1-S6 Figure S1. Alignment of SDRs investigated in this report. Figure S2. SDR activity with the simplified substrate analog 3β-hydroxy-syn- pimaradiene. Figure S3. SDR activity with the simplified substrate analog 2α-hydroxy-ent-cassadiene. Figure S4. SDR activity with the simplified substrate analog 3α-hydroxy-ent-cassadiene. Figure S5. SDR activity with the simplified substrate analog 3α-hydroxy-ent- sandaracopimaradiene. Figure S6. SDR activity with the simplified substrate analog 7β-hydroxy-ent- sandaracopimaradiene. This work was supported in part by a grant from the NIH (GM109773 to R.J.P.). Figure 1 Biosynthetic gene clusters and prospective pathways for phytocassanes A – F and momilactones A & B. Figure 2 The rice SDR110C family. Phylogenetic analysis of the two main clades (*, inducible transcript accumulation; underlined, biochemical characterization). Almost all of the members of the monocot I (MI) clade are found in a tandem gene array. Figure 3 Simplified substrate analogs of oxo group containing rice diterpenoids. Figure 4 SDR catalyzed oxidation of the precursor 3β-hydroxy-syn-pimaradien-19,6β-olide (Pre) to momilactone A (MA), as indicated by GC-MS chromatograms. Figure 5 GC-MS chromatograms and mass spectra demonstrating SDR activity with oryzalexin D. Figure 6 Biosynthesis of oryzalexins A – C from oryzalexin D by MI3 and MS3 indicated by the results reported here. Table 1 OsSDR110C activity against oxygenated diterpenes.a Substrate MS1 MS2 MS3 MI2 MI3 3β-hydroxy-syn-pimaradien-19,6β-olide +++ +++ ++ – +++ 3β-hydroxy-syn-pimaradiene +/− +++ + – +++ oryzalexin D +++ +++ +++ – +++ 3α-hydroxy-ent-sandaracopimaradiene +/− + ++ – +++ 7β-hydroxy-ent-sandaracopimaradiene – – + – – 2α-hydroxy-ent-cassadiene +++ +++ +++ – + 3α-hydroxy-ent-cassadiene ++ ++ +++ – ++ 11α-hydroxy-ent-cassadiene – – – – – a Conversion rates: +++, > 50%; ++, 16–49%; +, 1–15%; +/−, <1%; –, not detectable. Table 2 OsSDR110C kinetic parameters with biosynthetic intermediates. Substrate SDR110C- KM (μM) kcat (s−1) kcat/KM (s−1 M−1) 3β-hydroxy-syn-pimaradien-19,6β-olide MS1 900 ± 400 (8 ± 3) x10−2 9 x101 MS2 200 ± 100 (4 ± 1) x10−1 2 x103 oryzalexin D MS1 60 ± 15 (2 ± 1) x10−2 3 x102 MS2 60 ± 20 (2.3 ± 0.3) x10−2 4 x102 MS3 40 ± 20 1.9 ± 0.2 5 x104 MI3 14 ± 10 2.7 ± 0.4 2 x105 Atawong A Hasegawa M Kodama O 2002 Biosynthesis of rice phytoalexin: enzymatic conversion of 3β-hydroxy-9β-pimara-7,15-dien-19,6β-olide to momilactone A Biosci Biotechnol Biochem 66 566 570 12005050 Bednarek P Osbourn A 2009 Plant-microbe interactions: chemical diversity in plant defense Science 324 746 748 19423814 Hammerschmidt R 1999 PHYTOALEXINS: What Have We Learned After 60 Years? Annual review of phytopathology 37 285 306 Kallberg Y Oppermann U Jornvall H Persson B 2002 Short-chain dehydrogenases/reductases (SDRs) European journal of biochemistry / FEBS 269 4409 4417 Kato T Aizawa H Tsunekawa M Sasaki N Kitahara Y Takahashi N 1977 Chemical transformations of the diterpene lactones momilactones A and B J Chem Soc, Perkin Trans I 250 254 Kato-Noguchi H Peters RJ 2013 The role of momilactones in rice allelopathy J Chem Ecol 39 175 185 23385366 Kavanagh KL Jornvall H Persson B Oppermann U 2008 Medium- and short-chain dehydrogenase/reductase gene and protein families : the SDR superfamily: functional and structural diversity within a family of metabolic and regulatory enzymes Cellular and molecular life sciences : CMLS 65 3895 3906 19011750 Kikuchi S Satoh K Nagata T Kawagashira N Doi K Kishimoto N Yazaki J Ishikawa M Yamada H Ooka H Hotta I Kojima K Namiki T Ohneda E Yahagi W Suzuki K Li CJ Ohtsuki K Shishiki T Otomo Y Murakami K Iida Y Sugano S Fujimura T Suzuki Y Tsunoda Y Kurosaki T Kodama T Masuda H Kobayashi M Xie Q Lu M Narikawa R Sugiyama A Mizuno K Yokomizo S Niikura J Ikeda R Ishibiki J Kawamata M Yoshimura A Miura J Kusumegi T Oka M Ryu R Ueda M Matsubara K Kawai J Carninci P Adachi J Aizawa K Arakawa T Fukuda S Hara A Hashidume W Hayatsu N Imotani K Ishii Y Itoh M Kagawa I Kondo S Konno H Miyazaki A Osato N Ota Y Saito R Sasaki D Sato K Shibata K Shinagawa A Shiraki T Yoshino M Hayashizaki Y 2003 Collection, mapping, and annotation of over 28,000 cDNA clones from japonica rice Science 301 376 379 12869764 Kitaoka N Wu Y Xu M Peters RJ 2015 Optimization of recombinant expression enables discovery of novel cytochrome P450 activity in rice diterpenoid biosynthesis Appl Microbiol Biotechnol 99 7549 7558 25758958 Kono Y Takeuchi S Kodama O Sekido H Akatsuka T 1985 Novel phytoalexins (oryzalexins A, B and C) isolated from rice blast leaves infected with Pyricularia oryzae Part II: Structural studies of oryzalexins Agric Biol Chem 49 1695 1701 Miyamoto K Fujita M Shenton MR Akashi S Sugawara C Sakai A Horie K Hasegawa M Kawaide H Mitsuhashi W Nojiri H Yamane H Kurata N Okada K Toyomasu T 2016 Evolutionary trajectory of phytoalexin biosynthetic gene clusters in rice Plant J Moummou H Kallberg Y Tonfack LB Persson B van der Rest B 2012 The plant short-chain dehydrogenase (SDR) superfamily: genome-wide inventory and diversification patterns BMC Plant Biol 12 219 23167570 Nutzmann HW Osbourn A 2014 Gene clustering in plant specialized metabolism Curr Opin Biotechnol 26 91 99 24679264 Okada A Shimizu T Okada K Kuzuyama T Koga J Shibuya N Nojiri H Yamane H 2007 Elicitor induced activation of the methylerythritol phosphate pathway towards phytoalexin biosynthesis in rice Plant Mol Biol 65 177 187 17634747 Persson B Kallberg Y Bray JE Bruford E Dellaporta SL Favia AD Duarte RG Jornvall H Kavanagh KL Kedishvili N Kisiela M Maser E Mindnich R Orchard S Penning TM Thornton JM Adamski J Oppermann U 2009 The SDR (short-chain dehydrogenase/reductase and related enzymes) nomenclature initiative Chem Biol Interact 178 94 98 19027726 Peters RJ 2006 Uncovering the complex metabolic network underlying diterpenoid phytoalexin biosynthesis in rice and other cereal crop plants Phytochemistry 67 2307 2317 16956633 Prisic S Xu M Wilderman PR Peters RJ 2004 Rice contains two disparate ent-copalyl diphosphate synthases with distinct metabolic functions Plant Physiol 136 4228 4236 15542489 Schmelz EA Huffaker A Sims JW Christensen SA Lu X Okada K Peters RJ 2014 Biosynthesis, elicitation and roles of monocot terpenoid phytoalexins Plant J 79 659 678 24450747 Shimura K Okada A Okada K Jikumaru Y Ko KW Toyomasu T Sassa T Hasegawa M Kodama O Shibuya N Koga J Nojiri H Yamane H 2007 Identification of a biosynthetic gene cluster in rice for momilactones J Biol Chem 282 34013 34018 17872948 Swaminathan S Morrone D Wang Q Fulton DB Peters RJ 2009 CYP76M7 is an ent-cassadiene C11α-hydroxylase defining a second multifunctional diterpenoid biosynthetic gene cluster in rice Plant Cell 21 3315 3325 19825834 Toyomasu T 2008 Recent Advances Regarding Diterpene Cyclase Genes in Higher Plants and Fungi Biosci Biotechnol Biochem 72 1168 1175 18460786 Wang Q Hillwig ML Okada K Yamazaki K Wu Y Swaminathan S Yamane H Peters RJ 2012a Characterization of CYP76M5-8 indicates metabolic plasticity within a plant biosynthetic gene cluster J Biol Chem 287 6159 6168 22215681 Wang Q Hillwig ML Peters RJ 2011 CYP99A3: Functional identification of a diterpene oxidase from the momilactone biosynthetic gene cluster in rice Plant J 65 87 95 21175892 Wang Q Hillwig ML Wu Y Peters RJ 2012b CYP701A8: A rice ent-kaurene oxidase paralog diverted to more specialized diterpenoid metabolism Plant Physiol 158 1418 1425 22247270 Wilderman PR Xu M Jin Y Coates RM Peters RJ 2004 Identification of syn-pimara-7,15-diene synthase reveals functional clustering of terpene synthases involved in rice phytoalexin/allelochemical biosynthesis Plant Physiol 135 2098 2105 15299118 Wu X Knapp S Stamp A Stammers DK Jornvall H Dellaporta SL Oppermann U 2007 Biochemical characterization of TASSELSEED 2, an essential plant short-chain dehydrogenase/reductase with broad spectrum activities FEBS J 274 1172 1182 17298439 Wu Y Hillwig ML Wang Q Peters RJ 2011 Parsing a multifunctional biosynthetic gene cluster from rice: Biochemical characterization of CYP71Z6 & 7 FEBS Lett 585 3446 3451 21985968 Wu Y Wang Q Hillwig ML Peters RJ 2013 Picking sides: Distinct roles for CYP76M6 and -8 in rice oryzalexin biosynthesis Biochem J 454 209 216 23795884 Xu M Galhano R Wiemann P Bueno E Tiernan M Wu W Chung IM Gershenzon J Tudzynski B Sesma A Peters RJ 2012 Genetic evidence for natural product-mediated plant-plant allelopathy in rice (Oryza sativa) New Phytol 193 570 575 22150231 Zi J Mafu S Peters RJ 2014 To Gibberellins and Beyond! Surveying the Evolution of (Di)Terpenoid Metabolism Annu Rev Plant Biol 65 259 286 24471837

PMC005xxxxxx/PMC5118074.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7503062 4443 J Am Geriatr Soc J Am Geriatr Soc Journal of the American Geriatrics Society 0002-8614 1532-5415 27626617 5118074 10.1111/jgs.14347 NIHMS784760 Article Self-reported Sleep Problems Prospectively Increase Risk of Disability: Findings from the Survey of Midlife Development in the United States (MIDUS) Friedman Elliot M. PhD 1 1 Department of Human Development and Family Studies, Purdue University, West Lafayette, IN Corresponding Author: Elliot M. Friedman, PhD, Department of Human Development and Family Studies, 1202 West State Street, West Lafayette, IN 47907, Ph: 765-496-6378, FAX: 765-496-1144, efriedman@purdue.edu 10 5 2016 14 9 2016 11 2016 01 11 2017 64 11 22352241 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objectives To determine whether subjective poor sleep prospectively increases functional limitations and incident disability in a national sample of adults living in the United States. Design Prospective cohort Setting Longitudinal Survey of Midlife Development in the United States (MIDUS) Participants Young, middle aged, and older men and women (age 24–75) surveyed in 1995–1996 (MIDUS 1) and followed up in 2004–2006 (MIDUS 2). Complete data were available for 3,620 respondents. Measurements Data were from telephone interviews and self-administered questionnaires. Participant reported chronic sleep problems within the prior month; functional limitations were assessed using the Functional Status Questionnaire. Demographic (age, sex, race), socioeconomic (educational attainment), health (chronic conditions, depression), and health behavior (obesity, smoking) covariates were assessed to reduce potential confounding. Results Approximately 11% of the sample reported chronic sleep problems at both MIDUS waves. Average numbers of basic and intermediate ADL limitations increased significantly between MIDUS 1 (basic: .06; intermediate: .95) and MIDUS 2 (basic: .15; intermediate: 1.6; P<.001). Adjusted regression models estimating change in ADL scores showed that chronic sleep problems at MIDUS 1 predicted significantly greater increases in both basic (incident rate ratio (IRR) = 1.55, P<.001) and intermediate (IRR = 1.28, P<.001) ADL limitations. In those with no functional limitations at baseline, logistic regression models showed that chronic sleep problems significantly increased the odds of incident basic (OR = 2.33, CI:1.68,3.24, P<.001) and intermediate (OR = 1.70, CI: 1.21,2.42, P<.01) ADL disability. Conclusion Reports of chronic sleep problems predicted greater risk of both onset of and increases in functional limitations 9–10 years later. Poor sleep may be a robust and independent risk factor for disability in adults of all ages. subjective sleep ADLs MIDUS disability INTRODUCTION Disability rates in older adults have declined in recent years,1,2 due principally to medical and technological advances and to broad improvements in socioeconomic indicators, including poverty and educational attainment.3 Nonetheless, inability to perform daily activities remains common. Estimates from the National Health Interview Surveys and the National Long-Term Care Survey show that 15%–20% of adults over age 65 have at least one functional limitation.1,2 Further improvements in disability rates will largely rest on changes in lifestyle factors. Obesity4 and physical inactivity,5 for example, are robustly linked to increased risk of disability,1,6 and rates of both are high in aging adults; these factors threaten to slow or stall the downward trends in disability rates. Another lifestyle factor – sleep – has received relatively little attention as a predictor of disability, despite well-documented links between sleep problems and a range of adverse health outcomes, including mortality.7–10 This study examines the prospective associations between self-reported sleep problems and activities of daily living (ADL) in a national sample of middle-aged and older adults living in the United States. Two lines of research converge on sleep as a compelling focus for understanding disability risk. First, sleep quality tends to decline with age. Objective assessments show significant age-related declines in sleep duration, sleep efficiency, rapid eye movement (REM), and stage 3/4 sleep coupled with significant increases in sleep latency and stage 1–2 sleep.11 Subjective complaints of poor sleep also tend to increase with age.12,13 Second, both quantitative (e.g. number of hours slept per night) and qualitative (e.g. complaints of poor sleep) aspects of sleep are linked to morbidity and mortality in older adults.7,8,14 Subjective reports of impaired sleep, for example, significantly increased the risk of subsequent heart attack, stroke, and cardiac mortality in older men and women with primary acute myocardial infarction from the Stockholm Heart Epidemiology Program.14 A regional longitudinal study in Sweden found that sleep complaints at baseline predicted greater risk of coronary artery disease15 and diabetes16 in men 12 years later. Few studies have examined the prospective links between sleep complaints and disability. A recent study of 908 older Catholic clergy in the United States found that poor subjective sleep significantly increased risk of disability at follow-up 9.6 years later.17 The current study extends this earlier work by examining the prospective associations of sleep complaints and disability in a representative national sample of men and women in the US: the Survey of Mid-Life Development in the United States (MIDUS)18. We hypothesized that subjective report of sleep problems would increase the risk of both increases in functional limitations and incident disability in those with no functional impairments at baseline. The MIDUS sample is community dwelling, making the results applicable to the general adult population of the US. The broad age range of the MIDUS sample, spanning 5 decades from mid-20s to mid-70s, also makes it possible to determine whether the links between poor sleep and functional impairment are limited to older adults or extend to mid-life and younger adults as well. Finally, while the main analyses in this study involve self-reported functional limitations, these are bolstered by supplemental analyses using objective assessments of functional status available for sub-samples of MIDUS participants. METHODS Participants Data are from the first two waves of MIDUS, a longitudinal study of the physical and mental health of middle-aged and older adults.18 The first wave of data collection (MIDUS 1; N = 7,108) included a national probability sample of non-institutionalized English-speaking adults living in the contiguous United States recruited by random digit dialing (RDD; n = 3,487), a sample of monozygotic and dizygotic twin pairs from a national twin registry (n = 1,914), oversamples from five metropolitan areas (n = 757), and siblings of individuals from the RDD sample (n = 950). Respondents completed telephone interviews and self-administered questionnaires (SAQ). A follow-up study was completed 9–10 years later (MIDUS 2). Mortality-adjusted retention from the original study was 75%. Complete data for the present analyses were available for 3,620 participants. Compared to the full longitudinal sample, the analytical sample had fewer female respondents, was better educated, and had fewer functional limitations at baseline; they were comparable on all other key variables. Collection of data for both waves of MIDUS and analyses for the current study were approved by the Institutional Review Boards at the University of Wisconsin-Madison and Purdue University, respectively. Measures Sleep complaints were assessed using a single questionnaire item: “In the past 12 months, have you experienced or been treated for chronic sleeping problems?” A dichotomous variable was used in all models. Information on basic and intermediate activities of daily living came from the Functional Status Questionnaire.19 Respondents were asked how much health limited their ability to do a number of activities. Basic ADL limitations were determined from two items: “bathing or dressing yourself” and “walking one block.” Intermediate ADL limitations were determined from seven items: “lifting or carrying groceries,” “climbing several flights of stairs,” “bending, kneeling, or stooping,” “walking more than a mile,” “walking several blocks,” “vigorous activities (e.g., running, lifting heavy objects),” and “moderate activities (e.g., bowling, vacuuming).” Response ranged from 1=Not at all to 4=A lot. To determine the number of activities for which respondents reported at least some degree of limitation, responses of “Some” or “A lot” of limitation were scored ‘1’ and other responses ‘0.’ Responses were then summed into separate scores for basic and intermediate ADLs with possible scores of 0–2 for basic and 0–7 for intermediate scales. Total scores at both MIDUS 1 and MIDUS 2 were calculated and used to examine changes in numbers of functional limitations over time. Dichotomous variables for presence or absence of any limitations were also created for logistic regression models estimating incident functional limitations between MIDUS 1 and MIDUS 2. A set of demographic, socioeconomic, health, and health behavior covariates was included in all models to reduce the likelihood of confounding. These included age, sex, race, and educational attainment (dummy coded variables for high school degree or GED, some college, and college degree or more). To assess health, a variable for 12 chronic medical conditions was used in all analyses. Information on 9 of these conditions came from participant responses to self-administered questionnaire items; participants were asked whether they had experienced or received treatment for any of the following conditions in the prior 12 months: chronic obstructive pulmonary disease (COPD), arthritis or other joint conditions, AIDS, hypertension, diabetes, tuberculosis, neurological disorders, stroke, ulcer. Presence of heart problems and cancer were determined from single items in the telephone interview. Participants were asked whether they had had heart trouble suspected or confirmed by a doctor and whether they had ever had cancer. Possible scores ranged from 0–12. Variables for obesity and smoking were included to control for health behaviors. Body mass index (BMI) was calculated from participant measurement of height and weight and dummy coded variables for normal weight (BMI<24.99), overweight (BMI between 25.00 and 29.99) and obese (BMI>=30.00) were created. Smoking status was assessed using dummy-coded variables indicating non-smoker, ex-smoker, and current smoker. Finally, as depressed individuals are more likely to report both poor sleep20 and greater functional impairment,21 likely depression was determined using the short form of the Composite International Diagnostic Interview (CIDI).22 Respondents were scored as positive for depressed affect if they indicated 1) that they felt sad, blue, or depressed and that “The feeling of being sad, blue, or depressed lasted ‘All day long’ or ‘Most of the day’ and 2) that they felt this way “Everyday” or “Almost every day.” They were scored as positive for anhedonia if they reported a loss of interest in most things lasting “All day long” or “Most of the day” and that this feeling was “Everyday” or “Almost every day.” A dichotomous variable indicating likely clinical depression (i.e. positive scores for both depressed affect and anhedonia) was included in all models. Statistical analyses Longitudinal increases in ADL limitations were estimated in separate Poisson regression models. Poisson modeling was appropriate because the outcome measures (numbers of limitations) were counts rather than continuous variables. Incident rate ratios (IRR) comparing the rate of functional limitations in respondents with sleep problems to those without were determined; these are interpreted in a similar fashion as odds ratios. All models adjusted for age, sex, race, educational attainment, and functional limitations at MIDUS 1. To control for the possible influences of coincident illness, obesity, and depressive symptoms on functional abilities, MIDUS 2 measures of number of chronic medical conditions, smoking status, BMI, and depression were included in all models. Poisson models used data from the full longitudinal sample. Binary logistic regression models were used to estimate the odds of incident basic and intermediate ADL disability between MIDUS 1 and MIDUS 2, adjusted for covariates. In these models, the analytical samples were limited to respondents with no functional limitations at MIDUS 1 (n = 3,244 for basic ADLs, and n=1,329 for intermediate ADLs). Models were estimated using Stata 13.0 (Statacorp., College Station, TX). Given age-related changes in both sleep complaints and disability risk, and to determine whether the associations between sleep and ADL limitations varied with age, we conducted additional analyses that included interaction terms for sleep problems and age as predictors of longitudinal changes in ADL and incident ADL impairments. Clustered robust standard errors were applied to account for familial relatedness among the twins and siblings in the sample. A threshold for statistical significance was set at alpha = 0.05 in all models. RESULTS Descriptive statistics for the full analytical sample are shown in Table 1. Sociodemographic characteristics are from MIDUS 1. Mean age was 46.5 years, slightly more than half the sample was female, 6.3% were non-White, and 35.9% had completed a 4-year college degree or more. Data on other variables were from both MIDUS 1 and MIDUS 2. Between the two data collection points the average number of basic (MIDUS 1 = 0.06; MIDUS 2 = 0.15; t(3,619)=−12.2, p<.001) and intermediate ADL limitations (MIDUS 1 = 0.95; MIDUS 2 = 1.61; t(3,619)=−20.2, p<.001) rose significantly while the fraction of the sample reporting sleep problems did not change significantly (MIDUS 1 = 11.3; MIDUS 2 = 10.7; χ2=0.91, p=0.33). Of those who reported sleep problems at MIDUS 1, 38.9% continued to report sleep problems at MIDUS 2 (data not shown). The fraction of people who met criteria for depression on the CIDI-SF declined significantly between MIDUS 1 (12.0 %) and MIDUS 2 (10.1%; χ2=10.32, p=.001), and average number of chronic conditions increased significantly between MIDUS 1 and MIDUS 2 (t(3,619)=28.89, p<.001). Poisson regression models were used to estimate the increase in basic and intermediate ADL limitations associated with reporting poor sleep at MIDUS 1 adjusted for MIDUS 1 sociodemographic characteristics and initial levels of functional limitations and MIDUS 2 health, health behavior, and depression. As shown in Table 2, reporting chronic sleep problems at MIDUS 1 was associated with significantly increased rates of both basic and intermediate ADL limitations. Compared to those without sleep complaints, respondents who reported chronic sleep problems in the prior year showed a 55% increase in basic (p<.001) and a 28% increase in intermediate ADL limitations (p<.001). Greater risk of increases in functional impairments was also associated with greater age, being female, being White (intermediate ADLs only), not having completed a 4-year college degree or more, greater numbers of chronic medical conditions, obesity, higher scores on the CIDI depression scale, and smoking either currently or in the past. The likelihood of incident disability between MIDUS 1 and MIDUS 2 was estimated using logistic regression models; analytical samples were limited to those respondents without ADL limitations at MIDUS 1. As shown in Table 3, compared to those with no sleep complaints, respondents who reported chronic sleep problems at MIDUS 1 were more than twice as likely to develop basic ADL limitations (p<.001) and 70% more likely to develop intermediate ADL limitations (p<.05). Age, being female, low educational attainment, being overweight or obese, more chronic conditions, depression, and smoking were all associated with likelihood of incident functional limitations at MIDUS 2. Analyses that included age X sleep problems interaction terms showed that age significantly moderated the association of MIDUS 1 sleep problems and longitudinal increases in intermediate ADL limitations. Specifically, the rate of increases in limitations among younger and middle-aged adults (e.g. 45 year-olds) who reported sleep problems (increase of 1.04 limitations) was double that of peers with no sleep complaints (2.13 increase; interaction effect: IRR = 0.98, p<.001). This interaction is displayed in Figure 1. Age did not moderate the association between sleep problems and change in basic ADL limitations. Analytical refinements We probed these results in four ways to rule out potential alternative explanations of the observed associations. First, 39% of full analytical sample reported chronic sleep problems at both MIDUS 1 and MIDUS 2. It is possible that the observed longitudinal associations for sleep complaints are accounted for by coincident poor sleep and functional limitations at MIDUS 2. To test this possibility, all regression models were re-estimated with an additional adjustment for sleep complaints at MIDUS 2. Inclusion of MIDUS 2 sleep complaints slightly reduced the coefficients for MIDUS 1 sleep complaints, but the associations remained robust (p<.001 in both basic and intermediate ADL models). In the logistic regression models, the likelihood that someone with chronic sleep problems would develop basic ADL limitations at MIDUS 2 declined from 133% that of someone without sleep complaints to 78%, but the effect remained statistically significant (p=.001). For intermediate ADL limitations, the odds declined from 79% to 69% and became marginally significant (p=.05). In all of these analyses, the coefficients and odds ratios associated with sleep problems at MIDUS 1 were consistently larger than those associated with sleep problems at MIDUS 2 (data not shown). Second, as sleep complaints were assessed using a single item that was subjective and global in nature, we compared these responses to scores on a widely used measure of sleep quality, the Pittsburgh Sleep Quality Index23 (PSQI; this measure was only used in a sub-sample of MIDUS 2 respondents (n = 1,063), so it was unavailable for longitudinal analyses). The PSQI is a 24-item scale assessing 7 different sleep components: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Compared to respondents without sleep problems, those who reported sleep problems at MIDUS 2 had significantly higher scores on the PSQI (M = 9.47 vs. 5.47, t(997) = 153.81, p<.001), indicating significantly poorer sleep quality. Third, use of sleep medication in the context of sleep complaints may contribute to functional limitations. The only variable available at MIDUS 1 referred to the use of sedatives without a doctor’s prescription; 105 MIDUS 1 respondents (2.3% of the sample) reported taking such medication, and inclusion of this variable in longitudinal models did not affect the relationship between MIDUS 1 sleep problems and MIDUS 2 functional limitations (data not shown). MIDUS 2 data, in contrast, included a specific question about the use of sleeping pills as part of the PSQI. To gauge the association of sleep medication use and ADLs cross-sectionally at MIDUS 2, we regressed numbers of basic and intermediate ADL limitations (separate models) on the 7 individual PSQI components, adjusted for age, sex, race, and education. The results showed that only the sleep disturbance (assessing trouble sleeping) and daytime dysfunction (assessing sleepiness during waking hours) components were associated with ADL limitations (p<.001 for both); use of sleep medication was not independently associated with either type of ADL (data not shown). Finally, sleep problems and functional limitations were both assessed using self-report measures, raising the possibility that unobserved subjective factors may underlie both sets of ratings. To probe this issue, we examined an objective measure of functional status – a 50 foot timed walk – and potential longitudinal effects of sleep complaints; walk time was measured in a sub-sample of MIDUS 2 respondents (n = 1,063). Linear regression models predicting walk time adjusted for sociodemographic, health, and health behavior covariates showed that on average respondents who reported chronic sleep problems at MIDUS 1 took more than 1 second longer to walk 50 feet than those who slept better (15.4 vs. 14.2 sec; p<.05); this association remained significant after adjusting for sleep problems at MIDUS 2 (p<.05). DISCUSSION The prospect of disability is a signature concern among aging adults, and as the population ages a better understanding of the causes of disability is increasingly important for individual health and quality of life as well as policies designed to improve overall population health. These analyses focused on subjectively poor sleep as a potential risk factor for disability. Reports of chronic sleep problems were associated with increases in basic and intermediate ADL limitations and with significantly greater odds of developing functional limitations at follow-up. In the case of basic ADL limitations, reporting poor sleep more than doubled the risk of incident disability. These associations were observed after accounting for a set of established disability risk factors, including advancing age, low educational attainment, chronic medical conditions, obesity, depression, and smoking. Moreover, with the exception of cross-time increases in intermediate ADL limitations, the relationship between sleep problems and disability did not vary with age (and for intermediate ADLs, differences based on sleep problems were more evident among younger and middle aged adults than older adults). Collectively, these results suggest that poor subjective sleep is a robust and independent risk factor for functional limitations, and that this risk is not limited to later life. Impaired or insufficient sleep has been linked to a variety of diseases and associated risk factors. Population-based studies have consistently shown that routine sleep duration that is shorter or longer than the optimal amount (typically 7 hours a night) predicts greater risk of mortality.8–10,24 Objectively assessed low sleep quality, often associated with disruptions due to sleep-disordered breathing, increases the risk of hypertension, diabetes, cardiovascular disease, stroke, disability, and mortality.25–28 Subjective sleep ratings have also been linked to a range of adverse health outcomes, including diabetes, cardiovascular disease, and cardiac mortality.14–16 In spite of these associations, relatively few studies have focused on disability risk associated with poor sleep, and fewer still have involved representative population samples. Regional studies in Italy29 and China30 have linked low subjective sleep quality to greater risk of disability, and a recent study in the US reported greater risk of incident disability in members of the clergy who reported sleep problems.17 The current study of a larger, community-based, nationally representative sample with a broader age range now adds robust support to the conclusion that poor sleep is an independent risk factor for disability in aging adults. Poor sleep may lead to impaired function by way of a number of paths. Physical activity, for example, is protective against functional decline in aging men and women. Subjective reports of poor sleep are linked to fatigue31 that is often sufficient to limit daily activities in older adults.32 Physical activity in adults who sleep poorly may thus be reduced to levels that increase risk of functional limitations. Poor sleep is also linked to dysregulation of diverse physiological systems. For example, studies using both objective and subjective assessments show that naturally occurring poor sleep is associated with higher circulating levels of inflammatory proteins.33–37 In parallel experimental studies, sleep restriction reliably produces elevated levels of inflammation both acutely and chronically.38–41 Inflammation in turn is linked prospectively with increased risk of disability,42 and experimental studies highlight a role for inflammatory proteins in the loss of muscle tissue that can result in sarcopenia43,44 and associated functional limitations. Poor sleep has also been shown to increase risk of obesity prospectively,36,45,46 and obesity increases risk of disability.6 There are thus multiple behavioral and physiological routes by which sleep may be linked to subsequent disability risk, although specific mechanisms have yet to be elucidated. This study has several important limitations. Principally, sleep problems and functional limitations were both reported by MIDUS respondents rather than assessed using objective measures. This leaves open the possibility that their associations are explained by one or more unmeasured variables that capture a common subjective dimension. A number of observations from the current study make this possibility less likely. In the sub-sample of respondents who completed an objective assessment of functional status – the timed walk – those who reported sleep problems 9–10 years earlier were significantly slower, independent of current sleep problems. Moreover, sleep problems were claimed by many MIDUS 1 respondents who reported no functional limitations, suggesting some independence between these measures, and those who did report sleep problems were twice as likely to develop substantial limitations in the intervening years. Finally, those who reported sleep problems had an average score on the PSQI, a widely used measure of sleep and sleep pathology, that was almost double that of people who reported none. All of these observations increase confidence that the observed links between sleep problems and functional status are not spurious. Another issue worth noting is that subjective reports of poor sleep often do not match the results of objective sleep assessments.13,47,48 Nonetheless, subjective complaints of poor sleep are meaningfully linked to health outcomes independently of objectively determined sleep patterns,49 the implication being that subjective and objective assessments capture unique aspects of sleep that are both important for understanding how sleep affects health. Against these limitations are substantial strengths, including a large, nationally representative sample, a large time difference between the waves of data collections for assessing long-term change, and the availability of data with which to control for confounding and to probe for alternative explanations. The current results suggest that subjective reports of poor sleep significantly increase disability risk independently of demographic characteristics, socioeconomic status, health, or health behavior, and that such risk extends to middle-aged men and women as well as older adults. These results add to a growing literature citing the importance of sleep to good health and the resulting need for effective ways to promote good sleep in the general population.50 This research was supported by grant R01-AG041750 (to EMF) from the National Institute on Aging. The MIDUS I study (Midlife in the U.S.) was supported by the John D. and Catherine T. MacArthur Foundation Research Network on Successful Midlife Development. The MIDUS II research was supported by a grant from the National Institute on Aging (P01-AG020166) to conduct a longitudinal follow-up of the MIDUS I investigation. Figure 1 Age X sleep problems interaction predicting longitudinal change in intermediate ADLs. Sleep problems were more likely to produce greater increases in intermediate ADL limitations among younger and middle-aged adults (<65 years old) than among older adults. Table 1 Descriptive statistics for longitudinal sample (N = 3,620). MIDUS 1 (1995–1996) MIDUS 2 (2004–2006) Variable Mean (SD) Range % Mean (SD) Range % Age 46.5 (12.5) 20–75 -- -- -- Sex (% female) 55.2 -- -- -- Race (% non-White) 6.3 -- -- -- Educational attainment  High school or GED 35.0 -- -- --  Some college 29.1 -- -- --  College or more 35.9 -- -- -- BADLs  Number of limitations 0.06 (0.3) 0–2 0.15 (0.4) 0–2  % with 1 or more BADLs 5.01 12.18 IADLs  Number of limitations 0.95 (1.8) 0–7 1.61 (2.2) 0–7  % with 1 or more IADLs 35.53 48.94 CIDI-SF depression (% Positive) 12.0 10.1 Chronic conditions 0.8 (1.1) 0–8 1.4 (1.5) 0–10 BMI categories  <25.00 (Normal weight) 41.4 32.3  25.00–29.99 (Overweight) 37.6 39.4  >=30.00 (Obese) 21.0 28.3 Smoking status  Current smoker 25.2 19.5  Ex-smoker 40.7 48.9  Never smoked 34.1 31.6 Sleep problems (% yes) 11.3 10.7 Table 2 Basic and Intermediate ADL limitations at MIDUS 2 regressed on MIDUS 1 sleep problems, functional limitations, and covariates (N = 3,620). Estimates were from Poisson regression models, and incident rate ratios (IRR) are shown for ease of interpretation. Basic ADLs Intermediate ADLs Variable IRR [95% CI] IRR [95% CI] Age 1.03*** [1.02,1.04] 1.02*** [1.02,1.03] Sex (female=1) 1.30*** [1.08,1.56] 1.29*** [0.07,0.15] Race (non-White=1) 0.88 [0.64,1.21] 0.85** [0.76,0.94] Educational attainment  High school or GED 1.90*** [1.50,2.41] 1.30*** [1.22,1.39]  Some college 1.42** [1.10,1.85] 1.10* [1.02,1.18]  College or more REF. REF. ADLs at wave 1 1.88*** [1.62,2.19] 1.16*** [1.14,1.17] Chronic conditions (MIDUS 2) 1.30*** [1.24,1.37] 1.17*** [1.15,1.19] BMI categories (MIDUS 2)  <25.00 (Normal weight) REF. REF.  25.00–29.99 (Overweight) 1.13 [0.89,1.44] 1.12** [1.04,1.20]  >=30 (Obese) 1.80*** [1.42,2.28] 1.52*** [1.42,1.63] CIDI-SF depression (MIDUS 2) 1.42** [1.12,1.79] 1.19*** [1.10,1.28] Smoking status (MIDUS 2)  Never smoked REF. REF.  Ex-smoker 1.26* [1.04,1.53] 1.11*** [1.05,1.18]  Current smoker 1.52** [1.19,1.94] 1.41*** [1.31,1.52] Sleep problems (MIDUS 1) 1.55*** [1.26,1.90] 1.28*** [1.20,1.37] * p<.05; ** p<.01; *** p<.001 Table 3 Logistic regression models predicting functional limitations at MIDUS 2. Odds ratios and 95% confidence intervals are shown. Basic and Intermediate ADL limitations were estimated in separate models. Only cases with no limitations at MIDUS 1 were included. Basic ADLs (n = 3,244) Intermediate ADLs (n = 1,329) Variable Odds Ratio [95% CI] Odds Ratio [95% CI] Age 1.04*** [1.03,1.05] 1.04*** [1.03,1.05] Sex (female=1) 1.73*** [1.32,2.27] 1.49*** [1.22,1.82] Race (non-White=1) 0.82 [0.46,1.46] 0.90 [0.61,1.35] Educational attainment  High school or GED 2.29*** [1.64,2.46] 1.39** [1.10,1.76]  Some college 1.55* [1.09,2.22] 1.21 [0.96,1.53]  College or more REF. REF. Chronic conditions (MIDUS 2) 1.54*** [1.40,1.69] 1.40*** [1.29,1.52] BMI categories (MIDUS 2)  <25.00 (Normal weight) REF. REF.  25.00–29.99 (Overweight) 1.42* [1.01,1.99] 1.38** [1.11,1.73]  >=30.00 (Obese) 2.81*** [1.99,3.97] 2.32*** [1.78,3.02] Depression (MIDUS 2) 1.42# [0.97,2.08] 1.82** [1.30,2.57] Smoking status (MIDUS 2)  Never smoked REF. REF.  Ex-smoker 1.55** [1.17,2.06] 1.01 [0.82,1.25]  Current smoker 2.14*** [1.49,3.08] 2.26*** [1.69,3.01] Sleep problems (MIDUS 1) 2.33*** [1.68,3.24] 1.70** [1.21,2.42] # p=.07; * p<.05; ** p<.01; *** p<.001 Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: Lead author was responsible for all phases of the study, from conceptualization to data analysis to writing the manuscript. Sponsor’s Role: Sponsor had no role in the project. 1 Manton KG Recent declines in chronic disability in the elderly U.S. population: Risk factors and future dynamics Annu Rev Public Health 2008 29 91 113 18031222 2 Schoeni RF Martin LG Andreski PM Persistent and growing socioeconomic disparities in disability among the elderly: 1982–2002 Am J Public Health 2005 95 2065 2070 16254235 3 Schoeni RF Freedman VA Martin LG Why is late-life disability declining? Milbank Q 2008 86 47 89 18307477 4 Flegal KM Carroll MD Kit BK Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999–2010 JAMA 2012 307 491 497 22253363 5 Buchman AS Wilson RS Yu L Total daily activity declines more rapidly with increasing age in older adults Arch Gerontol Geriatr 2014 58 74 79 24007938 6 Alley DE Chang VW The changing relationship of obesity and disability, 1988–2004 JAMA 2007 298 2020 2027 17986695 7 Ayas NT White DP Manson JE A prospective study of sleep duration and coronary heart disease in women Arch Intern Med 2003 163 205 209 12546611 8 Cappuccio FP D’Elia L Strazzullo P Sleep duration and all-cause mortality: A systematic review and meta-analysis of prospective studies Sleep 2010 33 585 592 20469800 9 Gallicchio L Kalesan B Sleep duration and mortality: A systematic review and meta-analysis J Sleep Res 2009 18 148 158 19645960 10 Kripke DF Garfinkel L Wingard DL Mortality associated with sleep duration and insomnia Arch Gen Psychiatry 2002 59 131 136 11825133 11 Ohayon MM Carskadon MA Guilleminault C Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals: Developing normative sleep values across the human lifespan Sleep 2004 27 1255 1273 15586779 12 Ancoli-Israel S Sleep and aging: Prevalence of disturbed sleep and treatment considerations in older adults J Clin Psychiatry 2005 66 Suppl 9 24 30 quiz 42-23 13 Unruh ML Redline S An MW Subjective and objective sleep quality and aging in the sleep heart health study J Am Geriatr Soc 2008 56 1218 1227 18482295 14 Clark A Lange T Hallqvist J Sleep impairment and prognosis of acute myocardial infarction: A prospective cohort study Sleep 2014 37 851 858 24790263 15 Mallon L Broman JE Hetta J Sleep complaints predict coronary artery disease mortality in males: A 12-year follow-up study of a middle-aged Swedish population J Intern Med 2002 251 207 216 11886479 16 Mallon L Broman JE Hetta J High incidence of diabetes in men with sleep complaints or short sleep duration: A 12-year follow-up study of a middle-aged population Diabetes Care 2005 28 2762 2767 16249553 17 Park M Buchman AS Lim AS Sleep complaints and incident disability in a community-based cohort study of older persons Am J Geriatr Psychiatry 2014 22 718 726 23567404 18 Brim OG Ryff CD Kessler RC How Healthy Are We? A National Study of Well-Being at Midlife Chicago Chicago University Press 2004 The John D. and Catherine T. MacArthur Foundation Series on Mental Health and Human Development 19 Jette AM Cleary PD Functional disability assessment Physical Ther 1987 67 1854 1859 20 van den Berg JF Luijendijk HJ Tulen JH Sleep in depression and anxiety disorders: A population-based study of elderly persons J Clin Psychiatry 2009 70 1105 1113 19607762 21 Dunlop DD Manheim LM Song J Incidence of disability among preretirement adults: The impact of depression Am J Public Health 2005 95 2003 2008 16254232 22 Kessler RC Andrews A Mroczek DK The World Health Organization Composite International Diagnostic Interview Short Form (CIDI-SF) Int J Methods Psychiatr Res 1998 7 171 185 23 Buysse DJ Reynolds CF 3rd Monk TH Quantification of subjective sleep quality in healthy elderly men and women using the Pittsburgh Sleep Quality Index (PSQI) Sleep 1991 14 331 338 1947597 24 Cai H Shu XO Xiang YB Sleep duration and mortality: A prospective study of 113 138 middle-aged and elderly chinese men and women Sleep 2015 38 529 536 25348122 25 Marshall NS Wong KK Cullen SR Sleep apnea and 20-year follow-up for all-cause mortality, stroke, and cancer incidence and mortality in the Busselton Health Study cohort J Clin Sleep Med 2014 10 355 362 24733978 26 Morgenstern M Wang J Beatty N Obstructive sleep apnea: An unexpected cause of insulin resistance and diabetes Endocrinol Metab Clin North Am 2014 43 187 204 24582098 27 Parish JM Somers VK Obstructive sleep apnea and cardiovascular disease Mayo Clin Proc 2004 79 1036 1046 15301332 28 Spira AP Stone KL Rebok GW Sleep-disordered breathing and functional decline in older women J Am Geriatr Soc 2014 62 2040 2046 25376169 29 Stenholm S Kronholm E Bandinelli S Self-reported sleep duration and time in bed as predictors of physical function decline: Results from the InCHIANTI study Sleep 2011 34 1583 1593 22043129 30 Chien MY Chen HC Poor sleep quality is independently associated with physical disability in older adults J Clin Sleep Med 2015 11 225 232 25515275 31 Endeshaw YW Do sleep complaints predict persistent fatigue in older adults? 1532–5415 (Electronic) 32 Goldman SE Ancoli-Israel S Boudreau R Sleep problems and associated daytime fatigue in community-dwelling older individuals J Gerontol A Biol Sci Med Sci 2008 63 1069 1075 18948557 33 Friedman EM Hayney MS Love GD Social relationships, sleep quality, and interleukin-6 in aging women Proc Natl Acad Sci U S A 2005 102 18757 18762 16339311 34 Mills PJ von Kanel R Norman D Inflammation and sleep in healthy individuals Sleep 2007 30 729 735 17580594 35 Okun ML Coussons-Read M Hall M Disturbed sleep is associated with increased C-reactive protein in young women Brain Behav Immun 2009 23 351 354 19007876 36 Patel SR Zhu X Storfer-Isser A Sleep duration and biomarkers of inflammation Sleep 2009 32 200 204 19238807 37 von Kanel R Dimsdale JE Ancoli-Israel S Poor sleep is associated with higher plasma proinflammatory cytokine interleukin-6 and procoagulant marker fibrin D-dimer in older caregivers of people with Alzheimer’s disease J Am Geriatr Soc 2006 54 431 437 16551309 38 Frey DJ Fleshner M Wright KP Jr The effects of 40 hours of total sleep deprivation on inflammatory markers in healthy young adults Brain Behav Immun 2007 21 1050 1057 17524614 39 Irwin MR Wang M Campomayor CO Sleep deprivation and activation of morning levels of cellular and genomic markers of inflammation Arch Intern Med 2006 166 1756 1762 16983055 40 Meier-Ewert HK Ridker PM Rifai N Effect of sleep loss on C-reactive protein, an inflammatory marker of cardiovascular risk J Am Coll Cardiol 2004 43 678 683 14975482 41 van Leeuwen WM Lehto M Karisola P Sleep restriction increases the risk of developing cardiovascular diseases by augmenting proinflammatory responses through IL-17 and CRP PLoS One 2009 4 e4589 19240794 42 Ferrucci L Harris TB Guralnik JM Serum IL-6 level and the development of disability in older persons J Am Geriatr Soc 1999 47 639 646 10366160 43 Michaud M Balardy L Moulis G Proinflammatory cytokines, aging, and age-related diseases J Am Med Dir Assoc 2013 14 877 882 23792036 44 Peake J Della Gatta P Cameron-Smith D Aging and its effects on inflammation in skeletal muscle at rest and following exercise-induced muscle injury Am J Physiol Regul Integr Comp Physiol 2010 298 R1485 1495 20393160 45 Gangwisch JE Malaspina D Boden-Albala B Inadequate sleep as a risk factor for obesity: Analyses of the NHANES I Sleep 2005 28 1289 1296 16295214 46 Hasler G Buysse DJ Klaghofer R The association between short sleep duration and obesity in young adults: A 13-year prospective study Sleep 2004 27 661 666 15283000 47 Lauderdale DS Knutson KL Yan LL Self-reported and measured sleep duration: How similar are they? Epidemiology (Cambridge, Mass) 2008 19 838 845 48 van den Berg JF Miedema HM Tulen JH Sex differences in subjective and actigraphic sleep measures: A population-based study of elderly persons Sleep 2009 32 1367 1375 19848365 49 McCrae CS Rowe MA Tierney CG Sleep complaints, subjective and objective sleep patterns, health, psychological adjustment, and daytime functioning in community-dwelling older adults J Gerontol B Psychol Sci Soc Sci 2005 60 P182 P189 15980285 50 Irish LA Kline CE Gunn HE The role of sleep hygiene in promoting public health: A review of empirical evidence Sleep Med Rev 2015 22 23 36 25454674

PMC005xxxxxx/PMC5118090.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8912860 1403 Am J Reprod Immunol Am. J. Reprod. Immunol. American journal of reproductive immunology (New York, N.Y. : 1989) 1046-7408 1600-0897 27753461 5118090 10.1111/aji.12589 NIHMS818217 Article Up-regulation of miR-203 expression induces endothelial inflammatory response: potential role in preeclampsia Wang Yuping Dong Qin Gu Yang Groome Lynn J. Department of Obstetrics and Gynecology, Louisiana State University Health Sciences Center – Shreveport, Shreveport, LA 71130 Address correspondence to: Yuping Wang, MD, PhD., Department of Obstetrics and Gynecology, Louisiana State University Health Sciences Center, Shreveport, LA 71130, (318)-675-5379 (work) and (318)-675-4671 (fax), ywang1@lsuhsc.edu 7 10 2016 18 10 2016 12 2016 01 12 2017 76 6 482490 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Problem To determine if miR-203 mediates endothelial inflammatory response in preeclampsia. Method of study Maternal vessel miR-203 expression was assessed by in situ hybridization. Suppressor of cytokine signaling-3 (SOCS-3) and ICAM expression was determined by immunostaining. Subcutaneous fat tissue sections from normal and preeclamptic pregnant women were used. MiR-203 induced inflammatory response was evaluated by measurements of IL-6, IL-8, ICAM, and VCAM expression and production, and neutrophil adhesion in endothelial cells (EC) transfected with miR-203 precursor, pre-miR-203. SOCS3 expression was also determined. Results Up-regulation of miR-203 and ICAM expression and down-regulation of SOCS-3 expression was demonstrated in maternal vessel endothelium in preeclampsia. Over-expression of miR-203 resulted in down-regulation of SOCS-3 expression and increases in production of IL-6, IL-8, ICAM and VCAM, and neutrophil adhesion in ECs. Conclusion As miR-203 is an inflammatory microRNA, increased miR-203 production/ expression in ECs could diminish anti-inflammatory activity and increase endothelial inflammatory response in preeclampsia. miR-203 SOCS-3 endothelial cells adhesion molecule inflammatory response preeclampsia Introduction Preeclampsia is a hypertensive and multi-system disorder in human pregnancy. Although the cause of preeclampsia is not clear, increased systemic endothelial inflammatory responsiveness is believed to be a major pathophysiological event of maternal vascular dysfunction in preeclampsia. Elevated maternal levels of inflammatory cytokines such as IL-6, IL-8, IL-16, and TNFα [1, 2] and endothelial adhesion molecules ICAM and VCAM [3] are all considered hallmarks of increased vascular inflammatory response in preeclampsia [4]. Moreover, increased inflammatory response also contributes to increased endothelial solute permeability [5, 6] and oxidative stress [7] in this pregnancy disorder. The impact of preeclampsia on women to their long-term health outcome is further emphasized by the findings of increased risk of cardiovascular diseases and intensiveness of inflammatory responses later in life in women who had preeclampsia during their pregnancy [8, 9]. However, the underlying cellular and molecular mechanism of increased inflammatory response in preeclampsia remains unclear. Cytokine signaling is negatively regulated by a family of proteins named suppressor of cytokine signaling (SOCS). The SOCS family has 8 members, including SOCS-1 to SOCS-7 and cytokine-induced STAT inhibitor (CIS). These molecules modulate cytokine signaling through several mechanisms, which include inactivation of Janus kinases (JAKs), blocking access of STATs to receptor binding sites, and ubiquitination of signaling proteins and their subsequent targeting to the proteasome [10, 11]. Among the SOCS family members, SOCS-3 has been well characterized to regulate IL-6, leptin, and erythropoietin through binding to their receptors [12, 13]. We previously reported that SOCS-3 expression was down-regulated in both maternal leukocytes and vascular endothelium in women with preeclampsia [14]. SOCS-3 expression was also reduced in placental trophoblasts in preeclampsia [15]. Although down-regulation of SOCS-3 activity/expression may lead to alteration of cytokine signaling and increased inflammatory responses, little is known about the mechanism of SOCS-3 regulation in vascular endothelial cells. Emerging evidence has shown that numerous microRNAs (miRNAs) are able to regulate immune and inflammatory responses and dysregulation of miRNA expression closely links to increased inflammatory responses [16]. For example, miR-146 regulates IL-2 expression and activation [17]. MiR-155 promotes Th17 relevant cytokines and is induced during macrophage inflammatory response [18]. Increased miR-203 expression is found to be associated with down-regulation of SOCS-3 expression in infected gingival epithelial cells [19]. In contrast, silencing miR-203 reverses the inhibition of SOCS-3 expression [19]. Moreover, over-expression of miR-203 significantly increases IL-6 and MMP-1 production in synovial fibroblasts [20]. However, the role of miR-203 associated with endothelial inflammatory response has never been studied in preeclampsia. To test the hypothesis that increased miR-203 expression may lead to down-regulation of SOCS-3 expression and result in increased endothelial inflammatory responses in preeclampsia, we determined miR-203 expression in maternal vessels from normal and preeclamptic pregnancies and assessed anti-inflammatory effects of SOCS-3 on endothelial cells. The role of miR-203 mediated endothelial inflammatory response was further evaluated by IL-6, IL-8 and ICAM expression and production, and neutrophil-endothelial adhesion in endothelial cells. Materials and Methods Sample collection Maternal subcutaneous fat tissue was collected during cesarean section delivery from normal and preeclamptic pregnancies. Freshly obtained subcutaneous fat tissue was fixed immediately with 10% formalin and then embedded with paraffin. Tissue sections were used for detection of miR-203 expression, and SOCS-3 and ICAM expression. Collection of subcutaneous fat tissue was approved by IRB at Louisiana State University Health Sciences Center – Shreveport (LSUHSC-S), LA. Normal pregnancy was defined as pregnancy with blood pressure (<140/90mmHg), absence of proteinuria and obstetrical and medical complications. Diagnosis of preeclampsia was defined as follows: sustained systolic blood pressure of ≥ 140 mmHg or a sustained diastolic blood pressure of ≥ 90mmHg on two separate readings; proteinuria measurement of 1+ or more on dipstick, or 24 hour urine protein collection with ≥ 300mg in the specimen. Smokers were excluded. None of the study subjects had signs of infection. To avoid clinical phenotypic differences in preeclamptic patients, patients complicated with HELLP syndrome (hemolysis, elevated liver enzyme and low platelet count), diabetes and/or renal disease were excluded. Subcutaneous fat tissue from 5 normal and 5 preeclamptic women were used in this study and the clinical characteristics of study subjects are presented in Table 1. Detection of miR-203 expression by in situ hybridization (ISH) MiR-203 expression was determined by in situ hybridization on formalin-fixed, paraffin-embedded subcutaneous fat tissue sections using the miRCURY LNA™ microRNA ISH Optimization Kit (FFPE) (product number 90010) purchased from Exiqon (Vedbaek, Denmark). The Kit contains 5′-DIG and 3′-DIG labeled LNA™ scrambled miRNA control probe (5′-DIG/gtgtaacacgtctatacgccca /DIG-3′) that was used as negative control and 5′-DIG labeled LNA™ U6 snRNA control probe (5′-DIG/cacgaatttgcgtgtcatcctt/DIG-3′) that was used as positive control. The miRCURY LNA™ 5′-DIG and 3′-DIG labeled detection probe specific to hsa-miR-203 (probe number 88079-15, 5′-DIG/tagtggtcctaaacatttca/DIG-3′) was also purchased from Exiqon. The staining procedure was performed following the manufacturer’s instruction. A concentration of 100nM for hsa-miR-203 probe was used in the assay. Counter staining was carried out with Nuclear Fast Red (Vector Laboratories, Burlingame, CA). Stained slides were then reviewed under an Olympus microscope (Olympus IX71), and images were captured by PictureFrame computer software (Uptronics, Sunnyvale, CA) and recorded to a microscope linked PC computer. Immunohistochemistry Maternal vessel endothelium expression for SOCS-3 and ICAM were examined by immunohistochemistry (IHC) of subcutaneous tissue sections as previously described [14]. Primary monoclonal antibody against SOCS-3 (ab16030) was purchased from Abcam Inc. (Cambridge, MA) and antibody for ICAM (SC-7891) was from Santa Cruz (San Diego, CA). The antibody concentration of 1.0μg/ml was used for both SOCS3 and ICAM staining. Stained slides were counterstained with Gill’s formulation hematoxylin. Tissue sections stained with secondary antibody only were used as controls. All slides stained with the same antibody were processed at the same time. Stained tissue slides were reviewed under an Olympus microscope, and images were captured. Endothelial isolation and culture Human umbilical cord vein endothelial cells (HUVECs) from normal term placentas were isolated as previously described [21, 22] and used in the study. Cells were cultured with endothelial growth medium (Lonza, Allendale, NJ) supplemented with hydrocortisone, ascorbic acid, bovine brain extract, epidermal growth factor (EGF), gentamicin sulfate amphotericin (GA-1000), and fetal bovine serum (FBS). HUVECs at passage 2 were used for pre-mir-203 transfection or SOCS-3 gene transfer experiments. Pre-miR-203 transfection miR-203 over-expression was carried out by transfection of pre-miR-203 (miR-203 precursor, AM17100) (Ambion) in endothelial cells using LipofectAMINE RNAiMAX transfection reagent (Invitrogen). Pre-miR-203 at a concentration of 50nM was used in the transfection assay. Cells transfected with pre-miR negative control (a random sequence miRNA) served as control. Total RNA was extracted by TRIzol reagent approximately 40h after transfection. miR-203 expression and mRNA expression for IL-6, IL-8, and ICAM were then determined by RT-PCR. Expression of U6 snRNA was determined and served as an endogenous control for miRNA-203 expression and GADPH was determined and served as an endogenous control for IL-6, IL-8, and ICAM mRNA expression. The primer sequences and accession number for miR-203, U6, IL-6, IL-8, ICAM, and GADPH are presented in Table 2. Primers for miR-203 (HP300235) and U6 (HP300001) were purchased from OriGene Technologies, Inc. (Rockville, MD). Primers for IL-6, IL-8, ICAM, and GADPH were synthesized by Integrated DNA Technologies (IDT, www.idtdna.com). For SOCS-3 expression, cells were lysed 24, 48, and 72hrs after pre-miR-203 transfection using protein lysis buffer containing 50mmol/L Tris, 0.5% NP40, 0.5% Triton X-100 with protease and phosphatase inhibitors. Total cellular protein was then collected and SOCS-3 protein expression was then determined. Protein expression Protein expression for SOCS-3 was determined by Western blot. Briefly, 10μg of total cellular protein was subject to electrophoresis (Bio-Red, Hercules, CA) and then transferred to nitrocellulose membranes. After blocking, the membrane was probed with SOCS-3 antibody (same antibody used for IHC) and then followed by a matched secondary antibody. An enhanced chemiluminescent (ECL) detection kit (Amersham Corporation, Arlington Heights, IL) and X-ray film were used to visualize the bound antibody. The membrane was then stripped and re-probed with β-actin antibody (Sigma). After scanned, the density of bands was analyzed by NIH Image J analysis program. β-actin expression was used to normalize SOCS-3 expression. Data were presented as mean ± SE from 4 independent experiments. Endothelial IL-6, IL-8, and ICAM production Endothelial production of IL-6, IL-8, and ICAM were measured in cell culture medium by enzyme-linked immunosorbent assay (ELISA). ELISA kits of IL-6, IL-8, and ICAM were purchased from R&D Systems, Inc. (Minneapolis, MN). All samples were measured in duplicate in each assay. All assays were carried out according to the manufacturer’s instructions. Within assay variations were <7% for all the assays. Myeloperoxidase (MPO) assay for neutrophil-endothelial adhesion MPO activity was assessed as an index of neutrophil adhesion. Neutrophils were isolated from healthy non-pregnant volunteers as previously described [23]. Neutrophil-endothelial adhesion assay was performed approximately 40h after pre-mir-203 transfection or 48h after SOCS-3 gene transfer. Freshly isolated neutrophils were applied to endothelial cells 30min after treatment with TNFα. The plates were then washed twice with 1.0ml PBS, and reaction was initiated by adding following reagents to each well in order: 750μl 80mM K2HPO4 (pH=4.5), 50 μl fresh 10%HETAB, 100 μl 16mMTMB. MPO activity was then measured by a VERSAmax microplate reader (Molecular Devices, Walpole, MA) at a wavelength of 450nm. Construction of SOCS-3/ZsGreen1 GFP vector Plasmid pZsGreen1-N1 (Clontech, Mountain View, CA) was used to construct SOCS-3 vector, pSOCS-3. Briefly, open reading frame of human SOCS-3 was amplified from human cDNA by polymerase chain reaction (PCR) using oligonucleotide to create restriction sites for Nhe I and Kpn I at 5′ and 3′ end of SOCS-3 sequence using following primers: sense primer 5′-AGCGCTAGCACCATGGTCACCCACAGCAAGTTTCC-3′ and antisense primer 5′-GGTGGTACCCAAAGCGGGGCATCGTACTG-3′. Restriction sites for Nhe I (5′GCTAGC3′) and Kpn I (5′GGTACC3′) are underlined, respectively. PCR was performed using pfx Taq polymerase (Invitrogen). PCR product and the vector pZsGreen1-N1 were digested with Nhe I (New England BioLabs, Inc. Ipswich, MA) and Kpn I (New England BioLabs, Inc.). After ligation, competent Ecoli-Top10 (Invitrogen, Carlsbad, CA) was transformed with plasmid and selected positive clones were amplified. SOCS-3 sequence was verified by Mclab (South San Francisco, CA). SOCS-3 gene transfer Electroporation was performed for SOCS-3 gene transfer. HUVECs at passage 2 were used for all experiments. For SOCS-3 (pSOCS-3/ZsGreen1) gene transfer, an aliquot of 20μg of plasmid DNA mixed with 5×106 cells in 600μl buffer per 4mm electroporation cuvette. A fixed electroporation was given with a capacitance of 950uf and 250v using a Bio-Rad Gene Pulser instrument (Bio-Rad, Hercules, CA). Cells transfected with ZsGreen1 vector only were used as control. Cell surface adhesion molecule ICAM and VCAM expression (see below), and neutrophil-endothelial adhesion assay were performed 24 hours after SOCS-3 gene transfer. Cells were also lysed with lysis buffer and total cellular protein was also collected for SOCS-3 protein expression determined by Western blot. Endothelial surface molecule ICAM and VCAM expression Endothelial surface molecule ICAM and VCAM expression was determined in cells transfected with SOCS-3 gene as previously described [21]. Briefly, cells were grown in 24 well/plate and fixed with 1% paraformaldehyde and then incubated with a primary antibody (mouse anti-human) to ICAM-1 (CD54) or VCAM-1 (CD106). Horseradish peroxidase-goat anti-mouse immunoglobulin G (Sigma) was used as the secondary antibody. Hydrogen peroxide (0.003%) and 3,30,5,50-tetramethylbenzidine (TMB) (0.1mg/ml) were used as substrate and color generation. The reaction was terminated by addition of 100μl of 8N H2SO4 to each well. Cells that reacted with secondary antibody only were used as background. After reaction, plates were read at 450 nm by VERSAmax microplate reader (Molecular Devices, Sunnyvale, CA, USA). All samples were tested in triplicate. Statistical Analysis Statistical analysis was performed with ANOVA or paired t-test by computer software Prism 5 (GraphPad Software, Inc. La Jolla, CA). Student-Newman-Keuls test was used as post hoc tests. Data is expressed as mean ± SE. A probability level of less than 0.05 was considered statistically significant. Results Endothelial miR-203 expression is increased in preeclampsia To determine if endothelial cells express miR-203 and whether altered miR-203 expression is present in preeclampsia, miR-203 expression was examined by in situ hybridization in subcutaneous fat tissue sections. Our results showed that miR-203 expression was weakly expressed in vessel endothelium in tissues from normotensive pregnant women, but strongly expressed in vessel endothelium in 4 out of 5 tissues from women with preeclampsia. Figure 1A shows representative miR-203 expression in maternal vessel endothelium from 2 normal and 2 preeclamptic subjects. An imaging panel showing miR-203 expression in maternal vessel endothelium from each study subject (5 normal and 5 preeclampsia) is presented in supplement Figure 1. These pictorial data clearly show miR-203 expression is up-regulated in maternal vessel endothelium in preeclampisa. Since SOCS-3 is a target of miR-203 and ICAM is an indicator of increased inflammatory response in endothelial cells, endothelial expression of SOCS-3 and ICAM was examined. As shown in Figure 1B, endothelial SOCS-3 expression was markedly reduced, which was consistent with our previous findings of reduced SOCS-3 expression in women with preeclampsia [14]. In contrast, endothelial ICAM expression was markedly increased in maternal vessels from women with preeclampsia compared to those from normal pregnant controls. These results suggest that up-regulation of miR-203 expression was associated with reduced SOCS-3 expression and increased ICAM expression in maternal vessel endothelium in women with preeclampsia. Over-expression of miR-203 results in down-regulation of SOCS-3 expression in endothelial cells SOCS-3 is an important cellular anti-inflammatory regulator. miR-203 targets SOCS-3 [24]. To determine if increased miR-203 production/expression could down-regulate SOCS-3 expression in endothelial cells, endothelial cells were transfected with miR-203 precursor, pre-mir-203, and cells were collected 24, 48, and 72hrs after pre-mir-203 transfection. Total protein and RNA were extracted. As shown in Figure 2, mature miR-203 expression was markedly increased in cells transfected with pre-mir-203 compared to that in control cells (Figure 2A). In contrast, SOCS-3 expression was significantly reduced in cells transfected with pre-mir-203 and this inhibitory effect was in a time-dependent manner. The bar graph shows relative SOCS-3 protein expression in endothelial cells at 24, 48, and 72hrs after pre-mir-203 transfection, p<0.05, Figure 2B. Data are means ± SE from 4 independent experiments. Over-expression of miR-203 promotes endothelial inflammatory response As shown in Figure 1, increased miR-203 expression and increased ICAM expression were seen in maternal vessel endothelium. To further determine the consequences of increased miR-203 expression in endothelial cells, mRNA expression for IL-6, IL-8, and ICAM was examined in cells with or without transfection of pre-mir-203. Similar to miR-203 expression, mRNA expression for IL-6, IL-8 and ICAM was also significantly increased in cells transfected with pre-mir-203, Figure 3A. We further determined endothelial production of IL-6, IL-8, and ICAM by measuring IL-6, IL-8, and ICAM concentrations in cell culture medium. In this experiment, fresh medium was replaced 40 hours after pre-mir-203 transfection, and then medium was collected at 2, 4, and 8hrs and measured for IL-6, IL-8, and ICAM production by ELISA. Consistent with up-regulation of IL-6, IL-8 and ICAM mRNA expression, endothelial production of IL-6, IL-8 and ICAM were all significantly increased in cells transfected with pre-mir-203 compared to control cells and the increased IL-6, IL-8 and ICAM production was in a time-dependent manner, Figure 3B. These results clearly show that increase in miR-203 expression/production promotes endothelial expression and production of inflammatory cytokines and adhesion molecules. Figure 3C shows neutrophil adhesion assessed by measuring of neutrophil myeloperoxidase (MPO) activity in endothelial cells transfected with pre-mir-203. Cells transfected with scramble pre-mir served as control. 40hrs after pre-mir-203 transfection, cells were treated with or without TNFα at a concentration of 50ng/ml for 2hrs and then freshly isolated neutrophils (1×105 cells/well in 24 well plate) were added to the cell culture. MPO activity was then determined. Our results showed that neutrophil adhesion was significantly increased in cells transfected with pre-mir-203 vs. control cells, p<0.05, Figure 3C. The increased neutrophil adhesion to endothelial cells was further increased in cells treated with TNFα. These results provide further evidence that up-regulation of miR-203 expression promotes endothelial inflammatory response. SOCS-3 decreases endothelial inflammatory response As shown in Figure 2 and 3, down-regulation of SOCS-3 expression was relevant to increased IL-6, IL-8, and ICAM expression and production in endothelial cells transfected with pre-miR-203. SOCS-3 is a target of miR-203. Reduced SOCS-3 expression has been demonstrated in maternal vessel endothelium in preeclampsia [14]. To further investigate if SOCS-3 modulates inflammatory response in endothelial cells, we constructed a SOCS-3/ZsGreen1 vector and transferred it into endothelial cells to determine if increase in SOCS-3 expression could increase anti-inflammatory activity in endothelial cells. Cells transfected with empty GFP vector were used as control. Cell adhesion molecule ICAM and VCAM expression and neutrophil-endothelial adhesion assay were performed 24hrs after SOCS3 gene transfer. Our results showed that SOCS-3 expression was increased in endothelial cells transfected with SOCS-3 gene (Figure 4A). In contrast, ICAM and VCAM expression were significantly reduced in cells transfected with SOCS-3 compared to control cells, p<0.01 (Figure 4B). Moreover, increased neutrophil-endothelial adhesion induced by TNFα were also significantly suppressed in cells transfected with SOCS-3 gene compared to those of controls, p<0.05 and p<0.01, respectively, Figure 4C. Discussion In the present study, we, for the first time, showed increased miR-203 expression in maternal vessel endothelium in women with preeclampsia compared to that in normal pregnant controls. We further found that over-expression of miR-203 expression could down-regulate cytokine suppressor SOCS-3 expression in endothelial cells, which was accompanied by increased endothelial inflammatory response. This was demonstrated by increased endothelial expression/production of IL-6, IL-8 and ICAM and increased neutrophil-endothelial adhesion in cells with over-expression of miR-203. The observation of up-regulation of ICAM expression in maternal vessel endothelium from preeclampsia is consistent as previously reported by Leik et al [25]. Thus, our findings are significant and provide new evidence that altered miR-203 expression could contribute to increased endothelial inflammatory response in preeclampsia. Although the number of maternal vessel samples in each group was small with different gestational age at delivery and primigravida between normal and preeclamptic pregnancies, and range of BMI within each of the groups, we believe that our ISH finding of increased miR-203 expression in maternal vessel endothelium from preeclampsia is reliable because the consistency of miRNA expression tested by ISH and real-time PCR was demonstrated in our previous published placental miRNA works [26]. MiRNAs have been recognized as a class of novel inflammatory regulators by modification of target genes at different levels through anti-inflammatory or pro-inflammatory signaling cascades in the cardiovascular system. Studies have shown that many miRNAs are involved in endothelial inflammation by directly or indirectly targeting the genes that regulate leukocyte recruitment. For example, miR-126 could inhibit VCAM-1 expression and limit leukocyte adherence to endothelial cells [27], while TNF-induced endothelial miR-31 and miR-17-3p expression could in turn suppress endothelial E-selectin and ICAM-1 expression induced by TNF stimulation [28]. An animal study has also shown that miR-181b inhibition exacerbated endotoxin-induced NF-κB activity, leukocyte influx, and lung injury [29]. Moreover, up-regulation of miR-146a, miR-9, miR-204 and miR-367 expression in senescent endothelial cells was also reported, and their predicted target genes include Toll-like receptor signaling (TLR) pathway, which is well known to play a pivotal role in inflammatory response, a key feature of senescence (inflammaging) [30]. It seems very likely that alteration of anti-inflammatory miRNA expression/production could directly disturb protective machinery of anti-inflammatory adaptive inflammatory responses. MiR-203 was considered a skin- and keratinocyte-specific miRNA that was originally reported to be associated with common chronic inflammatory disorders in skin [31]. Up-regulation miR-203 expression was reported in psoriasis-affected skin plaques [32] and in diabetic foot ulcer skin tissues [33]. In the present study, we found increased miR-203 expression was in line with reduced SOCS-3 expression and increased ICAM expression in maternal vessel endothelium in women with preeclampsia compared to normal pregnant controls. Altered miR-203 expression was also reported in maternal plasma and placental tissue from preeclampsia [34]. It would be ideal to corroborate the findings of increased miR-203 in maternal vessel endothelium from preeclamptic pregnant women using a second approach such as real-time PCR. However, it is unlikely to obtain or isolate endothelial cells from systemic vasculature from pregnant women and to quantify the difference in miR-203 expression between normal and preeclamptic pregnancies, but our ISH results (supplement Figure 1) clearly showed that miR-203 expression is markedly increased in maternal vessel endothelium from preeclamptic pregnancies compared to that from normal pregnant women. Because SOCS-3 is a target of miR-203 [24] and SOCS-3 is an anti-inflammatory mediator, we then tested the consequences of miR-203 mediated inflammatory response in endothelial cells and determined effects of increased miR-203 expression on SOCS-3 and ICAM expression, and endothelial inflammatory response in cells transfected with miR-203 precursor, pre-miR-203. As we expected, miR-203 expression was dramatically increased in cells transfected with pre-mir-203. Moreover, cells transfected with pre-miR-203 resulted in a time-dependent down-regulation of SOCS-3 expression, which is inversely related to increased mRNA expression and production of inflammatory cytokine IL-6 and IL-8 and endothelial adhesion molecule ICAM in endothelial cells. These findings clearly showed consequences of miR-203 up-regulation mediated increased inflammatory responsiveness in endothelial cells. MiR-203 mediated increased endothelial inflammatory response was further assessed by neutrophil-endothelial adhesion assay in endothelial cells transfected with miR-203 precursor pre-miR-203 in the presence or absence of TNFα. We found significant increases in neutrophil adhesion to endothelial cells, in which cells were transfected with pre-miR-203 compared to cells transfected with control pre-miRNA. We further found that pre-miR-203 could promote increased neutrophil adhesion to endothelial cells induced by TNFα. These data further demonstrate that increase in endothelial miR-203 expression could stimulate endothelial inflammatory response. We also noticed that pre-miR-203 transfection could induce TNFα expression in endothelial cells (data not shown). Although we did not validate the targeting effects of miR-203 on TNFα in endothelial cells, a study conducted by Primo et al, by screening a panel of cytokines that are up-regulated in psoriatic skin induced by miR-203, demonstrated that miR-203 could regulate inflammatory cytokine production of TNFα, IL-24, IL-15, and IL-17A, etc [35]. Their study suggests that miR-203 could serve to fine-tune cytokine signaling and may dampen skin immune responses by altering key pro-inflammatory cytokines [35]. SOCS-3 is an important anti-inflammatory mediator. Down-regulation of SOCS-3 expression by miR-203 is associated with increased endothelial inflammatory response. To further study endothelial anti-inflammatory activity mediated by SOCS-3, SOCS-3 gene transfer was used as a testing model, and endothelial ICAM and VCAM expression and neutrophil-endothelial adhesion were then determined. Our results clearly showed that endothelial ICAM and VCAM expression was significantly reduced in cells transfected with SOCS-3 gene. Moreover, over-expression of SOCS-3 could also suppress TNFα-induced increased neutrophil adhesion to endothelial cells. This data implies that down-regulation of SOCS-3 and increased inflammatory response in maternal vasculature could be consequences of increased miR-203 expression in preeclampsia. Taken together, in this study we found that increased miR-203 expression is associated with reduced anti-inflammatory mediator SOCS-3 expression and increased endothelial adhesion molecule expression in maternal vessel endothelium in preeclampsia. We also demonstrated that over-expression of miR-203 resulted in increased inflammatory cytokines IL-6 and IL-8 and endothelial adhesion molecule ICAM expression/production and promoted neutrophil-endothelial adhesion. These findings provide considerable evidence that increased miR-203 expression could contribute to increased endothelial inflammatory response in preeclampsia. Supplementary Material Supp Fig S1 This study was supported in part by grants from National Institute of Health, NHLBI R01 HL65997 and NICHD R21HD076289 to Yuping Wang. Figure 1 Expression of miR-203, SOCS-3, and ICAM in maternal vessel endothelium in normal and preeclamptic pregnancies Subcutaneous fat tissue sections were used. A: Representative images for miR-203 expression in maternal vessels from 2 normal and 2 preeclamptic pregnancies. miR-203 expression was not expressed in maternal vessels from normal pregnancies. However, strong miR-203 expression was observed in vessel endothelium in preeclampsia. a and c: normal; b and d: preeclampsia; e: negative control; and f: positive control U6 staining. a–d: bar = 50micron and e–f: bar = 100 micron. B: Representative images for SOCS-3 and ICAM expression in maternal vessels from normal and preeclamptic pregnancies. SOCS-3 expression was markedly reduced and ICAM expression was markedly increased in maternal vessel endothelium from preeclampsia compared to normal pregnant controls. SOCS-3 and ICAM expression was undetectable in tissue sections stained with secondary antibody only (not shown). a and b: SOCS-3; c and d: ICAM. a and c: normal pregnancies; b and d: preeclampsia. bar = 50micron. Figure 2 Transfection of miR-203 precursor results in up-regulation of miR-203 expression and down-regulation of SOCS-3 expression in endothelial cells A: miR-203 expression was dramatically increased in endothelial cells transfected with pre-miR-203. U6 expression was determined as internal control for each sample. Lane 1–3: control cells were transfected with pre-mir control and lane 4–6: cells were transfected with pre-miR-203. B: The upper sequences show the miRNA response elements (MREs) of miR-203 to the 3′ UTR of SOCS-3 mRNA that was predicted by TargetScan and miRBase Target. Red lined indicates the “seed” regions. The blot shows that SOCS-3 protein expression in endothelial cells transfected with pre-miR-203 at 24, 48, and 72hrs compared to untransfected control cells (C). Down-regulation of SOCS-3 expression was time-dependent in endothelial cells transfected with pre-miR-203. The bar graph was expressed as mean ± SE from 4 independent transfection experiments. * p<0.05: pre-miR-203 transfected cells vs. C. Statistical analysis was done by ANOVA and Newman-Keuls test was used as post hoc tests. Figure 3 Increased IL-6, IL-8, and ICAM expression and production, and increased neutrophil-endothelial adhesion in endothelial cells transfected with pre-miR-203 A: mRNA expression for IL-6, IL-8, and ICAM was increased in endothelial cells transfected with pre-miR-203 compared to control cells. Lane 1–3: control cells transfected with pre-miR control and lane 4–6: cells transfected with pre-miR-203. B: Consistent with mRNA expression, IL-6, IL-8, and ICAM production was also significantly increased in cells transfected with pre-miR-203 compared to cells transfected with pre-miR controls. In this experiment, medium was changed 40hrs after transfection and newly added medium was then collected at 2, 4, and 8hrs and medium concentrations for IL-6, IL-8, and ICAM were measured by ELISA. IL-6, IL-8 and ICAM production was time-dependent increased in cells with or without pre-miR-203 transfection. However, cells transfected with pre-miR-203 produced significantly more IL-6, IL-8 and ICAM than cells transfected with control pre-miR. Data are presented as means ± SE from 6 independent experiments, * p<0.05 and ** p<0.01: pre-miR-203 transfected cells vs. control cells at each time point (tested by paired t-test). ˆ p<0.05 and ˆˆ p<0.01: control cells at 8hrs vs. 2hrs, and ## p<0.01: pre-miR-203 transfected cells at 4hrs or 8hrs vs. 2hrs (tested by ANOVA and Newman-Keuls test was used as post hoc tests), respectively. C: MPO activity in endothelial cells transfected with pre-miR-203 with or without TNFα stimulation. Cells transfected with pre-miR control served as control. Neutrophil-endothelial adhesion was significantly increased in cells transfected with pre-miR-203 and further increased when TNFα was present in the culture. Data are expressed as mean ± SE from 3 independent experiments each in triplicate. * p<0.05 and ** p<0.01: pre-miR-203 transfected cells vs. controls; # p<0.05: pre-miR-203 transfected cells treated with TNFα vs. without TNFα treatment (by paired t-test), respectively. Figure 4 Reduction of ICAM and VCAM expression and neutrophil-endothelial adhesion in endothelial cells transfected with SOCS-3 gene A: SOCS-3 protein expression is increased in ECs transfected with SOCS-3 gene compared to control cells. Total cellular protein was collected 48hrs after transfection. B: Adhesion molecule ICAM and VCAM expression in ECs with or without transfection of SOCS-3 gene. Cells transfected with SOCS-3 expressed less ICAM and VCAM compared to the control cells, Data are presented as means ± SE from 4 independent experiments, ** p<0.01: SOCS-3 transfected cells vs. control cells. C: Neutrophil-endothelial adhesion assessed by MPO activity in ECs with or without transfection of SOCS-3 gene. Data are presented as means ± SE from 4 independent experiments. In control cells, MPO activity was dose-dependently increased in cells treated with TNFα, ˆ p < 0.05 and ˆˆ p < 0.01: cells treated with TNFα vs. untreated cells. Data was analyzed by ANOVA with Student-Newman-Keuls test as post hoc test. MPO activity was significantly reduced when cells were transfected with SOCS-3 gene compared to the control cells treated with same dose of TNFα, * p<0.05 and ** p<0.01, respectively. Paired t-test was used for data analysis. MPO activity was not statistically significant in cells transfected with SOCS-3 with or without treatment of TNFα (analyzed by ANOVA). Table 1 Clinical information of study subjects from which maternal vessels were examined in the study Variables Normal (n=5) Preeclampsia (n=5) p value Maternal age (years) 29 ± 7 (20–37) 26 ± 5 (20–33)   0.4633 Racial Status  White 1 0 ND  Black 3 5 ND  Other 1 0 ND BMI 35 ± 11 (23–51) 34 ± 8 (26–44)   0.8683 Blood Pressure (mmHg)  Systolic   113 ± 11 (101–130) 164 ± 15 (148–182) 0.0003  Diastolic 70 ± 7 (60–78) 102 ± 7 (93–113)     <0.0001   Primigravida 0 4 ND Proteinuria N/A 1 – 4+ ND Uric acid N/A 6.7±1.9 (4.1–9.4) ND Gestational Age  at delivery (weeks)     37+3 ± 2+2 (34+3–39+5) 31+6 ± 3+4 (28–36+2) 0.0317 Data are expressed as mean ± SD (range). ND: not determined. Table 2 Primer sequences used in the study Gene Name Primer Sequences Accession # miR-203a-3p Forward: 5′-GTGAAATGTTTAGGACCAC-3′ MIMAT0000264 Reverse: 5′-GAACATGTCTGCGTATCTC-3′ U6 Forward: 5′-CTCGCTTCGGCAGCACAT-3′ NR_004394.1 Reverse: 5′-TTTGCGTGTCATCCTTGCG-3′ IL-6 Forward: 5′-AAAGAGGCACTGGCAGAAAA-3′ NM_000600 Reverse: 5′-AGCTCTGGCTTGTTCCTCCTCAC-3′ IL-8 Forward: 5′-TCTGCAGCTCTGTGTGAAGG-3′ NM_000584 Reverse: 5′-ACTTCTCCACAACCCTCTGC-3′ ICAM-1 Forward: 5′-GCTTTCCGGCGCCCAACGTGATTCTGA-3′ NM_000201 Reverse: 5′-ACTCACACAGGACACGAA GCTCCCGGGTCT-3′ GADPH Forward: 5′-CAAAAGGGTCATCATCTCTGC-3′ NM_002046 Reverse: 5′-AGTTGTCATGGATGACCTTGG-3′ 1 Lyall F Greer IA Boswell F Macara LM Walker JJ Kingdom JCP The cell adhesion molecule, VCAM-1, is selectively elevated in serum in pre-eclampsia: does this indicate the mechanism of leucocyte activation? Br J Obstet Gynaecol 1994 101 485 487 7517182 2 Haller H Ziegler E-M Homuth V Drab M Eichhorn J Nagy Z Busjahn A Vetter K Luft FC Endothelial adhesion molecules and leukocyte integrins in preeclamptic patients Hypertension 1997 29 291 296 9039117 3 Taylor RN Crombleholme WR Friedman SA Jones LA Casal DC Roberts JM High plasma cellular fibronectin levels correlate with biochemical and clinical features of preeclampsia but cannot be attributed to hypertension alone Am J Obstet Gynecol 1991 165 895 901 1951550 4 Redman CWG Sacks GP Sargent IL Preeclampsia: An excessive maternal inflammatory response to pregnancy Am J Obstet Gynecol 1999 180 499 506 9988826 5 Wang Y Gu Y Granger DN Roberts JM Alexander JS Endothelial junctional protein redistribution and increased monolayer permeability in HUVECs isolated during preeclampsia Am J Obstet Gynecol 2002 186 214 220 11854638 6 Zhang Y Gu Y Li H Lucas MJ Wang Y Increased endothelial monolayer permeability is induced by serum from women with preeclampsia but not by serum from women with normal pregnancy or that are not pregnant Hypertens Pregn 2003 22 121 131 7 Davidge ST Oxidative stress and altered endothelial cell function in preeclampsia Sem Reprod Endocrinol 1998 16 65 73 8 Bellamy L Casas JP Hingorani AD Williams DJ Pre-eclampsia and risk of cardiovascular disease and cancer in later life: systematic review and meta-analysis BMJ 2007 335 7627 974 17975258 9 Freeman DJ McManus F Brown EA Cherry L Norrie J Ramsay JE Clark P Walker ID Sattar N Greer IA Short- and long-term changes in plasma inflammatory markers associated with preeclampsia Hypertension 2004 44 708 714 15452036 10 Krebs DL Hilton DJ SOCS proteins: negative regulators of cytokine signaling Stem Cells 2001 19 378 387 11553846 11 Alexander WS Starr R Metcalf D Nicholson SE Farley A Elefanty AG Brysha M Kile BT Richardson R Baca M Zhang JG Willson TA Viney EM Sprigg NS Rakar S Corbin J Mifsud S DiRago L Cary D Nicola NA Hilton DJ Suppressors of cytokine signaling (SOCS): negative regulators of signal transduction J Leukoc Biol 1999 66 588 592 10534114 12 Alexander WS Hilton DJ The role of suppressors of cytokine signaling (SOCS) proteins in regulation of the immune response Annu Rev Immunol 2004 22 503 529 15032587 13 Howard JK Flier JS Attenuation of leptin and insulin signaling by SOCS proteins Trends Endocrinol Metab 2006 17 365 371 17010638 14 Wang Y Lewis DF Gu Y Zhao S Groome LJ Elevated maternal soluble gp130 and IL-6 levels and reduced gp130 and SOCS-3 expressions in women with preeclampsia Hypertension 2011 57 336 342 21173340 15 Zhao S Gu Y Dong Q Fan R Wang Y Altered IL-6 receptor, IL-6R and gp130, production and expression and decreased SOCS-3 expression in placentas from women with preeclampsia Placenta 2008 29 1024 1028 18986700 16 O’Connell RM Rao DS Baltimore D microRNA regulation of inflamatory responses Annu Rev Immunolm 2012 30 295 312 17 Sonkoly E Ståhle M Pivarcsi A MicroRNAs and immunity: novel players in the regulation of normal immune function and inflammation Semin Cancer Biol 2008 18 131 140 18291670 18 O’Connell RM Kahn D Gibson WS Round JL Scholz RL Chaudhuri AA Kahn ME Rao DS Baltimore D MicroRNA-155 promotes autoimmune inflammation by enhancing inflammatory T cell development Immunity 2010 33 607 619 20888269 19 Moffatt CE Lamont RJ Porphyromonas gingivalis induction of microRNA-203 expression controls suppressor of cytokine signaling 3 in gingival epithelial cells Infect Immun 2011 79 2632 2637 21536793 20 Stanczyk J Ospelt C Karouzakis E Filer A Raza K Kolling C Gay R Buckley CD Tak PP Gay S Kyburz D Altered expression of microRNA-203 in rheumatoid arthritis synovial fibroblasts and its role in fibroblast activation Arthritis Rheum 2011 63 373 381 21279994 21 Wang Y Adair CD Coe L Weeks JW Lewis DF Alexander JS Activation of endothelial cells in preeclampsia: Increased neutrophil-endothelial adhesion correlates with up-regulation of adhesion molecule P-selectin in human umbilical vein endothelial cells isolated from preeclampsia J Soc Gynecol Investig 1998 5 237 243 22 Jaffe EA Nachman RL Becker CG Minick CR Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria J Clin Invest 1973 52 2745 2756 4355998 23 Wang Y Adair CD Weeks JW Lewis DF Alexander JS Increased neutrophil-endothelial adhesion induced by placental factors is mediated by platelet-activating factor in preeclampsia J Soc Gynecol Investig 1999 6 136 141 24 Wei T Orfanidis K Xu N Janson P Ståhle M Pivarcsi A Sonkoly E The expression of microRNA-203 during human skin morphogenesis Exp Dermatol 2010 19 854 856 20698882 25 Leik CE Walsh SW Neutrophils infiltrate resistance-sized vessels of subcutaneous fat in women with preeclampsia Hypertension 2004 44 72 77 15148293 26 Gu Y Sun J Groome LJ Wang Y Differential miRNA expression profiles between the first and third trimester human placentas Am J Physiol Endocrinol Metab 2013 304 E836 843 23443922 27 Harris TA Yamakuchi M Ferlito M Mendell JT Lowenstein CJ MicroRNA-126 regulates endothelial expression of vascular cell adhesion molecule 1 Proc Natl Acad Sci U S A 2008 105 1516 1521 18227515 28 Suárez Y Wang C Manes TD Pober JS Cutting edge: TNF-induced microRNAs regulate TNF-induced expression of E-selectin and intercellular adhesion molecule-1 on human endothelial cells: feedback control of inflammation J Immunol 2010 184 21 25 19949084 29 Sun X Icli B Wara AK Belkin N He S Kobzik L Hunninghake GM Vera MP MICU Registry Blackwell TS Baron RM Feinberg MW MicroRNA-181b regulates NF-κB-mediated vascular inflammation J Clin Invest 2012 122 1973 1990 22622040 30 Olivieri F Lazzarini R Recchioni R Marcheselli F Rippo MR Di Nuzzo S Albertini MC Graciotti L Babini L Mariotti S Spada G Abbatecola AM Antonicelli R Franceschi C Procopio AD MiR-146a as marker of senescence-associated pro-inflammatory status in cells involved in vascular remodelling Age (Dordr) 2013 35 1157 1172 22692818 31 Sonkoly E Ståhle M Pivarcsi A MicroRNAs: novel regulators in skin inflammation Clin Exp Dermatol 2008 33 312 315 18419608 32 Sonkoly E Wei T Janson PC Sääf A Lundeberg L Tengvall-Linder M Norstedt G Alenius H Homey B Scheynius A MicroRNAs: novel regulators involved in the pathogenesis of psoriasis? PLoS One 2007 2 e610 17622355 33 Liu J Xu Y Shu B Wang P Tang J Chen L Qi S Liu X Xie J Quantification of the differential expression levels of microRNA-203 in different degrees of diabetic foot Int J Clin Exp Pathol 2015 8 13416 13420 26722550 34 Yang S Li H Ge Q Guo L Chen F Deregulated microRNA species in the plasma and placenta of patients with preeclampsia Mol Med Rep 2015 12 527 534 25738738 35 Primo MN Bak RO Schibler B Mikkelsen JG Regulation of pro-inflammatory cytokines TNFα and IL24 by microRNA-203 in primary keratinocytes Cytokine 2012 60 741 748 22917968

PMC005xxxxxx/PMC5118104.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9304420 2335 Nucl Med Biol Nucl. Med. Biol. Nuclear medicine and biology 0969-8051 1872-9614 27744117 5118104 10.1016/j.nucmedbio.2016.07.008 NIHMS822981 Article Targeted therapy of osteosarcoma with radiolabeled monoclonal antibody to an insulin-like growth factor-2 receptor (IGF2R) Geller David S. 123 Morris Jonathan 1 Revskaya Ekaterina 4 Kahn Mani 1 Zhang Wendong 1 Piperdi Sajida 2 Park Amy 2 Koirala Pratistha 237 Guzik Hillary 5 Hall Charles 6 Hoang Bang 123 Yang Rui 123 Roth Michael 23 Gill Jonathan 23 Gorlick Richard 237 Dadachova Ekaterina 348 1 Department of Orthopaedic Surgery, Montefiore Medical Center and The Children’s Hospital at Montefiore, Bronx, NY, USA 2 Department of Pediatrics, Montefiore Medical Center and The Children’s Hospital at Montefiore, Bronx, NY, USA 3 Albert Einstein College of Medicine, Bronx, NY, USA 4 Department of Radiology, Nuclear Medicine, Montefiore Medical Center, Bronx, NY, USA 5 Analytical Imaging Facility, Albert Einstein College of Medicine, Bronx, NY, USA 6 Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA 7 Department of Molecular Pharmacology, Albert Einstein College of Medicine, Bronx, NY, USA 8 Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, NY, USA Requests for reprints: David S. Geller, M.D., Department of Orthopaedic Surgery, Montefiore Medical Center, 3400 Bainbridge Avenue, Bronx, NY 10467. Phone: 718-920-4429; dgeller@montefiore.org 23 10 2016 30 7 2016 12 2016 01 12 2017 43 12 812817 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Introduction Osteosarcoma overall survival has plateaued around 70%, without meaningful improvements in over 30 years. Outcomes for patients with overt metastatic disease at presentation or who relapse are dismal. In this study we investigated a novel osteosarcoma therapy utilizing radioimmunotherapy (RIT) targeted to IGF2R, which is widely expressed in OS. Methods Binding efficiency of the Rhenium-188(188Re)-labeled IGF2R-specific monoclonal antibody (mAb) to IGF2R on OS17 OS cells was assessed with Scatchard plot analysis. Biodistribution studies were performed in heterotopic murine osteosarcoma xenografts. Tumor growth was compared over a 24-day period post-treatment between mice randomized to receive 188Re-labeled IGF2R-specific murine mAb MEM-238 (188Re-MEM-238) or one of three controls—188Re-labeled isotype control mAb, unlabeled MEM-238,or no treatment. Results Results demonstrate the radioimmunoconjugate had a high binding constant to IGF2R. Both 188Re-MEM-238 and the isotype control had similar initial distribution in normal tissue. After 48 hours 188Re-MEM-238 exhibited a 1.8 fold selective uptake within tumor compared to the isotype control (p=0.057). Over 24 days, the tumor growth ratio was suppressed in animals treated with RIT compared to unlabeled and untreated controls (p=0.005) as demonstrated by a 38% reduction of IGF2R expressing osteosarcoma cells in the RIT group (p=0.002). Conclusions In conclusion, given the lack of new effective therapies in osteosarcoma, additional investigation into this target is warranted. Advances in Knowledge High expression of IGF2R on osteosarcoma tumors, paired with the specificity and in vivo anti-cancer activity of 188Re-labeled IGF2R-specific mAb suggests that IGF2R may represent a novel therapeutic target in the treatment of osteosarcoma. Implications for Patient Care This targeted approach offers the benefits of being independent of a specific pathway, a resistance mechanism, and/or an inherent biologic tumor trait and therefore is relevant to all OS tumors that express IGF2R. OS Radioimmunotherapy Insulin-like Growth Factor-2 Receptor (IGF2R) Rhenium-188 INTRODUCTION Osteosarcoma (OS) is the most common non-hematologic primary bone malignancy [2, 3]. Additionally it is the most common primary malignant bone tumor and the fifth most common primary malignancy among adolescents and young adults [4]. Unfortunately, overall 5 year survival has plateaued at approximately 70%, with no meaningful improvement realized in over 30 years [3-6]. Patients with metastatic disease have poor outcomes, with pulmonary and osseous metastases portending an overall survival of less than 40% and 20%, respectively. Recent studies have demonstrated an inability to improve outcomes using conventional chemotherapy strategies, underscoring the need for novel treatment approaches [7]. Overexpression of the cation-independent mannose-6-phosphate/insulin-like growth factor-2 receptor (IGF2R) has been demonstrated across a panel of OS cell lines [1]. Additionally, a single nucleotide polymorphism (SNP) within a haplotype block in IGF2R has been associated with an increased risk of developing OS [8], however, the role of IGF2R in the pathogenesis of OS is still being investigated. Although IGF2R is expressed normally across a wide array of tissue types, the consistent overexpression of IGF2R in OS suggests it may serve as a valuable therapeutic target. Radioimmunotherapy (RIT) is a method of delivering cytotoxic radiation therapy in a targeted fashion whereby an antigen-specific antibody is bound to either an α- or β-emitting radioisotope [9, 10]. The technique has been successfully employed in a number of challenging oncologic settings [11, 12]. RIT delivers cytocidal radiation to the targeted cell, is less affected by the multidrug resistance ensuing mechanisms than chemotherapy, and does not depend on the immune status of a patient or microenvironmental fluctuations the way immunotherapy with naked antibodies or T cells does. It permits for systemic administration, antibody-mediated specificity, and physical cytocidal damage in a manner which is well-tolerated by the patient. The purpose of this study was to investigate a novel therapeutic approach to OS, using IGF2R-targeted RIT. We have selected 188-Rhenium (188Re), a powerful β-emitter (Emax=2.01 meV) with relatively short physical half-life of 16.9 hrs for radiolabeling the IGF2R-specific antibody. 188Re has been successfully used in both pre-clinical and clinical RIT with insignificant side effects [12 and references therein]. We hypothesized that 188Re-labeled IGF2R-specific mAb will serve to effectively target OS tumor cells and that 188Re will be able to deliver high tumoricidal doses to the tumors without toxicity to healthy tissues. EXPERIMENTAL DESIGN Cell lines OS-17 is a well-characterized pediatric preclinical testing program patient-derived OS xenograft model, obtained from a primary femoral tumor and grown continuously in SCID mice [13, 14]. Tumor authentication was performed on the cell morphologically and via differentiation markers. Cell lines were grown in Eagle’s Minimum Essential Medium and supplemented with 10% FBS and a combination of 100 U penicillin with 0.1 mg/mL streptomycin (P/S). Cells were grown in a humidified condition of 95% air and 5% CO2 at 37°C. Once confluent, cells were washed with PBS twice, trypsinized and re-suspended in media. Approximately 3 million cells in 150 μL of media were implanted per animal. Animal Model Experiments were performed with the approval of the Albert Einstein College of Medicine Institutional Animal Care and Use Committee and in accordance with the institutional animal welfare policy. Six- to eight-week old female CB17 SCID mice weighing approximately 20 g (Taconic Biosciences) were housed in a pathogen-free barrier facility until tumor implantation. Twenty mice per experiment underwent heterotopic implantation of OS-17 in the right flank [14, 15]. Tumors were allowed to grow until palpable, measuring approximately 5 mm in diameter. Radioisotopes and Radiolabeling of Antibodies The IGF2R-specific murine mAb MEM-238 and the isotype matching control mAb MOPC21 (Abnova, Novus Biologics) were reduced using 75 molar excess of dithiothreitol (DTT) over the antibody by adding 5-10 μL of 3 mg/mL DTT in water to the mAbs to generate SH groups on the mAbs [16]. The mixture was left to react for 40 minutes at 37°C. The mAb was washed to remove excess dithiothreitol with 2×1.5 mL of ammonium acetate buffer in a Centricon-30 microconcentrator (Amicon). The β-emitting isotope 188Re in the form of Na188ReO4, was eluted with normal saline from a 188W/188Re radionuclide generator (Isotope Technologies). The 188ReO4− was reduced for 60 minutes at 37°C with 20 mg/mL SnCl2 in 0.1 M HCl in the presence of 50 mg/mL sodium gluconate. The reduced 188Re was then incubated with the DTT-treated antibodies for 60 min at 37°C as in [16]. The radiolabeling yields for 188Re-mAbs were measured by instant thin layer chromatography (ITLC) by developing silica gel (SG) 10 cm strips in saline. In this system the radiolabeled antibodies stay at the point of application while free 188Re moves with the solvent front. The average radiolabeling yield for either 188Re-mAbs was 85±5%. The radiolabeled mAbs were purified by HPLC using a TSK gel G4000SW size exclusion column (Tosoh Biosciences, Japan) eluted with PBS at 1 ml/min using Waters HPLC system equipped with UV (Waters 2487) and radiation (Bioscan Flow-Count) flow-through detectors. Biodistribution As murine and human IGF2Rs are not homologous, the mAb MEM-238 to human IGF2R can only bind specifically to IGF2R in OS17 xenografts. Thus, the biodistribution was performed to assess whether IGF2R-specific mAb will localize in the tumor preferentially when compared to non-specific isotype matching control MOPC21. The in vivo stability of the radiolabeled mAbs could also be assessed in such experiment. . The IGF2R-specific murine mAb MEM-238 and isotype matching control mAb MOPC21 were radiolabeled with 188Re as above. Mice were randomized and administered 50 μCi via an intraperitoneal injection, with ten mice receiving 188Re-MEM-238 and ten receiving 188Re -MOPC21. At 24 hours post-injection, 5 mice from each group were sacrificed and their tumors, blood and major organs were harvested, weighed and counted for radioactivity in a gamma counter using a dosimetry approach developed specifically for the laboratory rodent models, accounting for both for γ- and β-radiation. [17, 18] At 48 hours post-injection, the remaining 5 mice were sacrificed and similar harvesting and measurements were performed. The percentage of the injected radiolabeled mAb dose per gram of tissue (ID/g, %) was calculated for each animal. Binding Sites The Scatchard transformation was used to assess the number of IGF2R binding sites expressed on OS-17 cells [19]. OS17 cells were cultured and for each concentration of radiolabeled mAb, approximately 8 million cells were included. The experiment was performed twice using 0.0025, 0.005, 0.010, and 0.020 nM concentrations of radiolabeled 188Re MEM-238 and MOPC21 mAbs. The calculation of association constant Ka and the number of IGF2R binding sites per OS-17 cell were performed as previously reported. [19] In Vivo RIT Studies Following OS-17 implantation and growth to 5 mm in diameter, the mice were randomized into 4 groups. Group 1 received 300 μCi of 188Re-MEM-238, group 2 received 300 μCi of 188Re-MOPC21, group 3 received unlabeled (cold) MEM-238, and group 4 was left untreated. This dose level was chosen as it proved to be effective for other 188Re-antibody combinations in RIT of experimental cervical cancer [20]. Intraperitoneal injections were performed as previously described. Tumor volume was calculated every 3 days. Electronic calipers were utilized to measure the 3 largest diameters of the mass. Volume was calculated utilizing the equation for spheroid volume (V=Π·d1·d2·d3/6). To normalize the variability in tumor size between each animal, tumor volume ratio was utilized. Tumor growth ratio (TGR) was calculated by dividing the difference between the initial tumor volume at the start of the experiment (Vo) and the calculated tumor volume (Vn), divided by the initial tumor volume: TGR=(Vo-Vn)/Vo [21]. The calculated TGRs were plotted against time. The experiment was performed twice, initially over a 14-day period (Supplemental Figure 1s.) and thereafter over a 24-day period. Histologic Analysis The tumors were harvested at the conclusion of the RIT study. Tumors were fixed in 4% paraformaldehyde, decalcified, and embedded in paraffin blocks. Paraffin-embedded tissue slides were heated at 60°C for 1 hour, de-paraffinized using xylenes and rehydrated using graded alcohols. Endogenous peroxidase activity was quenched using 0.3% hydrogen peroxide in methanol. Antigenic proteins were unmasked by heat induced antigen retrieval method using sodium citrate buffer. The tissue was blocked with 10% normal goat serum in 1% bovine serum albumin (BSA) in tris-buffered saline (TBS), and stained with 10 μg/mL MEM-238 IGF2R mAb diluted in 1% BSA in TBS overnight at 4°C. Commercially available paraffin-embedded placenta tissue (BioChain Institute, Hayward, CA) was used as the positive control, and MOPC-21 IgG1 diluted in the same diluent as above was used instead of the primary antibody as the isotype negative control. Detection of the antibody-binding reaction was performed with biotinylated secondary antibody coupled with streptavidin-horseradish peroxidase. Avidin biotinylated enzyme complex (Vectastain ABC System; Vector Laboratories) was used according to the manufacturer’s instructions. The tissue was treated with 3,3’-diaminobenzidine (Vector Laboratories) to visualize the antibody binding and counterstained with hematoxylin. The tissue was then dehydrated with alcohol, permeated with xylenes and mounted with Permount organic mounting solution (Fisher Scientific). Staining technique was optimized by comparison with positive and negative controls. Stained slides were viewed using a PerkinElmer P250 High Capacity Slide Scanner (SIG #1S10OD019961-01, TissueScope™, Huron Digital Pathology) and staining patterns were contrasted quantitatively utilizing 3-D imaging analysis software (Volocity® 6.3, PerkinElmer). The total pixels of the tissue and the total number of brown pixels on each slide were counted. The total number of brown pixels was divided by the total number of pixels of tissue on each slide. Three slides per group were evaluated and results for each group were averaged. Statistical Methodology Power analysis for the in vivo cytotoxicity studies was estimated using PASS version 11 (NCSS, Inc.) using simulations of different tumor volumes based on pilot data and conservative assumptions regarding the groups treated with the radiolabeled antibodies. All simulations showed power of at least 83% with only five animals per group because of the large differences between treated and untreated animals. Thus, 5 mice per group were utilized in the in vivo studies. The differences between the biodistribution groups were analyzed using the Kruskal-Wallis and/or the Mann-Whitney tests. Differences between the treatment groups were similarly analyzed using the Kruskal-Wallis and/or the Mann-Whitney tests. Immunohistochemical analysis was performed by comparing the fraction of stained pixels using an overdispersed logistic model. A two-sided p value <0.05 was considered statistically significant for all studies. RESULTS OS-17 cells demonstrate a high binding efficacy for IGF2R-specific mAb The results of the Scatchard plot used to characterize and validate the binding affinity of the IFG2R specific murine mAb and the isotype matching control mAb demonstrated that there are 1,800 binding sites for 188Re-MEM-238 mAb with an association constant, ka, of 5.8 × 1010 M−1 (Figure 1). The slope for 188Re-MOPC21 mAb was not significantly different from zero, indicating there was no observed specific binding to OS-17 cells. IGF2R labeled mAb preferentially localizes to tumor cells The distribution of 188Re -MEM-238 in normal tissue was not significantly different from that of the isotype control, indicating that it was stable in vivo and as a result did not preferentially pool within any specific anatomic location. Importantly, when compared with the isotype control, 188Re-MEM-238 preferentially localized within the tumor after 48 hours, at which point complete separation of the two groups was apparent. The radiolabeled IGF2R antibody exhibited a 1.8 fold selective uptake within tumor tissue compared to the isotype control. This difference approached statistical significance (p=0.057, Figure 2). The clearance of 188Re-MEM-238 from blood was faster than that of the control 188Re-MOPC21. Treatment with 188Re-labeled IGF2R-specific mAb results in tumor growth suppression in vivo The effect of both specific and non-specific radionuclide therapy was evident, with Group 1 (188Re-MEM-238) and Group 2 (188Re-MOPC-21) demonstrating a clear decrease in both absolute tumor volume (p<0.001) and tumor growth ratio (p=0.001) compared with that of the unlabeled control groups, (unlabeled MEM-238 and untreated) in which tumor grew aggressively. The pilot RIT experiment, performed over a 14-day period (Figure 1S), demonstrated a strong trend to suppress tumor growth in the 188Re-MEM-238 treated group compared to the control groups. In the follow-up 24-day RIT experiment (Figure 3) the average tumor growth ratio (TGR) was 4.8 for the 188Re-MEM-238 group, as compared to 13.7 for the 188Re-MOPC-21 group, 29.6 for the unlabeled MEM-238 group, and 30.4 for the untreated group. A whisker plot of day 24 TGR demonstrates the range of values for each treatment arm of the experiment (Figure 2S). The TGR of Group 1 (188Re-MEM-238) was compared to that of Group 3 (unlabeled MEM-238) and Group 4 (untreated), demonstrating suppression of tumor growth in Group 1 (188Re-MEM-238, p=0.005). The specific effect of 188Re-MEM-238 utilized in Group 1 appeared more profound than that of control Group 2 (188Re-MOPC-21), reaching a non-significant trend (p=0.057). Treatment with 188Re-labeled IGF2R-specific mAb induces cytotoxicity in OS cells expressing IGF2R Treatment with 188Re-MEM-238 (Group 1) resulted in 37.4% staining of total pixels after 24 days, (Figure 4A), suggesting a substantial tumoricidal effect on IGF2R-expressing OS cells. Treatment with 188Re-MOPC21 (Group 2) resulted in 44.4% staining of total pixels (Figure 4B), suggesting a partial effect on IGF2R-expressing OS cells. Robust IGF2R staining was visualized in tumor treated with unlabeled MEM-238 (Group 3, Figure 4C) and in untreated tumor (Group 4, Figure 4D), in which 60.8% and 59.6% of total pixels were stained respectively, suggesting minimal to no effect on IGF2R-expressing OS cells. Staining of 188Re-MEM-238 (Group 1) was significantly different from that of unlabeled MEM-238 (Group 3) and untreated (Group 4, p=0.002). The difference between the188Re-MOPC21 (Group 2) and the 188Re-MEM-238 (Group 1) did not reach significance. DISCUSSION This study demonstrates the feasibility of targeting IGF2R with RIT in a heterotopic murine OS model. Scatchard plot analysis demonstrated a moderate number of available IGF2R binding sites and tight Ab binding, reflected by the high binding constant [19]. Since efficacy of RIT is directly proportional to the mAb’s binding constant, high RIT efficacy was predicted. It should be noted that in RIT studies where similar antigen levels on the surface of the targeted cells were observed in vitro [22,23], the RIT was nevertheless successful in killing the targeted cells. In the future, use of αemitters should also be explored for IGF2R-targeted RIT of osteosarcoma to counteract the relatively low number of binding sites. The radioimmunoconjugate demonstrated preferential localization to tumor tissue at 48 hours when compare to the isotype control. Finally, the use of RIT for the treatment of OS resulted in the suppression of tumor growth within the described model. The summation of these findings suggests that IGF2R has the potential to be an effective target for the treatment of OS and that 188Re-labeled IGF2R-specific mAb may offer a useful approach for the treatment of OS, particularly in patients with metastatic or refractory disease. Currently, there is a paucity of therapeutic options available to patients who fail first-line therapy and the addition of conventional chemotherapeutic agents to the backbone of methotrexate, doxorubicin, and cisplatin is unlikely to significantly improve overall survival. Novel approaches are needed for patients who demonstrate local or distant relapse and arguably for patients who respond poorly to standard chemotherapy. Techniques such as stereotactic body radiotherapy, proton therapy, and immune-modulated radiotherapy have changed the landscape of radiation therapy and overcome some of its historical challenges [24]. To date, the gold standard treatment for OS unequivocally remains complete resection and cytotoxic chemotherapy [2, 5]. However, radiotherapy is utilized in patients for whom complete resection is not feasible, such as in patients with tumors in the axial skeleton. In such cases, radiotherapy, in combination with standard chemotherapy, has been reported to result in higher overall survival then with chemotherapy alone [25, 26]. Newer radiation modalities are proving to be important for palliative management of pulmonary and osseous disease [24, 27, 28]. Although radiation therapy is not currently a part of first-line therapy, there is precedence for its use in scenarios where conventional strategies are insufficient. Radionuclide therapy for the treatment of osseous metastases from carcinoma and OS has been previously explored. Samarium-153 and Radium-223 (Alpharadin) act as calcium mimetics targeting bone cortical surfaces and areas of high bone turnover. Samarium-153 lexidronan is approved as a palliative agent for pain associated with osteoblastic metastases in hormone resistant prostate carcinoma [29] while Alpharadin has improved median overall survival by 3.6 months in patients with metastatic prostate cancer [30]. Lastly, there is an open phase I trial in OS patients with osteoblastic tumors [29], underscoring its relevance as a therapeutic modality. The effectiveness of combining radionuclides and monoclonal antibodies [31] has been well-documented for both hematopoietic and solid tumors [32-34]. The incorporation of RIT in the treatment of OS is founded on its reliable surface overexpression of IGF2R. IGF2R’s consistent overexpression on the tumor’s surface and mAb localization within the tumor at 48 hours both suggest it is a useful target. Binding studies additionally demonstrate high efficacy of IGF2R, further supporting this contention. RIT acts through the direct-targeted mechanism and also exerts a “cross-fire” effect, whereby proximate non-IGF2R binding cells are impacted as well. This may prove very relevant in the setting of a genetically variable and unstable entity such as OS, as it obviates the need for all tumor cells to possess the target and could rely on a more heterogeneous expression pattern. The current study demonstrates the effect of utilizing IGF2R as a reliable target as demonstrated by 188Re-labeled IGF2R-specific mAb mediated growth suppression and decreased expression of IGF2R immunohistochemical staining. Regardless of whether RIT ultimately develops into a clinically relevant modality, the current study serves as a proof of concept for IGF2R-targeted therapy. Limitations of this study include the relatively short experimental time course. As an initial study, a 24-day time course was felt to be sufficient to characterize preliminary results. However, a longer time course may yield additional information relevant to toxicity, resistance, and durability. The evaluation of human OS was accomplished by using immunocompromised animals. Although the model is well-described and validated, it is difficult to know to what extent the immune response, or lack thereof, contributed to these results. While dictated by the described power analysis, the relatively small sample size utilized likely contributes to the lack of significance in the difference seen between targeted and non-specific RIT treatment, particularly since one animal in each group died at an early time-point. Future studies should incorporate more OS cell lines and should evaluate biodistribution across a greater number of time points. Finally, our antibody was specific only to human IGF2R and future use of an antibody with human and murine specificity will permit for a more comprehensive assessment of anti-tumor effect and associated toxicity. Based on the limitations of the current study, a broader and longer-term study is warranted. Current findings serve as proof-of-principles and the foundation for a more comprehensive investigation. This targeted approach offers the benefits of being independent of a specific pathway, a resistance mechanism, and/or an inherent biologic tumor trait and therefore is relevant to all OS tumors that express IGF2R. It is recognized that this is an entirely different approach to treating this malignancy, but given the lack of effective new treatments it may prove to be of tremendous value, in particular to patients with metastatic disease for whom there is often no useful therapy. The current frustration with unimproved survival rates and the absence of a clearly beneficial second line therapy strongly stand in favor of alternative approaches that may provide a subset of patients a meaningful, though incremental, improvement. Supplementary Material Figure 1 Scatchard plot demonstrating non-specific binding of MOPC21 and specific binding for MEM-238. Figure 2 The biodistribution of radioactivity post-injection across selected SCID mouse organs following the administration of 188Re-labeled IGF2R-specific mAb (MEM-238) and 188Re-labeled control mAb (MOPC21): A) 24 hrs; B) 48 hrs Figure 3 Tumor growth ratio over a 24 day-period demonstrated marked suppression of tumor growth by 188Re-MEM-238 (Group 1). The non-specific control 188Re-MOPC21 mAb (Group 2) produced a lesser effect on tumor growth suppression. The tumors treated with unlabeled mAb (Group 3) and untreated tumors (Group 4) demonstrate uninhibited growth. Figure 4 Immunohistochemical staining demonstrated minimal IGF2R-staining intensity following treatment with 188Re-MEM-238 (A), moderate IGF2R-staining following treatment with 188Re-MOPC21 (B), and robust IGF2R-staining in both tumors treated with unlabeled MEM-238 (C) or in tumor untreated (D). Findings suggest that OS cells expressing IGF2R have been killed by the RIT treatment. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Conflicts of Interest: Richard Gorlick is a member of the Scientific Advisory Board of Oncolytics, which has no relationship to the subject of the current paper. No potential conflicts of interest were disclosed by any of the other authors associated with this manuscript. REFERENCES 1 Hassan SE Cell surface receptor expression patterns in OS Cancer 2012 118 3 740 9 21751203 2 Mirabello L Troisi RJ Savage SA International OS incidence patterns in children and adolescents, middle ages and elderly persons Int J Cancer 2009 125 1 229 34 19330840 3 Mirabello L Troisi RJ Savage SA OS incidence and survival rates from 1973 to 2004: data from the Surveillance, Epidemiology, and End Results Program Cancer 2009 115 7 1531 43 19197972 4 Kaatsch P Epidemiology of childhood cancer Cancer Treat Rev 2010 36 4 277 85 20231056 5 Ferrari S Neoadjuvant chemotherapy with high-dose Ifosfamide, high-dose methotrexate, cisplatin, and doxorubicin for patients with localized OS of the extremity: a joint study by the Italian and Scandinavian Sarcoma Groups J Clin Oncol 2005 23 34 8845 52 16246977 6 Meyers PA OS: a randomized, prospective trial of the addition of ifosfamide and/or muramyl tripeptide to cisplatin, doxorubicin, and high-dose methotrexate J Clin Oncol 2005 23 9 2004 11 15774791 7 Whelan JS EURAMOS-1, an international randomised study for OS: results from pre-randomisation treatment Ann Oncol 2015 26 2 407 14 25421877 8 Savage SA Analysis of genes critical for growth regulation identifies Insulin-like Growth Factor 2 Receptor variations with possible functional significance as risk factors for OS Cancer Epidemiol Biomarkers Prev 2007 16 8 1667 74 17684144 9 Milenic DE Brady ED Brechbiel MW Antibody-targeted radiation cancer therapy Nat Rev Drug Discov 2004 3 6 488 99 15173838 10 Sharkey RM Goldenberg DM Perspectives on cancer therapy with radiolabeled monoclonal antibodies J Nucl Med 2005 46 Suppl 1 115S 27S 15653660 11 Kaminski MS 131I-tositumomab therapy as initial treatment for follicular lymphoma N Engl J Med 2005 352 5 441 9 15689582 12 Klein M Safety and efficacy of 188-rhenium-labeled antibody to melanin in patients with metastatic melanoma J Skin Cancer 2013 2013 828329 23365757 13 Whiteford CC Credentialing preclinical pediatric xenograft models using gene expression and tissue microarray analysis Cancer Res 2007 67 1 32 40 17210681 14 Houghton PJ The pediatric preclinical testing program: description of models and early testing results Pediatr Blood Cancer 2007 49 7 928 40 17066459 15 Peterson JK In vivo evaluation of ixabepilone (BMS247550), a novel epothilone B derivative, against pediatric cancer models Clin Cancer Res 2005 11 19 Pt 1 6950 8 16203787 16 Dadachova E Mirzadeh S The role of tin in the direct labelling of proteins with Rhenium-188 Nucl Med Biol 1997 24 6 605 8 9316092 17 Thompson S 166Ho and 90Y labeled 6D2 monoclonal antibody for targeted radiotherapy of melanoma: comparison with 188Re radiolabel Nucl Med Biol 2014 41 3 276 81 24533987 18 Quispe-Tintaya W Nontoxic radioactive Listeria(at) is a highly effective therapy against metastatic pancreatic cancer Proc Natl Acad Sci U S A 2013 110 21 8668 73 23610422 19 Dadachova E Interaction of radiolabeled antibodies with fungal cells and components of the immune system in vitro and during radioimmunotherapy for experimental fungal infection J Infect Dis 2006 193 10 1427 36 16619191 20 Wang X-G Treating cancer as an infectious disease - viral antigens as novel targets for treatment and potential prevention of tumors of viral etiology PLOS One 2007 2 10 e1114 17971877 21 Revskaya E Radioimmunotherapy of experimental human metastatic melanoma with melanin-binding antibodies and in combination with dacarbazine Clin Cancer Res 2009 15 7 2373 9 19293257 22 Vervoordeldonk SF Preclinical studies with radiolabeled monoclonal antibodies for treatment of patients with B-cell malignancies Cancer 1994 73 3 Suppl 1006 1011 8306242 23 Miederer M Comparison of the radiotoxicity of two alpha-particle-emitting immunoconjugates, terbium-149 and bismuth-213, directed against a tumor-specific, exon 9 deleted (d9) E-cadherin adhesion protein Radiat Res 2003 159 5 612 620 12710872 24 Lo SS Stereotactic body radiation therapy: a novel treatment modality Nat Rev Clin Oncol 2010 7 1 44 54 19997074 25 Ozaki T OS of the spine: experience of the Cooperative OS Study Group Cancer 2002 94 4 1069 77 11920477 26 Ozaki T OS of the pelvis: experience of the Cooperative OS Study Group J Clin Oncol 2003 21 2 334 41 12525527 27 Whelan JS A systematic review of the role of pulmonary irradiation in the management of primary bone tumours Ann Oncol 2002 13 1 23 30 28 Schwarz R The role of radiotherapy in oseosarcoma Cancer Treat Res 2009 152 147 64 20213389 29 Anderson PM Subbiah V Rohren E Bone-seeking radiopharmaceuticals as targeted agents of OS: samarium-153-EDTMP and radium-223 Adv Exp Med Biol 2014 804 291 304 24924181 30 Parker C Alpha emitter radium-223 and survival in metastatic prostate cancer N Engl J Med 2013 369 3 213 23 23863050 31 Palmedo H Repeated bone-targeted therapy for hormone-refractory prostate carcinoma: tandomized phase II trial with the new, high-energy radiopharmaceutical rhenium-188 hydroxyethylidenediphosphonate J Clin Oncol 2003 21 15 2869 75 12885803 32 Kraeber-Bodere F Radioimmunoconjugates for the treatment of cancer Semin Oncol 2014 41 5 613 22 25440606 33 Dadachova E Dead cells in melanoma tumors provide abundant antigen for targeted delivery of ionizing radiation by a mAb to melanin Proc Natl Acad Sci U S A 2004 101 41 14865 70 15469923 34 Jain M Emerging trends for radioimmunotherapy in solid tumors Cancer Biother Radiopharm 2013 28 9 639 50 23844555

PMC005xxxxxx/PMC5118106.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8712028 5753 Mol Microbiol Mol. Microbiol. Molecular microbiology 0950-382X 1365-2958 27542978 5118106 10.1111/mmi.13485 NIHMS816375 Article UAP56 is a conserved crucial component of a divergent mRNA export pathway in Toxoplasma gondii Serpeloni Mariana 123§ Jiménez-Ruiz Elena 3§ Vidal Newton Medeiros 4 Kroeber Constanze 3 Andenmatten Nicole 3 Lemgruber Leandro 3 Mörking Patricia 1 Pall Gurman S. 3 Meissner Markus 3* Ávila Andréa R. 1* 1 Instituto Carlos Chagas, FIOCRUZ, Curitiba, Brasil 2 Departamento de Biologia Celular e Molecular, Universidade Federal do Paraná, Curitiba, Brazil 3 Wellcome Trust Centre for Molecular Parasitology, Institute of Infection, Immunity & Inflammation, College of Medical, Veterinary and Life Sciences, University of Glasgow, United Kingdom 4 National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, USA * To whom correspondence should be addressed: ARA: Tel: +55 (41)33163230 aravila@fiocruz.br. MM: Tel: +44(0)141 3306201; Markus.Meissner@glasgow.ac.uk § Equally contribution 17 9 2016 14 9 2016 11 2016 01 11 2017 102 4 672689 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. SUMMARY Nucleo-cytoplasmic RNA export is an essential post-transcriptional step to control gene expression in eukaryotic cells and is poorly understood in apicomplexan parasites. With the exception of UAP56, a component of TREX (Transcription Export) complex, other components of mRNA export machinery are not well conserved in divergent supergroups. Here we use Toxoplasma gondii as a model system to functionally characterize TgUAP56 and its potential interaction factors. We demonstrate that TgUAP56 is crucial for mRNA export and that functional interference leads to significant accumulation of mRNA in the nucleus. It was necessary to employ bioinformatics and phylogenetic analysis to identify orthologs related to mRNA export, which show a remarkable low level of conservation in T. gondii. We adapted a conditional Cas9/CRISPR system to carry out a genetic screen to verify if these factors were involved in mRNA export in T. gondii. Only the disruption of TgRRM_1330 caused accumulation of mRNA in the nucleus as found with TgUAP56. This protein is potentially a divergent partner of TgUAP56, and provides insight into a divergent mRNA export pathway in apicomplexans. Graphical Abstract mRNA export pathway conditional Cas9 system Toxoplasma gondii INTRODUCTION Nucleo-cytoplasmic RNA export is an essential post-transcriptional pathway to control gene expression in eukaryotic cells. In Metazoan and Fungi, the nuclear export of most RNA species (such as microRNAs, ribosomal rRNAs, small nuclear RNAs and transfer RNAs) requires specific exportins and the small GTPase Ran. In contrast, nuclear export of bulk messenger RNAs (mRNAs) is Ran-exportin independent (Cullen, 2003, Kohler & Hurt, 2007). The mRNA export machinery is tightly coupled to mRNA splicing and includes different sets of mRNA binding proteins and nucleoporins (Kohler & Hurt, 2007, Rodriguez-Navarro et al., 2004, Wolyniak & Cole, 2008). In general proteins of THO complex associate with mRNA and recruit processing and export factors resulting in the formation of the Transcription/Export (TREX) complex (Strasser et al., 2002). TREX complex interacts with the spliceosome and mature mRNAs are exported through the nuclear pore complex (NPC) by binding to the heterodimeric receptor Mex67:Mtr2/TAP:p15, in an Ran-independent manner (reviewed by (Katahira, 2015, Wickramasinghe & Laskey, 2015)). The mRNAs are then released for translation into the cytoplasm by ATP-dependent helicase Dbp5/DDX19 (Kohler & Hurt, 2007, Nino et al., 2013). While mRNA export is well understood in opisthokonts, in the case of early divergent supergroups, such as Chromalveolata and Excavata, proteins involved in this pathway are not conserved. The only exception is Sub2 (UAP56), a DEAD-box helicase and member of TREX complex (Serpeloni et al., 2011b). While a role of UAP56 in mRNA processing has been suggested for the apicomplexa Plasmodium falciparum (Shankar et al., 2008), its role in mRNA export remains unknown. Using the apicomplexan Toxoplasma gondii as model, we show that knocking-out TgUAP56 led to a significant block of mRNA export, suggesting the presence of a specific mRNA export route akin to other eukaryotes. However, the identification of interaction partners of this crucial protein was not straightforward using standard search analysis due to low conservation of the components. Instead, we employed bioinformatics and phylogenetic analysis to identify T. gondii orthologs of factors that have been described as essential for mRNA export in opisthokonts. We identified some orthologs and discovered that these proteins sequences are very divergent in comparison with orthologs from other species. We predicted we would find the main mRNA export receptor in eukaryotes, Mex67, however our bioinformatics and phylogeny analysis failed to reveal its ortholog in T.gondii. Therefore, it may be possible that in apicomplexans, unlike in opisthokonts, mRNA is exported by a non-conserved export pathway in an Exportin/Ran-dependent manner. To dissect the factors involved in mRNA export in T. gondii, and to discriminate between Ran-dependent and -independent routes, we used reverse genetic strategies to interfere with specific components of both pathways. As part of these strategies we developed a conditional Cas9/CRISPR in T.gondii to target a subset of potential candidates in order to identify factors involved in mRNA export. Functional interference with GTPAse Ran and the exportin CRM1 does not lead to bulk mRNA export defects, suggesting that mRNA export is Exportin/Ran-independent in T. gondii, consistent with other eukaryotes. In addition other orthologs analyzed in this work, including Exportin/Ran-independent factors, were not crucial for mRNA export pointing to the presence of potential unique components for mRNA export in T. gondii. Functional interference with TgRRM_1330, which interacts with TgUAP56 and seems to be a divergent ortholog of Yra1/Aly, led to a phenocopy of TgUAP56 KO. Here we discuss that TgUAP56 and the divergent TgYra1 may be the first components of a divergent mRNA export pathway operating in apicomplexan parasites, and how these can be used empirically to identify other components that to date could not be identified through in silico sequence-based homology screens. RESULTS UAP56 is crucial for mRNA export in T. gondii We have previously shown that most components central to the Ran-independent mRNA export pathway are not conserved in organisms that divergent early during evolution (Serpeloni et al., 2011b). One exception that is conserved in all eukaryotes is the DEAD-box helicase UAP56, which is also involved in splicing of pre-mRNAs (Libri et al., 2001, Fleckner et al., 1997, Thakurta et al., 2007, Jensen et al., 2001, Strasser & Hurt, 2001). In order to reconstruct the phylogeny of UAP56 orthologs sequences across eukaryotes, we surveyed 43 representative species of different eukaryotic groups. The data presented in Figure S1 show that UAP56 protein phylogeny resembles the eukaryotic species phylogeny. The orthologs protein of UAP56 in T. gondii, TgUAP56, is annotated as a DEAD-box polypeptide DDX39 (ID TGME49_216860 at ToxoDB, GI 237843393 at NCBI) and is 78.6 and 72.8 % similar to orthologs in human cells (Hsa_UAP56) and yeast (Sce_Sub2), respectively. Multiple sequence alignments of TgUAP56 and representative sequences from other eukaryotic groups demonstrate high sequence conservation along the entire protein including two RNA helicase domains. To characterize the function of TgUAP56, we generated transgenic parasites expressing TgUAP56 N-terminally fused to ddFKBP-GFP (Breinich et al., 2009) (dd-GFP-TgUAP56). The ddFKBP domain allows conditional stabilization of the fusion protein in the presence of the ligand Shield-1 (Shld1) (Banaszynski et al., 2006). We verified that incubation of transgenic parasites with 1 μM results in stabilization of dd-GFP-TgUAP56. Furthermore, dd-GFP-TgUAP56 co-localizes with endogenous TgUAP56 in the nucleus, as shown by immunofluorescence analysis using a polyclonal antibody raised against the trypanosome ortholog (Serpeloni et al., 2011a), which specifically detects TgUAP56 (Figure 1A-i, ii). dd-GFP-TgUAP56 stabilization is very efficient and the protein can be detected as early as 6 hours after addition of 1 μM Shld1, reaching a peak at 36 h (Figure 1B-i). We found that overexpression of dd-GFP-TgUAP56 efficiently interferes with parasite growth indicating functional interference with endogenous TgUAP56 and an essential role for this protein (Figure S2A). We speculated that block of parasite growth is caused by interference with mRNA export. Indeed, the incubation of parasites for 48 hours in presence of Shld1 caused an obvious nuclear accumulation of mRNA that co-localizes with dd-GFP-TgUAP56 (Figure 1B-ii). To exclude that the phenotype observed after overexpression of dd-GFP-TgUAP56 is due to a non-specific interference with the mRNA export pathways, we used an inducible gene-swap strategy (Andenmatten N., 2013) to generate a conditional knockout for uap56 (cKOuap56). This strategy is based on site-specific recombination by DiCre, which catalyzes excision of DNA flanked by loxP sites after induction with rapamycin (Bargieri et al., 2013, Andenmatten N., 2013). Parental DiCre strain was transfected with cKOuap56 plasmid to generate a stable cell line for conditional knockout of uap56 gene (cKOuap56 strain). Upon induction, excision of uap56 results in replacement of the reporter gene mcherry under control of the endogenous promoter and hence knockout parasites can be identified based on red fluorescence (Figure 1C-i). We confirmed correct homologous recombination of the uap56-LoxP cassette in the original uap56 locus by gDNA PCR analysis (Figure S2B-i,primers set 1 and 2, A and C comparison). uap56 gene excision was confirmed at genomic level by gDNA PCR analysis after 24 hours induction with rapamycin (Figure S2B-ii, primers set 1 and 2, A and B comparison). Real-time quantitative PCR and immunoblot assays were performed to analyze uap56 mRNA and TgUAP56 protein levels, respectively, at specific time points after rapamycin induction. Non-induced and induced parasites of parental DiCre strain have normal levels of uap56 mRNA in the presence of rapamycin at all time points tested (Figure S2C-i). In the case of induced parasites of cKOUAP56 strain, uap56 mRNA levels decreased drastically 48 hours after addition of rapamycin (Figure S2C-ii). In good agreement, the levels of TgUAP56 protein significantly decreased after 24 hours of rapamycin incubation and after 48 hours it is virtually undetectable (Figure 1C-ii). The nuclear accumulation of poly(A)+ RNA is observed after 48 hours when levels of TgUAP56 are undetectable (Figure 1C-iii). The incubation of parasites in presence of 50 nM rapamycin has no toxic effect on parasites as demonstrated previously (Andenmatten et al., 2013). In our case, the treatment of rapamycin itself in the parental DiCre strain did not affect the expression levels of TgUAP56 (Figure S2E-i) or the export of mRNAs (Figure S2E-ii). In addition, the detection of mcherry expression after rapamycin incubation confirmed that uap56-loxP was excised successfully (Figure 1C-iii). Furthermore, we confirmed that TgUAP56 is essential, since uap56 excision caused a lethal phenotype (Figure S2D). Importantly, a similar mRNA accumulation in the nucleus was observed after dd-GFP-TgUAP56 stabilization, demonstrating that this phenotype is specific and therefore an essential role for TgUAP56 in mRNA export. Next we surveyed a selection of genes by a semi-quantitative PCR analysis developed by Suvorova et al, 2013 (Suvorova et al., 2013), using primers spanning an intron as listed in Table S2 to assess if TgUAP56 plays an important role in mRNA-splicing. Parasites of parental DiCre and cKOuap56 strains were incubated with 50 nM of rapamycin for 24 and 48 hours before extraction of total RNA. gDNA extracted from cKOuap56 strain was used as a reference for intron-containing pre-spliced mRNAs. PCR data did not show accumulation of pre-mRNA for any of the genes analyzed indicating that uap56 knockout does not affect mRNA splicing (Figure 1D). Bioinformatic and phylogenetic analysis provide the identification of ortholog proteins in T. gondii Since TgUAP56 is related to mRNA export in T. gondii, we decided to systematically probe the genome of the parasite in order to identify proteins that are potentially related to TgTREX (T. gondii TREX complex) and downstream events. Our first approach was based on protein sequence search and phylogenetic reconstruction. The general criteria to choose the candidates were: a) evidence of Sub2/UAP56 interaction partners and/or b) they are essential for mRNA export in humans or yeast. Sub2/UAP56 and Yra1/Aly are known partners that are part of TREX complex in opisthokonts and are loaded onto mRNAs in a splicing-dependent manner (for reviews, see (Masuda et al., 2005, Reed & Hurt, 2002, Custodio et al., 2004)). TREX can recruit the RNA export receptor Mex67:Mtr2 to spliced mRNAs (Hurt et al., 2004, Hackmann et al., 2014, Iglesias et al., 2010, Gilbert & Guthrie, 2004) and it is known that TAP:p15 can be targeted to spliced mRNPs directly to a spliceosome U2AF35 subunit and this interaction is conserved across metazoan species (Zolotukhin et al., 2002). In addition to these RNA binding proteins, Npl3 and Gbp2, the latter associated with TREX complex, can recruit mRNA export receptor Mex67:Mtr2 to mRNAs also and are crucial for formation of export-competent mRNP (Hurt et al., 2004, Hackmann et al., 2014, Iglesias et al., 2010, Gilbert & Guthrie, 2004). Besides, mutants of npl3 show strong mRNA-export defects (Lee et al., 1996, Krebber et al., 1999, Strasser & Hurt, 2000, Gilbert & Guthrie, 2004, Gwizdek et al., 2006, Gilbert et al., 2001) and overexpression of Gbp2 is toxic and causes a nuclear retention of bulk poly(A)+ RNA (Windgassen & Krebber, 2003). Our approaches allowed the identification of orthologs of most proteins cited above in T. gondii, although low sequence conservation is observed in Kinetoplastida and Apicomplexa (Table 1, Figures S4–7). However, this approach did not allow the identification of a T. gondii ortholog of Mex67, the specific receptor of mRNA export in yeast. Based on the fact that GTPase Ran and CRM1 are required for the export of proteins and a subset of mRNAs in some eukaryotes (Cullen, 2003, Wickramasinghe & Laskey, 2015) (reviewed by (Delaleau & Borden, 2015)), we used the T. gondii orthologs to assess if mRNA export was Ran-dependent in T. gondii. Exportin CRM1 and its cofactor GTPase Ran are highly conserved throughout eukaryotic phylogeny (Figures S8 and 9) (Serpeloni et al., 2011b)). To analyze the potential role of the candidates in mRNA export we took advantage of the conditional gene disruption by ddCas9, as described below. Conditional Cas9 expression allows identification of mRNA export mutants The recent adaptation of CRISPR/Cas9 in T. gondii parasites is a powerful technology to generate direct knockouts for non-essential genes in a rapid and reliable manner (Shen et al., 2014, Sidik et al., 2014). Transient expression of Cas9 might be helpful to rapidly identify certain phenotypes (Harding et al., 2016), but constitutive expression of Cas9 might lead to artifacts. To overcome this limitation we designed a conditional nuclear Cas9 fused to ddFKBP to allow precise regulation of Cas9 expression levels (Figure S3A). ddCas9 is detectable as early as 1 hour after addition of Shld1 (Figure 2A) and localizes mainly to a defined region within the nucleus in all parasites. Longer stabilization leads to Cas9 localization throughout the nucleus of the parasite (Figure 2B, Figure S3B). Importantly, while short incubation times, up to 4 hours, with Shld1 did not significantly affect parasite viability, longer stabilisation leads to accumulation of ddCas9 and appearance of aberrant parasites (Figure S3B). To minimise toxicity and off-target effects caused by over-stabilisation of ddCas9, conditional mutants were generated by incubation of parasites with 1 μM of Shld1 for 4 hours. To validate the efficiency and specificity of conditional ddCas9, we stably transfected RHddCas9 with 2 different sgRNA-expression vectors (Table S2, Figure S3C) targeting gap40 (gliding associated protein 40) and reproduced the highly specific phenotype in the inner membrane complex (IMC) biogenesis caused by deletion of gap40 through conventional KO strategies (Harding et al., 2016). In the absence of Shld1 RHddCas9-gap40sgRNA clonal strains generated from the 2 different vectors showed no deficit. Addition of Shld1 resulted in efficient disruption of gap40; presenting typical gap40KO phenotype in ~ 65% of the induced population for both clones (Figure S3D). We confirmed the essentiality of gap40 for parasite growth (Figure 2C) and the typical collapse of the IMC (Figure 2D), which is detected as early as 24 hours after induction. Using PCR analysis we confirmed that the RHddCas9 parental strain and non-induced RHddCas9-gap40 sgRNA strain had no mutations in gap40. In contrast, specific mutations in gap40 gene (deletions/insertions) were identified in gDNA from RHddCas9-gap40sgRNA incubated with Shld1 at the exact position targeted by the gap40 sgRNA sequence (Figure S3E). As a negative control we used a sgRNA targeting an exogenous sequence not present in the T. gondii genome (lacZ gene). Neither parental RHddCas9 strain nor RHddCas9-lacZsgRNA strain showed any alteration in parasite morphology or growth in the presence or absence of Shld1 (not shown). Disruption of uap56 led to the same poly(A)+ RNA nuclear accumulation (Figure 2E), as seen for TgUAP56 overexpression and uap56 conditional deletion using the DiCre system (see Figure 1). This phenotype is related to TgUAP56 function, since the absence of the protein leads to mRNA export blocking (Figure 2F-i, ii). This phenotype was not observed in RH-ddCas9 or RH-ddCas9gap40 gRNA strains (Figure 2E), demonstrating the specificity of gene disruption using the ddCas9 strategy. This strategy proves to be an useful tool to analyze the function of genes. However, we observed several caveats that make this tool cumbersome to use in a higher throughput scale (see discussion). Next, we stably introduced sgRNA-expression vectors targeting each potential mRNA export candidate into RHddCas9 (Table S2). Interestingly only the disruption of a RNA recognition motif-containing protein (TGME49_291330, named here as TgRRM_1330) resulted in a similar phenotype as observed for disruption of uap56 (Figure 3). In this case, we observed severe nuclear accumulation of poly(A)+ RNA ultimately leading to the death of the parasite (not shown). In RHddCas9-uap56 gRNA and RHddCas9-TgRRM_1330 gRNA strains mRNA export blocking events were up to 46% and 38% of vacuoles, respectively. In contrast, components of the Ran-dependent export pathway do not seem to be required for mRNA export in T. gondii (Figure 3). Identification of a RNA recognition motif-containing protein as a novel factor for mRNA export in T. gondii The results obtained by ddCas9 dependent disruption were further confirmed by overexpression studies, based on ddFKBP-GFP as described above for TgUAP56. For most of the candidates analyzed, overexpression also resulted in lethal phenotypes without any mRNA export defects (Figure 4). Concomitant with the results obtained for ddCas9-mediated gene disruption, the overexpression of TgRRM_1330 resulted in mRNA export defect (Figure 5A, B) and consequent death of parasites (Figure 5C), as observed for TgUAP56 overexpression. Even though TgRRM_1330 sequence is not conserved, the protein contains a RNA binding domain that is conserved in orthologs of Yra1/Aly, an essential component of mRNA export in yeast (Figure S4), and the experimental data suggest that TgRRM_1330 is potentially a highly divergent functional homologue of Yra1/Aly. In summary, disruption and overexpression of CRM1 and GTPase Ran (components of Ran-dependent pathway) did not show any mRNA export defect in T. gondii. However, TgUAp56 and TgRRM_1330 were found to be shown to be essential for bulk mRNA export, suggesting that T. gondii operate in a Ran independent pathway as found in higher eukaryotes. TgUAP56 and TgRRM_1330 form a functional complex To test the hypothesis that TgRRM_1330 is a partner of TgUAP56 we performed co-localization and immunoprecipitation analysis. For this purpose, we imaged the nucleus of tachyzoites with super-resolution and immune-electron microscopy. Maximum projection of SR-SIM image of TgUAP56 (in red) and dd-GFP-TgRRM_1330 (in green) show a high co-localization rate, with a similar distribution over the nucleus (stained with DAPI, in blue). The graph shows the same localization of the red signal over the green signal, both together with the localization of the DAPI signal (Figure 5D-i). Ultrastructural observation showed a labeling of TgUAP56 (arrows) together with TgRRM_1330 (arrowheads) near areas of dense chromatin in the nucleus of T. gondii tachyzoites (Figure 5D-ii). The interaction between dd-GFP-TgRRM_1330 and TgUAP56 was furthermore confirmed by immunoprecipitation (Figure 5D-iii). This interaction was not observed with dd-GFP (control strain) or with dd-GFP-TgCRM1, as expected (data not shown). DISCUSSION mRNA export is well studied in higher eukaryotes where TREX complex has an important role, including UAP56 and adaptor proteins (for review see (Muller-McNicoll & Neugebauer, 2013)). However, several proteins involved in this pathway are not conserved throughout the eukaryotic phylogeny with the exception of UAP56, including in early divergent eukaryotes (Serpeloni et al., 2011b). We previously demonstrated that this protein is a component of mRNA export pathway in Trypanosomes (Serpeloni et al., 2011a). Here we confirmed the UAP56 ortholog in T. gondii, named TgUAP56. The most divergent sequence is Tryp-Sub2 ortholog (66.1% similarity). Considering that Tryp-Sub2 is a component of the mRNA export pathway (Serpeloni et al., 2011a), it would be reasonable to hypothesize that TgUAP56 has the same role as a basic component of the mRNA export pathway. However, phylogenic relationship is not a guarantee for functional conservation. In this work we present experimental data supporting that UAP56 is an essential factor for mRNA export in T. gondii. TgUAP56 is exclusively nuclear and dispersed all over the nuclei in a punctuate pattern, similar to observations in other eukaryotes (Gatfield et al., 2001, Serpeloni et al., 2011a, Sahni et al., 2010). TgUAP56 overexpression results in a dominant negative phenotype leading to parasite death due to a block of mRNA export. Similarly, overexpression of UAP56 in Caenorhabditis elegans and human cells impairs mRNA export and leads to nuclear retention of poly(A)+ mRNAs (Strasser & Hurt, 2001, Luo et al., 2001) (MacMorris et al., 2003). The specificity of the overexpression effect was confirmed by conditional knockout for uap56, where nuclear accumulation of poly(A)+ RNAs was prominent after 48 hours and consequently parasites were unable to grow and died within the host cell. These results agree with previous observations in yeast, Drosophila melanogaster and trypanosomes where the depletion of UAP56 orthologs resulted in growth arrest and robust accumulation of poly(A)+ mRNAs within the nucleus (Gatfield et al., 2001, Strasser & Hurt, 2001) (Serpeloni et al., 2011a). Together these data demonstrate the essential role of TgUAP56 in mRNA export in T. gondii. We also investigated if TgUAP56 is involved in mRNAs splicing since this protein was originally associated with the splicing machinery (reviewed in Linder and Stutz, 2001 (Linder & Stutz, 2001)). Our approach was based on the strategies used previously for the characterization of the splicing factor TgRRM1 (Suvorova et al., 2013), where the authors showed that a selected list of genes of T. gondii were mis-spliced in the absence of TgRRM1. Importantly, we did not observe any interference in mRNA splicing of these targets in the absence of TgUAP56. Therefore, TgUAP56 appears to be exclusively required for mRNA export, potentially releasing spliced mRNAs from adaptor proteins. This idea is corroborated by studies in other model organisms that indicate that Sub2/UAP56 also plays an essential role in mRNA export (Gatfield et al., 2001, Jensen et al., 2001, Luo et al., 2001, Strasser & Hurt, 2001, Dias et al., 2010, Steckelberg & Gehring, 2014). Since UAP56 is a necessary and specific component of a specialized mRNA export complex in mammals, the identification of homologous proteins in parasites would point to the presence of a similarly specialized pathway in apicomplexans. For this purpose, we decided to use a genetic screen based on the Cas9/CRISPR system. In T. gondii, CRISPR/Cas9 technology has been used successfully in several occasions (Shen et al., 2014, Sidik et al., 2014). However, constitutive expression of Cas9 in T. gondii seems to be toxic. In higher eukaryotes, different strategies for controlling the activity of Cas9 have been employed including the tetracycline inducible system (Zhu et al., 2014), splitting the Cas9 in half (Wright et al., 2015, Zetsche et al., 2015) and, very recently, a conditional Cas9 system based on the fusion with a destabilization domain (FKBP12-L106P) (Geisinger et al., 2016). Here we succeeded to generate parasites that allow regulation of Cas9 activity using the ddFKBP-system (Herm-Gotz et al., 2007) and we show that specific conditional disruption of essential genes is feasible using this approach. Our phenotypic assays using ddCas9 demonstrated that it allows the identification of specific phenotypes. As a proof of concept, we reproduced the phenotype caused by deletion of gap40 (Harding et al., 2016) and confirmed that disruption of this gene using ddCas9 causes the typical collapse of the IMC, while mRNA export is not affected. In contrast, disruption of uap56 caused a block in mRNA export, corroborating our previous findings after conditional deletion of the gene using DiCre and overexpression analysis. However, while ddCas9 can be successfully employed to screen for specific phenotypes, such as mRNA export as shown here, it should be used with caution, when screening for general growth phenotypes. In our experience disruption of non-essential genes resulted sometimes in abnormal parasites (data not shown), a phenomenon we are currently investigating. Furthermore, overexpression of ddCas9 over extended periods results in parasites with aberrant morphology, further complicating employment of this system. Therefore, a thorough downstream analysis using additional reverse genetic tools is required to confirm identified phenotypes. Despite these disadvantages, the ddCas9 system allowed us to efficiently analyze different candidates. The selection of candidates for genetic analysis was based on searching for orthologs of proteins that have been described in mRNA export in opisthokonts. Among all the proteins analyzed, only the ortholog of Mex67 was not identified. In this case, PSI-BLASTP searches resulted in spurious hits corresponding to only the Leucine-rich region of Mex67, no hits corresponding to NTF2-Like and TAP C-terminal domains were found. It may not be surprising since Mex67 is not a highly conserved protein in divergent groups (Kramer et al., 2010, Serpeloni et al., 2011b). The only description of Mex67 in protozoa shows that it is a divergent protein with an essential motif that is absent from all other Mex67 orthologues known (Dostalova et al., 2013, Kramer et al., 2010). Consequently, all the candidates bar Mex67 were studied by functional interference. Our results showed that TgU2_6910, TgRRM_2620 and TgSF2_9530, that are T. gondii orthologs of factors involved in Ran-independent mRNA export pathway in opisthokonts, are not crucial for mRNA export although they are essential proteins for parasite survival. These data point to the lack of a conserved mRNA export pathway in T. gondii. Interestingly, the lack of a conserved pathway has also been observed for the protozoa Trypanosoma brucei albeit Mex67 is a functional mRNA receptor in these parasites. Some authors have proposed the hypothesis of an evolutionarily divergent mechanism for mRNA export (Schwede et al., 2009, Dostalova et al., 2013, Obado et al., 2016) and a shared platform for transport for rRNA and mRNA has been suggested (Neumann et al., 2010, Buhlmann et al., 2015). Based on this, we decided to address if mRNA export in T. gondii would be dependent on exportin (CRM1) and GTPase Ran since they are conserved proteins throughout eukaryote phylogeny and there are evidences of CRM1/Ran involvement in export of specific mRNA and rRNA (for review see (Kohler & Hurt, 2007)). The disruption of both genes did not block the export in bulk of mRNA, suggesting that it is potentially a Ran-exportin independent route. These results together with the identification of TgUAP56 indicate the presence of a specific mRNA export pathway as described for other organisms. We still cannot affirm that Mex67 is absent in T. gondii but undoubtedly the protein structure is very distinct of the receptors described so far. Our genetic approaches identified TgRRM_1330, a RNA binding protein that is essential and specific component of mRNA export in T. gondii. Interestingly, this protein contains a RNA-binding domain that is also present in Yra1/Aly, a component of TREX that interacts with UAP56 orthologs in opisthokonts (Strasser & Hurt, 2000, Luo et al., 2001, Dufu et al., 2010). In good agreement with this, we demonstrate that TgRRM_1330 interacts with TgUAP56 in T. gondii. TgRRM_1330 was not detected using standard BLAST searches, however using PSI-BLAST with 2 iterations it was possible to recover this yeast Yra1 ortholog candidate with a significant E-value (5e-17). The identity of TgRRM_1330 with the yeast (Yra1) and human (Aly) protein is only 21.5% and 24.9% respectively, supporting the idea that mRNA export pathway in T. gondii might contain divergent components. Indeed, apicomplexans and other parasites have acquired particular features in relation to mRNA metabolism during evolution and further evidence is required to check if the presence of divergent components would correspond to distinct mechanisms of mRNA export in comparison with the pathway of opisthokonts. Our results provide the first insights into components of mRNA export in apicomplexan parasites and the description of interacting partners that are crucial for the divergent pathway. These essential proteins can serve as a handle to identify interacting proteins in further investigations. EXPERIMENTAL PROCEDURES Identification of candidate genes in Toxoplasma gondii In order to identify putative ortholog proteins involved in the RNA export pathway in T. gondii, we used previously described proteins in the literature from yeast, human and P. falciparum as query sequences. The list of queries used is as following: Sub2 (GI: 6320119), Yra1 (GI: 6320589), Mex67 (GI: 6325088), GTPase Ran GSp1 (GI: 6323324) and CRM1 (GI: 398366207) from yeast; U2AF35 (GI: 68800128) from human; and Npl3 (PF10_0217) and Gbp2 (PF10_0068) from P. falciparum. We performed BLASTP searches with E-value threshold of 1e-03. To be considered orthologs in T. gondii, identified proteins should satisfy the reciprocal best hit criteria. Alternatively, PSI-BLAST with inclusion threshold of 0.005 was used to identify a candidate ortholog sequence of yeast Yra1 in T. gondii. Identification of orthologs genes in eukaryotes Putative ortholog proteins identified in T. gondii were used as query sequences in BLASTP searches (E-value threshold 1e-03) against the RefSeq database to identity orthologs in 43 representative species from the major eukaryotic groups as listed on Table S1: Metazoans, Fungi, Amoebozoa, Plants, Apicomplexans, Kinetoplastids and Parabasilids. To be considered orthologs, identified proteins should satisfy the reciprocal best hit criteria. Sequence and domain analysis Identity and similarity percentages were obtained using needle program from the EMBOSS package (Rice et al., 2000), which finds the optimal global alignment of two sequences. Protein domain searches were performed running hmmscan program from HMMER package (Eddy, 1998) against the collection of Hidden Markov models downloaded from the Pfam database version 27.0 (Finn et al., 2014). Phylogenetic analysis Multiple sequence alignments of orthologs sequences were done using MAFFT version 7 with the following parameters: -localpair -maxiterate 1000 –reorder (Katoh & Standley, 2013). Phylogenetic analysis was conducted using the approximately maximum likelihood method implemented in the FastTree 2.1 program (Price et al., 2010) with default parameters. The tree was rendered using FigTree v1.4 (http://tree.bio.ed.ac.uk/software/figtree/). Cloning of DNA constructs All oligonucleotides used in this study are listed in Table S2. The TgUAP56 gene swap-vector (loxPUap56loxP-mCherry-HX) was generated by cloning uap56 ORF (TGME49_216860), amplified directly from cDNA with F/R primers using ApaI and loxP-NsiI sites respectively. The amplified fragment was placed upstream of mcherry sequence, replacing tub8 of parental plasmid (unpublished vector). Uap56 5′UTR was amplified from genomic DNA (gDNA) with F/R primers using KpnI and loxP-ApaI sites respectively. The amplified fragment was placed upstream of the uap56 cDNA for the transcription of uap56 gene to be driven by the endogenous promoter present in 5′UTR. Finally, 3′UTR of uap56 was amplified from gDNA using F/R primers using SacI restriction site for cloning. The plasmid was linearized with KpnI before RH DiCre ΔKu80 strain transfection. To generate conditional ddCas9 plasmid (p5RT70DDmycFlagCas9) the synthetic cas9 expressing cassette from Lourido’s lab (Addgene ID 52694(Sidik et al., 2014)) was cloned into plasmid p5RT70ddmycGFPPfMyoAtailTy-HX (Hettmann et al., 2000) using NcoI-NotI restriction sites. The plasmid was linearized with NotI before transfection into the RH Δhxgprt strain. A single guide RNA (sgRNA) cassette for each candidate synthesized by GeneScript was inserted into a plasmid including the DHFR resistant cassette (pU6-gRNA-crisprRNA). The 20 nucleotides of the guide RNA (gRNA) were designed using the online software E-CRISP (http://www.e-crisp.org/E-CRISP/designcrispr.html) aiming for exons close to the 5′ end and maintaining NGG as PAM sequence. The plasmids were linearized with NotI before transfection into the ddCas9 strain. The Overexpression vectors were generated by cloning each ORF (TgUAP56 (TGME49_216860), TgCRM1 (TGME49_249530), TgSF2_9530 (TGME49_119530), TgRRM_2620 (TGME49_062620), TgRan (TGME49_248340), TgRRM_1330 (TGME49_291330), TgU2_6910 (TGME49_236910)), amplified directly from cDNA with F/R primers using appropriate restriction sites, into p5RT70ddmycGFPPfMyoAtailTy-HX (Hettmann et al., 2000). The plasmids were linearized with KpnI before transfection into the RH Δhxgprt strain. Parasite parental and transgenic strains T. gondii tachyzoites (RH Δhxgprt – RH strain (Donald R.G., 1996) and RH DiCre ΔKu80 strain - DiCre strain (Pieperhoff et al., 2015), and transgenic strains generated in this study) were cultured on human foreskin fibroblasts (HFF) and maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum, 2 mM glutamine and 25 μg/mL gentamicin. Freshly released parasites (~5x107) were transfected and selected in presence of mycophenolic acid and xanthine (Donald R.G., 1996) or pyrimethamine to generate stable lines as previously described (Donald & Roos, 1993). RH Δhxgprt parasites were transfected individually with overexpression plasmids to generate clonal lines for overexpression of all candidates. The RH DiCre ΔKu80 strain was transfected with loxPUap56loxP-mCherry-HX plasmid to generate TgUAP56 knockout strain (cKOuap56) based on a previously used gene-swap strategy (Andenmatten N., 2013). RHΔhxgprt parasites were transfected with p5RT70DDmycFlagCas9 plasmid to generate a conditional Cas9 expressing strain (ddcas9). This parental cell line was transfected with individual synthetic pU6-gRNA-crisprRNAs plasmids (listed on Table S2) to generate ddCas9 strains for each target candidate. Analysis of mutations caused by ddCas9/CRISPR Targeting Genomic DNA was extracted from induced and non-induced parasites using the DNeasy Blood and Tissue kit following manufacturer procedures (Qiagen cat# 69506). PCR primers flanking the predicted target site were used to amplify amplicons using High Fidelity Platinum® Taq DNA Polymerase (ThermoFisher cat# 11304-011), PCR primers listed in Table S2. Amplicons were cloned into pGEM-T Easy Vector System (Promega cat# A1360) using standard cloning and amplification procedures prior to sequencing (LIGHTRUN™ Sequencing Service at GATC, Germany) using T7 and SP6 primers (Table S2). Sequences were analyzed using BioEdit software version 7.2.5. (Tom Hall. Ibis Biosciences. Carlsbad, CA, USA). Immunoblot assays Parasites were incubated in culture media supplemented with or without 1 μM Shld1 (overexpression assays and Cas9-mediated disruption assays) or 50 nM of rapamycin (inducible knockout assay). Protein extracted from parasites were prepared for Western Blot analysis as described previously (Hettmann C., 2000), using 12% polyacrylamide gels under reducing condition with 100 mM DTT. Equal number of parasites was loaded per experiment. Polyclonal anti-Tryp-Sub2 (1:1000) (Serpeloni et al., 2011a), monoclonal anti-GFP (1:500) (Roche, #cat11814460001) and monoclonal anti-flag (1:500) (Fisher/Thermo Scientific, #cat 11525702) antibodies were used for specific protein detection respectively. Monoclonal anti-aldolase (1:10.000) (Starnes et al., 2006) was used as loading control. ImageJ software with the densitometry plugin (Version 1.6, National Institutes of Health, Bethesda, MD) was used for protein quantification. Fluorescent in situ hybridization (FISH) and Immunofluorescence assays Intracellular parasites were grown in the absence or presence of 0.1–2 μM Shld1 (overexpression assays), 1 μM Shld1 (Cas9-mediated disruption assays) or 50 nM of rapamycin (inducible knockout assay). To detect poly(A)+ RNA in T. gondii FISH assays were performed as previously described (Lirussi & Matrajt, 2011, Serpeloni et al., 2011a). Tachyzoite-infected HFF cells grown on glass coverslips were fixed with 4% formaldehyde in PBS for 20 minutes at room temperature and permeabilized with 0.2% Triton X-100 in 2× SSPE(SSPE 2X: 300 mM NaCl, 20 mM phosphate buffer pH 7.4, 2 mM EDTA) for 20 minutes. The parasites were washed 3 times with 2X SSPE and blocked with hybridization solution (HS) (10% Dextran, SSPE 2X, 35 % formamide, 0,5 mg/mL tRNA) for 30 minutes at 37 0C in a humidified chamber. Further, 1ng/μL of the probe (oligodT conjugated with Alexa488 or Alexa555, synthesized by Invitrogen) was added in HS and denatured for 3 minutes at 65 0C before hybridization with the cells. The cells were incubated with the probe for 16 hours at 37 0C in the humidified chamber. After the hybridization, the cells were washed with 2× SSC followed by 1× SSC for 15 minutes each (1× SSC : 0.15M NaCl, 0.015M Na3Citratex-2H2O pH 7.4). For IF assays, cells were blocked and permeabilized with PBS containing 3% bovine serum albumin and 0.2% Triton X-100 for 20 minutes before incubation with primary antibody for one hour at RT. Anti-TrypSub2, (Serpeloni et al., 2011a), anti-IMC1 or IMC3 (1:1000) and anti-flag (1:200) (Fisher/Thermo Scientific, #cat 11525702) antibodies were used for immunolocalization. Post primary antibody hybridization Cells were incubated with the secondary antibodies (Alexa 594-conjugated anti-rabbit (Invitrogen, #cat A11012); Alexa 488 or 594-conjugated goat anti-mouse antibodies (Invitrogen #cat A-11001 and A11005, respectively)) diluted to 1:3000 and incubated for 1 hour at RT. Cells were washed (buffer) several times and mounted onto slides with DAPI Fluoromount (Southern Biotech UK, #cat 0100–20). We combine immunofluorescence and FISH assays. For this, the immunofluorescence protocol was used and a fixation step with PFA 4% fixation followed by incubation oligodT conjugated with Alexa (as described for FISH) was included before the incubation with secondary antibodies. Image acquisition was conducted using a 100X or 63X oil objective on a Zeiss Axioskop 2 fluorescence microscope with an Axiocam MRm CCD camera using Zen software (Zeiss). Further image processing was performed using ImageJ 1.34r software and Photoshop CS6 (Adobe Systems Inc). Immunofluorescence signals quantification was performed with ImageJ software with densitometry plugin (Version 1.6, National Institutes of Health, Bethesda, MD). Super-resolution structure illumination microscopy was performed using a Zeiss Elyra PS.1 super-resolution microscope equipped with a sCMOS PCO camera. The Plan-Apochromat 63×/1.4 Oil DIC lens was used, and Z-stacks were acquired in 5 rotations using the ZEN Black Edition Imaging software (Zeiss). Images were then processed in ZEN Black Edition Imaging software (Zeiss), using the structural illumination manual processing tool, with a noise filter of −6.0, and an output as SR-SIM. Colocalization and fluorescence intensity analysis were carried out in FIJI software (Schindelin et al., 2012). Immunoelectron Microscopy For the immunocytochemistry, infected cells were fixed overnight in 4% paraformaldehyde and 0.2% glutaraldehyde in phosphate buffer. The samples were washed in phosphate buffer and dehydrated in ascending ethanol solutions. After a progressive infiltration process with LR White resin, the polymerization was carried out in gelatin capsules under ultraviolet light. Formvar-coated nickel grids with ultrathin sections were incubated with blocking solution (3% BSA, 0.02% Tween 20, in phosphate buffer) for 1 hour. The grids were incubated with the primary antibody anti-GFP (1:20, Roche, #cat11814460001) diluted in blocking buffer for 1 hour followed by several washes in blocking buffer. The grids were then incubated with 10 nm gold-conjugated anti-mouse secondary antibody (1:10, Aurion, Netherlands) for 1 hour, followed by several washes with blocking buffer. The grids were incubated with anti-TrypSub2 primary antibody (1:50, polyclonal (Serpeloni et al., 2011a)) for 1 hour, followed by several washes in blocking buffer. The grids were then incubated with 15 nm gold-conjugated Protein A (Aurion, Netherlands) for 1 hour, followed by washes in blocking buffer and phosphate buffer. The material was stained with uranyl acetate prior observation in a Tecnai T20 transmission electron microscope (FEI, Netherlands). Images were analyzed and processed in FIJI (Schindelin et al., 2012) and Adobe Photoshop. Real-time PCR (qPCR) Total RNA was isolated in duplicate from RH DiCre ΔKu80 and cKOuap56 strains incubated with 50 nM rapamycin at different points using RNeasy® Mini Kit (Qiagen) according to the manufacturer’s direction and MIQE criteria (Bustin et al., 2009). Contaminating DNA was digested with 1 U DNAse RNAse-free (Promega) per μg RNA. 1 μg RNA was reverse transcribed using random primers (Invitrogen) and ImProm-II™ Reverse Transcription System (Promega) as default protocol. To access uap56 mRNA levels, we performed real-time PCR reactions in triplicate using SYBR green master mix (Applied Biosystems) on AB7500 (Applied Biosystems, Invitrogen). mRNA levels were normalized to reference tubulin mRNA levels, using primers described previously (Dalmasso et al., 2009). The relative expression levels were calculated based on the Livak method (Livak & Schmittgen, 2001) and percentage of uap56 mRNA levels was analyzed by media of total mRNA per time. Growth assays Plaque assays were performed as described previously (Roos DS, 1994). Monolayers of HFF grown in 6 well plates were infected with 200 tachyzoites per well. Parasites were incubated with 0.1–2 μM of Shld1 (overexpression assays) or 50 nM of rapamycin (inducible knockout assay). For ddCas9 assays, cells were incubated with 1 μM of Shld1, after 24 hours this was replaced with fresh DMEM without Shld1. For inducible knockout assays, the cells were incubated with 50 nM of rapamycin for 24 hours. After this time, the cells were washed and maintained with DMEM. After 108 hours of incubation at 37 °C, 5% CO2, cells were fixed for 10 minutes with 100% methanol at −20 °C, stained with Giemsa for 10 minutes and washed once with PBS. Images were taken using a Zeiss microscope (Axiovert 200M) with a 4× objective and plaque size was determined using Axiovision software (Zeiss). Analysis of mRNA processing by splicing The analysis of mRNA splicing was performed by analytical PCR for seven selected genes as previously described (Suvorova et al., 2013). Total RNA was purified using RNeasy® Mini Kit (Qiagen) from RH DiCre ΔKu80 and cKOuap56 strains after 24 and 48 hours of incubation with rapamycin. The RNA was reverse transcribed using random primers (Invitrogen #cat 48190-011) and the target sequences were amplified by PCR using specific primers that span an intron, as listed on Table S2. gDNA was used as reference to distinguish between properly spliced (S) and pre-spliced (PS) species. mRNAs from the following genes were analyzed: RNA polymerase II p8.2 subunit (TGME49_217560), RNA polymerase II p19 subunit (TGME49_271300), RNA polymerase II p23 subunit (TGME49_240590), imc1 (TGME49_231640), imc15 (TGME49_275670), imc5 (TGME49_224530), and transcription factor iid (TGME49_258680). Tubulin was used as loading control and the primers were described previously (Dalmasso et al., 2009). Immunoprecipitation We performed immunoprecipitation experiments using the dd-GFPTgRRM_1330 strain to determine whether TgUAP56 co-immunoprecipitates with dd-GFP-TgRRM_1330. Intracellular tachyzoites were incubated with 0.5 μM of Shld1 for 2 hours and then homogenized in a salt buffer [10 mM Trisodium Citrate, 20 mM HEPES, pH 7.4, 1 mM MgCl2, 10 μM CaCl2, 0,1% CHAPS, 0,5% Nonidet P-40 (NP-40), phosphate inhibitor cocktail (P-5726; Sigma, St. Louis, MO)]. The samples were centrifuged at 15,000 xg for 10 min and the supernatant was incubated with α-GFP antibody (Roche) with protein G-agarose beads (CAT) for 2 hours at 4°C. The beads were washed four times in the same buffer solution and antigens were eluted from the beads with 60 μl of 2× Laemmili SDS buffer. Samples were boiled for 5 min before SDS-PAGE separation. Supplementary Material Supp info We would like to thank Prof. Gary Ward (University of Vermont, US) and Dr. Marc Jan Gubbels (Boston College, US) for kindly providing IMC antibodies. We would also like to thank Dr. Sebastian Lourido (Whitehead Institute, US) for the Toxoplasma gondii Cas9 vector (Addgene ID 52694). This work was supported by the Programa Estratégico de Apoio à Pesquisa em Saúde (PAPES/FIOCRUZ, Brazil) [407775/2012-9 –CNPq]; Fundação Araucária [145/2015]; CAPES Foundation, Ministry of Education of Brazil (CAPES, Brazil) [PDSE Fellowship 9445-12-9]; National Counsel of Technological and Scientific Development (CNPq, Brazil) [PDE Fellowship 249741/2013-0]; This work was supported by an ERC-Starting grant (ERC-2012-StG309255-EndoTox) and the Wellcome Trust 087582/Z/08/Z Senior Fellowship for MM. The Wellcome Trust Centre for Molecular Parasitology is supported by core funding from the Wellcome Trust (085349)).This research was partially supported by the Intramural Research Program of the National Library of Medicine, National Institutes of Health. NMV’s postdoctoral fellowship is funded by a partnership between Conselho Nacional de Desenvolvimento Cientifico e Tecnologico and the National Institutes of Health, Figure 1 Localization and functional analysis of TgUAP56 A. Analysis of dd-GFP-TgUAP56 expression. i) In non-induced parasites (- Shld1), only the endogenous protein (TgUAP56) is detected by Western blot using polyclonal antibody against Tryp-Sub2 (Serpeloni et al., 2011a), renamed here as α-TgUAP56. Both TgUAP56 and dd-GFP-TgUAP56 are detected by this antibody after incubation with 1 μM of Shld1 for 6 hours. Aldolase: Loading control. ii) In the immunofluorescence assay dd-GFP-TgUAP56 is nuclear and colocalizes with endogenous TgUAP56 protein, in red. Nuclear and apicoplast DNA staining with DAPI: in blue. Scale bar: 5 μm. B. Analysis of mRNA distribution after dd-GFP-TgUAP56 overexpression with 1 μM Shld1. i) Western blot to analyze induction of dd-GFP-TgUAP56. The protein was detected with anti-GFP. Aldolase was used as loading control. dd-GFP-TgUAP56 was detectable after 6 hours of incubation with 1 μM of Shld1. The numbers above the Western blot indicate the relative expression levels of dd-GFP-TgUAP56 at each indicated time point, normalized to the loading control aldolase using ImageJ software with the densitometry plugin (Version 1.6, National Institutes of Health, Bethesda, MD). ii) To check mRNA distribution, poly(A)+ mRNAs were detected by fluorescent in situ hybridization (FISH) using oligodT-Alexa594 as probe: in Red. Nuclear and apicoplast DNA staining with DAPI: in blue. dd-GFP-TgUAP56: in green. In the right: quantification of signals of immunofluorescence and DAPI in selected parasites, in yellow box. Scale bar: 10 μm. C. Analysis of mRNA distribution after knockout of uap56 by gene-swap strategy based on Di-Cre system in T. gondii. i) Endogenous uap56 gene is replaced with mcherry by Dicre after incubation with 50 nM of rapamycin. ii) uap56 knockout was analyzed in cKOuap56 strain western blot after incubation with 50 nM of rapamycin at different times. -, not induced; +, induced. Aldolase: Loading control. iii) mRNA distribution was analyzed in cKOuap56 strain after incubation with rapamycin at different times by fluorescent in situ hybridization (FISH) using oligodT-Alexa488 as probe, in green. Nuclear and apicoplast DNA was staining with DAPI: in blue. In the right: quantification of signals of immunofluorescence and DAPI in selected parasites. Scale bar: 10 μm. D. PCR analysis of mRNA splicing for selected genes. Total RNA was purified from DiCre strain and cKOuap56 strain both incubated with 50 nM of rapamycin, as indicated. -, not induced; +, induced. The total RNA was reverse transcribed and PCR amplified using primers that span an intron. PCR of gDNA was included as a reference to distinguish between properly spliced (S) and pre-spliced (PS) forms of each gene. In the top: Tubulin, used as loading control (Dalmasso et al., 2009). In the bottom: PCR analysis of mRNA splicing for selected genes: RNA polymerase II p8.2 subunit (TGME49_217560), RNA polymerase II p19 subunit (TGME49_271300), RNA polymerase II p23 subunit (TGME49_240590), imc1 (TGME49_231640), imc15 (TGME49_275670), imc5 (TGME49_224530), and transcription factor iid (TGME49_258680). gDNA: genomic DNA from non-induced parasites of cKOuap56 strain. Expected sizes for pre-mRNA (Pre-spliced) and mRNA (spliced) are shown for each selected gene. Figure 2 Establishment of a conditional Cas9 (ddCas9) A. Western blot showing ddCas9-FLAG overexpression. Parasites were induced for indicated times with 1μM of Shld1 prior protein extraction. RH-Δhxgprt strain was used as control (first and second lanes - wild type-wt) Aldolase: Loading control. B. i) Nuclear localization of Cas9 in RHddCas9 parasites after addition of 1 μM Shld1 for 24h. Nuclear and apicoplast DNA staining with DAPI: in blue. Scale bar: 10μm. ii) Co-localization map, showing in grey the areas where there was co-localization with an M2 of 0.8. Graph represents the areas where there was a signal for green (green line) and areas where there was signal for blue (blue line). Y axis: intensity level; X axis: distance in microns. C. Plaque assays for parental RHddCas9 and RHddCas9-gap40 gRNA strains. Both parasite strains were grown on human foreskin fibroblasts, in the presence or absence of 1 μM of Shld1, as indicated, for 108 hours. -, not induced; +, induced. Scale bar: 500 μm. D. Immunofluorescence assay of parental RHddCas9 and RHddCas9-gap40 sgRNA strains in the presence or absence of Shld1. Collapsing vacuoles with a significant reduction of GAP40 signal were only observed in induced parasites that expressed the specific sgRNA against gap40. Scale bar: 8μm. E. mRNA distribution was analyzed in RHddCas9, RHddCas9-gap40 gRNA and RHddCas9-uap56 gRNA strains after incubation for 4 hours with 1 μM Shld1 and further incubation for 48 hours in media. The analysis was performed by fluorescent in situ hybridization (FISH) using oligodT-Alexa594 as probe, in red. Nuclear and apicoplast DNA was staining with DAPI: in blue. Scale bar: 5 μm. F. mRNA distribution and TgUAP56 protein presence analysis in RHddCas9-uap56 gRNA strain after incubation for 4 hours with 1 μM Shld1 and further incubation for 48 hours in media. i) The analysis was performed by immunofluorescence assay, α-TgUAP56 in green, combined with fluorescent in situ hybridization (FISH) using oligodT-Alexa594 as probe, in red. mRNA export blocking was observed in absence of TgUAP56 in induced parasites that expressed the specific sgRNA against uap56. Nuclear and apicoplast DNA was staining with DAPI: in blue. Scale bar: 5μm. ii) Western blot to analyse TgUAP56 protein levels after 1 μM Shld1 incubation for 24 and 48 hours. The protein was detected with anti-TgUAP56. Aldolase was used as loading control. ddCas9-FLAG was detected using antibody α-Flag. -, not induced; +, induced. Figure 3 ddCas9 genetic screen for potential candidates related to mRNA export in T. gondii mRNA distribution was analyzed in RHddCas9, RHddCas9-candidate strains after incubation for 4 hours with 1 μM Shld1 and then 48 hours with fresh media. The analyses were performed by fluorescent in situ hybridization (FISH) using oligodT-Alexa594 as probe, in red. Nuclear and apicoplast DNA was stained with DAPI: in blue. Scale bar: 5 μm. Figure 4 Subcellular localization and phenotypic analysis after overexpression of T. gondii candidate proteins in T. gondii A. Plaque assays for overexpression strains. Both parasite strains were grown on human foreskin fibroblasts, in the presence or absence of 1 μM of Shield, as indicated, for 108 hours. -, not induced; +, induced. Scale bar: 500 μm. B–F) mRNA distribution in different times of candidates overexpression by incubation with 1 μM of Shld1. poly(A)+ mRNAs were detected by fluorescent in situ hybridization (FISH) using oligodT-Alexa594 as probe: in Red. Nuclear and apicoplast DNA staining with DAPI: in blue. dd-GFP-candidates: in green. Scale bar: 5 μm. Figure 5 Localization and functional analysis of TgRRM_1330. A. Analysis of mRNA distribution during dd-GFP- TgRRM_1330 overexpression after incubation with 1 μM Shld1 at different times To check mRNA distribution, poly(A)+ mRNAs were detected by fluorescent in situ hybridization (FISH) using oligodT-Alexa555 as probe: in Red. Nuclear and apicoplast DNA staining with DAPI: in blue. dd-GFP- TgRRM_1330: in green. Scale bar: 5 μm. B. dd-GFP-TgRRM_1330 protein was detected with anti-GFP and the overexpression levels were quantified by comparison with aldolase levels. dd-GFP-TgRRM_1330 was stabilized after 6 hours of incubation with 1 μM of Shld1. The numbers above the Western blot means the percentage of overexpression at each indicated time point, related to result after 6 hours, proportional to loading control. C. dd-GFPTgRRM_1330 strain growth assay. Parasites were grown on human foreskin fibroblasts in the presence of different concentration of Shld1. After 108 hours of incubation, the cells were fixed and stained with Giemsa. D. Colocalization analysis between TgUAP56 and dd-GFP- TgRRM_1330. i) Super-resolution microscopy of four tachyzoites nuclei. There is a clear co-stain with anti-TgUAP56 (in red) and the GFP signal of dd-GFP-TgRRM_1330 (in green) within the nucleus, stained in blue with DAPI. The fluorescence signals were analyzed and plotted, showing a co-localization of the fluorescences in the same areas within the nuclei highlighted in the white box. ii) Immunoelectron micrograph of a tachyzoite nucleus. Arrows indicate labelling of anti-TgUAP56 protein, and arrowheads indicate anti-GFP protein. iii) dd-GFP-TgRRM_1330 immunoprecipitation. dd-GFP-TgRRM_1330 was stabilized for 2 hours with 0,5 μM of Shld1. The membrane was incubated with anti-TgUAP56. I: Input. W1: First wash. W4: Fourth and last wash. E: Eluted. Table 1 mRNA export candidates in T. gondii based on sequence and phylogenetic analysis. S. cerevisiae Metazoans ID NCBI - Description References Length (aa) PFAM-Domains ToxoDB - Description ID Length (aa) PFAM - Domains Yra1p Aly/REF GI: 6320589 Nuclear polyadenylated RNA-binding protein; required for export of poly(A)+ mRNA from the nucleus. Is deposited onto mRNAs through its interaction partner Sub2/UAP56 and couple mRNA export with 3′ end processing via its interactions with Mex67p Lou et al (2001); Zhou (2000), Masuda et al (2005); Meinel et al (2013); Ma et al. (2013); Johnson et al. (2011) 226 RRM_Aly_REF RNA recognition motif-containing protein TGME49_291330 (TgRRM_1330) 228 RRM_1 Npl3 PF10_0217 (serine/arginine-rich splicing factor 4 (SRSF4)) In yeast: RS containing shuttling RNA-binding-protein recruited to the mRNAs co-transcriptionally early by RNA polymerase II and is required for pre-mRNA splicing. Phosphorylation is essential for efficient competent mRNP export Tuteja and Mehta, (2010 538 RRM1_RRM1 Splicing factor SF2 (SF2) TGME49_319530 (TgSF2_9530) 512 RRM_1/RRM_1 Gbp2 PF10_0068 (RNA-binding protein, putative) In yeast: Poly(A+) RNA-binding protein recruited to the mRNAs co-transcriptionally via THO complex involved in mRNA surveillance and nuclear mRNA quality control. Tuteja and Mehta, (2010) 246 RRM1_RRM1 RNA recognition motif-containing protein TGME49_262620 (TgRRM_2620) 293 RRM_1/RRM_1 U2AF35 GI: 68800128 RNA recognition motif in U2 small nuclear ribonucleoprotein auxiliary factor U2AF 35 kDa subunit, that directly binds to TAP, and this interaction is conserved across metazoan species. Wu et al (1999); Zolotukhin et al., (2002) 240 aa zf-CCCH/RRM_5/zf-CCCH U2 snRNP auxiliary factor, putative TGME49_236910 (TgU2_6910) 254 zf-CCCH/zf-CCCH/RRM_5 CRM1 Xpo1 GI: 398366207 Major karyopherin/Exportin involved in export of proteins, snRNAs, rRNAs, viral RNAs and a subset of endogenous mRNAs Hammell et al. (2002), Cullen (2003), Koyama e Matsuura (2010); Sun et al. (2013), Wickramasinghe and Laskey, 2015 1084 CRM1 _C/Xpo1/IBN_N Exportin 1, putative TGME49_249530 (TgCRM1) 1125 CRM1_C/Xpo1/IBN_N Ran GTPase GSP1 Ran GI: 6323324 Ran GTPase; GTP binding protein involved in the maintenance of nuclear organization, RNA processing and transport Cullen (2003), Wickramasinghe and Laskey, 2015 219 Ras GTP-binding nuclear protein ran/tc4 TGME49_248340 (TgRan) 229 Ras Abbreviated Summary mRNA export is well characterised in mammals but poorly understood in single-cell parasites and other divergent organisms. We have utilized sequence searching, phylogenetic analyses and reverse genetic tools to investigate this pathway in Toxoplasma gondii, a parasite that causes Toxoplasmosis in humans and is related to the malaria causing parasite. This study has revealed essential factors and point towards that this parasite may have undiscovered unique components for mRNA export. Author Contributions Experimental design: MS, EJR, MM and ARA; Generation of transgenic strains: MS, EJR, CK and NA; Conditional Cas9 system establishment: EJR and MS; Image acquisition, analysis and interpretation: MS, EJR and LL; Sequence and phylogenetic analysis: NMV; Splicing analysis: MS and PM; Manuscript writing: MS, EJR, NMV, GP, MM and ARA. Conflict of interest statement: None declared. Andenmatten N Egarter S Jackson AJ Jullien N Herman JP Meissner M 2013 Conditional genome engineering in Toxoplasma gondii uncovers alternative invasion mechanisms Nat Methods 10 125 127 23263690 Andenmatten N ES Jackson AJ Jullien N Herman JP Meissner M 2013 Conditional genome engineering in Toxoplasma gondii uncovers alternative invasion mechanisms Nature Methods 10 Banaszynski LA Chen LC Maynard-Smith LA Ooi AG Wandless TJ 2006 A rapid, reversible, and tunable method to regulate protein function in living cells using synthetic small molecules Cell 126 995 1004 16959577 Bargieri DY Andenmatten N Lagal V Thiberge S Whitelaw JA Tardieux I Meissner M Menard R 2013 Apical membrane antigen 1 mediates apicomplexan parasite attachment but is dispensable for host cell invasion Nat Commun 4 2552 24108241 Breinich MS Ferguson DJ Foth BJ van Dooren GG Lebrun M Quon DV Striepen B Bradley PJ Frischknecht F Carruthers VB Meissner M 2009 A dynamin is required for the biogenesis of secretory organelles in Toxoplasma gondii Curr Biol 19 277 286 19217293 Buhlmann M Walrad P Rico E Ivens A Capewell P Naguleswaran A Roditi I Matthews KR 2015 NMD3 regulates both mRNA and rRNA nuclear export in African trypanosomes via an XPOI-linked pathway Nucleic acids research 43 4491 4504 25873624 Bustin SA Benes V Garson JA Hellemans J Huggett J Kubista M Mueller R Nolan T Pfaffl MW Shipley GL Vandesompele J Wittwer CT 2009 The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments Clin Chem 55 611 622 19246619 Cullen BR 2003 Nuclear RNA export J Cell Sci 116 587 597 12538759 Custodio N Carvalho C Condado I Antoniou M Blencowe BJ Carmo-Fonseca M 2004 In vivo recruitment of exon junction complex proteins to transcription sites in mammalian cell nuclei RNA 10 622 633 15037772 Dalmasso MC Onyango DO Naguleswaran A Sullivan WJ Jr Angel SO 2009 Toxoplasma H2A variants reveal novel insights into nucleosome composition and functions for this histone family J Mol Biol 392 33 47 19607843 Delaleau M Borden KL 2015 Multiple Export Mechanisms for mRNAs Cells 4 452 473 26343730 Dias AP Dufu K Lei H Reed R 2010 A role for TREX components in the release of spliced mRNA from nuclear speckle domains Nat Commun 1 97 20981025 Donald RG CD Ullman B Roos DS 1996 Insertional tagging, cloning, and expression of the Toxoplasma gondii hypoxanthine-xanthine-guanine phosphoribosyltransferase gene. Use as a selectable marker for stable transformation J Biol Chem 271 Donald RG Roos DS 1993 Stable molecular transformation of Toxoplasma gondii: a selectable dihydrofolate reductase-thymidylate synthase marker based on drug-resistance mutations in malaria Proc Natl Acad Sci U S A 90 11703 11707 8265612 Dostalova A Kaser S Cristodero M Schimanski B 2013 The nuclear mRNA export receptor Mex67-Mtr2 of Trypanosoma brucei contains a unique and essential zinc finger motif Mol Microbiol 88 728 739 23560737 Dufu K Livingstone MJ Seebacher J Gygi SP Wilson SA Reed R 2010 ATP is required for interactions between UAP56 and two conserved mRNA export proteins, Aly and CIP29, to assemble the TREX complex Genes Dev 24 2043 2053 20844015 Eddy SR 1998 Profile hidden Markov models Bioinformatics 14 755 763 9918945 Finn RD Bateman A Clements J Coggill P Eberhardt RY Eddy SR Heger A Hetherington K Holm L Mistry J Sonnhammer EL Tate J Punta M 2014 Pfam: the protein families database Nucleic acids research 42 D222 230 24288371 Fleckner J Zhang M Valcarcel J Green MR 1997 U2AF65 recruits a novel human DEAD box protein required for the U2 snRNP-branchpoint interaction Genes Dev 11 1864 1872 9242493 Gatfield D Le Hir H Schmitt C Braun IC Kocher T Wilm M Izaurralde E 2001 The DExH/D box protein HEL/UAP56 is essential for mRNA nuclear export in Drosophila Curr Biol 11 1716 1721 11696332 Geisinger JM Turan S Hernandez S Spector LP Calos MP 2016 In vivo blunt-end cloning through CRISPR/Cas9-facilitated non-homologous end-joining Nucleic acids research Gilbert W Guthrie C 2004 The Glc7p nuclear phosphatase promotes mRNA export by facilitating association of Mex67p with mRNA Mol Cell 13 201 212 14759366 Gilbert W Siebel CW Guthrie C 2001 Phosphorylation by Sky1p promotes Npl3p shuttling and mRNA dissociation RNA 7 302 313 11233987 Gwizdek C Iglesias N Rodriguez MS Ossareh-Nazari B Hobeika M Divita G Stutz F Dargemont C 2006 Ubiquitin-associated domain of Mex67 synchronizes recruitment of the mRNA export machinery with transcription Proc Natl Acad Sci U S A 103 16376 16381 17056718 Hackmann A Wu H Schneider UM Meyer K Jung K Krebber H 2014 Quality control of spliced mRNAs requires the shuttling SR proteins Gbp2 and Hrb1 Nat Commun 5 3123 24452287 Harding CR Egarter S Gow M Jimenez-Ruiz E Ferguson DJ Meissner M 2016 Gliding Associated Proteins Play Essential Roles during the Formation of the Inner Membrane Complex of Toxoplasma gondii PLoS Pathog 12 e1005403 26845335 Herm-Gotz A Agop-Nersesian C Munter S Grimley JS Wandless TJ Frischknecht F Meissner M 2007 Rapid control of protein level in the apicomplexan Toxoplasma gondii Nat Methods 4 1003 1005 17994029 Hettmann C Herm A Geiter A Frank B Schwarz E Soldati T Soldati D 2000 A dibasic motif in the tail of a class XIV apicomplexan myosin is an essential determinant of plasma membrane localization Mol Biol Cell 11 1385 1400 10749937 Hettmann C HA Geiter A Frank B Schwarz E Soldati T Soldati D 2000 A dibasic motif in the tail of a class XIV apicomplexan myosin is an essential determinant of plasma membrane localization Mol Biol Cell 11 Hurt E Luo MJ Rother S Reed R Strasser K 2004 Cotranscriptional recruitment of the serine-arginine-rich (SR)-like proteins Gbp2 and Hrb1 to nascent mRNA via the TREX complex Proc Natl Acad Sci U S A 101 1858 1862 14769921 Iglesias N Tutucci E Gwizdek C Vinciguerra P Von Dach E Corbett AH Dargemont C Stutz F 2010 Ubiquitin-mediated mRNP dynamics and surveillance prior to budding yeast mRNA export Genes Dev 24 1927 1938 20810649 Jensen TH Boulay J Rosbash M Libri D 2001 The DECD box putative ATPase Sub2p is an early mRNA export factor Curr Biol 11 1711 1715 11696331 Katahira J 2015 Nuclear export of messenger RNA Genes (Basel) 6 163 184 25836925 Katoh K Standley DM 2013 MAFFT multiple sequence alignment software version 7: improvements in performance and usability Mol Biol Evol 30 772 780 23329690 Kohler A Hurt E 2007 Exporting RNA from the nucleus to the cytoplasm Nat Rev Mol Cell Biol 8 761 773 17786152 Kramer S Kimblin NC Carrington M 2010 Genome-wide in silico screen for CCCH-type zinc finger proteins of Trypanosoma brucei, Trypanosoma cruzi and Leishmania major BMC Genomics 11 283 20444260 Krebber H Taura T Lee MS Silver PA 1999 Uncoupling of the hnRNP Npl3p from mRNAs during the stress-induced block in mRNA export Genes Dev 13 1994 2004 10444597 Lee MS Henry M Silver PA 1996 A protein that shuttles between the nucleus and the cytoplasm is an important mediator of RNA export Genes Dev 10 1233 1246 8675010 Libri D Graziani N Saguez C Boulay J 2001 Multiple roles for the yeast SUB2/yUAP56 gene in splicing Genes Dev 15 36 41 11156603 Linder P Stutz F 2001 mRNA export: travelling with DEAD box proteins Curr Biol 11 R961 963 11728322 Lirussi D Matrajt M 2011 RNA granules present only in extracellular toxoplasma gondii increase parasite viability Int J Biol Sci 7 960 967 21850205 Livak KJ Schmittgen TD 2001 Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method Methods 25 402 408 11846609 Luo ML Zhou Z Magni K Christoforides C Rappsilber J Mann M Reed R 2001 Pre-mRNA splicing and mRNA export linked by direct interactions between UAP56 and Aly Nature 413 644 647 11675789 MacMorris M Brocker C Blumenthal T 2003 UAP56 levels affect viability and mRNA export in Caenorhabditis elegans RNA 9 847 857 12810918 Masuda S Das R Cheng H Hurt E Dorman N Reed R 2005 Recruitment of the human TREX complex to mRNA during splicing Genes Dev 19 1512 1517 15998806 Muller-McNicoll M Neugebauer KM 2013 How cells get the message: dynamic assembly and function of mRNA-protein complexes Nat Rev Genet 14 275 287 23478349 Neumann N Lundin D Poole AM 2010 Comparative genomic evidence for a complete nuclear pore complex in the last eukaryotic common ancestor PloS one 5 e13241 20949036 Nino CA Herissant L Babour A Dargemont C 2013 mRNA nuclear export in yeast Chem Rev 113 8523 8545 23731471 Obado SO Brillantes M Uryu K Zhang W Ketaren NE Chait BT Field MC Rout MP 2016 Interactome Mapping Reveals the Evolutionary History of the Nuclear Pore Complex PLoS biology 14 e1002365 26891179 Pieperhoff MS Pall GS Jimenez-Ruiz E Das S Melatti C Gow M Wong EH Heng J Muller S Blackman MJ Meissner M 2015 Conditional U1 Gene Silencing in Toxoplasma gondii PloS one 10 e0130356 26090798 Price MN Dehal PS Arkin AP 2010 FastTree 2--approximately maximum-likelihood trees for large alignments PloS one 5 e9490 20224823 Reed R Hurt E 2002 A conserved mRNA export machinery coupled to pre-mRNA splicing Cell 108 523 531 11909523 Rice P Longden I Bleasby A 2000 EMBOSS: the European Molecular Biology Open Software Suite Trends Genet 16 276 277 10827456 Rodriguez-Navarro S Fischer T Luo MJ Antunez O Brettschneider S Lechner J Perez-Ortin JE Reed R Hurt E 2004 Sus1, a functional component of the SAGA histone acetylase complex and the nuclear pore-associated mRNA export machinery Cell 116 75 86 14718168 Roos DS DR Morrissette NS Moulton AL 1994 Molecular tools for genetic dissection of the protozoan parasite Toxoplasma gondii Methods Cell Biol 45 Sahni A Wang N Alexis JD 2010 UAP56 is an important regulator of protein synthesis and growth in cardiomyocytes Biochem Biophys Res Commun 393 106 110 20116367 Schindelin J Arganda-Carreras I Frise E Kaynig V Longair M Pietzsch T Preibisch S Rueden C Saalfeld S Schmid B Tinevez JY White DJ Hartenstein V Eliceiri K Tomancak P Cardona A 2012 Fiji: an open-source platform for biological-image analysis Nat Methods 9 676 682 22743772 Schwede A Manful T Jha BA Helbig C Bercovich N Stewart M Clayton C 2009 The role of deadenylation in the degradation of unstable mRNAs in trypanosomes Nucleic acids research 37 5511 5528 19596809 Serpeloni M Moraes CB Muniz JR Motta MC Ramos AS Kessler RL Inoue AH daRocha WD Yamada-Ogatta SF Fragoso SP Goldenberg S Freitas LH Junior Avila AR 2011a An essential nuclear protein in trypanosomes is a component of mRNA transcription/export pathway PloS one 6 e20730 21687672 Serpeloni M Vidal NM Goldenberg S Avila AR Hoffmann FG 2011b Comparative genomics of proteins involved in RNA nucleocytoplasmic export BMC Evol Biol 11 7 21223572 Shankar J Pradhan A Tuteja R 2008 Isolation and characterization of Plasmodium falciparum UAP56 homolog: evidence for the coupling of RNA binding and splicing activity by site-directed mutations Arch Biochem Biophys 478 143 153 18722339 Shen B Brown KM Lee TD Sibley LD 2014 Efficient gene disruption in diverse strains of Toxoplasma gondii using CRISPR/CAS9 MBio 5 e01114 01114 24825012 Sidik SM Hackett CG Tran F Westwood NJ Lourido S 2014 Efficient Genome Engineering of Toxoplasma gondii Using CRISPR/Cas9 PLoS One 9 e100450 24971596 Starnes GL Jewett TJ Carruthers VB Sibley LD 2006 Two separate, conserved acidic amino acid domains within the Toxoplasma gondii MIC2 cytoplasmic tail are required for parasite survival J Biol Chem 281 30745 30754 16923803 Steckelberg AL Gehring NH 2014 Studying the composition of mRNPs in vitro using splicing-competent cell extracts Methods 65 342 349 24021717 Strasser K Hurt E 2000 Yra1p, a conserved nuclear RNA-binding protein, interacts directly with Mex67p and is required for mRNA export EMBO J 19 410 420 10722314 Strasser K Hurt E 2001 Splicing factor Sub2p is required for nuclear mRNA export through its interaction with Yra1p Nature 413 648 652 11675790 Strasser K Masuda S Mason P Pfannstiel J Oppizzi M Rodriguez-Navarro S Rondon AG Aguilera A Struhl K Reed R Hurt E 2002 TREX is a conserved complex coupling transcription with messenger RNA export Nature 417 304 308 11979277 Suvorova ES Croken M Kratzer S Ting LM de Felipe MC Balu B Markillie ML Weiss LM Kim K White MW 2013 Discovery of a splicing regulator required for cell cycle progression PLoS Genet 9 e1003305 23437009 Thakurta AG Selvanathan SP Patterson AD Gopal G Dhar R 2007 The nuclear export signal of splicing factor Uap56p interacts with nuclear pore-associated protein Rae1p for mRNA export in Schizosaccharomyces pombe J Biol Chem 282 17507 17516 17449473 Wickramasinghe VO Laskey RA 2015 Control of mammalian gene expression by selective mRNA export Nat Rev Mol Cell Biol 16 431 442 26081607 Windgassen M Krebber H 2003 Identification of Gbp2 as a novel poly(A)+ RNA-binding protein involved in the cytoplasmic delivery of messenger RNAs in yeast EMBO Rep 4 278 283 12634846 Wolyniak MJ Cole CN 2008 Harnessing genomics to explore the processes and evolution of mRNA export RNA Biol 5 68 72 18483471 Wright AV Sternberg SH Taylor DW Staahl BT Bardales JA Kornfeld JE Doudna JA 2015 Rational design of a split-Cas9 enzyme complex Proceedings of the National Academy of Sciences of the United States of America 112 2984 2989 25713377 Zetsche B Volz SE Zhang F 2015 A split-Cas9 architecture for inducible genome editing and transcription modulation Nature biotechnology 33 139 142 Zhu Z Gonzalez F Huangfu D 2014 The iCRISPR platform for rapid genome editing in human pluripotent stem cells Methods in enzymology 546 215 250 25398343 Zolotukhin AS Tan W Bear J Smulevitch S Felber BK 2002 U2AF participates in the binding of TAP (NXF1) to mRNA J Biol Chem 277 3935 3942 11724776

PMC005xxxxxx/PMC5118109.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9304420 2335 Nucl Med Biol Nucl. Med. Biol. Nuclear medicine and biology 0969-8051 1872-9614 27694058 5118109 10.1016/j.nucmedbio.2016.08.017 NIHMS820311 Article Monooxorhenium(V) complexes with 222-N2S2 MAMA ligands for bifunctional chelator agents: Syntheses and preliminary in vivo evaluation Demoin Dustin Wayne ad Dame Ashley N. ad Minard William D. a Gallazzi Fabio b Seickman Gary L. d Rold Tammy L. cd Bernskoetter Nicole d Fassbender Michael E. e Hoffman Timothy J. acd Deakyne Carol A. a Jurisson Silvia S. a a Department of Chemistry, University of Missouri, Columbia, MO 65211, USA b Department of Structural Biology Core, University of Missouri, Columbia, MO 65211, USA c Department of Medicine, University of Missouri, Columbia, MO 65211, USA d Research Division, Harry S. Truman Memorial Veteran’s Hospital, Columbia, MO 65201, USA e Chemistry Division, Los Alamos National Laboratory, PO Box 1663, Los Alamos, NM 87545 USA Corresponding author at: Department of Chemistry, 601 South College Avenue, University of Missouri, Columbia, MO 65211; jurissons@missouri.edu; 573-882-2107 (phone); 573-882-2754 (fax) 5 10 2016 31 8 2016 12 2016 01 12 2017 43 12 802811 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Introduction Targeted radiotherapy using the bifunctional chelate approach with 186/188Re(V) is challenging because of the susceptibility of monooxorhenium(V)-based complexes to oxidize in vivo at high dilution. A monoamine-monoamide dithiol (MAMA)-based bifunctional chelating agent was evaluated with both rhenium and technetium to determine its utility for in vivo applications. Methods A 222-MAMA chelator, 222-MAMA(N-6-Ahx-OEt) bifunctional chelator, and 222- MAMA(N-6-Ahx-BBN(7-14)NH2) were synthesized, complexed with rhenium, radiolabeled with 99mTc and 186Re (carrier added and no carrier added), and evaluated in initial biological distribution studies. Results An IC50 value of 2.0 ± 0.7 nM for natReO-222-MAMA(N-6-Ahx-BBN(7-14)NH2) compared to [125I]-Tyr4-BBN(NH2) was determined through competitive cell binding assays with PC-3 tumor cells. In vivo evaluation of the no-carrier added 99mTc-222-N2S2(N-6-Ahx-BBN(7-14)NH2) complex showed little gastric uptake and blockable pancreatic uptake in normal mice. Conclusions The 186ReO-222-N2S2(N-6-Ahx-BBN(7-14)NH2) complex showed stability in biological media, which indicates that the 222-N2S2 chelator is appropriate for chelating 186/188Re in radiopharmaceuticals involving peptides. Additionally, the in vitro cell studies showed that the ReO-222-N2S2(N-6-Ahx-BBN(7-14)NH2) complex (macroscopically) bound to PC3-tumor cell surface receptors with high affinity. The 99mTc analog was stable in vivo and exhibited pancreatic uptake in mice that was blockable, indicating BB2r targeting. Rhenium-186 rhenium(V) MAMA ligands quantum chemical studies bombesin radiotherapy 1. Introduction Two medically useful rhenium isotopes (186Re and 188Re) are beta emitters with different half-lives, nuclear production routes [1–3], particle energies, and thus tissue ranges (Table 1). The nuclear properties [4] of these two Re radioisotopes, both available at no carrier added (nca) concentrations, makes them suitable for various radiotherapeutic applications (i.e., 186Re for antibody labeling and smaller tumors and 188Re for peptide labeling and larger tumors) for personalized tumor treatment. Rhenium is the third row transition-metal congener of technetium, which is widely used in 99mTc-based diagnostic agents for SPECT (single-photon emission computed tomography) imaging [5–7]. A bifunctional chelating agent conjugated to a targeting vector for in vivo localization that delivers Tc and Re to the tumor site could provide complementary diagnostic and therapeutic agents. A number of NxS4-x ligand systems have been reported for Re(V) [5,6,8,9], with the N2S2- based monoamine-monoamide dithiol (MAMA) chelators of particular interest since the amine allows for conjugation of a single targeting vector. Kung et al. compared the 222-MAMA ligands for chelating oxotechnetium(V) with the bis(aminoethanethiol) (BAT) ligands and showed that the 222-MAMA chelators were more stable over time [10]. The MAMA-based complexes were more hydrophilic than their BAT analogues based on shorter HPLC retention times and lower rat brain uptake [10]. Subsequently, N-alkylated 222-MAMA derivatives were complexed with monooxotechnetium(V) [and to a lesser extent with monooxorhenium(V)] and shown to be stable under biological conditions [11–18]. The 222-MAMA-based bifunctional chelating agents were evaluated for forming stable 186/188Re complexes for targeted peptide-based radiotherapy agents, using the seventh through fourteenth amino acids (BBN(7-14)NH2) of the bombesin peptide sequence as a potential bombesin receptor (BB2r) targeting vector [19]. The BBN(7-14)NH2 sequence coupled to a variety of bifunctional chelating agents has been shown to specifically target the BB2r both in vitro and in vivo [6,20–26]. The synthesis, characterization, and radiolabeling of a new 222-MAMA derivatized bifunctional chelate for complexing 186/188/natRe(V) are reported with initial stability studies and in vivo biodistributions. 2. Methods 2.1. General methods All chemicals, unless otherwise indicated, were commercially available and used without further purification. Technetium-99m pertechnetate was available from saline elution of a 99Mo/99mTc generator (Mallinckrodt Medical, St. Louis, MO). Rhenium-186 was obtained using acid dissolved enriched Al(185ReO4)3 targets (94.6%; Isotech) from 185Re(n,γ)186Re production reactions at the University of Missouri Research Reactor (MURR, Columbia, MO) with a specific activity of 3.6–6.4 GBq/mg (0.098–0.172 Ci/mg). No carrier added 186Re was produced at the Los Alamos National Laboratory Isotope Production Facility (LANL-IPF) using an encapsulated target of enriched 186WO3 Powder (99.9%; Isoflex USA) in a 40 MeV incident proton beam. The 186Re was recovered as previously reported [27,28] with a slight modification: the anion exchange column was eluted with 8 M HNO3 (2 x 50 mL) for ReO4− desorption. The nitric acid was removed by evaporation and the residue reconstituted in 0.1 M HCl. The specific activity was determined to be 790 ± 90 GBq/mg (21 ± 2 Ci/mg) [27,28]. PC-3 tumor cells were obtained from ATTC (Manassas, VA) and cultured by the MU Cell and Immunobiology Core (CIC). NMR studies were performed with a Bruker DRX 300 or 500 MHz Spectrometer (as noted). ESI-MS was performed using a Finnigan TSQ7000 triple-quadrupole mass spectrometer. HPLC analysis was performed with a Shimadzu HPLC system outfitted with both UV-vis (254 and 358 nm) and radioisotope detectors. HPLC columns included a Betabasic-18 column (Thermo Scientific, Waltham, MA, 150 mm × 4.6 mm, 5 μm), a Jupiter C-18 column (Phenomenex, Torrance, CA, 250 mm × 4.60 mm, 5 μm, 300 A pore size), and a Nova-Pac semi-prep C-18 column (Waters, Milford, MA, 300 mm × 19 mm, 6 μm, 60 Å). All HPLC mobile phases used acetonitrile (AcN) in water with 0.1% trifluoroacetic acid (TFA); the specific conditions (i.e., column, gradient, flow rate) are indicated for purifications and analyses. Sep-Pak C18 Plus Light cartridges were purchased from Waters (Milford, MA). Radio-TLC strips (Saturation pads, Analtech, Newark, DE) were counted using a Bioscan 200 Imaging Scanner (gas ionization detector), or by cutting the TLC strips and counting them in a NaI(Tl) well detector (Harshaw Chemical Company, Serial Number EX-53, with Ortec electronics and a Canberra high-voltage supply) or a Tri-Carb 2900TR Liquid Scintillation Analyzer (PerkinElmer, Waltham, MA). A Beckman Coulter HPLC system in series with an ion trap mass analyzer (a LCQ FLEET instrument, Thermo Fisher Scientific, with positive ion ionization) was used to obtain LC/ESI-MS results (Betabasic-18, 10–50% AcN over 30 min, 1 mL/min). MS analysis was performed using the XCalibur software (Thermo Fisher Scientific). A Thermo Nicolet AEM FT-IR instrument was used for IR analysis (KBr pellets). UV-Vis analysis was performed with an OceanOptics USB2000 USB-ISS-UV/VIS spectrometer. All elemental analyses were performed by Atlantic Microlabs, Inc (Norcross, GA). 2.2. Synthesis of compounds and complexes Compounds 1 and 2 (Fig. 1) were synthesized with trityl protecting groups following a preparation similar to Ono et al. and verified by NMR [17]. Ethyl 6-((2-oxo-2-((2-(tritylthio)ethyl)amino)ethyl)(2-(tritylthio)ethyl)amino)hexanoate (3) [trityl protected 222-MAMA(N-6-Ahx-OEt)] Method 1 In oven-dried glassware under an Ar (g) atmosphere, 2 (1.15 g, 1.7 mmol) was added in 2 mL of anhydrous dimethylformamide (DMF). Anhydrous DMF (13 mL), ethyl-6-bromohexanoate (0.4 mL, 2.2 mmol), and diisopropylethylamine (0.33 mL, 1.9 mmol) were added in series. The reaction mixture was stirred under Ar (g) at 90 °C for 3 h. The product mixture was dried in vacuo, dissolved in ethylacetate (EtOAc; 10 mL) and extracted with a saturated NaCl solution (10 mL). The organic layer was dried over anhydrous Na2SO4, gravity filtered, rotary evaporated to an oil and placed under vacuum overnight. The product was purified on a silica gel column (1 in. × 6 in.) conditioned with 15% EtOAc in hexanes and eluted with 15% EtOAc in hexanes (100 mL), 30 % EtOAc in hexanes (100 mL), and 50% EtOAc in hexanes (300 mL), sequentially. The product, collected in two brown-colored fractions (eluted during the 30% EtOAc fraction and later in the 50% EtOAc fraction), were individually collected and dried to yield oils, which contained 3 and residual DMF. The two product fractions were combined and subsequently heated to 90 °C under vacuum for 24 h to obtain 0.367 g (0.4 mmol, 26.3% yield) of 3. 1H-NMR (500 MHz, CDCl3, δ (ppm), letters correspond to Fig. 1): 1.16–1.36 (m, 7H, n, h, i), 1.53 (quintet, J = 7.6, 2H, j), 2.18–2.28 (m, 6H, f, g, k), 2.33–2.41 (m, 4H, a, e), 2.83 (s, 2H, d), 2.96–3.05 (m, 2H, b), 4.07–4.15 (m, 2H, m), 7.13–7.32 (m, 18H, q, s), 7.34–7.50 (m, 12H, r). 13C-NMR (500 MHz, CDCl3, δ (ppm), letters correspond to Fig. 1): 14.3 (n), 24.7 (j), 26.8 (h & i), 29.9 (f), 32.0 (a), 34.2 (k), 37.9 (b), 53.8 (g), 54.6 (e), 58.2 (d), 60.2 (m), 66.7 (o), 126.7 (s), 127.9 (q), 129.5 (r), 144.7 (p), 171.2 (c), 173.5 (l). ESI-MS (m/z): 821.10 (calc. 821.38 [C52H57N2O3S2+][M+H+]). Anal. Calcd (Found) for C52H56N2O3S2·2HCl: C, 69.86 (69.88); H, 6.54 (6.84); N, 3.13 (3.22); S, 7.17 (7.49). Method 2 To a scintillation vial equipped with a stir bar, 2 (1.01 g, 1.49 mmol) was dissolved in dry AcN (2 mL; dried over molecular sieves). Diisopropylethylamine (0.25 g, 1.93 mmol) in AcN (3 mL) was added, followed by slow addition of ethyl-6-bromohexanoate (0.43 g, 4.67 mmol) in 3 mL of AcN. The reaction mixture was placed in an oil bath and refluxed at 85 °C with under N2 gas, and allowed to react for 24 h. The reaction vessel was brought to room temperature before removal of solvent in vacuo. EtOAc (5 mL) was added to the residue, and undissolved particulates were removed by filtration. The volume was reduced by half and 0.5 mL of hexanes was added. The crude product was purified on a silica gel column (10 g, 1.3 cm diameter) that had been pre-conditioned with 15% EtOAc in hexanes. The column was eluted with 30% EtOAc in hexanes (100 mL) followed by 50% EtOAc in hexanes (200 ml). Fractions (5 mL) were collected and analyzed by TLC analysis (silica gel; 50% EtOAc in hexanes. Fractions showing a single species with Rf = 0.4 were combined, and dried under vacuum overnight to obtain 1.08 g (1.32 mmol, 88% yield) of 3. 6-((2-oxo-2-((2-(tritylthio)ethyl)amino)ethyl)(2-(tritylthio)ethyl)amino)hexanoic acid (4) [trityl protected 222-MAMA(N-6-Ahx-OH)] 3 (0.14 g, 0.17 mmol) was dissolved in dry tetrahydrofuran (THF; 3 mL) and transferred to the reaction flask. Ethanol (EtOH; 3 mL) and aqueous 3 M NaOH (2 mL) were added. The mixture was refluxed for 9 h. The solvent was removed by rotary evaporation and the product left under vacuum overnight. 4 was dissolved in CHCl3 and loaded onto a silica gel column (1 in. × 6 in.), eluted with CHCl3 (200 mL) and, subsequently, 20% methanol (MeOH) in CHCl3 (100 mL). The brown-colored fraction (obtained from the 20% MeOH eluent) contained 0.10 g (0.13 mmol, 76.5% yield) of 4. 1H-NMR (500 MHz, CDCl3, δ (ppm), letters correspond to Fig. 1): 1.18–1.27 (m, 2H, i), 1.32 (tt, J = 7.4, 7.4 Hz, 2H, h), 1.54 (q, J = 7.5 Hz, 2H, j), 2.20–2.29 (m, 6H, f, g, k), 2.32–2.42 (m, 4H, e, a), 2.83 (s, 2H, d), 3.00 (quartet, J = 6.3 Hz, 2H, b), 7.14–7.30 (m, 18H, o, q), 7.35–7.42 (m, 12H, p), 7.44 (t, J = 5.8 Hz, 1H, r). 13C-NMR (500 MHz, CDCl3, δ (ppm), letters correspond to Fig. 1): 24.5 (j), 26.7 (i), 26.7(h), 30.0 (f), 32.0 (a), 33.6 (k), 37.9 (b), 53.8 (e), 54.5 (g), 58.2 (d), 66.7 (m), 126.7 (q), 127.9 (o), 129.5 (p), 144.7 (n), 171.3 (c), 177.5 (l). LC-ESI-MS (m/z, 10–50 % AcN over 30 min, 1 mL/min, Rt = 34.4 min): 793.8 (calc. 793.35 [C50H53N2O3S2+]); ESI-MS (m/z): 791.38 (calc. 791.33 [C50H51N2O3S2-1][M-H-]). Anal. Calcd (Found) for C50H52N2O3S2·HCl·H2O: C, 70.85 (70.86); H, 6.54 (6.30); N, 3.31 (3.34); S, 7.57 (7.53). N1-((5S,8S)-11-((1H-imidazol-4-yl)methyl)-5-carbamoyl-24-(1H-indol-3-yl)-8-isobutyl-20-isopropyl-17-methyl-7,10,13,16,19,22-hexaoxo-2-thia-6,9,12,15,18,21-hexaazatetracosan-23-yl)-2-(6-((2-mercaptoethyl)(2-((2-mercaptoethyl)amino)-2-oxoethyl)amino)hexanamido)pentanediamide (5) [222-MAMA(N-6-Ahx-BBN(7-14)NH2)] The peptide was synthesized in a model 396 multiple peptide synthesizer (AAPPTEC, Louisville, KY) on a Sieber resin using standard, solid-phase, Fmoc protection strategy for linear elongation. Fmoc protected amino acids with protected side-chains (where appropriate) were used to synthesize the amino acid sequence -Gln-Trp-Ala-Val-Gly-His-Leu-Met-NH2 [BBN(7-14)]. Coupling was achieved by in situ activation with HBTU and DIPEA at every elongation step. 4 was added as the final amino acid using these same conditions. The product was cleaved and side chains deprotected with 85% trifluoroacetic acid (TFA) and non-reducing scavengers (phenol/water/triisopropylsilane, 5% each). The synthesis yielded 300 mg of impure 5 (85% pure), which was HPLC purified (20-40% AcN (0.1% TFA) in H2O (0.1% TFA) over 40 min) resulting in 80 mg of 5 in >98% purity. LC/ESI-MS (m/z, 10–50 % AcN over 30 min, 1 mL/min, Rt = 15.6 min): 1230.3 (calc. 1230.59 [C55H88N15O11S3+][M+H+]), 615.6 (calc. 615.80 [C55H89N15O11S32+][M+2H2+]). N-(2-mercaptoethyl)-2-((2-mercaptoethyl)amino)acetamide (6) [deprotected 222-MAMA] 2 (0.1 g, 0.1 mmol) was deprotected with 0.4 mL of a TFA solution (95% TFA:5% triethylsilane [TES]) at room temperature for 2 h with constant shaking. The solvent was evaporated under N2 (g), and the resulting solids were extracted with hexanes (6 x 10 mL) to yield an oil (0.03 g, 0.1 mmol). 1H-NMR (500 MHz, d6-DMSO, δ(ppm), letters correspond to Fig. 1): 2.56 (dt, J = 7.2, 7.1 Hz, 2H, a), 2.73 (dt, J = 6.5, 6.5 Hz, 2H, f), 2.78–2.88 (m, 1H, g), 3.11 (t, J = 7.5 Hz, 2H, e), 3.30 (dt, J = 6.3, 6.4 Hz, 2H, b), 3.76 (s, 2H, d), 8.61 (s, 1H, h). 13C-NMR (500 MHz, d6-DMSO, δ (ppm), letters correspond to Fig. 1): 19.5 (f), 23.3 (a), 42.1 (b), 47.4 (d), 49.4 (e), 165.1 (c). ReO(222-MAMA(N-6-Ahx-OEt)) [ReO-3] Method 1 3 (0.9277 g, 1.1 mmol) was deprotected in a TFA solution (95% TFA:5% TES), washed with hexanes (6 × 10 mL), dissolved in MeOH (18 mL), and pH adjusted to 4.5–5 with 1 M NH4OAc in MeOH using pH strips. ReOCl3(PPh3)2 (0.98 g, 1.2 mmol) was added as a solid and the reaction mixture refluxed at 80 °C for 3.5 h. The product mixture was rotary evaporated to dryness, and dissolved in EtOAc (20 mL) and water (20 mL). The organic layer was extracted with saturated NaCl solution (20 mL), filtered to remove green solids, dried over anhydrous sodium sulfate, and taken to dryness in vacuo. The residue dissolved in CH2Cl2 was purified on a silica gel column (1 in. × 6 in.), eluted with 100% CH2Cl2 (250 mL), 1% MeOH in CH2Cl2 (200 mL), 2% MeOH in CH2Cl2 (300 mL), 4% MeOH in CH2Cl2 (200 mL), 7% MeOH in CH2Cl2 (200 mL), 10% MeOH in CH2Cl2 (200 mL), and 15% MeOH in CH2Cl2 (100 mL), sequentially. Colored bands were collected as separate fractions (i.e., a brown band, a red band, and a second brown band). Each fraction was rotary evaporated to dryness. The first (0.3735 g impure product) and second (0.0680 g impure product) brown bands were each HPLC purified (Betabasic-18, 10–40% AcN over 10 min, followed by 40-43.25% AcN over 13 min, and followed by 43.25–50% AcN over 2 min, 1 mL/min) in portions to collect the HPLC peaks at 18.5 min and 19.2 min together. The collected peaks were rotary evaporated to dryness, dissolved in CHCl3, dried over anhydrous Na2SO4, and rotary evaporated to dryness. Approximately 25% yield (0.15 g, 0.28 mmol) of solid ReO-3 was obtained. IR (KBr Pellet): 1653.0 cm−1 (C=O); 969.90 cm−1 (Re=O). 1H-NMR (500 MHz, CDCl3, δ (ppm), letters correspond to Fig. 1): 1.27 (t, J = 7.0 Hz, 3H, n), 1.38-1.48 (quint, J = 7.5 Hz, 2H, i), 1.64 (dd, J = 4.5, 12.8 Hz, 1H, e), 1.72 (quint, J = 7.6 Hz, 2H, j), 1.77–1.92 (m, 2H, h), 2.36 (t, J = 7.3 Hz, 2H, k), 2.90 (dd, J = 4.5, 13.5 Hz, 1H, f), 3.10–3.32 (m, 3H, a, b, e), 3.38 (dt, J = 3.2, 13.3 Hz, 1H, f), 3.53 (td, J = 3.7, 12.9 Hz, 1H, g), 3.95 (dt, J = 5.9, 12.8 Hz, 1H, g), 4.10 (dt, J = 3.3, 13.5 Hz, a), 4.15 (q, J = 7.2 Hz, 2H, m), 4.19 (d, J = 16.5 Hz, 1H, d), 4.54 (dd, J = 5.7, 11.8 Hz, 1H, b), 4.70 (d, J = 16.5 Hz, 1H, d). 13C-NMR (500 MHz, CDCl3, δ (ppm), letters correspond to Fig. 1): 14.2 (n), 23.5 (h), 24.4 (j), 26.3 (i), 33.8 (k), 39.1 (f), 47.7 (a), 59.6 (b), 60.7 (m), 63.6 (g), 64.2 (e), 67.0 (d), 173.5 (l), 188.1 (c). UV-Vis (ethyl acetate): 358 nm. LC/ESI-MS (m/z; Betabasic-18, 10-50% AcN over 30 min, 1 mL/min; Rt = 30.1 min): 537.1 (calc. 537.09, [C14H26N2O4ReS2+][M+H+]). Anal. Calcd (Found) for C14H25N2O4ReS2·0.36CH2Cl2: C, 30.46 (30.82); H, 4.58 (4.42); N, 4.95 (4.88); S, 11.32 (10.98). Method 2 Ammonium perrhenate (0.044 g, 0.16 mmol) and sodium citrate dihydrate (0.060 g, 0.20 mmol) were dissolved in 5 mL of DI water. Stannous tartrate (0.10 g, 0.37 mmol) was added and the reaction mixture was stirred at room temperature for 10 min to yield a blue solution. In a separate vial, 3 (0.050 g, 0.061 mmol) was deprotected in a TFA solution (95% TFA:5% TES), washed with hexanes (6 × 10 mL), and rotary evaporated to dryness to yield 6. The rhenium-citrate solution was added to the vial containing 6 and the reaction was refluxed at 85 °C for 4 h. Sodium phosphate (pH 5) was added and the reaction was refluxed for another 4 hr. The reaction mixture was cooled to room temperature, extracted with EtOAc (2 × 5 mL), the organic extracts were combined, dried over anhydrous Na2SO4 and rotary evaporated to dryness. Approximately 60% yield (0.2 g, 0.37 mmol) of solid ReO-3 was obtained. HPLC: (Betabasic-18, 10–50 % AcN over 30 min, 1 mL/min) Rt = 30.2 min. LC/ESI-MS (m/z; Betabasic-18, 10–50% AcN over 30 min, 1 mL/min; Rt = 29.4 min): 537.3 (calc. 537.09, [C14H26N2O4ReS2+][M+H+]). ReO(222-MAMA(N-6-Ahx-BBN(7-14)NH2)) [ReO-5] Method 1 5 (14.1 mg, 11.4 μmol) and ReOCl3(PPh3)2 (12.5 mg, 15.0 μmol) were combined in a 1.5 mL snap cap centrifuge tube, followed by MeOH (400 μL) and 1 M NH4OAc in MeOH (100 μL). The centrifuge tube was capped and the reaction mixture was heated at 80 °C for 2 h with stirring. The product mixture was centrifuged and the supernatant removed (~ 300 μL). A portion of the sample (50 μL) was removed for LC-ESI-MS. The solids were washed with 200 μL of MeOH, centrifuged, and the supernatant removed four times. The combined supernatants were rotary evaporated to yield 20.7 mg of impure product. The residue was dissolved in 1 mL of MeOH and HPLC purified (Nova-Pac semi-prep, 10–50% AcN over 60 min, 10 mL/min). The peaks from 47 to 50 min (38.5-40% AcN in water with 0.1% TFA) were collected. The product was further HPLC purified (Betabasic-18, 37–50 % AcN over 20 min, 1 mL/min; collected the 5.3 min peak) and the solvents removed in vacuo to obtain 2.0 mg of pure ReO-5 (12 % isolated yield). LC-ESI-MS (m/z; Betabasic-18, 10–50 % AcN over 30 min, 1 mL/min; Rt = 25.5 min): 1430.4 (calc. 1430.52 [C55H85N15O12ReS3+][M+H+]); 1452.5 (calc. 1452.5 [C55H84N15O12ReS3Na+]); 715.8 (calc. 715.8 [C55H86N15O12ReS32+][M+2H2+]). Method 2 Ammounium perrhenate (2.69 mg, 10.04 μmol), sodium citrate dihydrate (5.91 mg, 20.1 μmol) and tin(II) tartrate (8.0 mg, 30.0 μmol) were added to 200 μL of DI water and sonicated for 15 min, or until the light blue color of Re-citrate had formed. In a separate vial, 5 (9.27 mg, 0.648 μmol), the Re-citrate solution, and 100 μL AcN were combined in a 1.5 mL snap cap centrifuge tube, and heated at 70°C for 1 h. The reaction mixture was centrifuged and the supernatant collected. The solids were washed with 50:50 AcN:H2O (500 μL), centrifuged, and the second supernatant combined with the first. The combined supernatants were filtered through a 0.45 μM filter, purified by HPLC (Betabasic-18, 10–50% AcN over 30 min at 1 mL/min), and the 25 min product peak was collected. The product was rotary evaporated to dryness to yield ReO-5 (10 mg; 90%). LC-ESI-MS (m/z; Betabasic-18, 10–50 % AcN over 30 min, 1 mL/min; Rt = 25.4 min): 1430.4 (calc. 1430.52 [C55H85N15O12ReS3+][M+H+]); 1452.5 (calc. 1452.5 [C55H84N15O12ReS3Na+]). 99mTcO(222-MAMA(N-6-Ahx-OEt)) [99mTcO-3] A GAS solution (56.4 mM sodium glucoheptonate, 0.25 M HCl, 3.7 mM SnCl2 solution) was prepared. A mixture of 3 in EtOH (0.006 M, 0.3 mL), generator eluent (0.3 mL; 300 MBq (10 mCi)), and GAS (0.3 mL) was vortexed and allowed to react at 70 °C for 40 min. Phosphate buffer (0.9 mL of a 20 mM solution, pH 8–9) was added to the reaction mixture, which was then vortexed, and allowed to react at 70 °C for 30 min. The EtOH was removed from the reaction mixture under a stream of N2 and the resultant mixture was extracted with EtOAc (3 × 2.0 mL). The organic layer was analyzed by radio-HPLC (Betabasic-18, 10–50% AcN over 30 min, 1 mL/min; Rt = 30.4 min). Radiochemical purity: 82 ± 7%. 99mTcO(222-MAMA(N-6-Ahx-OH)) [99mTcO-4] The 99mTcO-4 was prepared using the same procedure described for 99mTcO-3, except with 0.3 mL of an EtOAc solution of 4 (4.97 mg in 0.32 mL EtOAc), with 23.2% isolated radiochemical yield. The organic layer was analyzed by radio-HPLC (Betabasic-18, 10-50% AcN over 30 min, 1 mL/min; Rt = 20.2 min). 99mTcO(222-MAMA(N-6-Ahx-BBN(7-14)NH2)) [99mTcO-5] A GAS solution (56.4 mM sodium glucoheptonate, 0.25 M HCl, 3.7 mM SnCl2 solution) was prepared. A mixture of 5 in a 50:50 EtOH:H2O solution (0.81 mM, 0.1 mL), 0.3 mL of generator [99mTc]-TcO41- eluent (~0.91 GBq (25 mCi)), and 0.3 mL of the GAS solution was vortexed, reacted in a 70 °C water bath for 40 min. Approximately, 0.3 mL of 20 mM phosphate buffer was added, and the resulting solution was pH 2. The product mixture was purified via a Waters Sep-Pak C18 Plus Light cartridge, which had been pretreated with 10 mL of EtOH followed by 10 ml of H2O. Three fractions were collected, the load volume, a 1 ml H2O wash, and a 1 ml EtOH elution. The EtOH fraction, contained ~0.21 GBq (5.7 mCi; 65.1% of the activity loaded). An aliquot of the concentrated 99mTcO-5 sample (7-8 μCi) was added to a sterile saline and checked for purity (Betabasic-18, 10–50% AcN over 30 min, 1 mL/min, Rt = 21.6 and 24.3 min in a 1:4 ratio). Radiochemical yield: 53 ± 6%, pre-purification; radiochemical purity: 88 ± 3%, post purification Initial cysteine challenge experiment The concentrated 99mTcO-5 product (50 μL, ~ 30 MBq (0.8 mCi) per vial) was added to approximately 450 μL of 1.1 mM cysteine in phosphate buffered saline (PBS) and incubated in a 37 °C water bath. The challenge experiment was performed twice (n = 3 each time), as was the experiment with pertechnetate blanks (using 99mTcO41− generator eluent instead of the 99mTcO-5 solution). RadioTLC analysis was accomplished using saturation pads that were developed in saline and 50% AcN in water. Pertechnetate moved with the solvent front in both cases, while 99mTcO-5 remained at the origin in saline and moved with the solvent front in 50% AcN in water. The strips were cut in half and counted using a NaI(Tl) well detector at 0, 1, 4, and 24 h for both 99mTcO41− and 99mTcO-5. Carrier Added 186ReO(222-MAMA(N-6-Ahx-OEt)) [ca 186ReO-3] A GA10S solution (55 mg sodium glucoheptonate, 50 μL HCl, 35 mg SnCl2·2H2O, and 3.5 mL H2O) was prepared. A mixture of 3 in EtOAc (4.5 mM, 0.3 mL), 10 μL of a [186Re]-ReO41− solution (8.25 MBq (223 μCi) of 186Re; 6.6 GBq/mg Re (0.179 Ci/mg)), and 0.3 mL of GA10S were combined, vortexed, and reacted in a 70 °C water bath for 40 min. Phosphate buffer (0.3 mL of a 20 mM solution, pH 8-9) was added, the solution was vortexed, and the reaction was continued in a 70 °C water bath for 30 min. The product mixture was extracted with EtOAc (2 × 2 mL) and the layers were counted using the dose calibrator (64.6% extracted yield). The first EtOAc collection vial contained 1.65 mg of gentisic acid (2,5-dihydroxybenzoic acid) to prevent radiolysis of the complex. Carrier Added 186ReO(222-MAMA(N-6-Ahx-BBN(7-14)NH2)) [ca 186ReO-5] 5 (0.8 mg, 0.7 μmol) in 0.3 mL of EtOH and 0.15 mL of a carrier added [186Re]-Al(ReO4)3 solution in water (32.9 MBq (889 μCi), 3.6 GBq/mg (0.098 Ci/mg), 0.0091 mg, 0.048 μmol) were combined with0.15 mL of deionized water and 0.3 mL of freshly prepared GA10S. The reaction mixture was vortexed, reacted at 75–80 °C in a dry bath for 40 min, and pH adjusted with pH 8–9 phosphate buffer (0.3 mL of a 20 mM phosphate buffer) to approximately 6.5. The product mixture was purified via a pretreated C-18 column (BondElut, Agilent Technologies, 1 in. × 2 in.) eluted with 15 mL of 50:50 AcN in H2O. Three fractions were collected with volumes of 4 mL, 6 mL, and 5 mL, respectively. The 6 mL fraction (45% isolated radiochemical yield) was collected with gentisic acid (8 mg), concentrated in a 70 °C water bath with a stream of N2 (g), and analyzed by radioTLC (0 and 1 h). No Carrier Added 186ReO(222-MAMA(N-6-Ahx-BBN(7-14)NH2)) [nca 186ReO-5] 5 (0.2 mg, 0.1 μmol) in 100 μL of absolute ethanol, 7.4 MBq (0.2 mCi) of nca 186ReO41- and 0.1 mL of GAS solution (56.4 mM sodium glucoheptonate, 0.25 M HCl, 3.7 mM SnCl2 solution) were combined, vortexed and heated in a sand bath at 80°C for 1 h. After removing from heat, 0.1 mL of 20 mM phosphate buffer was added. The reaction mixture was purified on a C-18 column (BondElut, Agilent Technologies, 0.5 in. × 0.5 in.) that had been pretreated with 5 mL of 50:50 AcN:H2O. The reaction mixture was loaded, washed with 3 mL of H2O, and eluted with 5 mL of 50:50 AcN:H2O. HPLC (Jupiter C-18 column, at 25 °C with a gradient of 10–70% AcN with 0.1% TFA in water with 0.1% TFA over 15 min) retention time: 13.1 min, analogous to the macroscopic ReO-5 standard, indicated formation of nca 186ReO-5. Initial stability studies in 0.88 mM cysteine The ca 186ReO-5 was prepared as described above, evaporated to near dryness and reconstituted in EtOH. An aliquot of the ethanolic solution (40 μL) was added to each of three tubes containing 160 μL of 1.1 mM cysteine, which were placed in a 37 °C water bath. Samples were analyzed by HPLC (Betabasic-18, 10–30% AcN over 15 min, followed by 30–50% AcN over 15 min, 1 mL/min) at 0, 1 and 17.4 h. The 17.4 h sample was also analyzed by TLC (paper chromatography, saline and 50% AcN in water eluents) and LSC. A sample prepared without cysteine was used as the standard for comparison. In vitro cell binding studies with ReO-5 and PC-3 tumor cells The protocol used was previously reported [24]. Specifically, [125I]-Tyr4-BBN(NH2) (PerkinElmer, Waltham, MA) was used as the competitor; cell media [500 mL RPMI 1640 cell media (Invitrogen, Carlsbad, CA) was supplemented with approximately 30 μM bovine serum albumin (BSA, 1 g, 15 μmol) and 20.6 mM 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES, 2.46 g, 10.3 mmol), filtered through a 0.22 μm sterile filter, and pH adjusted to 7.4] was prepared for the study. Approximately 30,000 PC-3 cells were incubated at 37 °C, 95% humidity, and 5% CO2 for 45 min, with approximately 30,000 cpm [125I]-Tyr4-BBN(NH2) and varying concentrations of ReO- 5 (prepared by Method 1) in the modified cell growth media. The cells were washed with cold modified media to remove unbound activity. An LTi Multi-Wiper gamma counter (LTI Laboratory Technologies; Elburn, IL) was used to count the activity bound to the cells. The entire process was performed three times in duplicate. The IC50 value was calculated using GraFit (Erithacus Software, West Sussex, UK). Pharmacokinetic studies in CF-1 mice All animal experiments were conducted in accordance with protocols approved by the Harry S. Truman Memorial Veteran’s Administration Hospital Subcommittee for Animal Studies. Normal mice (CF-1; 8 weeks old; Charles Rivers Laboratories; Wilmington, MA) were injected via the tail vein with approximately 0.37 MBq (10 μCi) of the radiolabeled material in approximately 100 μL of sterile PBS. The mice were sacrificed at 1 h (eight mice, four with 99mTcO41− and four with 99mTcO-5) post injection. The organs and tissues were isolated, weighed, counted using a NaI(Tl) well counter, and the percent injected dose (% ID) and percent injected dose per gram (% ID/g) of each were calculated. The entire blood mass was estimated to be 6.5% of the total body weight [26]. All urinary bladder contents excreted from the time of injection to the time of sacrifice were collected, counted, and reported as urine (% ID). 2.3. Computational Methods All calculations were performed using the Guassian09 program package [29,30]. ReO-2 and ReO-222-MAMA(N-methyl) were fully optimized in the gas and solution phases (SMD chloroform [31]) using the PBE0/6-31G**,LANL2TZ method and basis set as validated in our previous work [32]. (In this notation, the first basis set (6-31G**) was used for the non-metal atoms; the second basis set (LANL2TZ) was used for the metal atom.) All equilibrium structures were verified as minima (no imaginary frequencies). The NMR calculations discussed herein were performed using the GIAO method [29] on the lowest-energy PBE0/6-31G**,LANL2TZ gas- or solution-phase equilibrium structure. The calculated NMR shifts are referenced to the 13C- and 1H-NMR shifts for tetramethylsilane computed at the same level and in the same phase. 3. Results and Discussion The 222-MAMA ligand conjugated to BBN(7–14)NH2 via a six carbon spacer was investigated as a potential bifunctional chelate for 186/188Re(V), the most readily accessible oxidation state of Re from the available 186/188ReO41− [6]. Although the 222-MAMA ligands have been evaluated with 99mTc, none conjugated to a peptide for chelating 186/188Re have been reported. The overall scheme for preparing the oxorhenium(V) or oxotechnetium(V) complexes with the various 222-MAMA ligands is shown in Fig.. 3.1. Ligand and bifunctional chelator synthesis 2 was coupled to BBN(7–14)NH2 through a 6-carbon linker, which was complexed with rhenium for in vitro and technetium for in vivo studies. The target 222-MAMA(N-6-Ahx-BBN(7–14)NH2) (5) was prepared in three steps as shown in Schemes 1–3 using TFA in the final step to cleave 5 from the resin bead [10,33]. The trityl groups for the ligands (2 and 3) were removed using TFA and TES prior to reacting with ReOCl3(PPh3)2 (Method 1) or rhenium-citrate (Method 2), or in situ with 99mTc or 186Re on the radiotracer scale in the presence of Sn2+ and low pH conditions. The bifunctional chelator (4), synthesized from 3, was conjugated to the BBN(7–14) peptide via traditional solid-phase peptide synthesis utilizing the carboxylic acid derivatized product as the final amino acid-like reactant. 5 was cleaved from the Sieber resin bead. The crude product was 85% pure, and further purified by semi-preparative HPLC to obtain pure 5 (>98% pure). 3.2. Rhenium Complexes The oxorhenium(V) 222-MAMA complexes with 3 and 5 were synthesized from either ReOCl3(PPh3)2 or ReO(citrate)21−, and exhibit absorbance around 358 nm (λmax). The 1H-NMR spectra exhibit complex splitting patterns indicating that the backbone protons became unique on coordination to the monooxorhenium(V) core. All chemical shifts are observed downfield on complexation with the exception of the methylene protons on the carbon alpha to the amine (e in the free ligands and the complexes, Fig. 1), with one of the protons shifted upfield and the other downfield, and identified via 2-D NMR experiments. Gas-phase NMR computations of the syn and anti isomers of ReO-222-MAMA(N-methyl) at the PBE0/6-31G**,LANL2TZ level using the GIAO method with chloroform solvent effects (SMD method) were used to validate the assignment of the upfield shifted proton. The assignment of protons in ReO-3 is consistent with the 1H-NMR results published by Katzenellenbogen and co-workers for syn-ReO-222-MAMA(N-benzyl) [11] and Jones and coworkers for the ReO-222-MAMA(N-β-Ala-OCH2CH3) complex [12]. The dramatic splitting of the protons on carbon e [2.33–2.41 ppm in 3 to 1.64 and 3.10–3.32 ppm in ReO-3] is due to the proximity to the Re≡O bond. The quantum-mechanical NMR computations indicate that the complex isolated from our synthesis is the syn isomer. The 1.24 ppm difference in chemical shifts between the e protons determined computationally for the syn isomer is similar to the 1.6 ppm shift difference observed for ReO-3 experimentally. Additionally, the downfield shift of peak g (3) from 2.18–2.28 multiplet to 3.53 ppm triplet of doublets and 3.95 ppm triplet of doublets indicates that the syn isomer was isolated [12,13]. The e proton shifts from our quantum-mechanical NMR computations for the anti isomer show a difference of 0.5–0.7 ppm. According to our calculations, the syn and anti isomers are equally thermodynamically favorable, but the transition between the two requires inversion of the rhenium center (through an associative mechanism using water) or deprotonation (basic conditions) of the amine. All of the 13C-NMR carbon signals are observed downfield on complexation to the monooxorhenium(V) center. Again, the 13C-NMR peak shifts closely match those reported for ReO-222-MAMA(N-β-Ala-OCH2CH3) [12]. All Re complexes were characterized by reversed phase HPLC, as standards for the 99mTc and nca 186Re analogues prepared at the radiotracer level using several methods. Table 2 lists the retention times of the complexes by the two most commonly used methods. 3.3. Radiolabeling of 3 The radiolabeling of 3 was optimized using 99mTcO41− under a variety of reaction conditions (i.e., deprotected ligand or protected ligand, biphasic or aqueous reaction mixtures, reaction pH (1–9), reaction time (20–40 min), reaction temperature (30–70 °C), and ligand concentration (0.0025–6.7 mM in the total reaction mixture). These studies were monitored by EtOAc extraction (following the extraction method used by Oya et al. [10]). The optimal radiolabeling yield was observed with the protected ligand dissolved in EtOH (0.006 M, 0.3 mL), a 40 min reaction time at 70 °C with pH of ~1 (from the GAS solution), a pH adjustment with 20 mM pH 8–9 phosphate buffer, a 30 min heating at 70 °C, and an EtOAc extraction. A radiochemical yield of 85±4% was observed. HPLC analysis of the product mixture (Fig. 3) indicates that the product from the 99mTc labeling reaction with 3 (222-MAMA(N-hexanoate)) (Rt = 30.4 min) is analogous to the non-radioactive ReO-3 HPLC standard (Rt = 30.2 min). The 186ReO-3 complex, using ca 186Re, was prepared similarly to the 99mTcO-3 complex (by Method 1), utilizing a GAS solution with ten times the [Sn2+] (GA10S) and gave an extracted yield >60%. The increased Sn2+ concentration was required because rhenium is more difficult to reduce than technetium [34] and the [186Re]-ReO41− solution was carrier added. EtOAc extractions of [186Re]-ReO41− from saline contained 6.24% of the radioactivity. The EtOAc layer from the ca 186ReO-3 labeling-reaction product mixture extraction contained approximately 5 % 186ReO2 (from TLC analysis). The amount of carrier added 186ReO-3 produced was insufficient for HPLC analysis. 3.4. Radiolabeling of 5 The 99mTc labeled 222-MAMA(N-6-Ahx-BBN(7–14)NH2) complex, 99mTcO-5, was prepared as described above with 53±6% yield. The 99mTcO-5 was purified by Sep-Pak to remove 99mTcO41−, 99mTcO2, and excess 5. Unreacted ligand 5 was still present according to the HPLC analysis; however, 99mTcO-5 was not further purified. Approximately 30% of the unlabeled 5 (average 0.03 mg) remained in solution following Sep-Pak purification. Radio-HPLC analysis of the purified 99mTcO-5 product mixture showed 88 ± 3% purity in the combined 21.6 and 24.3 min peaks, with the remainder present as pertechnetate. When 99mTcO-5 was purified by Sep-Pak, two species were observed at 21.6 min and 24.3 min, presumably the syn and anti isomers. This isomerization was confirmed by collecting each individual product (peak) separately, reducing the volume, and reinjecting a sample (Fig. 4). The species at 21.6 min showed the growth of a peak at 24.3 min, which over time equilibrated to a 1:2 ratio favoring the later eluting species. The 24.3 min species changed very little over time, with <5% in-growth of the 21.6 min species. To verify this isomerization reaction, an aliquot of ReO-5 was dissolved in EtOH and analyzed by HPLC over time. Two species were observed at 22.1 and 24.8 min, with the 24.78 min species dominating (89%). In AcN only the later eluting species is observed. The 99mTcO-5 was relatively stable in sterile saline (without added gentisic acid) dropping from ≥90% initially to ≥75% remaining at 6 h, as confirmed by HPLC. The ca 186ReO-5 complex was prepared similarly to ca 186ReO-3 and isolated by C18 column chromatography, and collected into a vial containing gentisic acid. The isolated fraction (0.988 mCi/mL) contained the majority of the desired product (approximately 66% of the total radioactivity). The isolated ca 186ReO-5 co-eluted with the ReO-5 standard at 10.1 min (Jupiter C-18, 30–60% AcN in water without TFA over 15 min) or 20.5 min (Betabasic, 10–30% AcN in water with 0.1% TFA over 15 min and 30–50% AcN in water with 0.1% TFA over 5 min) and was unstable in the absence of gentisic acid. Previous results have shown that complexes containing beta emitters are less prone to radiolysis with the addition of a small amount of gentisic acid [35–38]. To assess the stability of ca 186ReO-5 in biological media, the complex was incubated with 0.88 mM cysteine at 37 °C in PBS. HPLC analysis showed only 186ReO-5 until 17.4 h, at which time TLC analysis indicated about 10–12% 186ReO41− and at most 1% 186ReO2 had formed. Thus, after 17.4 h approximately 86–89% of the original 186ReO-5 complex remained intact. This amount is approximately the same as the amount of the 99mTcO-5 complex intact after 1 h. The difference in stability could be due to the addition of gentisic acid in the purified ca 186ReO-5 sample or the higher concentration of the ca 186ReO-5 synthesis. These results are consistent with the stability of 186ReO-MAMA(N-bisphosphonate) derivatives reported by Ogawa et al. [39]. The nca 186ReO-5 was synthesized similarly to 99mTcO-5 and was analyzed by HPLC (Jupiter C-18 column, at 25°C with a gradient of 10–70% AcN in water with 0.1% TFA over 15 min) (Fig. 5). Initially, only one peak at 13.1 min was observed, which over time separated into two peaks (likely the syn and anti isomers) observed at 12.5 and 13.5 min. These results are consistent with those observed for 99mTcO-5 and indicate formation of the desired nca 186ReO-5 product. Two isomers are observed for ReO-5 only when EtOH is present. 3.5. In vitro Stability Studies Initial stability studies of HPLC purified 99mTcO-5 complex (without gentisic acid) in 1 mM cysteine at 37 oC showed that approximately 90% remained intact at 1 h and 48% at 24 h (Fig. ). Oya et al. reported that 70% of the 99mTcO-2 complex was intact after incubation in rat serum at 37 °C for 30 min [10]; Yamamura et al. reported that 87.1% of 99mTcO-4 remained intact after 1 h in 10 μM cysteine in 0.05 M phosphate buffer (pH 7.0) at 37 °C [40]. 3.5. In vitro competitive cell binding studies In vitro competitive cell binding studies in PC-3 prostate cancer cells, known to express GRP (BB2r) receptors, of ReO-5 (prepared by Method 1) against [125I]-Tyr4-BBN(NH2) gave a calculated IC50 value of 2.0 ± 0.7 nM (n = 6). This value is similar to previous results with other metalated and non-metalated BB2r targeting compounds [21,22]. Additionally, this value is nearly identical to that reported for In-DOTA-5-Ava-BBN(7–14)NH2 (1.7 ± 0.4 nM) [22] and is similar to the value reported for ReO-N3S-5-Ava-BBN(7–14)NH2 (1.0 ± 0.2 nM) [20]. 3.6. In vivo biodistribution studies Biodistribution studies with the 99mTcO-5 complex (prepared by Method 2, >95% pure) were performed in normal CF-1 mice, with four mice injected with TcO-5 (0.26–0.40 MBq; 7–8 μCi) and four mice injected with BBN[1–14] as a blocking agent 5 min prior to injection of 99mTcO-5 (0.26–0.40 MBq; 7–8 μCi). The biodistribution data obtained from these studies is tabulated in Table 3 and shown graphically in Figs. 7 and 8. Approximately 2 % ID/g uptake was observed in the pancreas; this lower pancreatic uptake may be due to the presence of unlabeled 5 competing for receptor sites. Thus, HPLC purification may be necessary to increase the specific activity of the tracer, which may improve tumoral uptake in future in vivo studies. The pancreatic uptake does suggest BB2r targeting because the uptake was blocked by injection of BBN[1–14] prior to injection of 99mTcO-5. Additionally, the complex is very hydrophobic, as seen with its long HPLC retention time and its high liver and intestinal uptake. Conclusions and future directions The 222-MAMA-based chelators were synthesized for potential use in radiopharmaceuticals that incorporate 186/188Re. The 222-MAMA chelator was conjugated to BBN(7–14)NH2 and analyzed via 99mTc, non-radioactive Re, and carrier-added and no carrier added 186Re studies. Cell surface receptor binding studies with the ReO-5 complex and in vivo studies with 99mTcO-5 indicate BB2r targeting by these constructs. Thus, these studies indicate that 222-MAMA(N-6-Ahx-BBN(7–14)NH2) chelates oxorhenium(V) and oxotechnetium(V) for delivery in vivo to BB2r-positive tissues. The bifunctional chelator (3) may be useful for conjugation to other biological targeting vectors and may have utility for delivering therapeutic doses of 186/188Re to tumor tissues using peptides and antibodies. The availability of nca 186Re gives impetus to develop 99mTc/186Re “matched pair” agents for imaging and treating cancer. Supplementary Material supplement Dr. Susan Lever provided helpful suggestions with the organic chemistry work. Drs. Tim Glass, Michael Harmata, and Vikram Gaddam were instrumental in obtaining 3 and 4. Ma-Guadalupe Ruvalcaba Andrade helped with corroborating the synthesis of these compounds. The authors acknowledge the University of Missouri Bioinformatics Consortium for the use of their High Performance Computing resources, University of Missouri Mass Spectroscopy Facility, National Science Foundation grant NSF-CHE-89-08304 for the use of the NMR facility, and Dr. Wei Wycoff for assistance with the NMR. Research was supported by the National Institute of Biomedical Imaging and Bioengineering Training Grant No. NIBIB 5 T32 EB004822 (DWD). The research was also supported in part by the U.S. Department of Energy, Office of Basic Energy Sciences, Heavy Element Chemistry program under grant No. DE-FG02-09ER16097 and Projects for Interrogations of Biological Systems No. DE-SC0002040 as well as VA Merit Award Number 1I01BX001699 and a Research Career Scientist Award (TJH) from the Biomedical Laboratory Research and Development Service of the U.S. Department of Veterans Affairs. Figure 1 1–6, and ReO-3 with lettering scheme used for 1H- and 13C-NMR. Figure 2 Overall schematic for rhenium and technetium complex formation (R is H, (CH2)5COOEt, (CH2)5COOH, or (CH2)5CO-QWAVGHLM-NH2). Figure 3 HPLC chromatograms (Betabasic, 10–50% AcN over 30 min, 1 mL/min) of A) 99mTcO-3 (radio-HPLC), and B) ReO-3 (UV-vis, 358 nm). The 21.9 min peak is presumably 99mTcO-4 (from ester hydrolysis during ligand deprotection). Figure 4 HPLC chromatograms (Betabasic, 10–50% AcN over 30 min, 1 mL/min) of A) 99mTcO-5, B) isolated 21.6 min peak re-injected (shows in-growth of 24.3 min species), and C) isolated 24.3 min peak (shows very little in-growth of 21.3 min species). Figure 5 HPLC chromatograms (Jupiter C-18, 10–70% AcN over 15 min, 1 mL/min) of A) ReO-5 and B) nca 186ReO-5. Figure 6 99mTcO-5 stability over 24 h in 1 mM cysteine at 37 °C. Fig. 7 Biodistributions in percent injected dose (%ID) for 99mTcO-5 (at 1 h) and 99mTcO-5 (at 1 h) with blocking agent administered 5 min prior. Fig. 8 Biodistributions in percent injected dose per gram (%ID/g) for 99mTcO-5 (at 1 h) and 99mTcO-5 (at 1 h) with blocking agent administered 5 min prior. Scheme 1 Two-step preparation of target trityl-protected MAMA ligands.13 Scheme 2 Attaching linking group to chelator and subsequent saponification to obtain the bifunctional chelator. Scheme 3 Solid-phase peptide synthesis (SPPS) and cleavage to give the bifunctional chelator conjugated to the targeting peptide. Table 1 Nuclear properties of 186/188Re [4]. Nuclide t1/2 Eβ− max γ decay energy (% abundance) Rangeβ− max (H2O) Production routes [1–3] 186Re 3.7 d 1.1 MeV 137 keV (8%) 4.8 mm 185Re(n,γ)186Re 186W(p,n)186Re 186W(d,2n)186Re 188Re 17 h 2.1 MeV 155 keV (15%) 10.4 mm 186W(2n,γ)188W → β − + 188Re + ν̄ 187Re(n,γ)188Re Table 2 HPLC retention times of compounds. Compound Betabasic-18, 10–50% AcN with 0.1% TFA over 30 min, 1 mL/min Jupitor-18, 10–70% AcN with0.1% TFA over 15 min, 1 mL/min ReO-3 (method 1) ReO-3 (method 2) 30.1 min 30.2 min 99mTcO-3 30.4 min 99mTcO-4 20.2 min 5 15.6 min ReO-5 (method 1) ReO-5 (method 2) 25.5 min 25.4 min 13.2 min 99mTcO-5 (method 2) 24.3 min nca 186ReO-5 13.1 min Table 3 Biodistribution in CF-1 normal mice for 99mTcO-5 and blocking study at 1 h post injection. %ID (n=4) %ID/g (n=4) 99mTcO-5 Blocked 99mTcO-5 Blocked Heart 0.02±0.02 0.05±0.07 0.11±0.16 0.40±0.51 Lung 0.13±0.16 0.15±0.07 0.55±0.69 0.80±0.41 Liver 20.61±4.47 12.47±4.27 10.38±1.88 6.51±1.98 Kidneys 2.00±0.23 1.60±0.39 4.24±0.60 3.57±89 Spleen 0.15±0.20 0.13±0.15 1.07±1.48 0.86±1.10 Stomach 1.64±0.71 1.44±0.33 3.60±2.15 1.87±0.29 S. Intestine 27.40±9.51 59.19±3.14 15.46±6.08 29.72±2.22 L. Intestine 23.20±10.47 0.39±0.05 18.04±8.25 0.37±0.11 Muscle 0.02±0.03 0.10±0.08 0.06±0.12 0.54±0.40 Bone 0.02±0.04 0.11±0.15 0.32±0.64 1.52±2.17 Brain 0.01±0.02 0.03±0.03 0.02±0.04 0.06±0.06 Pancreas 0.72±0.03 0.09±0.06 2.03±0.42 0.24±0.17 Bladder 2.46±1.12 1.40±0.16 0.05±0.02 0.3±0.06 Blood 0.83±0.23 1.09±2.80 0.38±0.09 0.49±0.52 Carcass 5.50±0.53 4.57±0.49 0.24±0.04 0.21±0.03 Urine - - 12.95±8.19 18.34±2.22 Cage Paper - - 4.76±9.25 0.23±0.42 Supporting Information Supporting information is available. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Hussain M Sudar S Aslam MN Malik AA Ahmad R Qaim SM Evaluation of charged particle induced reaction cross section data for production of the important therapeutic radionuclide 186Re Radiochim Acta 2010 98 385 95 2 Lapi S Mills WJ Wilson J McQuarrie S Publicover J Schueller M Production cross-sections of 181–186Re isotopes from proton bombardment of natural tungsten Appl. Radiat Isot 2007 65 345 9 3 Moustapha ME Ehrhardt GJ Smith CJ Szajek LP Eckelman WC Jurisson SS Preparation of cyclotron-produced 186Re and comparison with reactor-produced 186Re and generator-produced 188Re for the labeling of bombesin Nuc. Med Biol 2006 33 81 9 4 National Nuclear Data Center Brookhaven, NY Brookhaven National Laboratory 5 Jurisson SS Lydon JD Potential Technetium Small Molecule Radiopharmaceuticals Chem Rev 1999 99 2205 18 11749479 6 Carroll V Demoin DW Hoffman TJ Jurisson SS Inorganic chemistry in nuclear imaging and radiotherapy: current and future directions Radiochim Acta 2012 100 653 67 25382874 7 Cutler CS Hennkens HM Sisay N Huclier-Markai S Jurisson SS Radiometals for combined imaging and therapy Chem Rev 2013 113 858 83 23198879 8 Liu G Hnatowich DJ Labeling Biomolecules with Radiorhenium - A Review of the Bifunctional Chelators Anti-Cancer Agent Me 2007 7 367 77 9 Liu S Bifunctional coupling agents for radiolabeling of biomolecules and target-specific delivery of metallic radionuclides Adv Drug Deliver Rev 2008 60 1347 70 10 Oya S Plössl K Kung M-P Stevenson DA Kung HF Small and Neutral TcVO BAT, Bisaminoethanethiol (N2S2) Complexes for Developing New Brain Imaging Agents. Nuc. Med Biol 1998 25 135 40 11 O'Neil JP Wilson SR Katzenellenbogen JA Preparation and Structural Characterization of Monoamine-Monoamide Bis(thiol) Oxo Complexes of Technetium(V) and Rhenium(V) Inorg Chem 1994 33 319 23 12 Mahmood A Kuchma MH Freiberg E Goldstone J Davison A Jones AG Functionalized Tetradentate N2S2 Chelates and their Technetium-99 and Rhenium Complexes: Synthesis, Spectroscopy, and Structural Characterization Nicolini M Mazzi U Technetium, Rhenium and Other Metals in Chemistry and Nuclear Medicine Bressanone, Italy Academia Cusanus 1999 253 7 13 Mahmood A Kuchma MH Goldstone J Morse C Davison A Jones AG An N2S2 Tetradentate Chelate for Solid-Phase Synthesis: Evaluation in Solution and Solid Phase and Characterization of Technetium-99 Complexes Nicolini M Mazzi U Technetium, Rhenium and Other Metals in Chemistry and Nuclear Medicine Bressanone, Italy Academia Cusanus 1999 71 6 14 Friebe M Mahmood A Bolzati C Drews A Johannsen B Eisenhut M [99mTc]Oxotechnetium(V) Complexes of Amine-Amide-Dithiol Chelates with Dialkylaminoalkyl Substituents as Potential Diagnostic Probes for Malignant Melanoma J Med Chem 2001 44 3132 40 11543682 15 Ogawa K Mukai T Arano Y Hanaoka H Hashimoto K Nishimura H Design of a radiopharmaceutical for the palliation of painful bone metastases: rhenium-186-labeled bisphosphonate derivative J Label Compd Radiopharm 2004 47 753 61 16 Luo T-Y Tang I-C Wu Y-L Hsu K-L Liu S-W Kung H-C Evaluating the potential of 188Re-SOCTA-trastuzumab as a new radioimmunoagent for breast cancer treatment. Nuc. Med Biol 2009 36 81 8 17 Ono M Ikeoka R Watanabe H Kimura H Fuchigami T Haratake M Synthesis and Evaluation of Novel Chalcone Derivatives with 99mTc/Re Complexes as Potential Probes for Detection of β-Amyloid Plaques. ACS Chem Neurosci 2010 1 598 607 18 Francesconi LC Graczyk G Wehrli S Shaikh SN McClinton D Liu S Synthesis and characterization of neutral MVO (M = technetium, rhenium) amine-thiol complexes containing a pendant phenylpiperidine group. Inorg Chem 1993 32 3114 24 19 Coy DH Short-chain pseudopeptide bombesin receptor antagonists with enhanced binding affinities for pancreatic acinar and Swiss 3T3 cells display strong antimitotic activity J Biol Chem 1989 264 14691 7 2475489 20 Smith CJ Gali H Sieckman GL Higginbotham C Volkert WA Hoffman TJ Radiochemical investigations of 99mTc N3S-X-BBN[7–14]NH2: An in vitro/in vivo structure activity relationship study where X = 0-, 3-, 5-, 8-, and 11-carbon tethering moieties Bioconjugate Chem 2002 14 93 102 21 Smith CJ Gali H Sieckman GL Hayes DL Owen NK Mazuru DG Radiochemical investigations of 177Lu-DOTA-8-Aoc-BBN[7–14]NH2: an in vitro/in vivo assessment of the targeting ability of this new radiopharmaceutical for PC-3 human prostate cancer cells. Nuc. Med Biol 2003 30 101 9 22 Hoffman TJ Gali H Smith CJ Sieckman GL Hayes DL Owen NK Novel series of 111In-labeled bombesin analogs as potential radiopharmaceuticals for specific targeting of gastrin-releasing peptide receptors expressed on human prostate cancer cells. J. Nuc Med 2003 44 823 31 23 Prasanphanich AF Nanda PK Rold TL Ma L Lewis MR Garrison JC [64Cu-NOTA- 8-Aoc-BBN(7–14)NH2] targeting vector for positron-emission tomography imaging of gastrinreleasing peptide receptor-expressing tissues PNAS 2007 104 12462 7 17626788 24 Garrison JC Rold TL Sieckman GL Figueroa SD Volkert WA Jurisson SS In Vivo Evaluation and Small-Animal PET/CT of a Prostate Cancer Mouse Model Using 64Cu Bombesin Analogs: Side-by-Side Comparison of the CB-TE2A and DOTA Chelation Systems. J. Nuc Med 2007 48 1327 37 25 Retzloff LB Heinzke L Figureoa SD Sublett SV Ma L Sieckman GL Evaluation of [99mTc-(CO)3-X-Y-bombesin(7–14)NH2] conjugates for targeting gastrin-releasing peptide receptors overexpressed on breast carcinoma Anticancer Res 2010 30 19 30 20150613 26 Jackson AB Nanda PK Rold TL Sieckman GL Szczodroski AF Hoffman TJ 64Cu- NO2A-RGD-Glu-6-Ahx-BBN(7–14)NH2: A heterodimeric targeting vector for positron emission tomography imaging of prostate cancer. Nuc. Med Biol 2012 39 377 87 27 Fassbender ME Ballard B Birnbaum ER Engle JW John KD Maassen JR Nortier FM Lenz JW Cutler CS Ketring AR Jurisson SS Wilbur DS Proton Irradiation Parameters and Chemical Separation Procedure for the Bulk Production of High-Specific-Activity 186gRe using WO3 targets Radiochim Acta 2013 101 339 346 28 Gott M Ballard BD Redman LN Maassen JR Taylor WA Engle JW Nortier FM Birnbaum ER John KD Wilbur DS Cutler CS Ketring AR Jurisson SS Fassbender ME Radiochemical Study of Re/W Adsorption Behavior on a Strongly Basic Anion Exchange Resin Radiochim Acta 2014 102 325 332 29 Frisch MJ Trucks GW Schlegel HB Scuseria GE Robb MA Cheeseman JR Gaussian 09 Wallingford CT Gaussian, Inc 2010 30 Glendening ED Reed AE Carpenter JE Weinhold F NBO Version 3.1 31 Marenich AV Cramer CJ Truhlar DG Universal solvation model based on solute electron density and a continuum model of the solvent defined by the bulk dielectric constant and atomic surface tensions J Phys Chem B 2009 113 6378 96 19366259 32 Demoin DW Li Y Jurisson SS Deakyne CA Method and basis set analysis of oxorhenium(V) complexes for theoretical calculations Comput Theor Chem 2012 997 34 41 23087847 33 Wuts PGM Greene TW Protection for the Thiol Group Greene's Protective Groups in Organic Synthesis John Wiley & Sons, Inc 2006 647 95 34 CRC Handbook of Chemistry & Physics Haynes WM Taylor and Francis Group, LLC 2013 35 Liu S Edwards DS Stabilization of 90Y-Labeled DOTA-Biomolecule Conjugates Using Gentisic Acid and Ascorbic Acid Bioconjugate Chem 2001 12 554 8 36 Mohsin H Jia F Bryan JN Sivaguru G Cutler CS Ketring AR Comparison of pretargeted and conventional CC49 radioimmunotherapy using 149Pm, 166Ho, and 177Lu Bioconjugate Chem 2011 22 2444 52 37 Balkin ER Liu D Jia F Ruthengael VC Shaffer SM Miller WH Comparative biodistributions and dosimetry of [177Lu]DOTA-anti-bcl-2-PNA-Tyr3-octreotate and [177Lu]DOTA-Tyr3-octreotate in a mouse model of B-cell lymphoma/leukemia. Nuc. Med Biol 2014 41 36 42 38 Chen J Linder KE Cagnolini A Metcalfe E Raju N Tweedle MF Synthesis, stabilization and formulation of [177Lu]Lu-AMBA, a systemic radiotherapeutic agent for Gastrin Releasing Peptide receptor positive tumors. Appl. Radiat Isot 2008 66 497 505 39 Yamamura N Magata Y Arano Y Kawaguchi T Ogawa K Konishi J Technetium-99m-Labeled Medium-Chain Fatty Acid Analogues Metabolized by β-Oxidation: Radiopharmaceutical for Assessing Liver Function Bioconjugate Chem 1999 10 489 95 40 Ogawa K Mukai T Arano Y Otaka A Ueda M Uehara T Rhenium-186-monoaminemonoamidedithiol-conjugated bisphosphonate derivatives for bone pain palliation Nucl Med Biol 2006 33 513 20 16720243

PMC005xxxxxx/PMC5118111.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7605074 6087 Neuroscience Neuroscience Neuroscience 0306-4522 1873-7544 27693474 5118111 10.1016/j.neuroscience.2016.09.038 NIHMS822629 Article Direct Projections from Hypothalamic Orexin Neurons to Brainstem Cardiac Vagal Neurons Dergacheva Olga * Yamanaka Akihiro ** Schwartz Alan R. *** Polotsky Vsevolod Y. *** Mendelowitz David * * Department of Pharmacology and Physiology, The George Washington University, 2300 Eye Street, NW, Washington, DC, 20037, USA ** Research Institute of Environmental Medicine, Nagoya University, Nagoya 464-8601, Japan *** Johns Hopkins University School of Medicine, Baltimore, MD 21224, USA Corresponding author's address: Olga Dergacheva, Department of Pharmacology and Physiology, The George Washington University, 2300 Eye Street, NW, Washington, DC, 20037, olgad@gwu.edu, phone: 202-994-5029, fax: 202-994-2870. 15 10 2016 28 9 2016 17 12 2016 17 12 2017 339 4753 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Orexin neurons are known to augment the sympathetic control of cardiovascular function, however the role of orexin neurons in parasympathetic cardiac regulation remains unclear. To test the hypothesis that orexin neurons contribute to parasympathetic control we selectively expressed channelrhodopsin-2 (ChR2) in orexin neurons in orexin-Cre transgenic rats and examined postsynaptic currents in cardiac vagal neurons (CVNs) in the dorsal motor nucleus of the vagus (DMV). Simultaneous photostimulation and recording in ChR2-expressing orexin neurons in the lateral hypothalamus resulted in reliable action potential firing as well as large whole-cell currents suggesting a strong expression of ChR2 and reliable optogenetic excitation. Photostimulation of ChR2-expressing fibers in the DMV elicited short-latency (ranging from 3.2 ms to 8.5 ms) postsynaptic currents in 16 out of 44 CVNs tested. These responses were heterogeneous and included excitatory glutamatergic (63%) and inhibitory GABAergic (37%) postsynaptic currents. The results from this study suggest different sub-population of orexin neurons may exert diverse influences on brainstem CVNs and therefore may play distinct functional roles in parasympathetic control of the heart. Optogenetic neurons orexin brainstem cardiac parasympathetic Preganglionic cardiac vagal neurons (CVNs) in the dorsal motor nucleus of the vagus (DMV) project directly to the cardiac ganglia and play a substantial role in the cardiac regulation (Sullivan and Connors, 1981, Ciriello and Calaresu, 1982, Cheng et al., 1999). Electrical stimulation of the cardiac branches of the vagus nerve antidromically activates neurons in the DMV and electrical stimulation of the DMV reduces heart rate and myocardial contractility (Calaresu and Pearce, 1965, Nosaka et al., 1979, Ciriello and Calaresu, 1982). Increasing the activity CVNs in the DMV protects the heart from ischemia/reperfusion injury independent of changes in heart rate (Mastitskaya et al., 2012). The activity of CVNs in the DMV are strongly influenced by neurotransmission from other neurons in the brainstem, as well as pathways from the locus coeruleus and oxytocin neurons in the hypothalamus (DePuy et al., 2013, Dergacheva et al., 2014, Pinol et al., 2014, Wang et al., 2014). The results from immunohistochemical studies indicate orexin neurons could be another important source of innervation to neurons in the DMV (Peyron et al., 1998, Date et al., 1999). However, the DMV is a heterogeneous nucleus and it is unknown whether there are direct connections between orexin neurons and the selective population of CVNs in the DMV that play a major role in controlling heart rate and cardiac function. Compelling evidence indicates orexin neurons and receptors play an important role in the regulation of cardiovascular function (Peyron et al., 1998, Ciriello and de Oliveira, 2003, Ciriello et al., 2003, Dergacheva et al., 2005, Carrive, 2013, Dergacheva et al., 2013). Orexin is well known to exert sympathoexcitatory effects such as increases in heart rate, blood pressure and sympathetic nerve activity (Samson et al., 1999, Shirasaka et al., 1999, Chen et al., 2000, Antunes et al., 2001, Matsumura et al., 2001). However, little is known about the role of orexin neurons in parasympathetic control of the heart and the few studies that have examined this issues are conflicting. For example, microinjection of orexin-A into the rostral ventral medulla produces tachycardia mediated in part by inhibition of parasympathetic activity to the heart (Ciriello et al., 2003), whereas microinjection of orexin-A into the nucleus ambiguus elicits bradycardia mediated by excitation of parasympathetic activity to the heart (Ciriello and de Oliveira, 2003). Thus, this study was undertaken to identify and characterize the synaptic pathway from orexin neurons to CVNs in the DMV. To accomplish this goal, we utilized a transgenic strain of orexin-Cre rats that allows us to photoexcite channelrhodopsin-2 (ChR2) in orexin fibers in the brainstem DMV while recording synaptic events in fluorescently identified CVNs in the DMV in an in-vitro slice preparation. EXPERIMENTAL PROCEDURES Animals Orexin-EGFP-2A-Cre rats (both males and females) were used in this study. The generation of these transgenic animals in which Cre recombinase and green fluorescent protein are exclusively expressed in hypothalamic orexin neurons has been described previously (Dergacheva et al., 2016). Rats were housed in the George Washington University animal care facility. All animal surgeries and experiments were approved by the George Washington University Institutional Animal Care and Use Committee. We made all efforts to minimize number of rats used in this study and reduce animal discomfort. Cardiac neuron labeling and viral injections into the lateral hypothalamus Parasympathetic CVNs were labeled as described previously (Dergacheva et al., 2013, Dergacheva, 2015). At postnatal days 4–5 orexin-Cre rats were anesthetized with hypothermia, the heart was exposed and 0.05 ml of 1–5% rhodamine (XRITC; Molecular Probes, Eugene, OR) was injected into the pericardial sac. The previous study showed the specificity of the cardiac vagal labeling (Bouairi et al., 2004). ChR2 fused to enhanced yellow fluorescent protein (EYFP) was targeted to the plasma membrane of orexin neurons and axons as shown in Fig. 1. Adeno-associated viral vector with “FLEX-switch” ChR2 construct (AAV1-ChR2-EYFP, Penn Vector Core, Philadelphia, PA, catalog number AV-1-20298P) was injected into the lateral hypothalamus of orexin-Cre rats using a stereotactic apparatus with a neonatal adapter (Stoelting, Wood Dale, IL). A pulled calibrated pipette (VWR, Radnor, PA) containing viral vector was positioned at the following coordinates: 1.7-1.9 mm posterior and 0.4 mm lateral relative to bregma. The viral vector (60 nL) was slowly injected 5.2 mm lower the dorsal surface of the brain. The pipette was left in the lateral hypothalamus for 5 minutes after injection, then slowly retracted. To reduce pain and discomfort caused by the surgery, buprenorphine was administered after surgery, and animals were carefully monitored until ambulatory. Slice Preparation and Electrophysiology Animals (21-24 days old) were anesthetized with isoflurane and sacrifice by transcardial perfusion of glycerol-based artificial cerebrospinal fluid (aCSF, 4°C) that contained (in mM): 252 glycerol, 1.6 KCl, 1.2 NaH2PO4, 1.2 MgCl, 2.4 CaCl2, 26 NaHCO3, and 11 glucose. Coronal slices of the brainstem (300-μm) were obtained with a vibratome. In another set of experiments 300-μm-thick coronal slices of the hypothalamus that included orexin neurons were made using a vibratome. The slices were then allowed to recover for 15 min in a solution containing (in mM): 110 N-methyl-d-glucamine (NMDG), 2.5 KCl, 1.2 NaH2PO4, 25 NaHCO3, 25 glucose, 110 HCl, 0.5 CaCl2, and 10 mM MgSO4 equilibrated with 95% O2-5% CO2 (pH 7.4, at 34°C). For performing electrophysiological experiments, the slices were transferred to a recording chamber containing (in mM) 125 NaCl, 3 KCl, 2 CaCl2, 26 NaHCO3, 5 glucose, 5 HEPES and equilibrated with 95% O2-5% CO2 (pH 7.4, at 25°C). CVNs in the DMV and orexin neurons in the lateral hypothalamus were identified by the presence of the retrograde tracer and EYFP-expression, respectfully. Infrared-sensitive video detection cameras and differential interference contrast optics were then used to image these identified CVNs and orexin cells. We patched CVNs with patch pipettes (2.5–3.5 MΩ) filled with a solution consisting of (in mM) 150 KCl, 2 MgCl2, 2 EGTA, 10 HEPES, and 2 Mg-ATP, pH 7.3. In experiments that examined activity in orexin cells, the patch pipettes were filled with a solution consisting of 135 mM K-gluconic acid, 10 mM HEPES, 10 mM EGTA, 1 mM CaCl2, and 1 mM MgCl2, pH 7.3. We performed voltage clamp whole-cell recordings at a holding potential of −80 mV. Firing activity of orexin neurons was examined in current-clamp whole-cell configuration. All recordings were made with an Axopatch 200 B and pClamp 8 software (Axon Instruments, Union City, CA). Drugs used in this electrophysiological study were purchased from Sigma-Aldrich Chemical Co (St. Louis, MO). For selective photostimulation of ChR2 expressing fibers surrounding CVNs we used a CrystaLaser (473-nm blue light, Reno, NV, USA) which was attached to the microscope using a dual housing adapter (Nikon). A series of 60 consecutive single stimulations (3 ms, at a frequency of 1 Hz) were applied to each neuron. Laser light intensity was kept at an output of 10 mW across all experiments. Immunohistochemistry and Imaging Study We utilized immunostaining to determine the specificity of ChR2-EYFP expression in orexin neurons in the lateral hypothalamus. Three weeks after AAV1-ChR2-EYFP viral injections into the lateral hypothalamus, hypothalamic slices (100 μm) were prepared and soaked 3 hours in 10% formalin and were then processed for orexin immunoreactivity. Rabbit anti-orexin-A was used as a primary antibody (1: 15,000 dilution, overnight incubation, Phoenix Pharmaceuticals, Inc., Burlingame, CA) and anti-rabbit Alexa Fluor 633 was used as a secondary antibody (1:200, 4 hours, Life Technologies, Carlsbad, CA). For analysis of co-localization of orexin immunoreactivity and ChR2-EYFP-expression the Zeiss 710 confocal imaging system was used. Orexin immunoreactivity was found in 83±3% of ChR2-EYFP neurons whereas 68±3% of orexin-containing neurons expressed ChR2-EYFP, (n=4 rats, see Fig. 1). Data Analysis and Statistics Postsynaptic currents were measured with pClamp 8 software (Molecular Devices, Sunnyvale, CA). The time between the onset of the optogenetic stimulation and the onset of synaptic current evoked by the stimulation was determined as a synaptic latency of the responses. The mean value for each neuron was created by averaging responses to 60 consecutive photostimulations. A summary of results for each population was created by averaging the mean value from each neuron in the population. The results were statistically compared using GraphPad Prism 5 software and using Student's paired T-test. The data is presented as mean ± SE with significant difference set at p < 0.05. RESULTS Selective photostimulation of orexin neurons Optogenetic photostimulation (3 ms, 1 Hz) of orexin neurons fluorescently identified by ChR2-EYFP expression in the lateral hypothalamus generated large inward current (in voltage clamp configuration, 521±1842 pA, n=8, Fig. 2, B,) and reliable action potentials with each photostimulation (n=8 cells, Fig. 2, A). This robust excitation upon photoactivation indicates there is a strong expression of ChR2-EYFP (Schoenenberger et al., 2011). No electrophysiological responses in orexin neurons in the lateral hypothalamus were observed in orexin-Cre rats that did not receive AAV1-ChR2-EYFP viral injections (n=5 cells, Fig. 2, C and D), indicating that the responses were not due to non-specific cellular activation by light pulses. Direct projections from orexin-containing neurons to CVNs in the DMV To identify and characterize functional projection from orexin neurons to CVNs in the DMV optogenetic stimulation of ChR2-containing fibers from orexin neurons was combined with patch-clamp recordings from CVNs. Although brainstem slices that contained CVNs in the DMV did not include the cell bodies of ChR2-EYFP-expressing orexin neurons it has been previously shown that light-triggered transmitter release from ChR2-EYFP-expressing fibers occurs in the absence of the ChR2-EYFP-expressing cell bodies (Pinol et al., 2012, Schone et al., 2012, Dergacheva et al., 2014). Photostimulation of ChR2-EYFP fibers with brief light pulses (3 ms) generated fast post-synaptic responses in 16 out of 44 CVNs tested (36%). A majority of responses (63%, n=10 neurons) were blocked by application of CNQX (50 μM, from 14.9±1.6 pA to 2.9±0.4 pA, p< 0.001, Student's paired t-test, Fig. 3) indicating these excitatory postsynaptic currents (EPSCs) were mediated via glutamatergic AMPA receptor activation. The latency of excitatory currents was 5.5±0.5 ms (ranging from 3.2 ms to 8.5 ms) while the decay time constant was 4.9±1.0 ms (ranging from 2.8 ms to 11 ms, n=10, see Fig. 3). GABAergic inhibitory postsynaptic currents (IPSCs) were detected in 6 out of 16 CVNs (37%) as these responses were not abolished by CNQX, but were successfully blocked by application of gabazine at a concentration of 25 μM (pick amplitude diminished from 36.8±7.9 pA to 3.3±0.9 pA, p< 0.005, Student's paired t-test). The latency of these GABAergic IPSCs was 4.9±0.4 ms (ranging from 4.3 ms to 6.8 ms) and the decay time constant was 15.3±4.2 ms (ranging from 7.6 ms to 35.2 ms, n=6, see Fig. 4). In addition to establishing functional connectivity between orexin neurons and the DMV, we examined whether orexin neurons project to another important region of cardiac parasympathetic control – CVNs in the nucleus ambiguus. Although ChR2-EYFP fibers from orexin neurons were detectable within the nucleus ambiguus no direct electrophysiological responses were found in CVNs upon optogenetic stimulation of orexin fibers (n=17 cells). DISCUSSION To the best of our knowledge, this is the first report identifying and characterizing neurotransmission from orexin neurons to CVNs in the DMV. There are 2 major findings in this study. 1) Optogenetic stimulation of ChR2-EYFP-expressing orexin neurons in the lateral hypothalamus results in reliable excitation of these orexin neurons, observed both by action potential firing and large inward currents. 2) Optogenetic stimulation of ChR2-EYFP-expressing orexin fibers in the DMV evokes short-latency postsynaptic responses in 35% of CVNs tested in the DMV. Of the CVNs that responded to photostimulation of orexin fibers 63% received glutamatergic and 37% received GABAergic neurotransmission. Orexin has previously been shown to contribute substantially to central cardiovascular and respiratory regulation (Peyron et al., 1998, Ciriello and de Oliveira, 2003, Ciriello et al., 2003, Dergacheva et al., 2005, Iigaya et al., 2012, Carrive, 2013, Dergacheva et al., 2013). Intraventricular administration of orexin elicits tachycardia, hypertension and hyperventilation (Samson et al., 1999, Shirasaka et al., 1999, Chen et al., 2000, Matsumura et al., 2001, Zhang et al., 2005), while orexin knock-out animals have reduced blood pressure and diminished cardiovascular response to bicuculline injections in the hypothalamus (Kayaba et al., 2003). These data point to a sympathoexcitatory role of orexin, however microinjection of orexin into various regions of the central autonomic network has been shown to produce unexpected and differential effects. For instance, pressor and tachycardiac effects have been demonstrated after injections of orexin into medullary raphe area (Luong and Carrive, 2012), rostral ventrolateral medulla (Chen et al., 2000, Machado et al., 2002, Huang et al., 2010) and commissural nucleus of the solitary tract(Smith et al., 2002). However, in contrast, orexin administrated into caudal dorsolateral and medial subnuclei of the solitary tract elicits depressor and bradycardia responses (de Oliveira et al., 2003). Microinjection of orexin into the nucleus ambiguus, an area also involved in parasympathetic regulation of heart rate, results in a dose-related decrease in heart rate and activated arterial baroreflex (Ciriello and de Oliveira, 2003). The results from these previous studies, therefore, strongly point to involvement of orexin in the cardiovascular regulation, however these results suggest orexin may play complex differential (or even opposite) roles in different regions of the brain. One important target of orexin neurons are the CVNs in the DMV. The DMV is innervated by orexin-containing fibers (Peyron et al., 1998, Date et al., 1999) and superfusion of orexin depolarizes neurons in this nucleus (Hwang et al., 2001). However, the pathways and mechanisms underlying cardiac effects of orexin in CVNs in the DMV have not been previously established. The results from this study indicate that ChR2-EYFP expression in orexin neurons is sufficient for optogenetic stimulations of both soma of orexin neurons and their distal axons surrounding CVNs in the DMV. Photostimulation of each orexin neurons results in reliable action potential firing and large inward currents. Optogenetic stimulation of orexin neuron fibers in the brainstem evokes short-latency responses in CVNs in the DMV indicating a monosynaptic connection (Petreanu et al., 2007). These responses include both glutamatergic and GABAergic postsynaptic currents. A majority of the responses (63%) are mediated by glutamate release and postsynaptic AMPA receptor activation as these responses are completely blocked by application of the AMPA receptor antagonist CNQX (50 μM). These results provide an evidence that orexin neurons monosynaptically excite, via glutamatergic neurotransmission, CVNs in the DMV. Similar data have been reported that optogenetic stimulation of ChR2-containing fibers from orexin neurons elicits glutamatergic postsynaptic currents in histaminergic tuberomammillary neurons (Schone et al., 2012). In addition, our data support previous work from immunohistochemical studies indicating that orexin neurons use glutamate as their main neurotransmitter (Torrealba et al., 2003, Henny et al., 2010, Carrive, 2013). A minority (37%) of the responses include GABAergic currents as these responses are not abolished by CNQX, but blocked by application of gabazine (25 μM). In accordance with these data, GABA-like immunoreactivity has been found in 10-25% of orexin neurons (Apergis-Schoute et al., 2015). Thus, our data suggest there are different subpopulations of orexin neurons including those that differentially release glutamate and GABA, with excitatory and inhibitory inputs, respectively, to CVNs in the DMV. These subpopulations of orexin-containing cells could be differently influenced by various environmental stimuli and/or internal factors and may therefore play differential roles in parasympathetic control of cardiovascular function. Since the majority of the inputs from orexin neurons to CVNs are excitatory it is likely that activation of orexin cells would increase the activity of parasympathetic CVNs in the DMV and decrease heart rate. Orexin neuron inactivation, in turn, may lead to diminished cardioprotective parasympathetic activity from CVNs to the heart. Supporting this hypothesis, individuals with narcolepsy, the disease characterized by orexin neurons deficiency, have increased heart rate during wakefulness (Grimaldi et al., 2012, Sorensen et al., 2013). Similar, tachycardia has been reported in orexin-deficient mice (Bastianini et al., 2011, Silvani et al., 2014). In conclusion, this report is the first demonstration of heterogeneous projections from orexin neurons to CVNs in the DMV. This study elucidates the network mechanisms by which orexin neurons contribute to parasympathetic regulation of cardiovascular function. Further work is needed to understand the differences between subpopulations of orexin neurons and their pathways to CVNs and how they may serve disparate roles in cardiovascular regulation and cardiovascular abnormalities associated with narcolepsy. Acknowledgements Supported by NIH grant HL 72006 (D.M.) ABBREVIATIONS aCSF Artificial cerebrospinal fluid CVNs cardiac vagal neurons ChR2 channelrhodopsin-2 DMV dorsal motor nucleus of the vagus EYFP enhanced green fluorescent protein EPSCs excitatory postsynaptic currents IPSCs inhibitory postsynaptic currents NMDG 110 N-methyl-d-glucamine Figure 1 Localization of orexin-A immunoreactivity (red, left panel) with ChR2-EYFP (green, middle panel) driven by orexin-Cre selective expression in the lateral hypothalamus. Co-localization is shown in the right panel. Representative example of n=4 animals. Scale bar, 100 μm. Figure 2 Optogenetic stimulation of orexin cell bodies in the lateral hypothalamus evoked reliable action potentials (A) as well as large inward whole-cell currents (B) in all orexin neurons tested (n=8) in rats that received AAV1-ChR2-EYFP viral injections. In contrast, neither light-triggered action potential firing (C) nor large inward currents (D) were observed in animals that did not receive AAV1-ChR2-EYFP viral injections (n=5 orexin neurons). Black circles represent light pulses (3 ms, 1 Hz). Figure 3 Glutamatergic EPSCs in CVNs. Representative example of optogenetically-evoked glutamatergic postsynaptic current in an individual CVN shown in A, left. Abolishment of this this synaptic response by application of CNQX (50 μM) is demonstrated in A, right. Properties of glutamatergic synaptic events in CVNs are illustrated in B. Black circles represent average from 60 sweeps in individual CVNs (3 ms stimulation at 1 Hz) while horizontal bars are population means ± SEM. Amplitude, latency and decay time constant of the glutamatergic responses are shown from left to right. Figure 4 GABAergic IPSCs in CVNs. Representative example of optogenetically-evoked GABAergic postsynaptic current in an individual CVN shown in A, left. Abolishment of this this current by application of gabazine (25 μM) is demonstrated in A, right. Properties of GABAergic responses in CVNs are illustrated in B. Black circles represent average from 60 sweeps in individual CVNs (3 ms stimulation at 1 Hz) while horizontal bars are population means ± SEM. Amplitude, latency and decay time constant of the GABAergic responses are shown from left to right. Highlights ● Photoactivation of orexin neurons by channelrhodopsin-2 expression ● Orexin neurons monosynaptically project to brainstem cardiac vagal neurons ● Orexin cells release GABA and glutamate to modulate cardiac vagal neuron activity This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. REFERENCES Antunes VR Brailoiu GC Kwok EH Scruggs P Dun NJ Orexins/hypocretins excite rat sympathetic preganglionic neurons in vivo and in vitro. American journal of physiology Regulatory, integrative and comparative physiology 2001 281 R1801 1807 Apergis-Schoute J Iordanidou P Faure C Jego S Schone C Aitta-Aho T Adamantidis A Burdakov D Optogenetic evidence for inhibitory signaling from orexin to MCH neurons via local microcircuits. The Journal of neuroscience : the official journal of the Society for Neuroscience 2015 35 5435 5441 25855162 Bastianini S Silvani A Berteotti C Elghozi JL Franzini C Lenzi P Lo Martire V Zoccoli G Sleep related changes in blood pressure in hypocretin-deficient narcoleptic mice. Sleep 2011 34 213 218 21286242 Bouairi E Neff R Evans C Gold A Andresen MC Mendelowitz D Respiratory sinus arrhythmia in freely moving and anesthetized rats. J Appl Physiol (1985) 2004 97 1431 1436 15155710 Calaresu FR Pearce JW Effects on Heart Rate of Electrical Stimulation of Medullary Vagal Structures in the Cat. The Journal of physiology 1965 176 241 251 14286352 Carrive P Orexin, orexin receptor antagonists and central cardiovascular control. Frontiers in neuroscience 2013 7 257 24415993 Chen CT Hwang LL Chang JK Dun NJ Pressor effects of orexins injected intracisternally and to rostral ventrolateral medulla of anesthetized rats. American journal of physiology Regulatory, integrative and comparative physiology 2000 278 R692 697 Cheng Z Powley TL Schwaber JS Doyle FJ 3rd Projections of the dorsal motor nucleus of the vagus to cardiac ganglia of rat atria: an anterograde tracing study. The Journal of comparative neurology 1999 410 320 341 10414536 Ciriello J Calaresu FR Medullary origin of vagal preganglionic axons to the heart of the cat. Journal of the autonomic nervous system 1982 5 9 22 6173408 Ciriello J de Oliveira CV Cardiac effects of hypocretin-1 in nucleus ambiguus. American journal of physiology Regulatory, integrative and comparative physiology 2003 284 R1611 1620 Ciriello J Li Z de Oliveira CV Cardioacceleratory responses to hypocretin-1 injections into rostral ventromedial medulla. Brain research 2003 991 84 95 14575880 Date Y Ueta Y Yamashita H Yamaguchi H Matsukura S Kangawa K Sakurai T Yanagisawa M Nakazato M Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proceedings of the National Academy of Sciences of the United States of America 1999 96 748 753 9892705 de Oliveira CV Rosas-Arellano MP Solano-Flores LP Ciriello J Cardiovascular effects of hypocretin-1 in nucleus of the solitary tract. American journal of physiology Heart and circulatory physiology 2003 284 H1369 1377 12531738 DePuy SD Stornetta RL Bochorishvili G Deisseroth K Witten I Coates M Guyenet PG Glutamatergic neurotransmission between the C1 neurons and the parasympathetic preganglionic neurons of the dorsal motor nucleus of the vagus. The Journal of neuroscience : the official journal of the Society for Neuroscience 2013 33 1486 1497 23345223 Dergacheva O Chronic intermittent hypoxia alters neurotransmission from lateral paragigantocellular nucleus to parasympathetic cardiac neurons in the brain stem. Journal of neurophysiology 2015 113 380 389 25318765 Dergacheva O Boychuk CR Mendelowitz D Developmental changes in GABAergic neurotransmission to presympathetic and cardiac parasympathetic neurons in the brainstem. Journal of neurophysiology 2013 110 672 679 23657280 Dergacheva O Dyavanapalli J Pinol RA Mendelowitz D Chronic intermittent hypoxia and hypercapnia inhibit the hypothalamic paraventricular nucleus neurotransmission to parasympathetic cardiac neurons in the brain stem. Hypertension 2014 64 597 603 24958501 Dergacheva O Wang X Huang ZG Bouairi E Stephens C Gorini C Mendelowitz D Hypocretin-1 (orexin-A) facilitates inhibitory and diminishes excitatory synaptic pathways to cardiac vagal neurons in the nucleus ambiguus. The Journal of pharmacology and experimental therapeutics 2005 314 1322 1327 15947034 Dergacheva O Yamanaka A Schwartz AR Polotsky VY Mendelowitz D Hypoxia and hypercapnia inhibit hypothalamic orexin neurons in rats. Journal of neurophysiology jn 2016 00196 02016 Grimaldi D Calandra-Buonaura G Provini F Agati P Pierangeli G Franceschini C Barletta G Plazzi G Montagna P Cortelli P Abnormal sleep-cardiovascular system interaction in narcolepsy with cataplexy: effects of hypocretin deficiency in humans. Sleep 2012 35 519 528 22467990 Henny P Brischoux F Mainville L Stroh T Jones BE Immunohistochemical evidence for synaptic release of glutamate from orexin terminals in the locus coeruleus. Neuroscience 2010 169 1150 1157 20540992 Huang SC Dai YW Lee YH Chiou LC Hwang LL Orexins depolarize rostral ventrolateral medulla neurons and increase arterial pressure and heart rate in rats mainly via orexin 2 receptors. The Journal of pharmacology and experimental therapeutics 2010 334 522 529 20494957 Hwang LL Chen CT Dun NJ Mechanisms of orexin-induced depolarizations in rat dorsal motor nucleus of vagus neurones in vitro. The Journal of physiology 2001 537 511 520 11731582 Iigaya K Horiuchi J McDowall LM Lam AC Sediqi Y Polson JW Carrive P Dampney RA Blockade of orexin receptors with Almorexant reduces cardiorespiratory responses evoked from the hypothalamus but not baro- or chemoreceptor reflex responses. American journal of physiology Regulatory, integrative and comparative physiology 2012 303 R1011 1022 Kayaba Y Nakamura A Kasuya Y Ohuchi T Yanagisawa M Komuro I Fukuda Y Kuwaki T Attenuated defense response and low basal blood pressure in orexin knockout mice. American journal of physiology Regulatory, integrative and comparative physiology 2003 285 R581 593 Luong LN Carrive P Orexin microinjection in the medullary raphe increases heart rate and arterial pressure but does not reduce tail skin blood flow in the awake rat. Neuroscience 2012 202 209 217 22178985 Machado BH Bonagamba LG Dun SL Kwok EH Dun NJ Pressor response to microinjection of orexin/hypocretin into rostral ventrolateral medulla of awake rats. Regulatory peptides 2002 104 75 81 11830280 Mastitskaya S Marina N Gourine A Gilbey MP Spyer KM Teschemacher AG Kasparov S Trapp S Ackland GL Gourine AV Cardioprotection evoked by remote ischaemic preconditioning is critically dependent on the activity of vagal pre-ganglionic neurones. Cardiovascular research 2012 95 487 494 22739118 Matsumura K Tsuchihashi T Abe I Central orexin-A augments sympathoadrenal outflow in conscious rabbits. Hypertension 2001 37 1382 1387 11408381 Nosaka S Yamamoto T Yasunaga K Localization of vagal cardioinhibitory preganglionic neurons with rat brain stem. The Journal of comparative neurology 1979 186 79 92 457932 Petreanu L Huber D Sobczyk A Svoboda K Channelrhodopsin-2-assisted circuit mapping of long-range callosal projections. Nature neuroscience 2007 10 663 668 17435752 Peyron C Tighe DK van den Pol AN de Lecea L Heller HC Sutcliffe JG Kilduff TS Neurons containing hypocretin (orexin) project to multiple neuronal systems. The Journal of neuroscience : the official journal of the Society for Neuroscience 1998 18 9996 10015 9822755 Pinol RA Bateman R Mendelowitz D Optogenetic approaches to characterize the long-range synaptic pathways from the hypothalamus to brain stem autonomic nuclei. Journal of neuroscience methods 2012 210 238 246 22890236 Pinol RA Jameson H Popratiloff A Lee NH Mendelowitz D Visualization of oxytocin release that mediates paired pulse facilitation in hypothalamic pathways to brainstem autonomic neurons. PloS one 2014 9 e112138 25379676 Samson WK Gosnell B Chang JK Resch ZT Murphy TC Cardiovascular regulatory actions of the hypocretins in brain. Brain research 1999 831 248 253 10412003 Schoenenberger P Scharer YP Oertner TG Channelrhodopsin as a tool to investigate synaptic transmission and plasticity. Experimental physiology 2011 96 34 39 20562296 Schone C Cao ZF Apergis-Schoute J Adamantidis A Sakurai T Burdakov D Optogenetic probing of fast glutamatergic transmission from hypocretin/orexin to histamine neurons in situ. The Journal of neuroscience : the official journal of the Society for Neuroscience 2012 32 12437 12443 22956835 Shirasaka T Nakazato M Matsukura S Takasaki M Kannan H Sympathetic and cardiovascular actions of orexins in conscious rats. The American journal of physiology 1999 277 R1780 1785 10600926 Silvani A Bastianini S Berteotti C Cenacchi G Leone O Lo Martire V Papa V Zoccoli G Sleep and cardiovascular phenotype in middle-aged hypocretin-deficient narcoleptic mice. Journal of sleep research 2014 23 98 106 24033681 Smith PM Connolly BC Ferguson AV Microinjection of orexin into the rat nucleus tractus solitarius causes increases in blood pressure. Brain research 2002 950 261 267 12231252 Sorensen GL Knudsen S Petersen ER Kempfner J Gammeltoft S Sorensen HB Jennum P Attenuated heart rate response is associated with hypocretin deficiency in patients with narcolepsy. Sleep 2013 36 91 98 23288975 Sullivan JM Connors NA An orthograde labelling study of axons of the dorsal motor nucleus of the vagus to rat cardiac ganglia as demonstrated by autoradiography. Acta anatomica 1981 110 128 135 6174013 Torrealba F Yanagisawa M Saper CB Colocalization of orexin a and glutamate immunoreactivity in axon terminals in the tuberomammillary nucleus in rats. Neuroscience 2003 119 1033 1044 12831862 Wang X Pinol RA Byrne P Mendelowitz D Optogenetic stimulation of locus ceruleus neurons augments inhibitory transmission to parasympathetic cardiac vagal neurons via activation of brainstem alpha1 and beta1 receptors. The Journal of neuroscience : the official journal of the Society for Neuroscience 2014 34 6182 6189 24790189 Zhang W Fukuda Y Kuwaki T Respiratory and cardiovascular actions of orexin-A in mice. Neuroscience letters 2005 385 131 136 15941620

PMC005xxxxxx/PMC5118115.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9214478 2614 Mol Ecol Mol. Ecol. Molecular ecology 0962-1083 1365-294X 27696597 5118115 10.1111/mec.13872 NIHMS820626 Article The fungal cultivar of leaf-cutter ants produces specific enzymes in response to different plant substrates Khadempour Lily 123 Burnum-Johnson Kristin E. 4 Baker Erin S. 4 Nicora Carrie D. 4 Webb-Robertson Bobbie-Jo M. 4 White Richard A. III 4 Monroe Matthew E. 4 Huang Eric L. 4 Smith Richard D. 4 Currie Cameron R. 13* 1 Department of Bacteriology, University of Wisconsin-Madison, Madison, WI, USA, 53706 2 Department of Zoology, University of Wisconsin-Madison, Madison, WI, USA, 53706 3 Department of Energy Great Lakes Bioenergy Research Center, University of Wisconsin-Madison, Madison, WI, USA, 53706 4 Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA, USA, 99352 * Corresponding author: CR Currie, Department of Bacteriology, University of Wisconsin-Madison, 6155 Microbial Sciences Building, 1550 Linden Drive, Madison, WI 53706, USA Fax: (608) 262-9865 currie@bact.wisc.edu 9 10 2016 26 10 2016 11 2016 01 11 2017 25 22 57955805 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Herbivores use symbiotic microbes to help derive energy and nutrients from plant material. Leaf-cutter ants are a paradigmatic example, cultivating their mutualistic fungus Leucoagaricus gongylophorus on plant biomass that workers forage from a diverse collection of plant species. Here, we investigate the metabolic flexibility of the ants’ fungal cultivar for utilizing different plant biomass. Using feeding experiments and a novel approach in metaproteomics, we examine the enzymatic response of L. gongylophorus to leaves, flowers, oats, or a mixture of all three. Across all treatments, our analysis identified and quantified 1,766 different fungal proteins, including 161 putative biomass-degrading enzymes. We found significant differences in the protein profiles in the fungus gardens of sub-colonies fed different plant substrates. When provided with leaves or flowers, which contain the majority of their energy as recalcitrant plant polymers, the fungus gardens produced more proteins predicted to break down cellulose: endoglucanase, exoglucanase, and β-glucosidase. Further, the complete metaproteomes for the leaves and flowers treatments were very similar, while the mixed substrate treatment closely resembled the treatment with oats alone. This indicates that when provided a mixture of plant substrates, fungus gardens preferentially break down the simpler, more digestible substrates. This flexible, substrate-specific enzymatic response of the fungal cultivar allows leaf-cutter ants to derive energy from a wide range of substrates, which likely contributes to their ability to be dominant generalist herbivores. Introduction Herbivores are the most abundant and diverse animals on earth (Ricklefs & Miller 2000). Their success is shaped, at least in part, by different animal lineages evolving to specialize on different plant species and plant parts, each of which provide different barriers for herbivores to access stored carbon and other nutrients (Hansen & Moran 2013). Arguably, the most important strategy herbivores use to contend with these barriers to consumption is establishing symbiotic associations with microbes that broaden their physiological capacity (Dowd 1991). The microbial mediation of herbivory has been studied at length in substrate-specialized herbivore systems. Microbial symbionts, which include bacteria, fungi and other microorganisms, mediate herbivory in three main ways: helping their hosts overcome recalcitrant plant material, supplementing nutrient-poor diets, and reducing the impact of plant defense compounds (Hansen & Moran 2013). For example, termites break down the highly recalcitrant biomass in wood through their association with both eukaryotic and bacterial symbionts (Tartar et al. 2009). The plant sap feeding aphids house intracellular Buchnera aphidicola that compensate for the absence of essential amino acids in their diet (Hansen & Moran 2011). Finally, when attacking trees the mountain pine beetle vectors fungi and bacteria, which break down terpenes that would otherwise be toxic to the developing larvae that specialize on tree phloem as a food source (Wang et al. 2012; Boone et al. 2013). Unlike most herbivores, leaf-cutter ants are polyphagous, meaning that they occupy a generalist herbivore niche. These dominant herbivores belong to two genera, Acromyrmex and Atta, and forage on 2–17% of all the foliar biomass in some ecosystems in the Neotropics (Herz et al. 2007; Costa et al. 2008). Their success as herbivores can be attributed to their obligate mutualism with a fungus, Leucoagaricus gongylophorus, which they cultivate for food: they provide the fungus with leaf material and, in turn, the fungus provides specialized hyphal swellings called gongylidia, which the ants feed on (Holldobler & Wilson 1990; Mayhé-Nunes & Jaffe 1998; Holldobler & Wilson 2008). The types of plant material that a colony consumes depends on the ant species, the location, and the season in which the colony is observed (De Vasconcelos 1990; Wirth 2003). In general, they tend toward young leaves with soft cuticles, less-toxic plant defense compounds, fewer trichomes, fewer endophytes and higher nutritional value (Howard 1987; 1988; Van Bael et al. 2011). Within these constraints, leaf-cutter ants incorporate many different types of plants into their fungus gardens and have been observed foraging at least 20 different species of plants over three days (Wirth et al. 1997). Ants also incorporate a variety of plant parts into their gardens such as leaves, flowers, seeds, and fruit parts in the wild, and oats and parboiled rice in laboratory settings (Wirth et al. 1997; Kooij et al. 2011). Leaf-cutter ants tend to their mutualistic fungus in gardens, which can be viewed as an ‘external gut’. These gardens contain both the fungus itself and a low diversity community of bacteria. Through enzymatic, metagenomic and metaproteomic analyses, the microbial communities in the fungus gardens of leaf-cutter ants Atta sexdens and Atta cephalotes have been explored. Many fungal amylases (Silva et al. 2006b), pectinases (Silva et al. 2006a), carbohydrate-active enzymes (CAZy), fungal oxidative lignin enzymes (FOLy), and secreted proteases have been identified (Aylward et al. 2012; 2013a), demonstrating that the fungus in this system is primarily responsible for the breakdown of plant biomass. The bacterial community in the fungus gardens was identified using isolation, metagenomics and 16S sequencing (Suen et al. 2010; Aylward et al. 2012). While the bacterial community has the genetic capacity for biomass degradation (Suen et al. 2010), there is not yet evidence that this is actually occurring in the gardens. In this study, we explore microbial mediation in a generalist herbivore by combining feeding experiments with metaproteomic analyses. Specifically, we fed sub-colonies of leaf-cutter ants leaves, flowers, oats or a mixture of all three. Using a novel multidimensional platform, coupling liquid chromatography, ion mobility spectrometry and mass spectrometry (LC-IMS-MS), we determined the metaproteomic response of fungus gardens on the different diets. Our working hypothesis is that the fungal cultivar L. gongylophorus responds to different plant substrates integrated into the garden by worker ants by producing specific proteins that have the capacity to break down the substrate provided. Methods Experimental design Atta cephalotes fungus gardens were excised from colonies excavated in the secondary tropical moist forest surrounding the Smithsonian Tropical Research Institute (STRI) Gamboa research station in Panama between Dec. 27, 2012 and Jan 10, 2013. Five mature colonies were excavated. Since lab-reared sub-colonies without queens are unstable, five fungus chambers were excised from each colony to ensure that we would have sufficient numbers of replicates for proteomics. These fungus chambers were split into four sub-colonies each and were contained within a plastic container (10×10×8 cm) that was kept in a larger plastic container (14×19×9 cm). Care was taken to minimize disturbance to the fungus gardens and to ensure that a relatively even number of workers were distributed to each sub-colony. Each sub-colony was randomly assigned to one of four feeding treatments, and received different plant biomass to use as substrate for cultivating their fungal mutualist. The four feeding treatments were Lagerstroemia speciosa L. leaves, Hibiscus rosa-sinensis flowers, Quaker instant oatmeal, or a mixture of all three (Figure 1). The substrates that were selected were all readily available and were readily incorporated into the gardens by the ants, but they varied in terms of their energy availability. Leaves are the most recalcitrant substrate of the three. The flowers are similar to leaves in terms of cell wall structures but are more easily digestible (Amaglo et al. 2010). The oats are highly processed and have the most accessible energy in the form of sugars and starches (Cuddeford 1995; Welch 1995). The flowers and leaves were collected daily from plants in the immediate vicinity in Gamboa. The sub-colonies were fed ad libitum, typically every one or two days, depending on how quickly the ants would incorporate new substrate. The colonies were maintained at ambient temperature and humidity. After 15 days, the entire fungus garden from each sub-colony was frozen in PBS buffer at −20°C in a 50 mL conical tube, in preparation for further processing. One of the five colonies was excluded from metaproteomic analysis because it did not have surviving sub-colonies from all treatments but it was included it in the survivorship analysis. From the surviving sub-colonies we selected 16 samples for metaproteomics (four treatments and four colony replicates each). The sub-colonies that were selected for metaproteomics were all active and still incorporating new material into their gardens at the end of the 15 days of the experiment. Mass spectrometry instrumentation Analysis of the trypsin-digested peptide mixtures (Supplemental Methods) from the gardens was performed on both a Thermo Fisher Scientific LTQ Orbitrap mass spectrometer (MS) (San Jose, CA, USA) operated in tandem MS (MS/MS) mode and an in-house built ion-mobility MS (IMS-MS) instrument that couples a 1-m ion mobility drift cell (Baker et al. 2007; 2010) with an Agilent 6224 time-of-flight (TOF) MS that was upgraded to have a 1.5 m flight tube for resolution around 25,000. The same fully automated in-house built 2-column HPLC system (Livesay et al. 2008) equipped with in-house packed capillary columns was used for both instruments with mobile phase A consisting of 0.1% formic acid in water and B comprised of 0.1% formic acid in acetonitrile. A 100 min LC separation was performed on the Velos MS (using 60-cm long columns having an o.d. of 360 µm, i.d. of 75 µm, and 3 µm C18 packing material) while only a 60 min gradient with shorter columns (30-cm long columns with the same dimensions and packing) that was used with the IMS-MS since the additional IMS separation helps address detector suppression and also faster LC analyses. Both gradients were linear with mobile phase B increasing from 0 to 60% until the final 2 min of the run when B was purged at 95%. 5 µL of each sample was injected for both analyses and the HPLC was operated under a constant flow rate of 0.4 µL/min for the 100 min gradient and 1 µL/min for the 60 min gradient. The Velos MS data was collected from 400–2000 m/z at a resolution of 60,000 (automatic gain control (AGC) target: 1×106) followed by data dependent ion trap MS/MS spectra (AGC target: 1×104) of the twelve most abundant ions using a collision energy setting of 35%. A dynamic exclusion time of 60 s was used to discriminate against previously analyzed ions. IMS-TOF MS data was collected from 100–3200 m/z. Metaproteomic data processing and statistical analysis Identification and quantification of the detected peptide peaks were performed using the accurate mass and time (AMT) tag approach (Zimmer et al. 2006; Burnum et al. 2012). Peptide database generation utilized Velos tandem MS/MS data (Kim et al. 2008; Piehowski et al. 2013) from pooled fractionated samples (Supplemental Methods). Due to the greater sensitivity and dynamic range of measurements (Burnum et al. 2012) relative quantitation of the peptide peaks utilized the LC-IMS-MS data. Multiple in-house developed (Monroe et al. 2007; Jaitly et al. 2009) informatics tools were used to process the LC-IMS-MS data and correlate the resulting LC-IMS-MS features to the AMT tag database containing LC elution times, IMS drift times, and accurate mass information for each assigned peptide. Our in-house ion mobility mass spectrometry platform has previously provided novel insight into complex biological systems (Burnum et al. 2012; Baker et al. 2014; Cha et al. 2015; Baker et al. 2015; Kyle et al. 2016). Data filtering was performed to remove peptides with inadequate data for statistics and samples that are extreme outliers (Webb-Robertson et al. 2010; Matzke et al. 2011). This resulted in 6,676 peptides and 1,766 proteins across the sixteen samples (four feeding treatments and four biological replicates for each treatment). Normalization approaches were evaluated using a statistical procedure for the analyses of peptide abundance normalization strategies (SPANS) and normalization factors were generated as the mean of the datasets that were observed consistently across technical replicates (Webb-Robertson et al. 2011). Peptide statistics were performed by comparing all treatment groups to one another using Analysis of Variance (ANOVA) with a post-hoc Tukey test to define peptide signatures. A BP-Quant quantification (Webb-Robertson et al. 2014) approach was used to estimate abundance at the protein level. Proteins were also evaluated with a Tukey test and deemed significant at a p-value<0.05. Only fungal proteins identified by ≥ 2 peptides are discussed (see Supplemental Table 1 for the full list of all detected proteins). Non-metric multidimensional scaling (NMDS) was conducted on these data with Bray-Curtis dissimilarity, using the vegan package in the R statistical programming environment (Oksanen et al. 2013; R Core Team 2013). To determine if the fungus gardens from different treatments had significantly different protein profiles, function adonis was used to run a Permutational Multivariate Analysis of Variance Using Distance Matrices (PERMANOVA). Results Fungal proteomics With our metaproteomic analysis of the fungus gardens, we identified and quantified 1,766 different fungal proteins, including 161 putative biomass-degrading enzymes (Supplemental Table 1). NMDS analysis of the global proteome profiles across treatments and replicates revealed grouping according to treatment (Figure 2A). These differences according to treatment were significant (PERMANOVA p<0.001). Fungus garden proteomic profiles in both the leaves and flowers treatments showed low variability within-group and between-group, while the oats and mixed treatments had greater within-group variability and overlapped with each other. These groupings are evident when individual proteins are compared between treatments. To analyze the differential abundance of individual proteins, we conducted pair-wise comparisons of each protein in the four treatments. Numerous proteins with significantly different abundances were identified between the treatments (Supplemental Table 1). When individual protein differences are observed globally using heat maps, we can again see grouping according to treatment (Figures 3 and 4): the oats sub-colonies were most similar to the mixed sub-colonies, while the leaves sub-colonies were similar to the flowers. The significant changes for each protein pairwise comparison were identified by at least 2 peptides with: oats/mixed having 52 significantly changing proteins, leaves/flowers - 31, leaves/oats - 286, flowers/oats - 259, leaves/mixed - 135, and flowers/mixed - 125 (Supplemental Table 1). All biomass-degrading enzymes observed to be significantly different (p<0.05) between treatments are listed in Table 1, where individual proteins are compared between the mixed and other treatments. We compared to the mixed treatment since it most closely resembles the ants’ natural tendency to incorporate a mixture of substrates into their fungus gardens. In general, the leaves and flowers treatments had similar results with much higher abundances of CAZys, proteases and enzymes necessary for the breakdown of cellulose: endoglucanases (GH5 and GH6), exoglucanase (GH6), and β-glucosidases (GH3 and GH31), compared with the other two treatments. However, the oats treatment was very similar to the mixed treatment with a lower abundance of these proteins and proteases (Table 1, Figure 4). Bacterial proteomics We detected only 44 unique bacterial peptides and from these data we determined, through similar pairwise comparisons between treatments, that there were three bacterial proteins that differed significantly between treatments. Each of these proteins was identified with only a single peptide. These proteins were identified based on genomes of bacterial symbionts of leaf-cutter ants (Enterobacter strain FGI 35, Serratia strain FGI 94 (Aylward et al. 2013c), Enterobacteriaceae strain FGI 57 (Aylward et al. 2013b), Pseudomonas strain FGI 182, Klebsiella variicola strain AT-22 and Pantoea strain AT-9b (Aylward et al. 2014)). Malate dehydrogenase, which mapped equally to Cronobacter, Pantoea, Serratia, Enterobacter, and Klebsiella genomes, was more abundant in the leaf treatment. Periplasmic trehalase, which mapped to the Enterobacter genome, was more abundant in the flower treatments. ATP synthase subunit β, which mapped to all six bacterial genomes, was the least abundant in the leaf treatments. Overall, the global bacterial protein profiles did not differ between treatments (Supplemental Table 2, Figure 2B). Sub-colony Survivorship The fungus garden of some sub-colonies did not remain healthy throughout the experimental period, but instead dried out, were discarded by workers, or were overgrown by a pathogen. This was especially common for sub-colonies created from the gardens excised from the last two parent colonies. A sub-colony was considered failed when all the ants were dead or when the fungus garden was overtaken by a pathogen. Overall, sub-colonies fed exclusively on oats had significantly lower survivorship than the other colonies (Figure 5). Discussion The breakdown of plant biomass by L. gongylophorus is central to the success of leaf-cutter ant colonies and the function of this ant-fungus mutualism. Nevertheless, our understanding of the process of digesting leaves and other plant substrates within the fungus garden is limited. Specifically, the ability of L. gongylophorus to digest cellulose and other recalcitrant material has been debated. Some have argued that it does not effectively break down cellulose and instead relies on other plant components such as pectin for energy (De Siqueira et al. 1998; Silva et al. 2006a; Moller et al. 2011). In contrast to this, sugar composition analysis and microscopy shows a significant decrease in cellulose within fungus gardens and genomics and metaproteomics show a significant capacity of L. gongylophorus to degrade it (Suen et al. 2010; Nagamoto et al. 2011; Aylward et al. 2012; Grell et al. 2013; Aylward et al. 2013a). Our results here provide further support for the role of the fungus in recalcitrant biomass degradation. Specifically, our metaproteomic analysis detected 100 CAZys produced by L. gongylophorus, including 53 glycoside hydrolases (GH), 6 carbohydrate esterases (CE), 8 carbohydrate binding molecules (CBM), 4 polysaccharide lyases (PL), and 30 auxiliary activities enzymes (AA) (Figure 4, Supplementary Table 1). This suite of enzymes includes all the components necessary for the breakdown of cellulose (endoglucanases GH5, GH12 and GH6, exoglucanase GH6 and β-glucosidase GH31). Although our combination of proteomics and feeding experiments provide further evidence for the ability of L. gongylophorus to deconstruct cellulose, our findings indicate that this enzymatic response is context-dependent. Specifically, we found metabolic flexibility in the ants’ fungal cultivar to preferentially digest various substrates; instead of consuming recalcitrant materials, the fungus digests the more readily accessible carbon sources when available. This is most clearly observed when comparing the mixed and oat treatment metaproteomes. In the mixed treatment the fungus does not produce an abundance of biomass-degrading enzymes, despite the presence of recalcitrant biomass. It instead has a metaproteome that is more similar to that of the oat treatment, suggesting that when given a mixture of substrates, the fungus derives its energy from the oats. The flexible, substrate-specific response of the fungus is important in a system where the ants cut a large diversity of substrates, which vary between seasons and environments. For example, in the dry season substrates that are rich in easily accessible nutrients may be more limited, such that the fungal cultivar needs to respond to and to derive energy from more recalcitrant sources. In contrast, in the wet season when substrates such as fruits and young leaves are more readily available, the fungal cultivar would benefit from reducing the energy expended on digesting recalcitrant material when easily accessible sugars are available. Evidence supporting the substrate-specific response in the leaf-cutter ant fungus garden has been previously reported elsewhere. Kooij et al. (2011) manipulated the substrate for A. cephalotes fungus gardens and using Azurine-Crosslinked (AZCL) assays measured changes in specific enzymes of interest, observing an overall shift in enzyme activity between substrates. AZCL is a high throughput method used to detect enzyme activity, while metaproteomics provides accurate detection and quantification of the specific proteins present. Thus, our approach represents a more thorough enzymatic response of the fungus garden, as follows. First, AZCL is conducted with a limited suite of substrates and only shows activity of enzymes to those substrates. This excludes any non-enzymatic proteins and any enzymes that did not have the appropriate substrate to respond to. Second, AZCL does not allow us to characterize specific proteins, whereas metaproteomics does.. Other systems where microbes are responsible for biomass breakdown also show substrate-specificity through fluctuations in the community structure of multiple microbes (Thoetkiattikul et al. 2013; Miyata et al. 2014). Here, a single vertically transmitted cultivar, with little variability between isolates (Silva-Pinhati et al. 2004) is responsible for the flexible, substrate-specific response of the system. The leaf-cutter ant system, which is optimized for the extraction of energy from plant material then fine-tunes the enzymatic response of the fungal cultivar. Previous work has shown that the lignocellulases and laccases from gongylidia are transferred by the ants from the middle of the garden and defecated on the top, serving as a pretreatment step for beginning rapid biomass degradation and detoxification (Cherrett et al. 1989; Moller et al. 2011; De Fine Licht et al. 2013; Aylward et al. 2015). Recent work has identified the presence of an apparent consistent bacterial community in the fungus garden (Pinto-Tomás et al. 2009; Suen et al. 2010; Aylward et al. 2012). Although certain functional roles of the bacteria have been elucidated, such as nitrogen fixation (Pinto-Tomás et al. 2009) and the apparent capacity to provide vitamins (Aylward et al. 2012), our insights regarding the bacteria remain limited. Here, we did not observe a notable change in bacterial proteins, other than the three which are all part of central carbon metabolism and unlikely to play a direct role in substrate breakdown or detoxification (Bergmeyer & Gawehn 1974; Boos et al. 1987). Only 1% of the unique peptides that were detected in these analyses were identified as bacterial. This is likely due to a considerable difference in the amount of fungal and bacterial biomass in the fungus gardens. It could also indicate that bacteria play a more limited role in the fungus gardens. Interestingly, despite our finding that L. gongylophorus preferentially uses the simplest energy source (i.e., oats) when provided with a mixture of substrates, sub-colony survivorship dramatically decreased when this was the only substrate provided. This correlation between decreased health and feeding exclusively on a simple, energy rich diet has been observed in other animals. Cows that are fed a grain-rich diet gain weight quickly but suffer frequently from ruminal acidosis, which negatively impacts both production and animal welfare (Krause & Oetzel 2006). Ruminal acidosis results from different rates of fermentation in the standard grassy diet and has effects on the microbial community composition in the rumen (Steele et al. 2011; Hook et al. 2011). Humans also show a correlation between diet, the gut microbiome, and health (De Filippo et al. 2010; Martínez et al. 2013). While this experiment suggests that the fungus gardens of oat-fed sub-colonies are apparently less stable, colony health was not the focus of our study. However, we hypothesize that an exclusive diet of oats lacks required micronutrients that the ants, fungus or bacteria obtain from fresh plant material. While there have been thorough investigations into plant characteristics that are deterrents to leaf-cutter ant foraging and how this limits the diversity of plants they consume, no work has been done investigating whether a more diverse diet leads to higher fitness for leaf-cutter ants. Testing this hypothesis in future studies would help us to determine what minimum requirements exist for leaf-cutter ant forage and whether this is achieved more effectively with a diverse diet. The mutualism between leaf-cutter ants and their fungal cultivar has been described as an “unholy alliance” (Cherrett et al. 1989), where the tasks of mechanical and enzymatic breakdown of plant material are partitioned to the ants and fungal cultivar, respectively. Through this alliance, leaf-cutter ants are capable of utilizing a wide diversity of plant material, unlike most other herbivores. Polyphagy in this system necessitates metabolic flexibility on the part of the fungus, and is a key factor in making leaf-cutter ants dominant herbivores. In this study, we dissect this unholy alliance at a previously unattainable depth, demonstrating that the cultivar does indeed have a flexible, specific response to different plant substrates. Our study provides an important step in building toward understanding the microbial mediation of a generalist herbivore system. Supplementary Material Supp Methods Supp Table S1 Supp Table S2 This work was funded in part by the Department of Energy Great Lakes Bioenergy Research Center (DOE Office of Science BER DE-FC02-07ER64494) with support for LK and CRC. Proteomics measurements were supported by the DOE, Office of Biological and Environmental Research, Genomic Science Program under the Pacific Northwest National Laboratory (PNNL) Pan-omics Program, and were performed in the Environmental Molecular Science Laboratory, a U.S. DOE national scientific user facility at PNNL in Richland, WA. Battelle operates PNNL for the DOE under contract DE-AC05-76RLO01830. This work was supported in part by grants NIEHS/NIH (R01ES022190) to ESB, HHS NCI/NIH (U01CA184783-01) to BJWR. Figure 1 Leaf cutter ants carrying various substrates (A) a leaf, (B) a flower and (C) an oat. Ants tending to their fungus garden with newly incorporated leaf material (D) (photographs by Don Parsons). Figure 2 NMDS plot of (A) fungal and (B) bacterial whole-community metaproteomics. While the fungal results were significantly different between treatment groups, the bacterial metaproteomes were not possible to differentiate statistically. Figure 3 A heat map of the complete metaproteome. Columns represent each treatment and rows represent each protein. A clear division is visible between the two left columns (leaves and flowers) and the two right columns (oats and mixed). Figure 4 Heat map of higher or lower abundance of biomass degrading enzymes. A clear division can be seen between leaves and flowers on the left and oats and mixed on the right. GH – glycoside hydrolases, CE – carbohydrate esterases, CBM – carbohydrate binding molecules, PL – polysaccharide lyases, AA – auxiliary activities. Proteins in red text were significantly different between at least two treatments. Figure 5 Sub-colony survival by treatment. Sub-colonies that were fed oats survived significantly (*) less than the other sub-colonies, over the course of the experiment (ANOVA p<0.05). Table 1 Fungal biomass-degrading enzymes that differ significantly from the mixed treatment LAG Protein Family Annotation Leaves Flowers Oats CAZy 1450 CE8 Pectin methylesterase − 925 CBM57, CE15 Found attached to glycosidases − 1065 GH31 α-glucosidase, and others − 2832 GH6 Endoglucanase, exoglucanase, cellobiohydrolase + 1778 CBM32 Binding to galactose, lactose, polygalacturonic acid, LacNAc + − 4224 GH10 Xylan targeting + 3545 GH5 Endo-β-1,4-glucanase / cellulase and many others + 3581 CE5 Acetyl xylan esterase, cutinase + + 3843 GH10 Xylan targeting + + 830 GH105 Unsaturated rhamnogalacturonyl hydrolase; d- 4,5-unsaturated β-glucuronyl hydrolase + + 420 GH18 Lysozyme, chitinase, many others + + 5098 GH3 β-glucosidase, and others + + 811 GH3 β-glucosidase, and others + + 1724 GH31 α-glucosidase, and others + + 1811 GH92 Mannose targeting − − 11012 AA5 Glyoxal oxidase − 3543 AA5 Glyoxal oxidase − 1590 AA3 Glucose oxidase − − 3638 AA3 Alcohol oxidase 1 + + 2639 AA1 Laccase-1 + + 3464 AA1 Laccase-4 + + 5297 AA1 Laccase-2 + 2404 AA1 Laccase-1 + 5522 AA1 Laccase-2 + 3730 AA2 Chloroperoxidase + 5105 AA2 Chloroperoxidase − − 3594 AA3 Dihydrolipoyl dehydrogenase, mitochondrial + Proteases 3716 M36 Endopeptidase − 971 C44 Self-processing precursor of Amidophosphoribosyltransferase − 2519 M67A Isopeptidases that releases ubiquitin from ubiquitinated proteins + 3036 C01B Endopeptidases or exopeptidases + 3725 M28E Aminopeptidase − 439 A01A Pepsin A + + 100 M03A Thimet oligopeptidase + + 748 M13 Metalloendopeptidase + + 1996 M41 ATP-dependent metalloendopeptidase + + 15046 M67A Isopeptidases that release ubiquitin from ubiquitinated proteins − − 3735 S08A Subtilisin Carlsberg + + 2389 S08A Subtilisin Carlsberg + 3512 S08A Subtilisin Carlsberg + − 5096 S08A Subtilisin Carlsberg + − 2939 S10 Carboxypeptidase Y + + 4473 S10 Carboxypeptidase Y − 2743 S10 Carboxypeptidase Y − − − 924 S26B Signalase 21 kDa component + + + 2527 S53 Sedolisin + + A significant increase in abundance compared to the mixed treatment is indicated by + and a significant decrease is indicated by −. Data accessibility All of the metaproteomic data from this study is available in the Supplemental Materials (Supplemental tables 1 and 2). References Amaglo NK Bennett RN Curto Lo RB Profiling selected phytochemicals and nutrients in different tissues of the multipurpose tree Moringa oleifera L., grown in Ghana Food Chemistry 2010 122 1047 1054 Aylward FO Burnum KE Scott JJ Metagenomic and metaproteomic insights into bacterial communities in leaf-cutter ant fungus gardens The ISME Journal 2012 6 1688 1701 22378535 Aylward FO Burnum-Johnson KE Tringe SG Leucoagaricus gongylophorus produces diverse enzymes for the degradation of recalcitrant plant polymers in leaf-cutter ant fungus gardens Applied and Environmental Microbiology 2013a 79 3770 3778 23584789 Aylward FO Khadempour L Tremmel DM Enrichment and Broad Representation of Plant Biomass-Degrading Enzymes in the Specialized Hyphal Swellings of Leucoagaricus gongylophorus, the Fungal Symbiont of Leaf-Cutter Ants PLoS ONE 2015 10 e0134752 26317212 Aylward FO Suen G Biedermann PHW Convergent Bacterial Microbiotas in the Fungal Agricultural Systems of Insects mBio 2014 5 e02077 e02014 25406380 Aylward FO Tremmel DM Bruce DC Complete Genome of Enterobacteriaceae Bacterium Strain FGI 57, a Strain Associated with Leaf-Cutter Ant Fungus Gardens Genome Announcements 2013b 1 e00238 e00212 Aylward FO Tremmel DM Starrett GJ Complete genome of Serratia sp. strain FGI 94, a strain associated with leaf-cutter ant fungus gardens Genome Announcements 2013c 1 e00239 e00212 Baker ES Burnum KE Ibrahim YM Enhancing bottom-up and top-down proteomic measurements with ion mobility separations Proteomics 2015 15 2766 2776 26046661 Baker ES Burnum-Johnson KE Jacobs JM Advancing the High Throughput Identification of Liver Fibrosis Protein Signatures Using Multiplexed Ion Mobility Spectrometry Molecular & cellular proteomics 2014 13 1119 1127 24403597 Baker ES Clowers BH Li F Ion mobility spectrometry-mass spectrometry performance using electrodynamic ion funnels and elevated drift gas pressures Journal of the American Society for Mass Spectrometry 2007 18 1176 1187 17512752 Baker ES Livesay EA Orton DJ An LC-IMS-MS platform providing increased dynamic range for high-throughput proteomic studies Journal of proteome research 2010 9 997 1006 20000344 Bergmeyer H Gawehn K Bergmeyer H Gawehn K Malate Dehydrogenase Methods of enzymatic analysis 1974 New York Academic Press 613 618 Boone CK Keefover-Ring K Mapes AC Bacteria Associated with a Tree-Killing Insect Reduce Concentrations of Plant Defense Compounds Journal of Chemical Ecology 2013 39 1003 1006 23807433 Boos W Ehmann U Bremer E Middendorf A Postma P Trehalase of Escherichia coli. Mapping and cloning of its structural gene and identification of the enzyme as a periplasmic protein induced under high osmolarity growth conditions The Journal of biological chemistry 1987 262 13212 13218 2820965 Burnum KE Hirota Y Baker ES Uterine deletion of Trp53 compromises antioxidant responses in the mouse decidua Endocrinology 2012 153 4568 4579 22759378 Cha J Burnum KE Bartos A Muscle Segment Homeobox Genes Direct Embryonic Diapause by Limiting Inflammation in the Uterus The Journal of biological chemistry 2015 290 15337 15349 25931120 Cherrett JM Powell RJ Stradling D Wilding N Collins NM Hammond PM Webber J The mutualism between leaf-cutting ants and their fungus Insect-fungus interactions 1989 London Academic Press 93 120 Costa AN Vasconcelos HL Vieira-Neto EHM Bruna EM Do herbivores exert top-down effects in Neotropical savannas? Estimates of biomass consumption by leaf-cutter ants Journal of Vegetation Science 2008 19 849 854 Cuddeford D Welch RW Oats for animal feed The oat crop: production and utilization 1995 New York Chapman Hall 321 368 De Filippo C Cavalieri D Di Paola M Impact of diet in shaping gut microbiota revealed by a comparative study in children from Europe and rural Africa Proceedings of the National Academy of Sciences 2010 107 14691 14696 De Fine Licht HH Schiøtt M Rogowska-Wrzesinska A Laccase detoxification mediates the nutritional alliance between leaf-cutting ants and fungus-garden symbionts Proceedings of the National Academy of Sciences 2013 110 583 587 De Siqueira C Bacci M Jr Pagnocca F Bueno OC Hebling M Metabolism of plant polysaccharides by leucoagaricus gongylophorus, the symbiotic fungus of the leaf-cutting ant atta sexdens L Applied and Environmental Microbiology 1998 64 4820 4822 9835568 De Vasconcelos HL Foraging activity of two species of leaf-cutting ants (Atta) in a primary forest of the central Amazon Insectes sociaux 1990 37 131 145 Dowd PF Symbiont-mediated detoxification in insect herbivores Microbial Mediation of Plant-Herbivore Interactions 1991 New York Wiley-Interscience 411 440 Grell MN Linde T Nygaard S The fungal symbiont of Acromyrmex leaf-cutting ants expresses the full spectrum of genes to degrade cellulose and other plant cell wall polysaccharides BMC Genomics 2013 14 928 24373541 Hansen AK Moran NA Aphid genome expression reveals host-symbiont cooperation in the production of amino acids PNAS 2011 108 2849 2854 21282658 Hansen AK Moran NA The impact of microbial symbionts on host plant utilization by herbivorous insects Molecular Ecology 2013 23 1473 1496 23952067 Herz H Beyschlag W Hölldobler B Assessing herbivory rates of leaf-cutting ant (Atta colombica) colonies through short-term refuse deposition counts Biotropica 2007 39 476 481 Holldobler B Wilson EO The Ants 1990 Cambridge, MA Belknap (Harvard University Press) Holldobler B Wilson EO The superorganism: the beauty, elegance, and strangeness of insect societies 2008 New York, NY W.W. Norton & Company Hook SE Steele MA Northwood KS Impact of subacute ruminal acidosis (SARA) adaptation and recovery on the density and diversity of bacteria in the rumen of dairy cows FEMS Microbiology Ecology 2011 78 275 284 21692816 Howard JJ Leafcutting ant diet selection: the role of nutrients, water, and secondary chemistry Ecology 1987 58 503 515 Howard JJ Leafcutting and diet selection: relative influence of leaf chemistry and physical features Ecology 1988 69 250 260 Jaitly N Mayampurath A Littlefield K Decon2LS: An open-source software package for automated processing and visualization of high resolution mass spectrometry data BMC bioinformatics 2009 10 87 19292916 Kim S Gupta N Pevzner PA Spectral probabilities and generating functions of tandem mass spectra: a strike against decoy databases Journal of proteome research 2008 7 3354 3363 18597511 Kooij PW Schiøtt M Boomsma JJ De Fine Licht HH Rapid shifts in Atta cephalotes fungus-garden enzyme activity after a change in fungal substrate (Attini, Formicidae) Insectes sociaux 2011 58 145 151 21475686 Krause KM Oetzel GR Understanding and preventing subacute ruminal acidosis in dairy herds: A review Animal Feed Science and Technology 2006 126 215 236 Kyle JE Zhang X Weitz KK Uncovering biologically significant lipid isomers with liquid chromatography, ion mobility spectrometry and mass spectrometry The Analyst 2016 141 1649 1659 26734689 Livesay EA Tang K Taylor BK Fully automated four-column capillary LC-MS system for maximizing throughput in proteomic analyses Analytical chemistry 2008 80 294 302 18044960 Martínez I Lattimer JM Hubach KL Gut microbiome composition is linked to whole grain-induced immunological improvements The ISME Journal 2013 7 269 280 23038174 Matzke MM Waters KM Metz TO Improved quality control processing of peptide-centric LC-MS proteomics data Bioinformatics (Oxford, England) 2011 27 2866 2872 Mayhé-Nunes A Jaffe K On the biogeography of Attini (Hymenoptera: Formicidae) Ecotropicos 1998 11 45 54 Miyata R Noda N Tamaki H Influence of feed components on symbiotic bacterial community structure in the gut of the wood-feeding higher termite Nasutitermes takasagoensis Bioscience, Biotechnology, and Biochemistry 2014 71 1244 1251 Moller IE De Fine Licht HH Harholt J Willats WGT Boomsma JJ Chave J The dynamics of plant cell-wall polysaccharide decomposition in leaf-cutting ant fungus gardens PLoS ONE 2011 6 e17506 Monroe ME Tolić N Jaitly N VIPER: an advanced software package to support high-throughput LC-MS peptide identification Bioinformatics (Oxford, England) 2007 23 2021 2023 Nagamoto NS Garcia MG Forti LC Microscopic evidence supports the hypothesis of high cellulose degradation capacity by the symbiotic fungus of leaf-cutting ants Journal of Biological Research-Thessaloniki 2011 16 308 312 Oksanen J Blanchet FG Kindt R vegan: Community ecology package CRAN.R-project.org 2013 Piehowski PD Petyuk VA Sandoval JD STEPS: a grid search methodology for optimized peptide identification filtering of MS/MS database search results Proteomics 2013 13 766 770 23303698 Pinto-Tomás AA Anderson MA Suen G Symbiotic nitrogen fixation in the fungus gardens of leaf-cutter ants Science 2009 326 1120 1123 19965433 R Core Team R: A language and environment for statistical computing R Foundation for Statistical Computing 2013 Ricklefs RE Miller GL Ecology 2000 New York, NY W. H. Freeman and Company Silva A Bacci M Jr Pagnocca FC Bueno OC Hebling MJA Production of polysaccharidases in different carbon sources by Leucoagaricus gongylophorus Möller (Singer), the symbiotic fungus of the leaf-cutting ant Atta sexdens Linnaeus Current microbiology 2006a 53 68 71 16775790 Silva A Bacci M Jr Pagnocca FC Bueno OC Hebling MJA Starch metabolism in Leucoagaricus gongylophorus, the symbiotic fungus of leaf-cutting ants Microbiological Research 2006b 161 299 303 16380244 Silva-Pinhati ACO Bacci M Jr Hinkle G Low variation in ribosomal DNA and internal transcribed spacers of the symbiotic fungi of leaf-cutting ants (Attini: Formicidae) Brazilian journal of medical and biological research 2004 37 1463 1472 15448866 Steele MA Croom J Kahler M Bovine rumen epithelium undergoes rapid structural adaptations during grain-induced subacute ruminal acidosis AJP: Regulatory, Integrative and Comparative Physiology 2011 300 R1515 R1523 Suen G Scott JJ Aylward FO An insect herbivore microbiome with high plant biomass-degrading capacity PLoS genetics 2010 6 e1001129 20885794 Tartar A Wheeler MM Zhou X Parallel metatranscriptome analyses of host and symbiont gene expression in the gut of the termite Reticulitermes flavipes Biotechnology for biofuels 2009 2 25 19832970 Thoetkiattikul H Mhuantong W Laothanachareon T Comparative analysis of microbial profiles in cow rumen fed with different dietary fiber by tagged 16S rRNA gene pyrosequencing Current microbiology 2013 67 130 137 23471692 Van Bael SA Estrada C Wcislo WT Fungal-fungal interactions in leaf-cutting ant agriculture Psyche: A Journal of Entomology 2011 2011 1 9 Wang Y Lim L DiGuistini S A specialized ABC efflux transporter GcABC-G1 confers monoterpene resistance to Grosmannia clavigera a bark beetle-associated fungal pathogen of pine trees New Phytologist 2012 197 886 898 23252416 Webb-Robertson B-JM Matzke MM Datta S Bayesian proteoform modeling improves protein quantification of global proteomic measurements Molecular & cellular proteomics 2014 13 3639 3646 25433089 Webb-Robertson B-JM Matzke MM Jacobs JM Pounds JG Waters KM A statistical selection strategy for normalization procedures in LC-MS proteomics experiments through dataset-dependent ranking of normalization scaling factors Proteomics 2011 11 4736 4741 22038874 Webb-Robertson B-JM McCue LA Waters KM Combined statistical analyses of peptide intensities and peptide occurrences improves identification of significant peptides from MS-based proteomics data Journal of proteome research 2010 9 5748 5756 20831241 Welch RW Welch RW The chemical composition of Oats The oat crop: production and utilization 1995 New York Chapman Hall 279 320 Wirth R Wirth R Herz H Ryel RJ Beyschlag W Hölldobler B Herbivory of leaf-cutting ants: a case study on Atta colombica in the tropical rainforest of Panama 2003 Germany Springer-Verlag Berlin Heidelberg Wirth R Beyschlag W Ryel RJ holldobler B Annual foraging of the leaf-cutting ant Atta colombica in a semideciduous rain forest in Panama Journal of Tropical Ecology 1997 13 1–741–75717 Zimmer J Monroe ME Qian WJ Smith RD Advances in proteomics data analysis and display using an accurate mass and time tag approach Mass Spectrometry Reviews 2006 25 450 482 16429408

PMC005xxxxxx/PMC5118126.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9214478 2614 Mol Ecol Mol. Ecol. Molecular ecology 0962-1083 1365-294X 27718295 5118126 10.1111/mec.13877 NIHMS822311 Article Mosquitoes host communities of bacteria that are essential for development but vary greatly between local habitats Coon Kerri L. Brown Mark R. Strand Michael R. Department of Entomology, The University of Georgia, Athens, GA, 30602, USA Corresponding author: Michael R. Strand, Department of Entomology, University of Georgia, 120 Cedar Street, 420 Biological Sciences, Athens, GA 30602, USA, Fax: 706-542-2279, mrstrand@uga.edu 13 10 2016 31 10 2016 11 2016 01 11 2017 25 22 58065826 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Mosquitoes are insects of interest because several species vector disease-causing pathogens to humans and other vertebrates. We previously reported that mosquitoes from long-term laboratory cultures require living bacteria in their gut to develop, but development does not depend on particular species of bacteria. Here, we focused on three distinct but interrelated areas of study to better understand the role of bacteria in mosquito development by studying field and laboratory populations of Aedes aegypti, Aedes albopictus, and Culex quinquefasciatus from the Southeastern United States. Sequence analysis of bacterial 16S rRNA gene amplicons showed that bacteria community composition differed substantially in larvae from different collection sites, whereas larvae from the same site shared similarities. Although previously unknown to be infected by Wolbachia, results also indicated that Ae. aegypti from one field site hosted a dual infection. Regardless of collection site or factors like Wolbachia infection, however, each mosquito species required living bacteria in their digestive tract to develop. Results also identified several concerns in using antibiotics to eliminate the bacterial community in larvae in order to study its developmental consequences. Altogether, our results indicate that several mosquito species require living bacteria for development. We also hypothesize these species do not rely on particular bacteria because larvae do not reliably encounter the same bacteria in the aquatic habitats they colonize. microbiota insect Wolbachia development Introduction Most multicellular animals host communities of microbes in their digestive tract. One important area of study is to understand how animals acquire these microbes and the factors that influence their composition. This is because several studies show that the gut microbiota and its composition have wide-ranging effects on the physiology of different vertebrates and invertebrates (Buchon et al. 2013; Lee & Brey 2013; Sommer & Backhed 2013; Blaser 2016; Goodrich et al. 2016). Among insects, some groups including termites (Blattodea), bees and ants (Hymenoptera), and select Heteroptera have specialized gut microbiota that are maintained by direct transfer between individuals (Fukatsu & Hosokawa 2002; Ohkuma & Brune 2011; Anderson et al. 2012; Martinson et al. 2012). Others including certain moths (Lepidoptera) and fruit flies (Diptera: Drosophilidae) acquire most if not all of their gut microbiota each generation from the environment (Robinson et al. 2010; Chandler et al. 2011; Tang et al. 2012). In the case of mosquitoes (Diptera: Culicidae), all species are aquatic as larvae, feeding on detritus, microorganisms and small invertebrates, while adults are terrestrial with both sexes usually feeding on nectar (Merritt et al. 1992; Clements 1992). Adult females of most species also must feed on vertebrate blood to produce eggs, which underlies why many mosquitoes are important disease vectors (Clements 1992). Deep sequencing of bacterial 16S rRNA gene amplicons from select species shows that larvae contain a subset of the operational taxonomic units (OTUs) in their aquatic environment, while some but not all of the OTUs in larvae are also in adults (Wang et al. 2011; Boissiere et al. 2012; Coon et al. 2014; Gimonneau et al. 2014). Controlled experiments further indicate that mosquito larvae contain no bacteria if they hatch from surface sterilized eggs and are maintained in a sterile environment, while also showing that most OTUs in adults overlap with those in larvae (Coon et al. 2014). Altogether, these data support that mosquito larvae acquire their gut microbiota from the water in which they feed and that some community members in larvae persist in adults. Variability in the microbiota of mosquitoes has been detected within and between species from different geographic locations (Boissiere et al. 2012; Osei-Poku et al. 2012; Gimonneau et al, 2014; Duguma et al. 2015; Buck et al. 2016; Muturi et al. 2016). In contrast, less is known about community composition on finer geographic scales, which is relevant to many vector species that develop as larvae in small volume containers, which are often located in close proximity to one another in urban habitats. The issue of variability in bacterial community composition is also functionally important from the perspective of mosquito development. Most mosquitoes molt through four instars before metamorphosis into adults when fed a nutritionally complete diet under conventional (non-sterile) rearing conditions (Clements 1992). However, our own recent studies of three species (Aedes aegypti, Aedes (=Georgecraigius) atropalpus, and Anopheles gambiae) from long-term laboratory cultures show that axenic larvae with no bacteria die as first instars when fed a nutritionally complete diet (Coon et al. 2014). Axenic larvae of each species also die as first instars if fed dead bacteria plus diet or diet preconditioned by living bacteria before feeding. Re-colonization with a mixed community of bacteria from conventional cultures, however, rescues development. Experiments with Ae. aegypti further show that several members of the laboratory bacterial community including an Acinetobacter, Aeromonas, Aquitalea, and Chryseobacterium or the non-community member Escherichia coli individually colonize the gut, which produces monoassociated, gnotobiotic larvae that develop normally (Coon et al. 2014). Resulting adults also show no morphological defects or reductions in fitness as measured by development time, size and fecundity (Coon et al. 2014; 2016). Thus, each of the species studied by Coon et al. (2014) required living bacteria for development, yet development did not depend on a particular species or community assemblage because several different bacteria rescued development of gnotobiotic larvae. In contrast, a study that used antibiotics to eliminate the gut microbiota from Anopheles mosquitoes reported delays in growth but larvae nonetheless survived and developed into adults (Chouaia et al. 2012). Older studies also report variable effects on mosquito development after perturbing the microbial communities in larval habitats (Hinman 1930; Rozeboom 1935; Ferguson & Micks 1961; Jones & DeLong 1961; Chao et al. 1963). Thus, unlike our own results, some studies in the literature suggest living bacteria in the gut are not required for mosquito larvae to develop. One explanation for these differing conclusions is that mosquitoes vary in their requirement for bacteria as a function of species or rearing conditions that could also affect the communities of bacteria that are naturally present. Another factor of potential importance is that vertically transmitted intracellular bacteria like Wolbachia infect some mosquitoes, and can affect fitness both positively and negatively (Brownlie et al. 2009; Ross et al. 2016). However, since Coon et al. (2014) examined laboratory populations of mosquitoes that were not infected by Wolbachia or other intracellular bacterial symbionts, this study could not assess whether Wolbachia potentially affect the importance of gut bacteria for development. The third possibility is that surface sterilization of eggs and antibiotic treatment do not equivalently eliminate bacteria from mosquitoes, which underlies why these two approaches have led to differing conclusions about the functional role of the gut microbiota in larvae. The overall goal of this study, therefore, was to determine if any of the factors discussed above affect whether mosquito larvae require living bacteria in their digestive tract for development. To meet this goal, we studied field and laboratory populations of Aedes aegypti, Aedes albopictus, and Culex quinquefasciatus from the Southeastern United States (US). Each of these species has a worldwide distribution in subtropical and tropical regions (Rey et al. 2006; Leonel et al. 2015). Ae. aegypti and C. quinquefasciatus were introduced into North America from Africa centuries ago, while Ae. albopictus was introduced from Asia in the 1980s (Huang et al. 2011; Bargielowshi & Lounibos 2014). Ae. albopictus has also altered the range of Ae. aegypti, which along the east coast of the US is infrequently collected outside of Florida (Bargielowshi & Lounibos 2014; Hahn et al. 2016). In the first part of our study, we used 16S rRNA gene analysis to characterize bacterial diversity in larvae from field versus laboratory cultures. We also examined larvae from different field sites to assess the variability of bacterial community composition between breeding sites that were in close proximity to one another. Second, we assessed whether larvae from each species differed in their requirement for living bacteria as a function of being from the lab or field, the composition of the bacterial community where larvae were collected from, or Wolbachia infection. Third, we compared the efficacy of surface sterilizing eggs and antibiotic treatment in eliminating bacteria from larvae and their effects on development. Results showed that bacterial diversity in larvae differed substantially between collection sites. However, each species required living bacteria in their digestive tract for development regardless of where they came from or whether they were infected with Wolbachia. Antibiotic treatment in contrast did not fully eliminate bacteria, which results in larvae that develop. High concentrations of some antibiotics also adversely affected larvae independent of their impact on bacteria. Materials and methods Field collections and laboratory cultures We assessed bacterial diversity in Ae. aegypti, Ae. albopictus and C. quinquefasciatus by collecting samples from six field sites and laboratory cultures. One field site was located in Athens, GA and contained Ae. albopictus and C. quinquefasciatus (Fig. 1). Five others were located in Jacksonville, FL USA with one containing both Ae. aegypti and C. quinquefasciatus, three containing only Ae. aegypti, and one containing only C. quinquefasciatus (Fig. 1). Each site was a small container habitat that was sampled 2 or 3 times over the breeding season by collecting: (1) 50 ml of water in a sterile Falcon tube (BD Diagnostics) that was transferred to ice, and (2) mosquito larvae in 300 ml of water that were placed in a 500 ml container. Samples were taken to the laboratory where larvae were counted in a laminar hood and identified to species using defined morphological characters (Breeland & Loyless 1989). Water samples were centrifuged at 5,000 x g for 30 min at 4° C and the resulting pellet was stored at −20° C until DNA isolation. Almost all larvae (>98%) collected from the field sites we sampled were Ae. aegypti, Ae. albopictus, or C. quinquefasciatus. As previously noted, results from prior studies showed that laboratory cultures of three species (Ae. aegypti, Ae. atropalpus, An. gambiae) contained no Wolbachia, and eggs that hatched from surface sterilized eggs contained no bacteria as measured using culture and PCR-based assays (Coon et al. 2014). This latter result indicated that all of the OTUs identified by deep sequencing in laboratory-reared Ae. aegypti were acquired from the environment through larval feeding, while also suggesting most if not all of the OTUs in conventionally reared larvae were gut community members (Coon et al. 2014). Prior studies also showed that the small size of mosquito larvae and corresponding small number of bacteria that are present (~103–4) (Coon et al. 2016) often do not yield sufficient DNA template for library construction. Thus for this study, pooled samples were prepared by collecting 15 larvae (fourth instars) per species, site and collection date. Each pool was then rinsed in 70% EtOH, dried, and stored at −20° C until use for DNA isolation. The remaining larvae were kept in their habitat water until pupation. Pupae were then surface sterilized and allowed to emerge in sterile cages (Coon et al. 2014). Three adult female Ae. aegypti from site 1 in Jacksonville, FL were rinsed in 70% EtOH, dried and stored at −20° C for DNA isolation. Ribosomal internal transcribed spacer region 1 (ITS1) from these adults was PCR amplified using ITS1A and ITS1B primers (Beebe et al. 2000) followed by comparison to the known ITS1 sequence for Ae. aegypti (Wesson et al. 1992). Other adults were maintained at 27° C and a 16 h light: 8 h dark photoperiod and fed 10% sterile sucrose in water. Day 3 females were blood fed on a surface-sterilized rat and then allowed to oviposit onto moist filter paper. Eggs were collected within 24 h of being laid and either stored in humidified containers (Ae. aegypti, Ae. albopictus) or immediately surface-sterilized and hatched (C. quinquefasciatus) for use in developmental assays. Laboratory cultures for each species were maintained in an insectary that was 2 km from the Athens, GA field site. The Ae. aegypti culture was established in 1974 from offspring collected in Athens, GA (UGAL strain) (Foster & Lea 1975). The Ae. albopictus and C. quinquefasciatus cultures were obtained in 2012 from the Centers for Disease Control (Atlanta, GA, USA) from offspring collected in Johannesburg, South Africa and Keyport, NJ, USA respectively (Cornel et al. 2003; Marcombe et al. 2014). Larvae were maintained in covered rearing pans containing distilled water and fed a nutritionally complete, standard diet of ground rat chow (Purina): lactalbumin: brewers yeast (1:1:1). Pupae were transferred to cages for adult emergence, while females were blood fed as described above to produce eggs. Water and larvae from the laboratory were collected for DNA isolation as described above. Bacterial 16S rRNA library construction and sequencing DNA was isolated from samples using the Gentra Puregene Yeast/Bacteria Kit (Qiagen). All samples were processed at the same time to eliminate potential batch contamination effects. Since our primary goal was to assess bacterial diversity within and between sites, bacterial 16S rRNA gene V3–V4 variable regions were PCR-amplified using the universal primers 341F and 785R (Faircloth & Glenn 2012; Klindworth et al. 2013) containing multiplex identifier sequences (Table S1). PCR amplifications were performed in triplicate 20 ul reactions containing 10 ng of template DNA, 0.4 U of iProof High-Fidelity polymerase (Bio-Rad), 1X polymerization buffer, 200 μM dNTPs and 0.2 μM of each primer. Reactions using template from blank DNA extractions served as a negative control. Reaction conditions were: initial denaturation cycle of 98° C for 30 s, followed by 25 cycles at 98° C for 10 s, 55° C for 15 s and 72° C for 15 s, and a final extension step at 72° C for 5 min. Products were visualized on 1.2% agarose gels, pooled, purified with the MinElute PCR purification kit (Qiagen), and eluted in 11 μl of EB buffer. Pooled amplicons served as template for a second round of PCR to attach Nextera indices and adapters (Faircloth & Glenn 2012). Products were again visualized on agarose gels and repurified before quantification using a Qubit dsDNA HS assay and fluorometer (Invitrogen). Three libraries were constructed and sequenced per sample (i.e. three technical replicates) using different barcode sequence combinations to prevent PCR bias. For each replicate PCR, negative controls produced no amplicon indicating no contamination of reagents. Altogether, 6 field sites sampled multiple times and 3 laboratory cultures resulted in 36 samples. Three technical replicates per sample thus resulted in 108 sequencing libraries. Barcoded products were combined in equimolar amounts with libraries repurified using a SPRI plate and Sera-Mag Speed-beads prior to paired-end sequencing (2 x 300 bp) on half of an Illumina MiSeq lane by the University of Georgia Genomics Facility. Triplicate libraries from three adult Ae. aegypti females that emerged from larvae collected at site 1 in Jacksonville, FL on 7/21/2014 were also paired-end sequenced. Amplicon data analyses Sequenced reads were trimmed at any site of two sequential bases receiving a quality score <Q30 and paired-end reads were merged using PEAR (Zhang et al. 2014). Read filtering criteria included: no ambiguous base calls or barcode/primer errors, all bases must have a PHRED equivalent score of 30 or higher (per base error rate of 0.1%), and reads must be between 440 and 484 bp in length. Quality-filtered reads were demultiplexed using an in-house Perl script and assigned to operational taxonomic units (OTUs) at a 97% identity threshold using the Quantitative Insights Into Microbial Ecology (QIIME) version 1.7.0 (Caporoso et al. 2010b) USEARCH 6.1 wrapper (Edgar 2010). Chimeric sequences were identified and removed using UCHIME (Edgar et al. 2011). Taxonomic classification was assigned against the latest Greengenes database (http://greengenes.lbl.gov/cgi-bin/nph-index.cgi) using the QIIME-based RDP Bayesian classifier with a 0.80 confidence threshold. OTUs making up less than 0.005% of sequence libraries were filtered out of our analysis (Bokulich et al. 2013). Sequence alignment was performed by PyNAST (Caporoso et al. 2010a), and phylogenetic tree construction was performed using FastTree (Price et al. 2009). For each sample, we first compared calculated indices for species richness and diversity (Chao1, Shannon’s H) within each technical replicate library to assess whether between replicate variation could affect alpha diversity estimates. One-way ANOVA detected no significant differences between replicated libraries for each measure (P > 0.05) (Table S2). We also compared beta diversity using the unweighted UniFrac distance and Bray-Curtis index of dissimilarity between pairs of samples within a given replicate. One-way ANOVA again detected no differences between technical replicates for any sample or either metric (P > 0.05) (Table S2). We therefore combined the three technical replicate libraries for each sample for diversity analyses, which were conducted on data rarefied to 38,735 sequences per sample. Separate analyses of variance (ANOVAs) were performed using R (http://www.r-project.org/) to compare alpha diversity (species richness, Chao1, and Shannon’s H diversity indices) between sites and mosquito species. We performed a principal coordinates analysis (PCoA) on Bray-Curtis distances using QIIME (Lozupone et al. 2007). A multivariate analysis of variance (MANOVA) was run on the first three PCoA vectors across all larval libraries to determine whether site, mosquito species, or collection date were significantly associated. We then performed univariate ANOVAs on significant MANOVA factors to examine which axis or axes drove the differences visible in the PCoA plots. Separate MANOVAs followed by univariate ANOVAs were also run using only the larval libraries for the sites in Jacksonville, FL or the two sites that contained more than one mosquito species. We assessed whether unweighted UniFrac distances were smaller for larval libraries of species within versus between sites using Monte Carlo t-tests (9,999 simulations). The same approach was used to assess whether UniFrac distances were smaller between water and larval libraries from a given site at late versus early sampling dates. We tested for differences in taxon abundance across sample groups using the QIIME script otu_category_significance.py. All MANOVAs and ANOVAs were run in R (http://www.r-project.org/) and analysed with Pillai’s Trace as the test statistic (Sullam et al. 2012). We used oligotyping of 16S ribosomal RNA gene amplicons to further investigate bacterial diversity within the six dominant phyla detected by OTU clustering. This was performed separately at the phylum level for five taxa (Actinobacteria, Bacteroidetes, Firmicutes, Cyanobacteria, Verrucomicrobia) while for Proteobacteria each class (Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, and Deltaproteobacteria) was analyzed separately. Taxonomic assignments were made using quality filtered reads against the latest Greengenes database using the RDP classifier with a 0.80 confidence threshold. Sequences corresponding to each phylum or class were extracted using an in-house python script and an o-pad-with-gaps script to eliminate any length variation between reads. Oligotyping, which uses Shannon (1948) entropy to detect the amount of diversity at each nucleotide position, was conducted as outlined by Eren et al. (2013). Entropy values for nucleotide positions ranged from 0 to 2 with values ≥ 1.0 indicating greater bacterial diversity. Oligotypes were generated using the first position with the highest entropy value followed by further decomposition using user-defined nucleotide selection. Between 25 and 85 positions were chosen for each of the 9 datasets until no peaks were left unresolved. Peaks ≤ 0.2 were considered background noise. The value for parameter ‘s’ (the minimum number of samples in which an oligotype is present) was set to 2 to correct for technical errors due to sequencing, parameter ‘A’ was set to 0.0005% (the minimum total abundance of an oligotype in all samples) of the total reads for each data set to match the parameter set for OTU clustering, and parameter ‘M’ (the minimum count of the most abundant unique sequence in an oligotype) was set to 137 which corresponded to the average number of reads per sample divided by 1000 (Eren et al. 2013). Oligotype representative sequences (i.e. the most abundant sequence within an oligotype) were classified using the RDP classifier with a 0.8 confidence threshold. For determining percent alignment scores, the sequences corresponding to individual oligotypes were extracted from the OLIGO-REPRESENTATIVES directory and aligned using MAFFT (Katoh et al. 2005). Multilocus sequence typing (MLST) of Wolbachia from Ae. aegypti Wolbachia from Ae. aegypti was genotyped by multilocus sequence typing (MLST) of five housekeeping genes (gatB, coxA, hcpA, fbpA and ftsZ) (Baldo et al. 2006). Each gene was amplified using primer sets and reaction conditions for double AB-infected individuals as outlined by the Wolbachia MLST website (http://pubmlst.org/wolbachia/info/amp_seq_double.shtml). PCR products were visualized on 1% agarose gels using ethidium bromide followed by purification (PCR Purification Kit, Qiagen) and bidirectional sequencing (Eurofins MWG Operon). Sequences were concatenated, aligned and trimmed followed by analysis against the Wolbachia MLST database (http://pubmlst.org/wolbachia/). PhyML version 3.0 with the SPR search option was used for phylogeny construction, while branch support was estimated by bootstrapping the dataset 1,000 times (Guindon et al. 2010). Development Assays First instars with no gut bacteria were produced from surface sterilized eggs as previously described (Coon et al. 2014; Supplementary Methods). The status of larvae was confirmed using culture and PCR assays (Supplementary Methods). Standard mosquito larval rearing diet was sterilized by cobalt 60 gamma irradiation with sterility of diet confirmed by culture-based assays (Supplementary Methods). Water for sterile rearing cultures was autoclaved. Ten first instars were placed into 25 cm2 cell culture flasks (Corning) containing 20 ml of sterile water and sterilized standard diet (2 mg). Some flasks were maintained this way while to others we added 100 μl of water from laboratory trays containing conventionally reared larvae, 100 μl of water from site 1 in Jacksonville, FL, or 2 μl of an overnight E. coli (K12 MG1655 strain) culture. Note that inoculating cultures with water from conventional laboratory cultures or field site 1 provided the community of bacteria that was present. Each of the above treatments was replicated a minimum of three times with the proportion of first instars that developed into adults and total development time (days) determined by inspecting cultures daily. Survival data were analysed by Fisher’s Exact tests followed by post-hoc Bonferroni-corrected pairwise tests to compare treatments. Development times were tested for normality and equality of variances before analysis by one-way ANOVA and post-hoc Tukey-Kramer Honest Significant Difference (HSD) tests using R. Antibiotic assays We previously isolated, identified and cultured several members of the bacterial community in UGAL Ae. aegypti (Coon et al. 2014). Several members of the community in Ae. albopictus fourth instars collected at the Athens, GA field site were isolated in the current study by the same methods (Table S3). Five community members from UGAL Ae. aegypti (Aquitalea, Aeromonas, Chryseobacterium, Comamonas, and Sphingobacterium spp.) and two from field collected Ae. albopictus (Bacillus and Aeromonas spp.) were then used to assess their antibiotic susceptibility by standardized dilution (Wiegand et al. 2008). In brief, single colonies from overnight culture plates were suspended to 5 x 105 CFU/ml in Luria Bertani (LB) broth containing serially diluted ampicillin (Sigma), kanamycin (Roche), streptomycin (Sigma), chloramphenicol (Sigma), penicillin (Sigma), tetracycline (Sigma), or gentamicin (Gibco) (200 μg - 2 ng/ml). Suspensions were incubated at 28° C for 18 h. The minimal inhibitory concentration (MIC) was recorded as the lowest amount of antibiotic where no visual growth of bacteria was detected. We then placed newly hatched first instars (N=12) from the conventionally reared UGAL Ae. aegypti culture into 1 ml of distilled water in a culture plate containing non-sterilized standard diet (300 μg) and the same dilution series of antibiotics. From 2–6 wells and a total of 24–72 larvae per antibiotic dose were bioassayed. The number of larvae that died or molted to the second instar after 2 or 5 days was then recorded with 2 day molting data further analyzed by Fisher’s Exact tests. In other replicates we selected 6 larvae at 2 days post-hatching in cultures that contained either 200 μg/ml of ampicillin, kanamycin streptomycin, chloramphenicol and penicillin or 20 μg/ml of tetracycline or gentamicin. Larvae cultured for the same period with no antibiotic served as a positive control while axenic larvae served as a negative control. Larvae were then individually homogenized in Luria broth (LB) followed by serial dilution on LB plates which were incubated at 27° C for 72 h. The number of living bacteria per individual that could be cultured were then determined by colony count assay (Coon et al. 2016). Lastly, axenic first instars were produced and cultured in different concentrations of the same antibiotics followed by determination of the proportion of larvae that died after 5 days. Axenic larvae cultured with no antibiotics served as the positive control with mortality data across all treatments analyzed by Fisher’s Exact tests. 3. Results Clustering analyses of amplicon data identify variable bacterial communities Multiplex sequencing of 16S rRNA gene amplicons for all field and laboratory samples generated a total of 8.3 million reads of which 5.8 million were retained after quality filtering. Reads per sample ranged from 38,735 to 191,711, and grouped by percentage similarity into 1,253 operational taxonomic units (OTUs) at a cut-off threshold of 97% (Fig. 1, Table S4). However, 68 OTUs accounted for more than 50% of the total quality filtered reads. Rarefaction curves saturated or near saturated at 3,000 sequences, which indicated that most bacteria in each sample were captured (Fig. S1). Sample complexity varied with OTUs ranging from 211 to 628 for water and 85 to 556 for larvae. Bacterial species diversity differed by site as measured by richness (F6,29 = 10.63, P < 0.0001), Chao1 (F6,29 = 9.672, P < 0.0001), and Shannon’s H indices (F6,29 = 3.68, P < 0.01) (Fig. 1). The highest richness values were from sites in Jacksonville, FL, while the lowest were from laboratory cultures (Fig. 1). Richness was higher in water than larvae at most sites while alpha diversity in water and larvae at each site significantly correlated (r =0.72, P=0.0004). Twenty nine bacterial phyla were identified across all samples but 6 accounted for 81% of the OTU groups at the 97% cut-off threshold: Proteobacteria (49%), Actinobacteria (11%), Bacteroidetes (12%), Firmicutes (5%), Cyanobacteria (2%), and Verrucomicrobia (2%). Classification into orders showed that larvae contained the same taxa present in the water they were collected from but relative abundance differed. To visualize these trends we merged sample times for each collection site to produce the bar graphs shown in Fig. 2. This showed that laboratory reared Ae. aegypti, Ae. albopictus, and C. quinquefasciatus contained a greater proportion of Actinomycetales than field collected larvae (P=0.0003, after FDR-correction for multiple comparisons), while Ae. albopictus and C. quinquefasciatus from the Athens, GA field site contained a higher proportion of Burkholderiales than larvae from other sites (P=0.005) (Fig. 2). Considerable heterogeneity at the ordinal level was also observed between the field sites in Jacksonville, FL, which was striking given the proximity of most sites to one another (Fig. 1). PCoA analysis using the Bray-Curtis dissimilarity index showed clustering by collection site, mosquito species, and sampling date (Fig. 3). A MANOVA across all larval libraries indicated that differences along the PCoA axes were significant for site but not species or collection date (Table 1A). Separate MANOVAs for only the larval libraries from Jacksonville, FL (Table 1B) or the three sites that contained more than one mosquito species (site 1 in Jacksonville, the field site in Athens, GA, and the laboratory) (Table 1C) likewise indicated the bacterial communities in larvae differed by site but not collection date or species. We next calculated the unweighted UniFrac distances between samples. Pairwise analysis of all larval sequencing libraries using Monte Carlo resampling indicated that UniFrac distances differed between the same species from different sites (P≤0.002), whereas no significant differences were detected between mosquito species from the same site (P>0.05) (Fig. 4). Comparing UniFrac distances between the larval and aquatic samples from each field site also showed that distances were significantly larger at the early versus late sampling date (P≤0.03). Overall, these results indicated that the bacterial communities in water and mosquito larvae strongly differed between sites, while also suggesting that the larval and water communities within each site were more dissimilar to one another at the early collection dates than at the late collection dates. A final feature of note was detection of low abundance reads corresponding to Wolbachia in some but not all larval samples. Two Wolbachia OTUs were detected in all field and laboratory samples for Ae. albopictus while one Wolbachia OTU was detected in all field and laboratory samples for C. quinquefasciatus. No Wolbachia were detected in the Ae. aegypti laboratory culture (UGAL), which we previously reported (Coon et al. 2014), or in Ae. aegypti larvae from sites 2–4 in Jacksonville, FL. However, two Wolbachia OTUs were detected in Ae. aegypti larvae from site 1 in Jacksonville, FL. Mean prevalence of Wolbachia across larval samples for all species was 0.35%. Detection of Wolbachia in Ae. albopictus and C. quinquefasciatus was expected since both are known to harbor natural infections in North America (Kittayapong et al. 2002; Zouache et al. 2009; Almeida et al. 2011; Atyame et al. 2011; Minard et al. 2014). This was not expected though for Ae. aegypti because no natural infection of this species by Wolbachia has been reported anywhere in the world (Ross et al. 2016). We therefore reared and analyzed Ae. aegypti adults from larvae collected at Jacksonville site 1 on 7/21/2014 to assess whether they too were Wolbachia infected. Sequencing ITS1 domains from three adult females confirmed each was correctly identified to species. Sequencing of 16S rRNA gene amplicons from triplicate libraries produced from the same females also showed that the abundance of Wolbachia reads was much higher than in larvae but grouped into the same two OTUs (Fig. S2). MLST genotyping indicated one of these strains belonged to supergroup A and was related to a strain in Ae. albopictus, while the second was a supergroup B member present in a number of other insects including Ae. albopictus that were collected in Thailand (Fig. S3). Non-Wolbachia reads in these adults grouped into 27–43 OTUs that were either identical or belonged to the same genera detected in larvae but whose relative abundance differed (Fig. S2). For example Rhodocyclales that were abundant in larvae were reduced while Rhodospirillales and Burkholderiales (adults 1 and 2), or Burkholderiales and Pseudomonadales (adult 3) were elevated (Fig. S2). Oligotyping identifies additional bacterial diversity Oligotyping provides a computational method for identifying subpopulations (oligotypes) within OTUs. We therefore generated oligotyping datasets for the six most abundant bacterial phyla we identified (Proteobacteria, Actinobacteria, Bacteroidetes, Firmicutes, Cyanobacteria, and Verrucomicrobia), which as noted above accounted for 81% of the OTU groups we identified across all samples. A total of 2.6 million quality-filtered reads were used, which ranged from 3,301 to 164,841 per sample, and grouped into 4,321 unique oligotypes (Table S5). Sequence identities within an oligotype ranged from 99.5% to 100%, while relative abundance of individual oligotypes ranged from 0.006% to 0.3%. We used oligotyping to further assess bacterial diversity by focusing on two locations: Jacksonville field site 1, which on 7/21/2014 contained Ae. aegypti and C. quinquefasciatus larvae, and our laboratory culture of Ae. aegypti which was sampled on a single date. We first determined the total number of oligotypes for each of the bacterial phyla or classes (Proteobacteria) that contained >5% of the OTU groups in Ae. aegypti larvae from these collections. For Jacksonville field site 1, these taxa were Actinobacteria, Bacteroidetes, and four classes of Proteobacteria, while for the laboratory culture they were Actinobacteria, Bacteroidetes, Firmicutes, Betaproteobacteria and Verrucomicrobia (Fig. 5A). We then plotted the proportion of these oligotypes that were present in water from the same site. A majority of the oligotypes detected in Ae. aegypti larvae from field site 1 were detected in the water for some taxa (Bacteroidetes) but not others (Deltaproteobacteria) (Fig. 5A). In contrast, almost all oligotypes in Ae. aegypti larvae from the laboratory culture were detected in the water habitat (Fig. 5A). We next selected the most abundant OTU for each of the taxa shown in Fig. 5A, and determined the number of oligotypes for each in Ae. aegypti larvae. The taxonomic assignments for these OTUs are summarized in Table S6, which also presents the sequences for the two most abundant oligotypes, and the full-length sequences for the corresponding 16s rRNA gene amplicons. We also separately examined the number and identity of the oligotypes for each of the OTUs in the three technical sequencing replicates for these samples. This showed that the number and identity of the oligotypes for each OTU per replicate were identical. Comparison of these OTUs and oligotypes to the C. quinquefasciatus larvae that were also present at Jacksonville field site 1 on 7/21/2014 showed almost complete overlap, whereas only partial overlap was detected in the water (Fig. 5B). Comparison to Ae. aegypti larvae collected from site 1 on 9/15/2014 also showed that some OTUs and corresponding oligotypes near fully overlapped (Leucobacter, Dysgonomonas), while others (C39, Desulfovibrio cuneatus) were absent although some oligotypes were detected in the water (Fig. 5B). In contrast, almost no overlap in oligotypes was detected in Ae. aegypti larvae collected from Jacksonville field sites 3 and 4 on 7/21/2014 or the laboratory, because either these OTUs were absent or very rare (Fig. 5B). The same exercise with the most abundant OTUs in Ae. aegypti larvae from the laboratory showed near complete overlap with the oligotypes detected in the water (Fig. 5C). In contrast, little overlap was detected with Ae. aegypti larvae from different Jacksonville field sites (Fig. 5C). The one exception to these trends was the unclassified Microbacteriaceae where a number of oligotypes overlapped with Ae. aegypti larvae from Jacksonville field site 1. However, this OTU was also very rare in site 1 larvae with almost all of the overlapping oligotypes detected as single sequences (singletons). For most of the OTUs discussed above, the two most abundant oligotypes differed from each other by only 1 nucleotide with corresponding full-length amplicons exhibiting ≥99% identity (Table S6). Exceptions were the unnamed genus C39 (Betaproteobacteria) detected in Ae. aegypti larvae from Jacksonville field site 1 where the two most abundant oligotypes derived from full-length amplicons that differed by 29 nucleotides over 463 nt (92% sequence identity) and the Chryseobacterium (Bacteroidetes) in Ae. aegypti larvae from the laboratory where the two most abundant oligotypes derived from full-length amplicons that differed by 96 nucleotides over the 453 nt (79% sequence identity). The strongest matches identified after BlastN searches in most cases exhibited ~99% sequence identity with known bacteria (Table S6). Most matches were to uncultured bacterial clones identified from different parts of the world in freshwater, rainwater, or soil (Table S6). Notably though, oligotype 1 for C39 shared 99% sequence identity with an uncultured bacterium identified in the larval gut of Anopheles stephensi while oligotype 1 for the Chryseobacterium in laboratory Ae. aegypti larvae shared 99% sequence identity with an uncultured clone from a cranefly, Tipula abdominalis, which resides in the sister family (Tupulidae) of the Culicidae (Table S6). Mosquitoes from the field and laboratory both require gut bacteria for development The preceding results indicated the microbiota in Ae. aegypti, Ae. albopictus, and C. quinquefasciatus larvae differ between local field sites and collection dates as well as between field sites and the laboratory. We therefore asked whether these differences affected the requirement of mosquitoes from the field and laboratory for gut bacteria to develop. Examining progeny from every field site for these experiments was not technically feasible because of the effort involved in rearing field-collected mosquitoes, producing eggs, and generating progeny with no gut bacteria. We therefore focused these experiments on comparing the laboratory population of Ae. albopictus to Ae. albopictus from the field site in Athens GA, and Ae. aegypti plus C. quinquefasciatus from the laboratory to the same species collected at Jacksonville field site 1. These choices were justified because each of these populations differed in terms of deriving from sites with distinctly different bacterial communities, having different culture histories, and in the case of Ae. aegypti also differing in regard to Wolbachia infection. We first verified that previously developed methods for surface sterilizing eggs (Coon et al. 2014) produced larvae with no gut bacteria. For each species, culture-based assays showed that numerous bacterial colonies grew when homogenates of first instars from non-sterilized eggs were plated on LB or BHI plates. PCR assays also generated amplicons using universal 16S rRNA bacterial gene primers and DNA templates from non-sterilized eggs. In contrast, no bacteria grew when homogenates of first instars from surface sterilized eggs were placed on culture plates. No bacterial amplicons were detected using universal primers and templates from laboratory (UGAL) Ae. aegypti first instars that hatched from surface sterilized eggs but weak amplicons were detected in the other samples that we assumed were due to Wolbachia. This interpretation was supported using primers for the Wolbachia wsp gene, which generated amplicons from each sample except UGAL Ae. aegypti. Thus, surface sterilization of eggs produced axenic larvae with no detectable bacteria in the case of UGAL Ae. aegypti, whereas larvae for all other samples likely had no gut bacteria but, as expected, remained infected with Wolbachia, which are vertically transmitted and intracellular. We then placed larvae from surface sterilized eggs in sterile water plus sterile food. All ultimately died as first instars after hatching (Fig. 6). Larvae in this condition usually died 6–9 days after hatching but some lived up to 14 days as first instars that never molted before finally dying. In contrast, a majority of larvae from each collection site and species developed into adults if inoculated with water from laboratory trays containing conventionally reared (non-sterile) Ae. aegypti larvae or water from field site 1 in Jacksonville, FL (Fig. 6). A majority of larvae from surface sterilized eggs also developed into adults if inoculated with only E. coli, which was absent from all field and laboratory samples we analyzed (Fig. 6). Antibiotics do not fully eliminate bacteria from conventionally reared Ae. aegypti larvae and adversely affect axenic larvae that contain no bacteria The preceding results indicated that Ae. aegypti, Ae. albopictus and C. quinquefasciatus require gut bacteria for development regardless of origin, prior rearing history, or Wolbachia infection. These results were also near identical to our previous results that examined only long-term laboratory cultures of Ae. aegypti, Ae. atropalpus and An. gambiae that were not Wolbachia infected (Coon et al. 2014). However, they clearly differed from results showing that mosquito larvae treated with antibiotics only exhibit delays in development (Chouaia et al. 2012). We thus assessed whether antibiotic treatment results in developmental delays because bacteria that colonize mosquitoes are not fully eliminated. As with our developmental assays, it was not logistically feasible to examine the antibiotic susceptibility of every OTU we identified across the species and breeding sites we sampled. We therefore selected 7 abundant community members that we isolated from UGAL Ae. aegypti and field collected Ae. albopictus. Like E. coli, each individually colonizes Ae. aegypti and Ae. albopictus from surface sterilized eggs to produce gnotobiotic larvae that develop normally into adults. We first tested the susceptibility of these bacteria to several different antibiotics over a concentration range used in other studies to eliminate bacteria from the mosquito digestive tract (Dong et al. 2009; Chouaia et al. 2012; Ramirez et al. 2012; Minard et al. 2013). The purpose of this was to assess the sensitivity of each bacterium to these antibiotics in the absence of any confounding effects from being in a mosquito. Results showed that these bacteria exhibited highly variable sensitivities with the minimal inhibitory concentration (MIC) for ampicillin exceeding 200 μg/ml for each species while MICs for kanamycin, streptomycin, and penicillin exceeded 200 μg/ml for most (Table S7). Average MICs were lower for tetracycline, chloramphenicol, and gentamicin but were ≥ 200 μg/ml for a Chryseobacterium sp. isolated from UGAL Ae. aegypti (Table S7). Thus, in vitro each bacterium showed differential sensitivity to the antibiotics we tested and in some cases were largely unaffected at high doses. Given these findings, we next assessed the effects of each antibiotic on larvae across the same range of concentrations. Given the highly variable microbial communities we identified across species and populations but the inability of each to develop without gut bacteria, we selected conventionally reared (non-sterile) UGAL Ae. aegypti for these assays, which was also the source of most bacterial isolates for the preceding MIC assays. Results indicated these larvae molted to the second instar in two days in the absence of antibiotics and across all concentrations of ampicillin (Fig. 7A). The other antibiotics had dose-dependent effects on mortality, timing of molting, or both (Fig. 7A). Chloramphenicol and penicillin had no mortality effects but all larvae molted to the second instar within 5 days. All or a large proportion of larvae treated with 200 μg/ml of kanamycin, streptomycin or tetracycline or ≥ 20 μg/ml of gentamicin died within 5 days, but all surviving larvae molted within 5 days (Fig. 7A). Larvae treated with lower concentrations of these antibiotics exhibited no mortality and molted within 2 days (Fig. 7A). To quantify bacterial clearance, individual larvae from cultures containing the highest concentration of each antibiotic that did not cause larvae to die were homogenized at 2 days post-hatching and cultured on LB plates. Results showed that each antibiotic except ampicillin significantly reduced the number of culturable bacteria when compared to conventional larvae that were cultured with no antibiotic (Fig. 7B). However, all antibiotic treated larvae still contained viable bacteria, whereas axenic larvae produced from surface sterilized eggs contained, as expected, no culturable bacteria (Fig. 7B). We therefore assessed whether the mortality tetracycline and gentamicin caused in conventional larvae when used at 200 μg/ml also occurred in axenic larvae. Dose-response assays showed that these antibiotics as well as kanamycin and streptomycin dose-dependently increased the mortality of axenic larvae when compared to axenic control larvae that were cultured in the absence of antibiotics (Fig. 7C). Altogether, these results indicated that several antibiotics failed to fully eliminate bacteria from UGAL Ae. aegypti while also showing that antibiotic treatment adversely affects first instars when used at high concentrations regardless of whether or not they contained culturable bacteria. Discussion Our previous results (Coon et al. 2014; 2016) indicated that laboratory populations of three mosquito species required living bacteria to develop into adults, but development did not depend on a particular bacterial species or community assemblage. In this study, we generated additional information in three areas of study to better understand the functional role of bacteria in mosquito development. This included the characterization of bacterial community diversity within and between three species from breeding sites in the field and laboratory. This information was essential in order to know whether or not species or larvae from different locations host similar or different bacterial communities of potential importance in development. With this knowledge in hand, we then could assess whether: 1) larvae differ in their requirement for living bacteria to develop as a function of species, rearing history, overall bacterial community composition, or Wolbachia infection that has previously been suggested to nutritionally benefit mosquitoes during periods of stress (Brownlie et al. 2009). These data also fully positioned us to assess whether the approach of producing axenic larvae by surface sterilizing eggs versus treating larvae with antibiotics yield similar or different outcomes. Our analyses of bacterial community diversity indicate that Ae. aegypti, Ae. albopictus and C. quinquefasciatus larvae primarily contain a subset of the OTUs in the water they were collected from, which supports that larvae are colonized by a subset of the bacteria they ingest during feeding. Since our data derive from pooled samples of whole larvae, we cannot conclude that the OTUs we identified are all restricted to the gut. However, prior results as well as data in this study showing that axenic larvae contain no bacteria unless infected by Wolbachia strongly support that larvae acquired the bacteria we identified from the environment by feeding which in turn also suggests most of the OTUs we identified are present in the digestive tract. However, it is also possible some bacteria present in the gut over time colonize other tissues given findings in the literature that identify some species of bacteria present in the gut that also colonize other organs (see Chouaia et al. 2012; Gimonneau et al. 2014). We recognize that additional sampling could further improve upon our bacterial diversity studies. However, our data set was sufficiently robust to compare within-site versus between-site beta diversity, which strongly supported that community composition varied greatly between sites, while strong similarities were found between larvae of different species that were developing in the same site. That no species effect was detected in sites containing two species of mosquitoes supports that a similar subset of bacteria colonized larvae of each. That identical oligotypes were recovered from water and larvae within but not between sites is additional strong evidence for between site variability for the bacterial communities in water and larvae. Interestingly, our within site comparisons between different sampling dates also indicate that the bacterial communities in larvae and their aquatic habitats become more similar over the breeding season. At this time, we are uncertain what accounts for this. Certainly bacteria from diverse sources could colonize the sites we sampled. However, it is also possible that mosquitoes modify their aquatic habitat through cyclic transmission, which has been proposed for other aquatic organisms (McFall-Ngai 1998; Sullam et al. 2012). This could occur through larvae feeding and excreting bacteria that are selected for in their own digestive tract. We also previously showed that some abundant members of the larval and adult gut communities in laboratory-reared Ae. aegypti are present on the surface of eggs females lay (Coon et al. 2014). Thus, adult mosquitoes may transmit some gut community members to habitats where they oviposit, which over time could affect the microbial community in the water where larvae develop. The ability of mosquitoes to alter the composition of bacterial communities in larval habitats would be strongest in small volume containers like those we sampled. Comparison of sites persistently colonized by mosquito larvae to neighboring sites where no mosquitoes can colonize would be one approach to assessing whether mosquitoes significantly alter the bacterial communities in small aquatic habitats. Our community diversity data share some parallels with studies of terrestrial insects including Drosophila and certain Lepidoptera, which show that the food larvae consume overrides the effect of species in defining the composition of the gut microbiota (Chandler et al. 2011; Wong et al. 2011; Priya et al. 2012). However, it has also been noted that the complex relationships between bacteria and their environments may require finer scale information about microbial diversity than can be provided by the classification of OTUs or clustering approaches (Eren et al. 2013). Our use of oligotyping allowed us to address this concern and assess whether concealed diversity within identified OTUs reveals additional patterns. We generated oligotyping data sets for the six most abundant bacterial phyla detected across our entire sequencing dataset. However, we focused on Jacksonville field site 1, which contained both Ae. aegypti and C. quinquefasciatus larvae, and our Ae. aegypti laboratory culture to assess whether oligotyping revealed similar or different trends to our clustering analyses. In terms of colonization, almost all oligotypes in laboratory reared Ae. aegypti were present in the water they were collected from. This was also true for some bacterial taxa in Ae. aegypti from field site 1 but for others like Deltaproteobacteria most of the oligotypes detected in larvae were not detected in the water they came from. More interesting though were our results that focused on the most abundant OTUs for each oligotyped phylum or class in Ae. aegypti larvae from Jacksonville field site 1. These data showed that each OTU consisted of several oligotypes but nearly all oligotypes detected in Ae. aegypti larvae were also present in C. quinquefasciatus larvae from the same site and date. This provided strong evidence that the same subset of bacteria in the local habitat colonized both species. In contrast, almost none of these OTUs or oligotypes were detected in Ae. aegypti larvae collected on the same date from other sites in Jacksonville or the laboratory, which underscores the high variation that existed between the sites we sampled. As previously noted, the presence of Wolbachia in Ae. albopictus and C. quinquefasciatus was expected but identification of A and B group Wolbachia in Ae. aegypti from Jacksonville site 1 was not becaue no natural infections of this species have previously been reported (Ross et al. 2016). We thus were very careful to document these mosquitoes were correctly identified and that we genotyped the Wolbachia strains. We sequenced only a small number of adults from this site because we did not anticipate encountering any Wolbachia in Ae. aegypti, and characterizing these bacteria was not an original goal of the study. As a result we had only a small number of frozen adults from this site that could be used for sequence analysis when we learned from the larval data set that Wolbachia reads were present. Nonetheless the data we generated confirmed each adult contained the same Wolbachia OTUs detected in larvae. At this time, we need to be very cautious about these results and their possible implications because our data also clearly indicate that Ae. aegypti at most of the sites we sampled contained no Wolbachia. That both the A and B group members detected in Ae. aegypti from Jacksonville, FL field site 1 are related but not identical to Wolbachia that have been identified from Ae. albopictus suggests the possibility of past lateral transfer, which would also be consistent with Ae. aegypti and Ae. albopictus sharing partially overlapping ranges in several parts of the world including the Southeastern US. The presence of a natural Wolbachia infection in Ae. aegypti also has applied implications given the use of introduced Wolbachia in Ae. aegypti for use in population replacement to alter vector competency (reviewed in Minard, 2013). While Wolbachia abundance in larval stage mosquitoes has not previously been reported, our finding that titers are very low relative to adults is similar to a recent study, which noted that D. melanogaster larvae also contain a low abundance infection relative to adult females (Stevanovic et al. 2015). On the other hand, the high abundance of Wolbachia in adult female mosquitoes has been reported in other studies and has also been noted to hinder characterization of gut microbiota diversity in pyrosequencing data sets (Minard et al. 2014; Muturi et al. 2016). Our Illumina data in contrast provided sufficient coverage to discern that adults contained a subset of the OTUs detected in larvae, which circumstantially supports results from previous laboratory experiments that adults acquire a subset of the gut community in larvae by transstadial transmission (Coon et al. 2014). Collectively, our bacterial diversity and Wolbachia data positioned us to address the second goal of this study by assessing whether larvae differed in their developmental requirement for gut bacteria as a function of species, origin (laboratory versus field), or Wolbachia infection status. Results showed they did not with all larvae failing to develop when gut bacteria were absent but a high proportion of larvae developing into adults when recolonized with a mixed community of bacteria in water from the field or lab, or E. coli, which was absent across all samples and collection sites. In contrast, the third goal of our study revealed two important outcomes in regard to antibiotics. First, while antibiotics are commonly used to eliminate resident microbiota from insects including mosquitoes, our results indicate they do not fully eliminate the microbiota from Ae. aegypti larvae, which is one reason why antibiotic therapy does not prevent larvae from developing. A recent study similarly revealed that antibiotic treatment also does not fully eliminate the microbiota from adult Anopheles mosquitoes (Hughes et al. 2014). Second, some of the antibiotics we tested adversely affected the survival of Ae. aegypti which contained no bacteria. While no published work has previously noted antibiotic sensitivity in mosquito larvae, some antibiotics including tetracyclines are known to adversely affect other eukaryotes including insects independent of their effects on the microbiota (Ridley et al. 2013; Moullan et al. 2015). We thus conclude that surface sterilization of eggs is much more effective for producing mosquito larvae with no gut bacteria and that antibiotic therapy has several major shortcomings for studying the function of bacteria in mosquito development. We have now examined a total of five mosquito species from field and laboratory populations that each exhibit a requirement for living bacteria in their gut to develop, but also show no evidence that development depends on a particular species or community assemblage. While several studies have noted variable communities of bacteria in mosquitoes from different geographic locations, this study shows that bacteria are consistently present in mosquito larvae but community composition is also highly variable between sites on a fine geographic scale. Together, these data suggest the hypothesis that the mosquito species we have studied to date rely on bacteria for development quite generally because of the high unpredictability in the community of bacteria between sites. This lack of specialization could also reflect that most of the mosquitoes we have studied have been introduced into habitats outside of their native range. More specialized mosquito species or these species in their native habitat in contrast may exhibit more specific associations with bacteria. The obvious question going forward is what do living bacteria provide that mosquito larvae require? This issue is also very interesting from the perspective of the model organism literature where Drosophila and mice show roles for gut bacteria in maturation of the immune and digestive systems but also indicate bacteria are not required for development into adults because axenic cultures of both can be maintained for generations if fed a nutritionally complete diet (Buchon et al. 2013; Lee & Brey 2013; Sommer & Backhed 2013; Blaser 2016; Goodrich et al. 2016). One option is that many bacteria share conserved metabolic features that facilitate breakdown of factors in the larval diet or could produce nutrients that supplement the mosquito diet. Several studies have experimentally demonstrated or provide circumstantial support for gut bacteria having digestive or nutritional functions in some insects (summarized by Engel & Moran, 2013). However, most of these examples focus on species that feed on specialized diets with nutrient constraints, whereas laboratory reared mosquitoes consume nutritionally rich and complete diets. Alternatively, several different types of bacteria could potentially affect physiochemical parameters such as pH or redox state that alter the gut environment in ways that mosquito larvae depend upon to grow and molt. A third option is that most bacteria share cell wall components and/or generate conserved metabolites that could function as signaling molecules. The former are well known pattern recognition factors that the insect immune system recognizes, which could activate different signaling pathways in the gut that are essential for development. Contact-dependent interactions between bacteria and the gut epithelium or diffusible bacterial metabolites that gut cells perceive have also been identified to mediate gut homeostasis or systemic growth of mice and Drosophila (Shin et al. 2011; Storelli et al. 2011; Nicholson et al. 2012). Understanding the mechanism(s) underlying why mosquitoes require living bacteria is of translational significance, because the processes involved may be amenable to disruption, which could be used as a tool for controlling species that transmit human disease. Supplementary Material Supp Fig S1 Supp Fig S2 Supp Fig S3 Supp MethodsS1 Supp Table S1 Supp Table S2 Supp Table S3 Supp Table S4 Supp Table S5 Supp Table S6 Supp Table S7 We thank K. J. Vogel and G. R. Burke for suggestions during the study, J. A. Russell who provided considerable feedback after initial submission of the paper including the suggestion for oligotyping, and comments from an impressive number of anonymous reviewers. We thank J. Wright with the US Navy Entomology Center of Excellence for assistance in identifying larval habitats in Jacksonville, FL. J. Brandt assisted with field collections, M. Brandt provided technical support, and S.R. Robertson and A. Elliot assisted with rearing. This work was supported by the National Science Foundation (Graduate Research Fellowship 038550-04 (KLC)), University of Georgia Graduate School (KLC), Sigma Xi (KLC), and National Institutes of Health R01AI106892 and T32GM007103 (MRS). Fig. 1 Collection sites and sequencing statistics for the 16S rRNA libraries prepared from water and mosquito larvae. The upper portion of the figure shows the locations of the collection sites in Georgia and Florida. The lower portion of the figure summarizes the sites, sampling dates, and mosquito species that were collected. For each collection site, latitude and longitude coordinates are indicated along with a general description of the habitats, which were all containers of variable size. Numbers in parentheses in the column named Mosquito species indicate the approximate number of larvae that were present per 300 ml of water. This provides a sense of larval density between sites and collection dates. Number of reads for each library after quality filtering, number of OTUs, and indices for species richness and diversity (Chao 1, Shannon’s H) are also presented. Fig. 2 Bacteria at the levels of phylum and order in water and mosquito larvae from each collection site. Water and larval (fourth instar) libraries for each collection date were pooled for the bar graphs presented. Each bar presents the proportion of sequencing reads assigned to a given bacterial order. Only categories >2% are presented. Fig. 3 Principal coordinates analysis based on pairwise Bray-Curtis distances. Symbols are coloured by collection site (laboratory: green, Athens, GA: red, Jacksonville, FL: orange (site 1), dark blue (site 2), light blue (site 3), yellow (site 4), purple (site 5). The legends in the upper right of each plot designate sample type (water or species of larva) by symbol shape. Water and larval samples for each collection date are presented but are not given unique identifiers because date was not statistically significant (see Table 1). Fig. 4 Within and between-site variation in the bacteria in mosquito larvae. Within-site variation was determined by average unweighted UniFrac distances between different mosquito species from the same site, while between-site variation was determined by distances between the same mosquito species collected from different sites. Mean values ± 95% confidence intervals are shown. **, P < 0.01; ***, P < 0.0001 (Student’s t-test with 10,000 Monte Carlo permutations). Fig. 5 Oligotype analyses. (A) Oligotype overlap between Aedes aegypti larvae and water collected from site 1 in Jacksonville, FL (7/21/2014) or the laboratory. Bars indicate the proportion of oligotypes present in larvae that were also present in water. Numbers above the bars indicate the total number of oligotypes present in larvae. (B) Oligotype distribution between larvae collected from site 1, 3 or 4 in Jacksonville FL (7/21/2014) and the laboratory. The column on the left indicates the number of oligotypes present in Ae. aegypti larvae from site 1 whose taxonomic assignment matched the dominant OTU corresponding to a given phylum (or class). Subsequent bars indicate the proportion of these oligotypes present in either Culex quinquefasciatus larvae or water collected from the same site, or Ae. aegypti larvae collected from other sites on the same collection date. Bars on the far right indicate the proportion of these oligotypes present in Ae. aegypti larvae in the laboratory. The lower panel displays the proportion of oligotypes present in Ae. aegypti larvae from site 1 that were present in either Ae. aegypti larvae or water collected from the same site on 9/15/2014. Bars in (C) present oligotype overlap between Ae. aegypti larvae collected from the laboratory and the same field sites. The column on the left indicates the number of oligotypes present in Ae. aegypti larvae from the laboratory corresponding to the dominant taxa identified by OTU clustering. Subsequent bars indicate the proportion of these oligotypes present in either water collected from the same site, or Ae. aegypti larvae collected from site 1, 3 or 4 in Jacksonville, FL (7/21/2014). Only phyla (or classes) that accounted for >5% of the total number of reads are included in the figure. Fig. 6 Survival to adulthood and development time of Ae. aegypti, Ae. albopictus, and C. quinquefasciatus larvae that hatched from eggs laid by adult, mated females reared from field-collected larvae or larvae from laboratory cultures. Field derived Ae. aegypti and C. quinquefasciatus came from site 1 in Jacksonville, FL, while Ae. albopictus came from the field site in Athens, GA. All laboratory-reared females came from our long-term laboratory cultures (see Methods). All eggs were surface sterilized and the first instars that hatched were confirmed to contain no gut bacteria. The bars in the upper graph show the proportion of larvae that developed into adults when fed sterilized diet in sterile water, sterile water inoculated with 100 μl of non-sterile laboratory water, sterile water inoculated with 100 μl of non-sterile field water, and sterile water inoculated with Escherichia coli. Fisher’s Exact tests were conducted separately for each species with post-hoc Bonferroni-corrected pairwise tests to compare treatments (NS, treatments not significantly different; ***, sterile water only treatment significantly differs from the lab culture water, field water, and E. coli treatments (P < 0.0001). The bars of the lower graph show the development time from egg hatching until adult emergence for larvae reared in water inoculated with lab culture water, field water and E. coli. Development time for larvae cultured in sterile water only is not shown in the lower graph because no individuals survived beyond the first instar. Columns indicate mean values with 95% confidence intervals. ANOVA and post-hoc comparison tests were used to compare development times separately for each species (*** laboratory derived Ae. aegypti and Ae. albopictus reared in cultures inoculated with field water developed faster than larvae in cultures inoculated with laboratory water or E. coli (P < 0.05). Figure 7 Effect of antibiotics on UGAL strain Ae. aegypti larvae and associated bacteria. (A) Mortality and molting of conventionally reared (non-sterile) first instars treated with different antibiotics that were added to cultures. Columns from left to right indicate: antibiotic type, dose, total number of larvae treated per dose, the number of larvae that died within 5 days (d), the number of larvae that molted after 2 d, and the number of larvae that molted after 5 days. Asterisks (*) in the column labeled Molting after 2 d indicate the treatment significantly differed from the negative control which was larvae treated with no antibiotic (0 μg/ml) (P < 0.05; Fisher’s Exact Test, pairwise comparison against control). (B) Bacterial loads in 2 day old larvae collected from treatment dishes as described in (A). Non-sterile first instars reared conventionally without antibiotics served as the positive control. First instars from surface-sterilized eggs reared under sterile conditions (Axenic) served as the negative control. In most cases, a total of 6 larvae were sampled from the highest dose for each antibiotic treatment (200 μg/ml). In the case of tetracycline and gentamicin, larvae were sampled from the 20 μg/ml treatment due to 100% mortality at 200 μg/ml (see A). Bars show mean bacterial load per larva with 95% confidence intervals as determined by colony forming units (CFUs). An asterisk (*) above a bar indicates the treatment significantly differed from the positive control (*, P < 0.05; ***, P < 0.0001; Dunnett’s test). (C) Mortality of axenic first instars reared in the presence of antibiotic. Antibiotic treatment and corresponding dose are indicated below each bar. Axenic first instars reared under sterile conditions without antibiotic treatment served as the control. A total of 24 larvae were assayed per treatment. Numbers above the bars indicate the number of larvae that died within 5 d while an asterisk (*) following this number indicates significantly more larvae died relative to the control (P < 0.05; Fisher’s Exact Test, pairwise comparison against control). Table 1 Effects of collection site, species, and collection date on larval bacterial communities in mosquitoes (A) Multivariate analysis of all larval libraries Source of variation Pillai’s trace Hypothesis df Error df F P Site 2.882 18 24 32.630 <0.0001* Species 0.705 6 14 1.269 0.332 Sampling date 0.560 9 24 0.612 0.775 Univariate analysis of site (all larval libraries) Type III SS df F P Site  PCoA1 1.160 6 29.905 <0.0001*  PCoA2 1.097 6 28.223 <0.0001*  PCoA3 0.729 6 60.672 <0.0001* (B) Multivariate analysis of larval libraries from only Jacksonville, FL Source of variation Pillai’s trace Hypothesis df Error df F P Site 2.624 12 15 8.718 <0.0001* Species 0.786 3 3 3.673 0.157 Sampling date 1.043 6 8 1.453 0.305 Univariate analysis of site (Jacksonville, FL) Type III SS df F P Site  PCoA1 0.409 4 10.417 <0.01*  PCoA2 0.189 4 4.752 <0.05*  PCoA3 0.461 4 48.283 <0.0001* (C) Multivariate analysis of larval libraries from only multi-species sites Source of variation Pillai’s trace Hypothesis df Error df F P Site 1.926 6 6 26.036 <0.001* Species 0.804 6 6 0.672 0.679 Sampling date 0.847 9 12 0.525 0.831 Univariate analysis of site (multi-species) Type III SS df F P Site  PCoA1 0.968 2 56.247 <0.0001*  PCoA2 0.718 2 44.357 <0.0001*  PCoA3 0.0535 2 10.745 <0.01* P-values from MANOVA (multivariate) and ANOVA (univariate) analyses are based on Bray-Curtis distances (*significant at the ≤ 0.05 level). Data accessibility Sanger-sequenced 16S reads for bacterial isolates used for MIC determination are available under GenBank accession no. KJ192338-KJ192343 and KX260132-KX260135. Sanger-sequenced reads used for MLST of Wolbachia are available under GenBank accession no. [awaiting accession no. assignment]. Raw Illumina reads were deposited in the NCBI Sequence Read Archive (http://www.ncbi.nlm.nih.gov/sra) under project ID PRJNA342829. Input files for the QIIME and oligotyping pipelines as well as raw data files and R code for statistical analyses have been deposited in the Dryad digital repository under accession number doi:10.5061/dryad.qs43p. Almeida FD Moura AS Cardoso AF 2011 Effects of Wolbachia on fitness of Culex quinquefasciatus (Diptera; Culicidae) Infection, Genetics and Evolution 11 2138 2143 Anderson KE Russell JA Moreau CS Kautz S Sullam KE Hu Y 2012 Highly similar microbial communities are shared among related and trophically similar ant species Molecular Ecology 21 2282 2296 22276952 Atyame CM Delsuc F Pasteur N Weill M Duron O 2011 Diversification of Wolbachia endosymbiont in the Culex pipiens mosquito Molecular Biology and Evolution 28 2761 2772 21515811 Baldo L Dunning Hotopp JC Jolley KA 2006 Multilocus sequence typing system for the endosymbiont Wolbachia pipientis Applied and Environmental Microbiology 72 7098 7110 16936055 Bargielowski I Lounibos LP 2014 Rapid evolution of reduced receptivity to interspecific mating in the dengue vector Aedes aegypti in response to satyrization by invasive Aedes albopictus Evolutionary Ecology 28 193 203 24563572 Beebe NW Cooper RD Foley DH Ellis JT 2000 Populations of the south-west Pacific malaria vector Anopheles farauti s.s. revealed by ribosomal DNA transcribed spacer polymorphisms Heredity 84 244 253 10762395 Blaser MJ 2016 Antibiotic use and its consequence for the normal microbiome Science 352 544 5 27126037 Boissiere A Tchioffo MT Bachar D 2012 Midgut microbiota of the malaria mosquito vector Anopheles gambiae and interactions with Plasmodium falciparum infection PLoS Pathogens 8 e1002742 22693451 Bokulich NA Subramanian S Faith JJ 2013 Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing Nature Methods 10 57 59 23202435 Breeland SG Loyless TM 1989 Illustrated keys to the mosquitoes of Florida: adult females and fourth stage larvae Journal of the Florida Anti-Mosquito Association 53 63 84 Brownlie JC Cass BN Riegler M 2009 Evidence for metabolic provisioning by a common invertebrate endosymbiont, Wolbachia pipientis, during periods of nutritional stress PLoS Pathogens 5 e1000368 19343208 Buchon N Broderick NA Lemaitre B 2013 Gut homeostasis in a microbial world: insights from Drosophila melanogaster Nature Reviews Microbiology 11 615 26 23893105 Buck M Nilsson LKJ Brunius C Dabire RK Hopkins R Terenius O 2016 Bacterial associations reveal spatial population dynamics in Anopheles gambiae mosquitoes Scientific Reports 6 22806 26960555 Caporaso JG Bittinger K Bushman FD DeSantis TZ Andersen GL Knight R 2010a PyNAST: a flexible tool for aligning sequences to a template alignment Bioinformatics 26 266 267 19914921 Caporaso JG Kuczynski J Stombaugh J 2010b QIIME allows analysis of high-throughput community sequencing data Nature Methods 7 335 336 20383131 Chandler JA Lang JM Bhatnagar S Eisen JA Kopp A 2011 Bacterial communities of diverse Drosophila species: Ecological context of a host-microbe model system PLoS Genetics 7 e1002272 21966276 Chao J Wistreich GA Moore J 1963 Failure to isolate microorganisms from within mosquito eggs. Microbial isolations from the mid-gut of Culex tarsalis Coquillet Annals of the Entomological Society of America 56 559 561 Chouaia B Rossi P Epis S 2012 Delayed larval development in Anopheles mosquitoes deprived of Asaia bacterial symbionts BMC Microbiology 12 S2 22375964 Clements AN 1992 The Biology of Mosquitoes, Vol. 1. Development, Nutrition, and Reproduction Chapman & Hall New York Coon KL Vogel KJ Brown MR Strand MR 2014 Mosquitoes rely on their gut microbiota for development Molecular Ecology 23 2727 2739 24766707 Coon KL Brown MR Strand MR 2016 Gut bacteria differentially affect egg production in the anautogneous mosquito Aedes aegypti and facultatively autogenous mosquito Aedes atropalpus Parasites & Vectors 9 375 27363842 Cornel AJ Mcabee RD Rasgon J Stanich MA Scott TW Coetzee M 2003 Differences in extent of genetic introgression between sympatric Culex pipiens and Culex quinquefasciatus (Diptera: Culicidae) in California and South Africa Journal of Medical Entomology 40 36 51 12597651 Dong YM Manfredini F Dimopoulos G 2009 Implication of the mosquito midgut microbiota in defense against malaria parasites PLoS Pathogens 5 1000423 Duguma D Hall MW Rugman-Jones P 2015 Developmental succession of the microbiome of Culex mosquitoes BMC Microbiology 15 140 26205080 Edgar RC 2010 Search and clustering orders of magnitude faster than BLAST Bioinformatics 26 2460 2461 20709691 Edgar RC Haas BJ Clemente JC Quince C Knight R 2011 UCHIME improves sensitivity and speed of chimera detection Bioinformatics 27 2194 2200 21700674 Engel P Moran NA 2013 The gut microbiota of insects-diversity in structure and function FEMS Microbiology Reviews 37 699 735 23692388 Eren AM Maignien L Sul WJ Murphy LG Grim SL Morrison HG 2013 Oligotyping: differentiating between closely related microbial taxa using 16S rRNA gene data Methods in Ecology and Evolution 4 1111 1119 Faircloth BC Glenn TC 2012 Not all sequence tags are created equal: designing and validating sequence identification tags robust to indels PLoS One 8 e42543 Ferguson MJ Micks DW 1961 Microorganisms associated with mosquitoes: I. Bacteria isolated from the mid-gut of adult Culex fatigans Wiedemann Journal of Insect Pathology 3 112 119 Foster WA Lea AO 1975 Renewable fecundity of male Aedes aegypti following replenishment of seminal vesicles and accessory glands Journal of Insect Physiology 21 1083 1090 Fukatsu T Hosokawa T 2002 Capsule-transmitted gut symbiotic bacterium of the Japanese common plataspid stinkbug, Megacopta punctatissima Applied and Environmental Microbiology 68 389 396 11772649 Gimonneau G Tchioffo MT Abate L 2014 Composition of Anopheles coluzzii and Anopheles gambiae microbiota from larval to adult stages Infection, Genetics and Evolution 28 715 724 Goodrich JK Davenport ER Water JL Clark AG Ley RE 2016 Cross-species comparisons of host genetic association with the microbiome Science 352 532 538 27126034 Guindon S Dufayard JF Lefort V Anisimova M Hordijk W Gascuel O 2010 New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML Systematic Biology 59 307 321 20525638 Hahn MB Eisen RJ Eisen L Boegler KA Moore CG McAllister J Savage HM Mutebi JP 2016 Reported distribution of Aedes (Stegomyia) aegypti and Aedes (Stegomyia) albopictus in the United Syayes, 1995–2016 (Diptera: Culicidae) Journal of Medical Entomology 0 1 7 Hinman EH 1930 A study of the food of mosquito larvae American Journal of Hygiene 12 238 270 Huang S Molaei G Andreadis TG 2011 Reexamination of Culex pipiens hybridization zone in the Eastern United states by ribosomal DNA-based single nucleotide polymorphism markers American Journal of Tropical Medicine and Hygiene 85 434 441 21896800 Hughes GL Dodson BL Johnson RM Murdock CC Tsujimoto H Suzuki Y Patt AA Cui L Nossa CW Barry RM Sakamoto JM Hornett EA Rasgon JL 2014 Native microbiome impedes vertical transmission of Wolbachia in Anopheles mosquitoes Proceedings of the National Academy of Sciences USA 111 12498 12503 Jones WL DeLong DM 1961 A simplified technique for sterilizing the surface of Aedes aegypti eggs Journal of Economic Entomology 54 813 814 Katoh K Juma K Toh H Miyata T 2005 MAFFT version 5: improvement in accuracy of multiple sequence alignment Nucleic Acid Research 33 511 518 Kittayapong P Baimai V O’Neill SL 2002 Field prevalence of Wolbachia in the mosquito vector Aedes albopictus The American Journal of Tropical Medicine and Hygiene 66 108 111 12135259 Klindworth A Pruesse E Schweer T 2013 Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies Nucleic Acids Research 41 e1 22933715 Lee W-J Brey PT 2013 How microbiomes influence metazoan development: insights form history and Drosophila modeling of gut-microbe interactions Annual Review of Cell Biology 29 571 592 Leonel BF Koroiva R Hamada N Ferreira-Keppler RL Roque FO 2015 Potential effects of climate change on ecological interaction outcomes between two disease-vector mosquitoes: a mesocosm experimental study Journal of Medical Entomology 52 866 872 26336208 Lozupone CA Hamady M Kelley ST Knight R 2007 Quantitative and qualitative beta diversity measures lead to different insights into factors that structure microbial communities Applied Environmental Microbiology 73 1576 1585 17220268 Martinson VG Moy J Moran NA 2012 Establishment of characteristic gut bacteria during development of the honeybee worker Applied and Environmental Microbiology 78 2830 2840 22307297 Marcombe S Farajollahi A Healy SP Clark GG Fonseca DM 1992 Insecticide resistance status of United States populations of Aedes albopictus and mechanisms involved PLoS One 9 e101992 McFall-Ngai MJ 1998 The development of cooperative associations between animals and bacteria: establishing détente among domains American Zoologist 38 593 608 Merritt RW Dadd RH Walker ED 1992 Feeding behavior, natural food, and nutritional relationships of larval mosquitoes Annual Review of Entomology 37 349 376 Minard G Mavingui P Moro CV 2013 Diversity and function of bacterial microbiota in the mosquito holobiont Parasites & Vectors 6 146 23688194 Minard G Tran G Dubost A Tran-Van V Mavingui P Moro CV 2014 Pyrosequencing 16S rRNA genes of bacteria associated with wild tiger mosquito Aedes albopictus: a pilot study Frontiers in Cellular and Infection Microbiology 4 59 24860790 Moullan N Mouchiroud L Wang X 2015 Tetracyclines disturb mitochondrial function across eukaryotic models: a call for caution in biomedical research Cell Reports 10 1681 1691 Muturi EJ Kim C Bara J Bach EM Siddappaji MH 2016 Culex pipiens and Culex restuans mosquitoes harbor distinct microbiota dominated by few bacterial taxa Parasites & Vectors 9 18 26762514 Nicholson JK Holmes E Kinross J Burcelin R Gibson G Jia W Pettersson S 2012 Host-gut microbiota metabolic interactions Science 336 1262 1267 22674330 Ohkuma M Brune A 2011 Diversity, structure, and evolution of the termite gut microbial community Biology of Termites: A Modern Synthesis Bignell DE Roisin Y Lo N 413 438 Springer New York Osei-Poku J Mbogo CM Palmer WJ Jiggins FM 2012 Deep sequencing reveals extensive variation in the gut microbiota of wild mosquitoes from Kenya Molecular Ecology 21 5138 5150 22988916 Price MN Dehal PS Arkin AP 2009 FastTree: computing large minimum evolution trees with profiled instead of a distance matrix Molecular Biology and Evolution 26 1641 1650 19377059 Priya NG Ojha A Kajla MK Raj A Rajagopal R 2012 Host plant induced variation in gut bacteria of Helicoverpa armigera PLoS ONE 7 e30768 22292034 Ramirez JL Souza-Neto J Cosme RT 2012 Reciprocal tripartite interactions between the Aedes aegypti midgut microbiota, innate immune system and dengue virus influences vector competence PLoS Neglected Tropical Diseases 6 e1561 22413032 Rey JR Nishimura N Wagner B Braks MAH O'Connell SM Lounibos LP 2006 Habitat segregation of mosquito arbovirus vectors in south Florida Journal of Medical Entomology 43 1134 1141 17162945 Ridley EV Wong AC Douglas AE 2013 Microbe-dependent and nonspecific effects of procedures to eliminate the resident microbiota from Drosophila melanogaster Applied and Environmental Microbiology 79 3209 3214 23475620 Robinson CJ Schloss P Ramos Y Raffa K Handelsman J 2010 Robustness of the bacterial community in the cabbage white butterfly larval midgut Microbial Ecology 59 199 211 19924467 Ross PA Endersby NM Hoffmann AA 2016 Costs of three Wolbachia infections on the survival of Aedes aegypti larvae under starvation conditions PLoS Neglected Tropical Diseases 10 e0004320 26745630 Rozeboom LE 1935 The relation of bacteria and bacterial filtrates to the development of mosquito larvae American Journal of Hygiene 21 167 179 Shannon CE 1948 A mathematical theory of communication Bell System Technical Journal 27 379 423 Shin SC Kim SH You H 2011 Drosophila microbiome modulates host developmental and metabolic homeostasis via insulin signaling Science 334 670 674 22053049 Sommer F Backhed F 2013 The gut microbiota-masters of host development and physiology Nature Reviews Microbiology 11 227 237 23435359 Stevanovic AL Arnold PA Johnson KN 2015 Wolbachia-mediated antiviral protection in Drosophila larvae and adults following oral infection Applied and Environmental Microbiology 81 8215 8223 26407882 Storelli G Defaye A Erkosar B Hols P Royet J Leulier F 2011 Lactobacillus plantarum promotes Drosophila systemic growth by modulating hormonal signals through TOR-dependent nutrient sensing Cell Metabolism 12 403 414 Sullam KE Essinger SD Lozupone CA 2012 Environmental and ecological factors that shape the gut bacterial communities of fish: a meta-analysis Molecular Ecology 21 3363 3378 22486918 Tang X Freitak D Vogel H Ping L Shao Y Cordero EA 2012 Complexity and variability of gut commensal microbiota in polyphagous lepidopteran larvae PLoS ONE 7 e36978 22815679 Wang Y Gilbreath TM Kukutla P Yan GY Xu JN 2011 Dynamic gut microbiome across life history of the malaria mosquito Anopheles gambiae in Kenya PLoS One 6 e24767 21957459 Wesson DM Porter CH Collins FH 1992 Sequence and secondary structure comparisons of ITS rDNA in mosquitoes (Diptera: Culicidae) Molecular Phylogenetics and Evolution 1 253 269 1364170 Wiegand I Hilpert K Hancock RE 2008 Agar and broth dilution methods to determine the minimal inhibitory concentration (MIC) of antimicrobial substances Nature Protocols 3 163 175 18274517 Wong CAN Ng P Douglas AE 2011 Low-diversity bacterial community in the gut of the fruitfly Drosophila melanogaster Environmental Microbiology 13 1889 1900 21631690 Zhang J Kobert K Flouri T Stamatakis A 2014 PEAR: a fast and accurate Illumina Paired-End reAd mergeR Bioinformatics 30 614 620 24142950 Zouache K Voronin D Tran-Van V Mousson L Failloux A Mavingui P 2009 Persistent Wolbachia and cultivable bacteria infection in the reproductive and somatic tissues of the mosquito vector Aedes albopictus PLoS One 4 e6388 19633721

PMC005xxxxxx/PMC5118135.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0132144 7838 Transplantation Transplantation Transplantation 0041-1337 1534-6080 27495762 5118135 10.1097/TP.0000000000001406 NIHMS800569 Article Graft versus Host Disease After Liver Transplantation in Adults: A Case series, Review of Literature, and an Approach to Management Murali Arvind Rangarajan MD *Division of Gastroenterology & Hepatology, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA arvind-murali@uiowa.edu Chandra Subhash MD *Division of Gastroenterology & Hepatology, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA subhash-chandra@uiowa.edu Stewart Zoe MD, PhD Department of Surgery, Division of Transplantation & Hepatobiliary Surgery, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA zoe-stewart@uiowa.edu Blazar Bruce R MD Division of Blood and Bone Marrow Transplantation, University of Minnesota, Minneapolis, Minnesota, USA blaza001@umn.edu Farooq Umar MD Division of Hematology and Oncology, Bone Marrow Transplant, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA umar-farooq@uiowa.edu Ince M Nedim MD Division of Gastroenterology & Hepatology, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA m-nedim-ince@uiowa.edu Dunkelberg Jeffrey MD, PhD Department of Gastroenterology & Hepatology, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA jeffrey-dunkelberg@uiowa.edu Corresponding author Dr. Jeffrey Dunkelberg, MD, PhD, Clinical Professor, Division of Gastroenterology and Hepatology, Department of Internal Medicine, University of Iowa Hospitals and Clinics, Iowa city, Iowa, USA 52246, jeffrey-dunkelberg@uiowa.edu, Telephone number: 3193562132 * Murali AR and Chandra S have contributed equally and are both first authors. 17 7 2016 12 2016 01 12 2017 100 12 26612670 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background Graft-versus-host-disease (GVHD) after liver transplantation (LT) is a deadly complication with very limited data on risk factors, diagnosis and management. We report a case series and a comprehensive review of the literature. Methods Data was systematically extracted from reports of GVHD after LT, and from the United Network for Organ Sharing (UNOS) database. Group comparisons were performed. Results 156 adult patients with GVHD after LT have been reported. Median time to GVHD onset was 28 days. Clinical features were skin rash (92%), pancytopenia (78%) and diarrhea (65%). 6-month mortality with GVHD after LT was 73%. Sepsis was the most common cause of death (60%). Enterobacter bacteremia, invasive aspergillosis and disseminated Candida infections were frequently reported. Recipient age over 50-years is a risk factor for GVHD after LT. Hepatocellular carcinoma (HCC) was over-represented, while chronic hepatitis C was under-represented, in reported United States GVHD cases relative to all UNOS database LT cases. Mortality rate with treatment of GVHD after LT was 84% with high-dose steroids alone, 75–100% with regimens using dose increases of calcineurin inhibitors (CNI), and 55% with IL-2 antagonists. Mortality was 25% in small case series using the CD2-blocker alefacept or tumor necrosis factor-α (TNF-α) antagonists. Conclusions Age over 50-years and HCC appear to be risk factors for GVHD. Hepatitis C may be protective. High-dose steroids and CNI are ineffective in the treatment of GVHD after LT. CD2-blockers and TNF-α antagonists appear promising. We propose a diagnostic algorithm to assist clinicians in managing adults with GVHD after LT. INTRODUCTION Graft-versus-host-disease (GVHD) is an infrequent complication after liver transplantation (LT), with an incidence of 0.5–2% 1–3. GVHD occurs as a result of donor immunocompetent cells recognizing recipient antigens as foreign and mounting an immune response. Grafts containing more immunocompetent donor lymphocytes, such as hematopoietic stem cell, bone marrow or peripheral blood stem cell transplantations, are associated with a high incidence of GVHD. Among solid organ transplants, intestinal transplantation has the highest incidence of GVHD, followed by LT; with lower rates of GVHD after kidney, heart or pancreas transplantation. The mortality rate for GVHD after LT has been reported to be up to 85%.1 In this article, we report a case series and a comprehensive review of the literature on GVHD after LT. The epidemiology, risk factors, clinical features, and treatment outcomes are described; and a diagnostic algorithm is proposed. MATERIALS and METHODS University of Iowa Hospitals and Clinics Case Series Medical records of all patients diagnosed with GVHD after LT at the University of Iowa Hospitals and Clinics (UIHC) were reviewed. GVHD cases were identified from a prospectively maintained list of complications after LT. Diagnosis of GVHD was established with skin and gastrointestinal biopsies. Histologic grading of skin and gastrointestinal GVHD was also performed (see supplement). 4, 5 GVHD was documented histologically in all patients. Donor chimerism, using a quantitative assay of short tandem DNA repeats (STR) in the cells of the skin, gastrointestinal mucosa, peripheral blood and/or bone marrow, was recorded when available. This study was approved by the University of Iowa Institutional Review Board. Review of literature on GVHD after LT A comprehensive search of the databases of biomedical literature (Medline and Embase) was performed from 1988 (first case report of GVHD after LT) to 2014. The search strategy is described in Appendix 1. All case reports and case series of GVHD after LT in adults were reviewed. Reference lists of published reports were searched to find additional reports. Data on patient demographics, clinical findings, management and outcomes were extracted. Data was extracted from the United Network for Organ Sharing database, until January 2015, for group comparisons between reported United States cases of GVHD after LT and all US LT patients. Statistical Analysis Observations are reported as frequencies, and central tendencies are expressed as mean or median with standard deviation and interquartile range, based on the distribution of data. Categorical variables are reported as number and percent frequency of occurrence. Categorical data were compared using Pearson’s chi-square or Fisher’s exact test, where appropriate. All statistical testing was 2-sided and assessed for significance at the 5% level using SAS v9.4 (SAS Institute, Cary, NC). RESULTS UIHC Case Series A total of 762 LT were performed in adults from 1988 to January 31st 2015. Five recipients (0.7%) developed GVHD. A summary of the case series at the University of Iowa has been provided in Table 1. Case 1 A 73 year-old diabetic man, with cirrhosis due to nonalcoholic steatohepatitis (NASH), underwent LT in June 2014. He presented 60 days after LT with fever, nausea, vomiting, diarrhea, and maculopapular rash (Figure 1). White blood cell (WBC) count was 1600/mm3, absolute neutrophil count (ANC) 1280/mm3, hemoglobin 6.8 g/mm3, platelet count 108*103/mm3; and ferritin 2225 ng/mL. Skin biopsy showed vacuolar interface alteration of the dermal-epidermal junction, overlying lymphocytic infiltration, and scattered apoptotic keratinocytes (GVHD grade 2) (Figure 2). Colonoscopy showed normal colonic mucosa, but mucosal biopsies showed increased crypt epithelial apoptosis (GVHD grade 1) (Figure 3). STR analysis revealed 21%, 6% and 3% lymphocyte macrochimerism in the skin, colon and peripheral blood, respectively. He was started on methylprednisolone 2 mg/kg/day, and his tacrolimus level was kept at 8–12 ng/mL. His course was complicated by multiple infections (cytomegalovirus viremia, cryptosporidiosis of the gastrointestinal tract, and lobar pneumonia), all managed medically, while continuing methylprednisolone at 2 mg/kg/day. Rash, gastrointestinal symptoms, and pancytopenia resolved, and peripheral blood macrochimerism decreased to 1%. He was discharged on prednisone 80 mg/day for 2 weeks and 60 mg/day for 2 more weeks, before presenting with fever, vomiting and pancytopenia (WBC count 1,000/mm3). Colonoscopy showed inflamed ulcerated mucosa; biopsies showed abundant apoptotic crypt epithelial cells and crypt drop-out (GVHD grade 3). He was again treated with high-dose methylprednisolone, 2 mg/kg, while maintaining the same dose of calcineurin inhibitor (CNI). He developed pneumonia and septic shock and died 220 days after LT. Case 2 A 65 year-old diabetic man underwent LT for alcoholic cirrhosis and hepatocellular carcinoma (HCC) in March 2011. He presented 46 days after transplantation with maculopapular skin rash, fatigue, fever, diarrhea, weight loss, and modest leucopenia (WBC count 2300/mm3, ANC 1350/mm3). Ferritin was 799 ng/mL. Skin biopsy confirmed GVHD (grade 2). He was treated with methylprednisolone and tacrolimus dose increase. Skin and gastrointestinal symptoms improved, and WBC count rose to 6000/mm3. He was discharged on prednisone taper. Four weeks later, he returned with diarrhea, skin rash, septic arthritis (ankle), bacteremia, and WBC count of 900/mm3. Colonoscopy with biopsies showed extensive crypt dropout and denudation of epithelium (grade 4 GVHD). He had 41% donor lymphocytes in peripheral blood and 31% in the bone marrow. Ferritin was 20,333 ng/mL. Tacrolimus dose was decreased. Clinical features of GVHD worsened. Methylprednisolone was restarted; IVIG and the IL-1 antagonist, Anakinra, were added. He developed vancomycin-resistant enterococcal (VRE) bacteremia with septic shock and died 125 days after LT. Case 3 A 60 year-old diabetic man, with cirrhosis and HCC from alcoholic liver disease (ALD) and chronic hepatitis C, underwent LT in November 2011. He presented 117 days after LT with maculopapular skin rash (GVHD grade 2), fever, altered mental status, pancytopenia (WBC 200/mm3, ANC 0, hemoglobin 8.1 g/dL, platelet count 111*103cells/mm3) and ferritin 7232 ng/mL. STR analysis of peripheral blood revealed 78% macrochimerism. Bone marrow biopsy was hypocellular, with 89% macrochimerism. He was treated with an interleukin-2 (IL-2) receptor blocker (basiliximab), anti-thymocyte globulin (ATG), methylprednisolone, and intravenous immunoglobulin (IVIG); and tacrolimus was increased to 12–16 ng/mL. Due to nonresponse to initial treatment, Anakinra was added, without improvement. He developed VRE bacteremia, septic shock and died 128 days after LT. Case 4 A 60 year-old diabetic man, with cirrhosis from ALD and hemochromatosis, underwent LT in 2002. He presented with maculopapular skin rash (GVHD histologic grade 2) and diarrhea (GVHD histologic grade 2) 85 days after LT. CBC was normal. Peak serum ferritin was 733 ng/mL. He was treated with methylprednisolone, IVIG, and topical tacrolimus. Systemic tacrolimus and mycophenolate mofetil were continued at the same dosage. Skin rash and diarrhea resolved. Serum ferritin normalized. He was doing well at follow-up 10 years after LT. Case 5 A 65 year-old diabetic woman with primary biliary cholangitis (PBC) underwent LT in February 1996. Induction immunosuppression regimen included methylprednisolone and tacrolimus. Fourteen days after LT, she developed maculopapular skin rash, altered mental status, pancytopenia (WBC 100/mm3, platelets 17,000/mm3) and septic shock. Skin biopsy was consistent with GVHD. Histologic grading and STR analysis were not performed. She succumbed to septic shock within 5 days of presentation, before she could be started on treatment for GVHD. A summary of the case series has been provided in Table 1. Review of Literature on GVHD after LT A total of 80 articles reported 1 or more case, with a total of 156 cases of GVHD in adult LT recipients. Characteristics of reported cases are summarized in Table 2. Mean age at LT was 55 years, and 67.3% were male. Median time to GVHD onset from LT was 28 days (interquartile range 21–38 days). The most common clinical features in patients with GVHD were skin rash (92%), followed by cytopenias (78%) and diarrhea (65%). HCC (34.7%) was the most common indication for LT in patients who developed GVHD, followed by alcoholic liver disease (22.9%) and acute or chronic hepatitis B (19.5%). The presenting organ involvement for all patients with GVHD in the world and in the US is reported in Table 2. Data on outcome of GVHD management was reported in 138 patients; 73.2% died within 6 months of GVHD onset. There have been no prospective trials of treatment of GVHD after LT. This review found reports of treatment of GVHD after LT in 130 patients. In 8 patients (6.2%) immunosuppression was decreased6–11, while immunosuppression was intensified in 122 patients (93.8%). Six-month mortality was 70.5% in patients who had increased immunosuppression and 62.5% in patients with decreased immunosuppression. This difference was not statistically significant (p=0.68). The most frequently reported treatment regimen for GVHD after LT was high-dose steroids (ranging from 2 mg/kg/day to 20 mg/kg/day). The number of patients treated and mortality rate associated with various treatments regimens are provided in Table 3. The common causes of death in patients with GVHD after LT were sepsis, multi-organ failure and gastrointestinal bleeding. In 61 cases (60.4%), sepsis was documented as cause of death. The causative organism was reported in 25 cases (41%); invasive aspergillosis was noted in 9 cases1, 9, 12–17 (36%), disseminated candidiasis in 7 (28%) (3 albicans, 1 kruseii, 1 glabrata and 2 unspecified species) 1, 9, 18, 19, enterococci in 7 (28%)1, 20–23 and Enterobacter in 2 (8%)1, 24. There were 66 reported cases of GVHD after LT from the US and these were compared to all other LT recipients in the US, as accessed through the UNOS database (Table 2). A significant association between age and GVHD was evident, where patients with GVHD were older than 50-years (p<0.01). Gender was not a risk factor for development of GVHD. Higher GVHD incidence was noted in patients transplanted for HCC (21.6 % vs 13.0%), while chronic hepatitis C infection (HCV) was associated with a lower incidence of GVHD after LT, as compared to all other US patients in UNOS database (11.8% vs 29.9%) (Table 2). There are 37 patients reported in the literature who survived GVHD after LT. The mean age of patients was 56.1 years, 83.3% were males, and mean time from LT to diagnosis of GVHD was 43.3 days (range 13–80). The etiology of liver disease was alcoholic liver disease in 43% of patients; HBV in 27%; NASH in 11%; HCV, PSC and A1AT in 5% each; with PBC and acute liver failure in 2% each. 50% of these patients presented with skin involvement only, 21% with bone marrow involvement only, and 18% with both skin and gastrointestinal involvement. The treatment regimens of patients who survived GVHD after LT are provided in Table 4. DISCUSSION Risk Factors This systematic review demonstrates an association between GVHD after LT with recipient age over 50-years. Additional risk factors reported in the literature include donor-recipient age difference greater than 20 years, younger donor age, any HLA class I match, and glucose intolerance.2, 25 Based on our results, GVHD may occur more frequently in patients transplanted for HCC and less frequently in patients transplanted for hepatitis C. Immune dysregulation plays a major role in the pathogenesis of HCC. Alterations in innate or adaptive immunity, for example a decrease in the CD4+ T lymphocyte function due to chronic inflammation (alcoholic or non-alcoholic steatohepatitis), chronic infection (viral hepatitis), or suppression of immunity (cirrhosis), may tolerance to tumor antigen and promote the development of HCC.26 Furthermore, HCC itself may cause immune system dysfunction.26 It is possible that the immune dysregulation in the recipient that originally led to the development of HCC, or alterations in the immune system caused by HCC, may predispose to alloreactivity and development of GVHD after LT.27, 28 HCV is known to inhibit T cell receptor-mediated signaling required for activation and effector functions of T cells.29 Whether HCV demonstrates the same effect on donor T-lymphocytes, thereby decreasing the incidence of GVHD, is unclear. Clinical Features Our review shows that GVHD usually develops 3 to 5 weeks after LT. Skin rash is erythematous, maculopapular, and can involve any part of the body including palms, soles, and the volar surfaces of extremities and trunk. Skin rashes may be subtle, nonpruritic, and not noticed by the patient. A very careful total-body skin exam in a well-lit room is recommended. Characteristic histologic features are vacuolar alteration at the dermo-epidermal junction, apoptosis of keratinocytes in the epidermis, and lymphocyte exocytosis (Figure 2). GVHD can affect all 3 hematopoietic cell lineages. The alloreactive donor lymphocytes engraft and proliferate in the recipient bone marrow, with subsequent immune-mediated attack on hematopoietic stem cells. Cytopenia in the first few months after LT is common; infection (Herpes virus, cytomegalovirus, Epstein Barr virus and parvovirus B19) and medications (mycophenolate mofetil, valganciclovir, trimethoprim-sulfamethoxazole) are the usual culprits. In GVHD, the presence of cytopenia may be a poor prognostic indicator, and sepsis associated with leucopenia is a commonly reported cause of death. Gastrointestinal manifestations are common in GVHD. Diarrhea is a common symptom in solid organ transplant recipients; up to 10–13% of patients have diarrhea in the first 4 post-transplant months.30, 31 The common etiologies for diarrhea are infection (Clostridium difficile and cytomegalovirus colitis) and medications (mycophenolate mofetil, everolimus, sirolimus and tacrolimus).32 Endoscopic evidence for GVHD is provided by the presence of erythema, exudates and superficial ulceration of gastrointestinal mucosa. However, the sensitivity and specificity of endoscopic findings are sub-optimal to rule in or rule out GVHD; histopathology is necessary. Rectosigmoid biopsies are most sensitive.33, 34 In GVHD, histopathology shows increased crypt epithelial apoptosis, crypt loss and neutrophilic infiltration. Apoptosis of epithelial cells is induced by activated donor cytotoxic T-lymphocytes. It is, however, important to note that epithelial apoptosis can be seen after LT from etiologies other than GVHD, such as cytomegalovirus (CMV) colitis, mycophenolate-induced colitis and non-steroidal anti-inflammatory drugs. CMV colitis can be diagnosed by immunohistochemical demonstration of CMV viral inclusions. Mycophenolate-induced colitis can be differentiated from GVHD by presence of more than 15 eosinophils per high power field, lack of endocrine cell aggregates in lamina propria, and lack of apoptotic microabscesses.35 Donor lymphocyte microchimerism (<1% donor lymphocyte chimerism) is often seen in liver transplant recipients; it is postulated that microchimerism is important for immune tolerance and graft acceptance by the host.3, 36–38 In contrast, patients with GVHD have donor lymphocyte macrochimerism (>1% donor lymphocyte chimerism) in recipient tissues (skin, gastrointestinal mucosa, peripheral blood), ranging from 1% to 80%.3, 19, 38-45 However, donor lymphocyte macrochimerism in peripheral blood alone doesn’t confirm the diagnosis of GVHD.46 Macrochimerism in a patient with clinical and histological features suggestive of GVHD (involvement of the skin, bone marrow and/or gastrointestinal tract) should be considered diagnostic of GVHD. Confirmation of macrochimerism should not be required to start treatment for GVHD, as it may take several days for results to be obtained. Monitoring donor lymphocyte macrochimerism in target organs and peripheral blood may be helpful, even after resolution of symptoms, as persistence of macrochimerism may suggest incomplete resolution of GVHD and a high risk of relapse with tapering of immunosuppression. Ferritin level was checked in 4 of the UIHC GVHD patients and was markedly elevated in all. The mean peak ferritin in patients who died of GVHD in our case series was 9930 ng/mL (range: 2225–20,333) while the peak ferritin in the surviving patient was 733 ng/mL. Marked hyperferritinemia in GVHD after LT has not been previously reported. Though serum ferritin is a nonspecific acute phase reactant, an extreme elevation of ferritin level is seen only in a few conditions.47 Cytokines released by activated donor lymphocytes, and the associated inflammatory response, is the likely mechanism behind hyperferritinemia in GVHD. Treatment and Outcome Fourteen of the 17 reported treatment regimens for GVHD after LT were associated with mortality rates over 70%, including all regimens that included high dose intravenous steroids only or an increase in CNI dose (with both tacrolimus and cyclosporine). Only 3 reported treatment regimens for GVHD after LT yielded mortality rates less than 60%. These regimens, used in a small number of patients, included IL-2 antagonists (basiliximab or daclizumab), the CD2 inhibitor alefacept, or TNF-α inhibitors. The efficacy of IL-2 antagonists in case series of patients with GVHD after hematopoietic stem cell transplantation (HSCT) has shown promise, with survival of 40–60%.48, 49 Though the survival rate may be better with IL-2 antagonists compared to other reported regimens for GVHD after LT, mortality rate is still substantial. Starting high-dose steroids upon diagnosis of GVHD after LT, with addition of Alefacept and ATG when the patient develops pancytopenia 50, 51, was shown to result in the immediate rebound of bone marrow function. Alefacept is a fusion protein that binds to the lymphocyte antigen CD2, inhibits the interaction of CD2 and human leukocyte function antigen-3 (LFA-3), thereby preventing the activation of CD4 and CD8 T-lymphocytes; while ATG eliminates the activated effector T cells. Alefacept also showed potential benefit for treatment of GVHD in bone marrow transplantation recipients.52 Unfortunately, Alefacept, has been discontinued by the manufacturers, without any safety or FDA regulation concerns.53 Sipilizumab is a similar agent that targets the CD2 receptor on T-lymphocytes and natural killer cells. A phase I trial of sipilizumab for treatment of GVHD after bone marrow transplantation in children reported a good response, but a higher incidence of post-transplant lymphoproliferative disorder was noted, raising safety concerns.54 Initial studies of patients with steroid-refractory GVHD post-HSCT showed promising results with the use of the TNF α inhibitor, infliximab.55, 56 Recent studies, including a phase-III study in patients with GVHD after HSCT, showed no benefit of addition of infliximab to methylprednisolone compared to methylprednisolone alone.57 However, higher response rates have been reported with etanercept in patients with GVHD after HSCT.58–62 Etanercept, unlike infliximab, does not lead to antibody-dependent cytotoxicity and induction of apoptosis of TNF-α-positive monocytes, possibly decreasing risk of infection compared to infliximab. The literature on the use of TNF-α inhibitors in GVHD after LT is limited. However, based on the high mortality with the majority of reported regimens, the 75% survival among the 4 patients treated with TNF-α-inhibition, and the data on etanercept in HSCT patients with GVHD, etanercept or other TNF-α antagonists could be useful as a second line agent in patients with GVHD after LT who are steroid-refractory or dependent. The major drawback of increasing immunosuppression in patients with GVHD after LT is the high risk of death from sepsis. Enterobacter septicemia, invasive aspergillosis and disseminated Candida infection are common causes of death with GVHD. A study performed on patients with GVHD post-HSCT showed a significant increase in the risk of non-Candida invasive fungal infections with the use of infliximab.63 Thus, vigilance for development of infection and timely use of antibiotics and antifungal agents is very important. Empiric use of antibiotics to cover gram negatives and anaerobic bacteria, especially VRE, and antifungal prophylaxis, appears reasonable. CMV and Pneumocystis prophylaxis is advised during high-level immunosuppression. The role of granulocyte-monocyte colony stimulating factor in GVHD is unclear, but may be given in patients with marked neutropenia. In patients with gastrointestinal GVHD after HSCT, a step-wise oral “GVHD diet” may be beneficial.64 Severe protein-calorie malnutrition as a result of protein losing enteropathy and malabsorption is treated with 1.5 g/kg/day of protein. In addition, these patients may develop vitamin, micronutrient and essential trace element (including magnesium and zinc) deficiencies. In patients with massive diarrhea, total parenteral nutrition may be needed. When diarrhea is less than 500 mL/day, oral foods are introduced in a step-wise manner.64 This approach may be beneficial in patients with GVHD after LT, though no data is available. Extracorporeal photopheresis, an apheresis and photodynamic therapy, has shown promising results in the treatment of patients with GVHD after HSCT.65 It is an immunomodulator therapy which involves collection of peripheral blood mononuclear cells, irradiation of these leucocyte cells in-vitro by ultraviolet A in the presence of the drug 8-methoxypsoralen, followed by re-infusion of the cells into the patient. The main advantage of this therapy is the absence of generalized immunosuppression, thereby decreasing the risk of developing life-threatening infections. Further trials are needed before establishing extracorporeal photopheresis as a treatment option for GVHD. Proposed diagnostic algorithm and treatment recommendations How then do we diagnose and treat our next patient with GVHD after LT? Based on our interpretation of currently available data, we propose a diagnostic algorithm (Figure 4) for GVHD after LT. Patients who have symptoms involving the most commonly involved organ systems in GVHD, namely skin, gastrointestinal tract and bone marrow, should be evaluated for GVHD. Patients with maculopapular skin rash post-LT should undergo skin biopsy, as it is a simple procedure with low morbidity, and the histology can be diagnostic of GVHD. In patients who present with diarrhea or pancytopenia after LT, the most common causes of these symptoms should be ruled out. If symptoms persist, a colonoscopy with mucosal biopsies should be performed to screen for GVHD changes. Strong treatment recommendations cannot be made due to the absence of prospective studies and due to the high mortality rates for the majority of reported treatment regimens. Multidisciplinary involvement, with hematologists, infectious disease specialists and immunologists, is essential in the management of this complex condition. Study limitations The proposed diagnostic algorithm is based on limited evidence. Comparisons of treatment regimen mortality rates are based on small cohort sizes and do not take into consideration other patient- or disease-related factors which may affect mortality rates. With only summary data available from the UNOS database, comparisons with reported US patients with GVHD are limited to univariate tests of association. Clearly, all US cases of GVHD after LT in the US have not been reported, and occurrence of GVHD after LT is not reported in the UNOS database, limiting the interpretation of statistical analysis. CONCLUSIONS GVHD after LT is infrequent, but is associated with a very high mortality rate. Most patients develop GVHD in 3–5 weeks after LT. GVHD may occur more often in LT patients over 50 years of age and who have diabetes. When reported US GVHD cases were compared to all LT patients in the UNOS database, HCC appeared to be over-represented, and HCV was under-represented. High-dose steroids alone, or combined with increasing CNI dose, are not effective treatment regimens. High-dose steroids combined with IL-2 antagonists or TNF-α inhibitors may be more promising approaches, though experience is limited. Donor macrochimerism and serum ferritin may be helpful for monitoring response to treatment. The participation of multiple institutions in a working group to prospectively study GVHD after LT, along with obligatory reporting of GVHD cases to UNOS, is needed. Supplementary Material Supplemental File Grants and financial support: This project is in part supported by R56 AI 116715 from the NIAID (National Institute of Allergy and Infectious Diseases). No other funding received for the manuscript. List of Abbreviations ALD Alcoholic liver disease ANC Absolute neutrophil count ATG Anti-thymocyte globulin CMV Cytomegalovirus CNI Calcineurin inhibitor GVHD Graft versus host disease HBV Hepatitis B virus HCC Hepatocellular carcinoma HCV Hepatitis C virus HSCT Hematopoietic stem cell transplant IL Interleukin IVIG Intravenous immunoglobulin LT Liver transplantation NASH Nonalcoholic steatohepatitis PBC Primary biliary cholangitis STR Short tandem repeats TNF Tumor necrosis factor UIHC University of Iowa Hospitals and Clinics UNOS United Network for Organ Sharing US United States VRE Vancomycin resistant enterococci WBC White blood cell Figure 1 Maculopapular skin rash in a patient with graft versus host disease after liver transplantation. Figure 2 Skin biopsy demonstrating perivascular mononuclear infiltrate (arrow) in the superficial dermis as well as vacuolar interface change, including scattered apoptotic keratinocytes (Grade 2 Graft Versus Host Disease). Figure 3 Ileal biopsy demonstrates apoptotic crypt epithelial cells (arrow) and degenerating crypts suggestive of graft versus host disease. Figure 4 Proposed diagnostic algorithm for GVHD in liver transplantation recipients. Table 1 Characteristics of patients with GVHD after liver transplantation at UIHC Year Age Sex Etiology DM GVHD onset (days) Involved systems (grade) GVHD related death Survival (days) Treatment for GVHD Peak serum ferritin (ng/ml) 2014 73 M NASH Yes 65 Skin (II) GI (I) BM Yes 161 High dose steroids, continued tacrolimus. 2225 2011 66 M ALD, HCC Yes 46 Skin (I) GI (IV) BM Yes 79 High dose steroids, basiliximab, IVIG and increased tacrolimus. 20,333 2011 60 M ALD, HCV, HCC Yes 117 Skin (II) BM Yes 11 Added ATG, increased tacrolimus, and continued mycophenolate. 7232 2002 60 M ALD, Hemo chromatosis Yes 85 Skin (II) GI (II) No >10 years High dose steroids, continued tacrolimus and mycophenolate. 733 1996 65 F PBC Yes 15 Skin BM Yes 5 High dose steroids, and continued tacrolimus. NA UIHC: University of Iowa Hospitals and Clinics, GVHD: Graft-versus-host-disease, ALD: alcoholic liver disease; ATG: Anti-thymocyte globulin; BM: Bone marrow; DM: Diabetes mellitus; F: Female; GI: Gastrointestinal tract; HCC: Hepatocellular cancer; HCV: Hepatitis C; M: Male; F: Female, NA: Not available; NASH: nonalcoholic steatohepatitis; PBC: Primary biliary cholangitis. Table 2 Patient characteristics of reported cases of acute GVHD after LT in the world literature, US literature, and of all adult liver transplants in UNOS database GVHD after LT (World literature) GVHD after LT (US) UNOS database1 p N 156 66 119,701 Age: n (%)  18 – 35 28 (17.9) 5 (7.6) 9,133 (7.6) <0.01  35 – 49 30 (19.2) 7 (10.6) 34,329 (28.7)  50 – 64 77 (49.4) 43 (65.2) 64,061 (53.5)  ≥65 21 (13.5) 11 (16.7) 12,178 (10.2) Gender, male: n (%) 105 (67.3) 44 (66.7) 84,030 (70.2) 0.53 Etiology of liver disease*: n (%)  1. Acute liver failure 3 (2.5) 1 (2.0) 6,824 (5.7)  2. Alcoholic liver disease 27 (22.9) 12 (23.5) 21,495 (18.0)  3. Chronic Hepatitis C 16 (13.6) 6 (11.8) 35,812 (29.9)  4. Chronic hepatitis B 23 (19.5) 5 (9.8) 5,168 (4.3)  5. Primary Sclerosing cholangitis 9 (7.6) 7 (13.7) 7,153 (6.0)  6. Primary Biliary Cirrhosis 12 (10.2) 5 (9.8) 5,754 (4.8)  7. Nonalcoholic steatohepatitis 13 (11.1) 9 (17.6) 13,415 (11.2)  8. Alpha-1 antitrypsin deficiency 2 (1.7) 0 (0) 1,578 (1.3)  9. Autoimmune hepatitis 4 (3.4) 2 (3.9) 3,802 (3.2)  10.Hemochromatosis 3 (2.5) 3 (5.9) 749 (0.6) Presence of Hepatocellular Carcinoma: n (%) 41 (34.7) 11 (21.6) 15,542 (13.0) Presenting organ involvement+: n(%)  1. Skin only 33 (31) 11 (23)  2. Bone Marrow only 17 (16) 10 (21)  3. Skin + GI + BM 17 (16) 9 (19)  4. Skin + GI 14 (13.2) 6 (13)  5. Skin + BM 14 (13.2) 7 (14.5)  6. GI only 11 (10.4) 5 (10.4) GVHD: graft-versus-host-disease; LT: liver transplantation; US: United States; UNOS: United network for organ sharing. * Etiology of liver disease was reported in 118 total cases, 51 of them were from USA. + Presenting organ involvement was reported in 106 cases. 1 Reported US GVHD cases were subtracted from the respective age and etiology categories in UNOS database for statistical analysis. Table 3 Combination immunosuppression regimens and outcomes of patients with GVHD after liver transplantation Treatment regimen Number of patients Mortality n (%) A. Steroid containing regimens  Steroids only*1, 9, 14, 27, 50, 66–75 25 21 (84)  Steroids + CNI dose increase13, 23, 76–80 8 6 (75)  Steroids + IVIG22, 80, 81 4 4 (100)  Steroids + Azathioprine12, 82, 83 3 3 (100)  Steroids ± OKT31, 11, 84–88 7 5 (71.4) B. ATG containing regimens  ATG only*1, 14, 89, 90 4 3 (75)  ATG + steroids*1, 11, 27, 91–97 22 18 (81)  ATG + Steroids + CNI98, 99 3 3 (100) C. IL-2 antagonist containing regimens  IL 2 antagonist + Steroids*17, 27, 72, 96, 100–103 12 7 (58)  IL 2 antagonist + Steroids + CNI23, 104 2 2 (100)  IL 2 antagonist + Steroids + ATG14, 15, 21, 105 4 4 (100) D. Other treatment regimens  Alefacept + steroids + ATG16, 50, 51 7 2 (28)  TNF alpha inhibitors + steroids + ATG17, 106–108 4 1 (25)  Rituximab Steroids ± ATG ± IL 2 antagonist109–112 4 3 (75) * No addition or dose increase of CNI was reported. GVHD: graft versus host disease; ATG: Anti-thymocyte globulin; CNI: calcineurin inhibitors; IL-2: interleukin-2; IVIG: Intravenous immunoglobulin; MLAG: Minnesota anti-lymphocyte globulin; OKT3: Muromonab-CD3; TNF: tumor necrosis factor. Table 4 Treatment regimen for patients who survived GVHD after Liver transplantation. Treatment regimen Number of patients (n) A. Steroid containing regimens  Steroids only7, 68, 69, 75 4  Steroids + CNI dose increase23, 75, 113 3  Steroids + OKT384, 87 2 B. ATG containing regimens  ATG only1 1  ATG + steroids92, 94, 97 4  ATG + Steroids + CNI98 1 C. IL-2 antagonist containing regimens  IL 2 antagonist + Steroids27, 101–103 5  IL-2 antagonist + Steroids + ATG105 1 D. Other treatment regimens  Alefacept + steroids + ATG50, 51 5  TNF alpha inhibitors + steroids + ATG106–108 3  Rituximab + Steroids + CNI dose increase + IL 2 antagonist109 1 * Immunosuppression was decreased in 3 patients who survived GVHD after liver transplantation. GVHD: graft versus host disease; ATG: Anti-thymocyte globulin; CNI: calcineurin inhibitors; IL-2: interleukin-2; OKT3: Muromonab-CD3; TNF: tumor necrosis factor. No other financial disclosures. No conflicts of interest. Presented in part at DDW 2015, Washington D.C, USA Author contributions 1. Arvind R Murali: study concept and design, acquisition of data, analysis and interpretation of data, drafting of manuscript, statistical analysis 2. Subhash Chandra: study concept and design, acquisition of data, analysis and interpretation of data, drafting of manuscript, statistical analysis 3. Zoe Stewart: critical revision of manuscript 4. Bruce R Blazar: critical revision of manuscript 5. Umar Farooq: critical revision of manuscript 6. M Nedim Ince: obtained funding, critical revision of manuscript 7. Jeffrey Dunkelberg: study concept and design, analysis and interpretation of data, drafting of manuscript, critical revision of manuscript, study supervision. 1 Smith DM Agura E Netto G Liver transplant-associated graft-versus-host disease Transplantation 2003 75 118 26 12544883 2 Chan EY Larson AM Gernsheimer TB Recipient and donor factors influence the incidence of graft-vs.-host disease in liver transplant patients Liver Transpl 2007 13 516 22 17394149 3 Taylor AL Gibbs P Sudhindran S Monitoring systemic donor lymphocyte macrochimerism to aid the diagnosis of graft-versus-host disease after liver transplantation Transplantation 2004 77 441 6 14966423 4 Lerner KG Kao GF Storb R Histopathology of graft-vs.-host reaction (GvHR) in human recipients of marrow from HL-A-matched sibling donors Transplant Proc 1974 6 367 71 4155153 5 Sale GE Shulman HM McDonald GB Gastrointestinal graft-versus-host disease in man. A clinicopathologic study of the rectal biopsy Am J Surg Pathol 1979 3 291 9 44107 6 Chinnakotla S Smith DM Domiati-Saad R Acute graft-versus-host disease after liver transplantation: role of withdrawal of immunosuppression in therapeutic management Liver Transpl 2007 13 157 61 17192857 7 Gao PJ Leng XS Wang D Graft versus host disease after liver transplantation: a case report Frontiers of medicine in China 2010 4 469 72 21125347 8 Nah YW Nam CW Park SE Graft versus host disease following liver transplantation. Liver Transplantation Conference: 16th Annual International Congress of the International Liver Transplantation Society Hong Kong China. Conference Start 2010 16 9 Roberts JP Ascher NL Lake J Graft vs. host disease after liver transplantation in humans: a report of four cases Hepatology 1991 14 274 81 1860684 10 Kotru A Gupta M Maloo M Graft versus host disease following liver transplantation - A diagnostic dilemma HPB 2010 12 379 11 Kohler S Pascher A Junge G Graft versus host disease after liver transplantation - a single center experience and review of literature Transplant Int 2008 21 441 51 12 Romagnuolo J Jewell LD Kneteman NM Graft-versus-host disease after liver transplantation complicated by systemic aspergillosis with pancarditis Can J Gastroenterol 2000 14 637 40 10978951 13 Schrager JJ Vnencak-Jones CL Graber SE Use of short tandem repeats for DNA fingerprinting to rapidly diagnose graft-versus-host disease in solid organ transplant patients Transplantation 2006 81 21 5 16421472 14 Perri R Assi M Talwalkar J Graft vs. host disease after liver transplantation: a new approach is needed Liver Transpl 2007 13 1092 9 17663410 15 Ghali MP Talwalkar JA Moore SB Acute graft-versus-host disease after liver transplantation Transplantation 2007 83 365 6 16 Rogulj IM Deeg J Lee SJ Acute graft versus host disease after orthotopic liver transplantation Journal of Hematology and Oncology 2012 5 22397365 17 Blank G Li J Kratt T Treatment of liver transplant graft-versus-host disease with antibodies against tumor necrosis factor-alpha Exp Clin Transplant 2013 11 68 71 23387543 18 Pollack MS Speeg KV Callander NS Severe, late-onset graft-versus-host disease in a liver transplant recipient documented by chimerism analysis Hum Immunol 2005 66 28 31 15620459 19 Kohler S Pascher A Junge G Graft versus host disease after liver transplantation - a single center experience and review of literature Transpl Int 2008 21 441 51 18266778 20 Connors J Drolet B Walsh J Morbilliform eruption in a liver transplantation patient Arch Dermatol 1996 132 1161 3 8859025 21 Meves A el-Azhary RA Talwalkar JA Acute graft-versus-host disease after liver transplantation diagnosed by fluorescent in situ hybridization testing of skin biopsy specimens Journal of the American Academy of Dermatology 2006 55 642 6 17010745 22 Post GR Black JS Cortes GY The utility of fluorescence in situ hybridization (FISH) analysis in diagnosing graft versus host disease following orthotopic liver transplant Ann Clin Lab Sci 2011 41 188 92 21844579 23 Chandra SMA Dunkelberg J Stewart Z Graft Versus Host Disease Post-Liver Transplantation: A Single Center Gastroenterology Iran J Med Sci 2015 148 S1042 1043 24 Au WY Lo CM Hawkins BR Evans' syndrome complicating chronic graft versus host disease after cadaveric liver transplantation Transplantation 2001 72 527 8 11502987 25 Elfeki MGP Pungpapong S Nguyen J Harnois D Graft-Versus-Host Disease After Orthotopic Liver Transplantation: Multivariate Analysis of Risk Factors Am J Transplant 2015 15 Abstract 81 26 Sachdeva M Chawla YK Arora SK Immunology of hepatocellular carcinoma World J Hepatol 2015 7 2080 90 26301050 27 Taylor AL Gibbs P Bradley JA Acute graft versus host disease following liver transplantation: the enemy within Am J Transplant 2004 4 466 74 15023138 28 McDonald-Hyman C Turka LA Blazar BR Advances and challenges in immunotherapy for solid organ and hematopoietic stem cell transplantation Sci Transl Med 2015 7 280rv2 29 Bhattarai N McLinden JH Xiang J Conserved Motifs within Hepatitis C Virus Envelope (E2) RNA and Protein Independently Inhibit T Cell Activation PLoS Pathog 2015 11 e1005183 26421924 30 Altiparmak MR Trablus S Pamuk ON Diarrhoea following renal transplantation Clin Transplant 2002 16 212 6 12010146 31 Wong NA Bathgate AJ Bellamy CO Colorectal disease in liver allograft recipients -- a clinicopathological study with follow-up Eur J Gastroenterol Hepatol 2002 14 231 6 11953686 32 Ginsburg PM Thuluvath PJ Diarrhea in liver transplant recipients: etiology and management Liver Transplantation 2005 11 881 90 16035068 33 Ross WA Ghosh S Dekovich AA Endoscopic biopsy diagnosis of acute gastrointestinal graft-versus-host disease: rectosigmoid biopsies are more sensitive than upper gastrointestinal biopsies Am J Gastroenterol 2008 103 982 9 18028511 34 Thompson B Salzman D Steinhauer J Prospective endoscopic evaluation for gastrointestinal graft-versus-host disease: determination of the best diagnostic approach Bone Marrow Transplant 2006 38 371 6 16915225 35 Star KV Ho VT Wang HH Histologic features in colon biopsies can discriminate mycophenolate from GVHD-induced colitis Am J Surg Pathol 2013 37 1319 28 24076772 36 Wood K Sachs DH Chimerism and transplantation tolerance: cause and effect Immunol Today 1996 17 584 7 discussion 588 8991291 37 Starzl TE Demetris AJ Murase N Cell migration, chimerism, and graft acceptance Lancet 1992 339 1579 82 1351558 38 Domiati-Saad R Klintmalm GB Netto G Acute graft versus host disease after liver transplantation: patterns of lymphocyte chimerism Am J Transplant 2005 5 2968 73 16303012 39 Rogulj IM Deeg J Lee SJ Acute graft versus host disease after orthotopic liver transplantation J Hematol Oncol 2012 5 50 22889203 40 Meves A el-Azhary RA Talwalkar JA Acute graft-versus-host disease after liver transplantation diagnosed by fluorescent in situ hybridization testing of skin biopsy specimens J Am Acad Dermatol 2006 55 642 6 17010745 41 Mawad R Hsieh A Damon L Graft-versus-host disease presenting with pancytopenia after en bloc multiorgan transplantation: case report and literature review Transplant Proc 2009 41 4431 3 20005417 42 Key T Taylor CJ Bradley JA Recipients who receive a human leukocyte antigen-B compatible cadaveric liver allograft are at high risk of developing acute graft-versus-host disease Transplantation 2004 78 1809 11 15614155 43 Gulbahce HE Brown CA Wick M Graft-vs-host disease after solid organ transplant Am J Clin Pathol 2003 119 568 73 12710129 44 Chinnakotla S Smith DM Domiati-Saad R Acute graft-versus-host disease after liver transplantation: role of withdrawal of immunosuppression in therapeutic management Liver Transpl 2007 13 157 61 17192857 45 Chaib E Silva FD Figueira ER Graft-versus-host disease after liver transplantation Clinics (Sao Paulo) 2011 66 1115 8 21808887 46 Yao GL Li W Liu AB The experimental results of GVHD following orthotropic liver transplantation Chung Hua Kan Tsang Ping Tsa Chih 2009 17 856 60 19958648 47 Rosario C Zandman-Goddard G Meyron-Holtz EG The hyperferritinemic syndrome: macrophage activation syndrome, Still's disease, septic shock and catastrophic antiphospholipid syndrome BMC Med 2013 11 185 23968282 48 Przepiorka D Kernan NA Ippoliti C Daclizumab, a humanized anti-interleukin-2 receptor alpha chain antibody, for treatment of acute graft-versus-host disease Blood 2000 95 83 9 10607689 49 Massenkeil G Rackwitz S Genvresse I Basiliximab is well tolerated and effective in the treatment of steroid-refractory acute graft-versus-host disease after allogeneic stem cell transplantation Bone Marrow Transplant 2002 30 899 903 12476283 50 Eghtesad B Askar M Bollinger J Graft-versus-host disease (GVHD) following liver transplantation (LT): Comparison of two eras Am J Transplant 2012 12 126 21920020 51 Stotler CJ Eghtesad B Hsi E Rapid resolution of GVHD after orthotopic liver transplantation in a patient treated with alefacept Blood 2009 113 5365 6 19470441 52 Shapira MY Resnick IB Dray L A new induction protocol for the control of steroid refractory/dependent acute graft versus host disease with alefacept and tacrolimus Cytotherapy 2009 11 61 7 19191054 53 National Psoriasis foundation APUSI Amevive (alefacept) voluntarily discontinued in the US Astellas Medical Information Department 2012 54 Brochstein JA Grupp S Yang H Phase-1 study of siplizumab in the treatment of pediatric patients with at least grade II newly diagnosed acute graft-versus-host disease Pediatr Transplant 2010 14 233 41 19671093 55 Yang J Cheuk DK Ha SY Infliximab for steroid refractory or dependent gastrointestinal acute graft-versus-host disease in children after allogeneic hematopoietic stem cell transplantation Pediatr Transplant 2012 16 771 8 22905718 56 Kobbe G Schneider P Rohr U Treatment of severe steroid refractory acute graft-versus-host disease with infliximab, a chimeric human/mouse antiTNFalpha antibody Bone Marrow Transplant 2001 28 47 9 11498743 57 Couriel DR Saliba R de Lima M A phase III study of infliximab and corticosteroids for the initial treatment of acute graft-versus-host disease Biol Blood Marrow Transplant 2009 15 1555 62 19896079 58 Kennedy GA Butler J Western R Combination antithymocyte globulin and soluble TNFalpha inhibitor (etanercept) +/− mycophenolate mofetil for treatment of steroid refractory acute graft-versus-host disease Bone Marrow Transplant 2006 37 1143 7 16699531 59 Busca A Locatelli F Marmont F Recombinant human soluble tumor necrosis factor receptor fusion protein as treatment for steroid refractory graft-versus-host disease following allogeneic hematopoietic stem cell transplantation Am J Hematol 2007 82 45 52 16937391 60 Park JH Lee HJ Kim SR Etanercept for steroid-refractory acute graft versus host disease following allogeneic hematopoietic stem cell transplantation Korean J Intern Med 2014 29 630 6 25228839 61 Uberti JP Ayash L Ratanatharathorn V Pilot trial on the use of etanercept and methylprednisolone as primary treatment for acute graft-versus-host disease Biol Blood Marrow Transplant 2005 11 680 7 16125638 62 Levine JE Paczesny S Mineishi S Etanercept plus methylprednisolone as initial therapy for acute graft-versus-host disease Blood 2008 111 2470 5 18042798 63 Marty FM Lee SJ Fahey MM Infliximab use in patients with severe graft-versus-host disease and other emerging risk factors of non-Candida invasive fungal infections in allogeneic hematopoietic stem cell transplant recipients: a cohort study Blood 2003 102 2768 76 12855583 64 van der Meij BS de Graaf P Wierdsma NJ Nutritional support in patients with GVHD of the digestive tract: state of the art Bone Marrow Transplant 2013 48 474 82 22773121 65 Calore E Marson P Pillon M Treatment of Acute Graft-versus–Host Disease in Childhood with Extracorporeal Photochemotherapy/Photopheresis: The Padova Experience Biol Blood Marrow Transplant 2015 21 1963 72 26183078 66 Collins RH Jr Cooper B Nikaein A Graft-versus-host disease in a liver transplant recipient Ann Intern Med 1992 116 391 2 1736772 67 Mazzaferro V Andreola S Regalia E Confirmation of graft-versus-host disease after liver transplantation by PCR HLA-typing Transplantation 1993 55 423 5 8434393 68 Knox KS Behnia M Smith LR Acute graft-versus-host disease of the lung after liver transplantation Liver Transpl 2002 8 968 71 12360443 69 Nemoto T Kubota K Kita J Unusual onset of chronic graft-versus-host disease after adult living-related liver transplantation from a homozygous donor Transplantation 2003 75 733 6 12640319 70 Schoniger-Hekele M Muller C Kramer L Graft versus host disease after orthotopic liver transplantation documented by analysis of short tandem repeat polymorphisms Digestion 2006 74 169 73 17341849 71 Sun B Zhao C Xia Y Late onset of severe graft-versus-host disease following liver transplantation Transplant Immunol 2006 16 250 3 72 Wang B Lu Y Yu L Diagnosis and treatment for graft-versus-host disease after liver transplantation: two case reports Transplant Proc 2007 39 1696 8 17580224 73 Guo ZY He XS Wu LW Graft-verse-host disease after liver transplantation: A report of two cases and review of literature World J Gastroenterol 2008 14 974 979 18240363 74 Jeanmonod P Hubbuch M Grunhage F Graft-versus-host disease or toxic epidermal necrolysis: diagnostic dilemma after liver transplantation Transplant Infect Dis 2012 14 422 6 75 Mataluni G Mangiardi M Rainone M Central and peripheral nervous system involvement as a manifestation of graft versus host disease after liver transplantation: Case report J Peripher Nerv Syst 2013 18 S20 76 Dumortier J Souraty P Bernard P Graft versus host disease following liver transplantation: Favorable outcome after conversion to tacrolimus. [French] Gastroenterol Clin Biol 2003 27 561 562 12843926 77 Soejima Y Shimada M Suehiro T Graft-versus-host disease following living donor liver transplantation Liver Transpl 2004 10 460 4 15004778 78 Hossain MS Roback JD Pollack BP Chronic GvHD decreases antiviral immune responses in allogeneic BMT Blood 2007 109 4548 56 17289817 79 Yilmaz M Ozdemir F Akbulut S Chronic graft-versus-host disease after liver transplantation: a case report Transplant Proc 2012 44 1751 3 22841262 80 Chen XB Yang J Xu MQ Unsuccessful treatment of four patients with acute graft-vs-host disease after liver transplantation World J Gastroenterol 2012 18 84 9 22228975 81 Whalen JG Jukic DM English JC 3rd Rash and pancytopenia as initial manifestations of acute graft-versus-host disease after liver transplantation J Am Acad Dermatol 2005 52 908 12 15858489 82 Schmuth M Vogel W Weinlich G Cutaneous lesions as the presenting sign of acute graft-versus-host disease following liver transplantation Br J Dermatol 1999 141 901 4 10583176 83 Au WY Ma SK Kwong YL Graft-versus-host disease after liver transplantation: documentation by fluorescent in situ hybridisation and human leucocyte antigen typing Clin Transplant 2000 14 174 7 10770425 84 Redondo P Espana A Herrero JI Graft-versus-host disease after liver transplantation J Am Acad Dermatol 1993 29 314 7 8340506 85 Sanchez-Izquierdo JA Lumbreras C Colina F Severe graft versus host disease following liver transplantation confirmed by PCR-HLA-B sequencing: report of a case and literature review Hepato-Gastroenterol 1996 43 1057 61 86 Burt M Jazwinska E Lynch S Detection of circulating donor deoxyribonucleic acid by microsatellite analysis in a liver transplant recipient Liver Transpl Surg 1996 2 391 4 9346682 87 PAIZIS G TAIT BD ANGUS PW Successfid resolution of severe graft versus host disease after liver transplantation correlating with disappearance of donor DNA fkorn the peripheral blood Aust NZ J Med 1998 28 830 832 88 Neumann UP Kaisers U Langrehr JM Fatal graft-versus-host-disease: a grave complication after orthotopic liver transplantation Transplant Proc 1994 26 3616 7 7998294 89 DePaoli AM Bitran J Graft-versus-host disease and liver transplantation Ann Intern Med 1992 117 170 1 1605435 90 Hara H Ohdan H Tashiro H Differential diagnosis between graft-versus-host disease and hemophagocytic syndrome after living-related liver transplantation by mixed lymphocyte reaction assay J Invest Surg 2004 17 197 202 15371161 91 Chiba T Yokosuka O Goto S Clinicopathological features in patients with hepatic graft-versus-host disease Hepatogastroenterology 2005 52 1849 53 16334791 92 Burdick JF Vogelsang GB Smith WJ Severe graft-versus-host disease in a liver-transplant recipient N Engl J Med 1988 318 689 91 3278235 93 Pageaux GP Perrigault PF Fabre JM Lethal acute graft-versus-host disease in a liver transplant recipient: relations with cell migration and chimerism Clin Transplant 1995 9 65 9 7742585 94 Aziz H Trigo P Lendoire J Successful treatment of graft-vs-host disease after a second liver transplant Transplant Proc 1998 30 2891 2 9745613 95 Merhav HJ Landau M Gat A Graft versus host disease in a liver transplant patient with hepatitis B and hepatocellular carcinoma Transplant Proc 1999 31 1890 1 10371987 96 Hanaway M Buell J Musat A Graft-Versus-Host Disease in Solid Organ Transplantation Graft 2001 4 205 97 Lehner F Becker T Sybrecht L Successful outcome of acute graft-versus-host disease in a liver allograft recipient by withdrawal of immunosuppression Transplantation 2002 73 307 10 11821752 98 Schulman JM Yoon C Schwarz J Absence of peripheral blood chimerism in graft-vs-host disease following orthotopic liver transplantation: case report and review of the literature Int J Dermatol 2014 53 e492 8 24372059 99 Calmus Y Graft-versus-host disease following living donor liver transplantation: High risk when the donor is HLA-homozygous J Hepatol 2004 41 505 507 100 Merle C Blanc D Flesch M A picture of epidermal necrolysis after hepatic allograft. Etiologic aspects Ann Dermatol Venereol 1990 117 635 9 2260804 101 Rinon M Maruri N Arrieta A Selective immunosuppression with daclizumab in liver transplantation with graft-versus-host disease Transplant Proc 2002 34 109 10 11959211 102 Sudhindran S Taylor A Delriviere L Treatment of graft-versus-host disease after liver transplantation with basiliximab followed by bowel resection Am J Transplant 2003 3 1024 9 12859540 103 Husova L Mejzlik V Jedlickova H Graft-versus-host disease as an unusual complication following liver transplant. [Czech] Gastroenterologie a Hepatologie 2012 66 109 115 104 Chaman JC Padilla PM Rondon CF Graft-versus-host disease after liver transplantation in a partial monosomy 11q of chromosoma 11: A case report Liver Transpl 2013 1 S181 105 Lu Y Wu LQ Zhang BY Graft-versus-host disease after liver transplantation: successful treatment of a case Transplant Proc 2008 40 3784 6 19100490 106 Piton G Larosa F Minello A Infliximab treatment for steroid-refractory acute graft-versus-host disease after orthotopic liver transplantation: a case report Liver Transpl 2009 15 682 5 19562700 107 Xu X Zhou L Ling Q Successful management of graft-versus-host disease after liver transplantation with Korean red ginseng based regimen Liver Transpl Conference: 16th Annual International Congress of the International Liver Transplantation Society Hong Kong China. Conference Start 2010 16 108 Thin L Macquillan G Adams L Acute graft-versus-host disease after liver transplant: novel use of etanercept and the role of tumor necrosis factor alpha inhibitors Liver Transp 2009 15 421 6 109 Grosskreutz C Gudzowaty O Shi P Partial HLA matching and RH incompatibility resulting in graft versus host reaction and Evans syndrome after liver transplantation Am J Hematol 2008 83 599 601 18470884 110 Mawad R Hsieh A Damon L Graft-versus-host disease presenting with pancytopenia after en bloc multiorgan transplantation: case report and literature review Transplant Proc 2009 41 4431 3 20005417 111 Young AL Leboeuf NR Tsiouris SJ Fatal disseminated Acanthamoeba infection in a liver transplant recipient immunocompromised by combination therapies for graft-versus-host disease Transplant Infect Dis 2010 12 529 37 112 Cheung CYM Leung AYH Chan SC Fatal graft-versus-host disease after unrelated cadaveric liver transplantation due to donor/recipient human leucocyte antigen matching Inten Med J 2014 44 425 426 113 Hassan G Khalaf H Mourad W Dermatologic complications after liver transplantation: a single–center experience Transplant Proc 2007 39 1190 4 17524929

PMC005xxxxxx/PMC5118144.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7503062 4443 J Am Geriatr Soc J Am Geriatr Soc Journal of the American Geriatrics Society 0002-8614 1532-5415 27641829 5118144 10.1111/jgs.14455 NIHMS793377 Article Psychological Well-being of Grandparents Caring for Grandchildren among Older Chinese Americans: Burden or Blessing? Tang Fengyan PhD 1 Xu Ling PhD 2 Chi Iris DSW 3 Dong XinQi MD 4 1 University of Pittsburgh, Pittsburgh, Pennsylvania 2 The University of Texas at Arlington, Arlington, Texas 3 University of Southern California, Los Angeles, California 4 Rush Medical College, Chicago, Illinois Contact: Fengyan Tang, University of Pittsburgh - School of Social Work, 2217C Cathedral of Learning, Pittsburgh Pennsylvania 15260, fet7@pitt.edu, T: 4126489356, F: 4126246323 10 6 2016 19 9 2016 11 2016 01 11 2017 64 11 23562361 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The rapid increase in grandparents caring for grandchildren has received growing attention; however, relatively little research has focused on Chinese American grandparents and their caregiving experiences. Drawing on cross-sectional data from the Population Study of Chinese Elderly (PINE) – a community-engaged, epidemiological study of Chinese American adults aged 60 and older, we examined the relationships between caregiving experiences and psychological well-being. Among 2,365 older adults who answered the question about grandparent caregiving, 818 (35%) were designated as caregivers, spending an average of 12 hours a week on childcare. About one in five caregivers reported caregiving burden, pressure, or negative health effect of caregiving. Compared with noncaregivers, caregivers had better psychological well-being, with significantly lower levels of depressive symptoms, anxiety, stress, and loneliness. But increased levels of caregiving burden, pressure from adult children, and perceived negative effect were related to increased rates for psychological distress. With a strong cultural expectation for family care, grandparent caregiving is generally associated with positive psychological well-being; but it is also a stress process, especially when older adults feel pressured and/or burdened to provide childcare. The study implies that cultural values and life transitions may shape grandparent caregiving experiences and well-being, pointing to the importance of respecting cultural differences in family caregiving. Understanding both positive and negative aspects of grandparent caregiving and the underlying mechanisms will help healthcare professionals identify caregivers at risk of psychological distress and provide proper interventions to attenuate negative outcomes while maximizing positive experience in Chinese American older adults. grandparent caregiver psychological well-being older Chinese Americans PINE Introduction The rapid increase in grandparents caring for grandchildren has received growing attention in research, policy, and practice since the 1990s.1 However, relatively little research has focused on Chinese American grandparents and their caregiving experiences. With a strong cultural expectation of grandparent caregiving, many Chinese older adults have immigrated to meet their adult children’s need for childcare assistance;2 however, migration links to the erosion of traditional culture and intergenerational value differences, which may further complicate caregiving experience.3 Chinese grandparents are expected to engage in co-parenting and care for grandchildren on an extensive daycare basis; 4,5 meanwhile, they have to cope with various practical challenges, including cultural adaptation, language barriers, financial and housing difficulty.6 As a result, they may experience both positive and negative health outcomes related to childcare, as documented in the literature. Yet the mechanisms to health outcomes may differ across racial/ethnic groups. Using the first population-based, epidemiological study of older Chinese Americans, this study examines the mechanisms of the associations between caregiving experiences and psychological well-being. Findings will help healthcare professionals identify caregivers at risk of psychological distress and provide proper interventions to attenuate negative outcomes while maximizing positive experience in Chinese American older adults. Probably due to cultural expectations, Chinese grandparents may not consider caring for grandchildren as a burden, but part of parenthood or family obligation.1.6 Childcare provided by grandparents is viewed as a family adaptive strategy to enhance family well-being by alleviating child mothers’ burden and increasing education and employment opportunities, because caregiving is assumed or considered as a women’s role in Chinese families.7 In fact, assistance with childcare is frequently cited as the main reason of immigration for many Chinese older adults,8 who are valuable resources to their adult children in America and contribute to the family as caregivers.2 If grandparents accept the caregiver role or the changed role in a given situation, e.g., immigration, they likely perceive better health linked with caregiving, as shown in a study of Filipino American grandparents.9 Studies in the U.S. and Hong Kong indicate that Chinese grandparents feel appreciated and loved, and are less depressive when they perceive grandparent role can enhance familial relationships.10,11 Among grandparents in Taiwan, long-term multigenerational caregiving is related to improvement in self-rated health and mobility.12 Caregiving is also associated with negative health outcomes. The incidence of depression, diabetes, hypertension, and insomnia is high among grandparent caregivers, as documented in Western literatue.13 Particularly for older Chinese grandparents, caregiving may exacerbate the immigration-related stresses, increasing the risks for social isolation and intergenerational conflicts.10 According to the role strain theory, grandparents who provide extensive care may experience role strain and the associated effects on their employment, self-care, or relationship with spouse or others.14 Consequently, the elevated level of caregiving stress may negatively affect grandparents’ mental and physical health.14 Yet social support from family networks may mitigate the negative health effects of caregiving, as postulated from the stress-buffering model.15 Grandchild’s parents can become a source of support, conflict/burden, or both.16 Support from adult children may buffer the negative effects of role strain; whereas pressure from them may increase caregiving stress. According to the Census Bureau, the older Chinese American population has increased almost four times as rapidly as the U.S. general older population from 2000 to 2010, with a large share of foreign-born and recent immigrants.17 They have been actively involved in assisting their adult children in raising the next generation. By contrast, the occurrence of grandparent caregiving in other racial/ethnic groups is usually the result of a crisis situation that impairs the ability of birth parents to adequately care for their children.18 Older Chinese Americans may experience struggles with cultural values and conflicts with family members, which further exacerbate childcare burden and the associated psychological distress. To understand how different aspects of caregiving experiences are related to psychological well-being in older Chinese Americans, we examine the mechanisms through which caregiving is associated with well-being and hypothesize that caregiving burden, pressure, and perceived negative effect are related to caregivers’ psychological well-being. The study will illustrate the importance of childcare in grandparents’ lives and the potential negative aspects of caregiving, thus providing implications on how to maximize caregiving benefits and minimize the drawbacks through proper interventions from healthcare professionals. Methods Sample We used the data from the Population-based Study of Chinese Elderly (PINE), a community-engaged, epidemiological study of Chinese Americans aged 60 and older. The PINE study was conducted between 2011 and 2013 in the greater Chicago area. Under the guidance of a community-based participatory research approach, the study was implemented by a team of the Rush Institute for Healthy Aging, Northwestern University Medical Center, and many community-based social services agencies and organizations. To enhance community participation and ensure study relevance, the research team applied culturally and linguistically appropriate community recruitment strategies, including engaging community-based agencies, integrating recruitment with routine services, recruiting through family members, neighbors and various social media, and involving family members in participation decision-making. After participants provided consent, trained bilingual research assistants conducted face-to-face interviews in participants’ preferred language/dialect.3 The study was approved by the Institutional Review Board of the Rush University Medical Center. Among 3,542 eligible adults who were approached, 3,159 participated in the study, reaching a response rate of 92%. According to the U.S. Census 2010 and a random block census project, the PINE study was representative of the Chinese aging population in the greater Chicago area.3 Out of these participants, about 94% were born in China; the majority live in ethnic enclaves where Chinese is used as the main language, and actually 99% chose Chinese in completing the interview. In the current study, we selected the respondents who answered the question about grandparent caregiving (N=2,365) to test the differences between caregivers and noncaregivers, and selected the caregivers (n=818) to assess the mechanisms to negative psychological well-being. Measures Dependent variables of psychological well-being included depressive symptoms, anxiety, stress, and loneliness. Independent variables included grandparent caregiver status in all respondents; and caregiving time, burden, pressure, and perceived negative effect in caregivers. Depressive symptoms Participants completed the Patient Health Questionnaire-9 (PHQ-9) that is appropriate for screening late-life depression with nine validated questions.19 The PHQ-9 contains the somatic domains that are common in Asian older adults with depressive symptoms.20 Respondents were asked how often they had been bothered during the past two weeks by feelings such as little interest in doing things and feeling tired. Responses were scaled from 0 (not at all) to 3 (nearly every day). A summary score were used, with higher scores indicating more symptoms (Cronbach’s α =.82). Anxiety We used the Hospital Anxiety and Depression Scale-Anxiety (HADS-A)21, which has been tested in Chinese populations and shown good inter-rater reliability.22,23 Respondents were asked whether they had experienced the symptoms such as feeling tense or wound up. Responses were scaled from 0 (not at all) to 3 (most of the time). A summary score was used; higher score indicated more anxiety (Cronbach’s α =.80). Stress We used the 10-item version of the Perceived Stress Scale (PSS), a valid and reliable instrument that was designed to measure the degree to which one’s life situations are appraised as stressful.24 The PSS asks about feelings and thoughts during the last month, particularly how unpredictable, uncontrollable, and overloaded respondents find their lives to be and current levels of experienced stress.24 Responses were scaled from 0 (never) to 4 (very often). A summary score was used, and higher scores indicate higher levels of stress (Cronbach’s α =.86). Loneliness We assessed loneliness with the 3-item R-UCLA Loneliness Scale, with good reliability and internal validity in the general population.25 Three questions asked about the feelings of lacking companionship, being left out of life, and being isolated from others. The response categories were coded 1 (hardly ever), 2 (sometime), or 3 (often). A summary score was used; higher scores indicated higher levels of loneliness (Cronbach’s α =.77). Caregiver status Caregiver status was defined if respondents reported any caregiving hours greater than zero, and non-caregiver status was designated if reporting zero hours. Caregiving time This continuous variable was measured by self-reported weekly hours that respondents spent caring for grandchildren. Caregiving pressure It was measured by the response to the question “How often do you feel pressured by your sons/daughters to take care of their children?” Responses ranged from 0 (never) to 4 (always). Caregiving burden It was indicated by the respondent’s feeling of burden to take care of grandchildren, with responses from 0 (never) to 4 (always). Perceived negative effect It was indicated by the thought that health was negatively affected by caring for grandchildren, with responses from 0 (never) to 4 (always). Socio-demographic characteristics and social support were controlled in the regression analyses. Socio-demographics included age (in years), gender (male or female), education (in years), personal income (1=less than $5,000 to 10=$45,000 or more), marital status (married or not), number of children alive, number of grandchildren, years living in the U.S., and self-rated health (1=poor to 4=very good). Social support was assessed by the frequency of receiving support from spouse, family members, and friends, including positive support and negative strain. Positive support was measured by the extent to which respondents opened up to family or friends, and the frequency of relying on them for help. Negative support was measured by how often respondents believed that too much was demanded and they had been criticized. Responses were given on a 3-point scale from 1 (hardly ever) to 3 (often). Positive support and negative support were calculated as the sum of the six items within each category. Higher scores indicated greater positive support (Cronbach’s α =.73), and more negative strain (Cronbach’s α =.63), respectively. Data Analysis We applied bivariate analyses, including independent t-tests, Wilcoxon signed-rank tests, and Chi-square tests to compare caregivers with noncaregivers in psychological well-being and socio-demographics. Then we estimated negative binominal regression models to test the relationships between caregiving experiences and psychological well-being, because four dependent variables were all discretely distributed with a large proportion reporting zeros and only a few cases reaching the critical points (e.g., major depression). In this case, a continuous version of negative binomial model is appropriate to improve the model fit to the data and account for over-dispersion.26 Model results were reported as Incidence Rate Ratios (IRR), which indicate the change in the incident rate of the outcome variable per unit change in the independent variable, while controlling for covariates. Five independent variables, i.e., caregiver status (caregivers being the reference group), caregiving time, pressure, burden, and perceived negative effect were entered and estimated in the models, respectively, after controlling for socio-demographic and social support covariates. Statistical analyses were conducted using SAS version 9.2 (SAS Institute Inc., Cary, NC). Results Among 2,365 respondents who answered the question about grandparent caregiving, 35% (n=818) were designated as caregivers, spending a weekly average of 12 hours on childcare. Over 80% reported no burden in caring for grandchildren, never felt pressured by adult children, or never perceived negative effect (Table 1). Compared to noncaregivers, caregivers had better psychological well-being, with significantly lower levels of depressive symptoms, anxiety, stress, and loneliness. Caregivers reported higher levels of positive and negative support, probably because they received not only more support but more demands from their family. In addition, caregivers were more likely to be younger, married, with less personal income but better self-rated health. They had fewer children and grandchildren but more members living in the household, with fewer years living in the U.S. relative to noncaregivers. No gender and education differences were observed. In four negative binominal regression models, five measures of grandparent caregiving experiences (i.e., caregiver status, caregiving time, burden, pressure, and perceived negative effect) were regressed on one of psychological well-being measures, respectively. Results showed that noncaregivers were 40% more likely to increase depressive symptoms, 20% more likely to feel anxious, 10% more likely to have stress, and 60% more likely to feel lonely than caregivers (Table 2). Generally, caregiving time was not associated with well-being, except a tiny but statistically significant relationship with stress. Caregiving pressure was associated with depressive symptoms, anxiety, and stress, but not with loneliness. In particular, one-unit increase in caregiving pressure was associated with about 40% increase rates in depressive symptoms, anxiety, and stress, respectively, after controlling for socio-demographics and social support. Caregiving burden was associated with about 10% increase rates in all outcomes. One-unit increase in perceived negative effect was related to 50% increase rate in depressive symptoms, 30% increase in stress, and 70% increase in loneliness, respectively. Discussion The study contributes to the limited research on grandparent caregiving among Chinese Americans. Cultural backgrounds and traditions shape expectations and values about grandparent caregiving, and therefore shape grandparent well-being.27 In addition, immigration has immense influence on family dynamics and interactions.28 Different from other ethnic groups, especially those in Western countries, Chinese grandparents provide childcare to maximize family benefits, rather than to solve the caregiving problems in the middle generation. This may explain why the majority of grandparents reported no burden in this study. For Chinese grandparent, caregiving is a family obligation that brings in family solidarity and perhaps future older-age support.5 Some immigrant families even send their children back to China for grandparents to rear. With a strong cultural expectation, caregiving in general is related to positive psychological well-being, probably because grandparents may feel fulfilled and satisfied from involvement in the lives of their adults children and grandchildren, helping adult children who struggle to balance work and childrearing, providing instrumental support, and transferring knowledge to younger generations.27 In addition, caregiving usually occurs in a multigenerational household, increasing the opportunities for family connections and favorable psychological outcomes.28 Caregiving, however, is not always rewarding and some are likely to experience stress.3 Our study shows that those with negative experiences are vulnerable to psychological distress. When caregiving is not a choice and older adults have to take care of their grandchildren under the pressure from adult children, they tend to have negative appraisals about caregiving. Caring for small children is definitely a demanding job, and many older immigrants take on full-time responsibility as the involved grandparents, a practice that is not common in Western culture. 2 It is indeed a burden and may have negative effects on grandparents’ health. Furthermore, some of them have to quit their professional work in China and contribute heavily to the interest of their adult children and grandchildren.2 The loss of primary community and sacrifice of self-interest may incur mental health issues, which further aggravate caregiving stress. Despite similar findings having been documented in the literature on the relationship between caregiving and well-being, this study implies the unique mechanisms of psychological distress in Chinese Americans. Particularly, caregiving pressure from adult children may further intensify problems with family relationships and intergenerational conflicts. Although the exchange of instrumental assistance in the form of grandparent caregiving is normally expected in many Chinese families, too many care demands and tasks are definitely detrimental to older grandparents’ well-being. Study findings imply the importance of balancing reciprocal assistance and the central role of family relationships in maintaining Chinese older adults’ well-being. This study also shows that positive social support plays an important role in coping with caregiving distress, while negative strains from families and friends may increase depressive symptoms and loneliness. In clinical practice, understanding the cultural norms of grandparent caregiving and the underlying mechanism to negative health outcomes is important for healthcare professionals to identify caregivers at risk of psychological distress and provide proper interventions to attenuate negative outcomes while maximizing positive experience. A family-centered perspective is often useful when working with Chinese older adults and culturally specific and relevant care are needed to improve adaptive coping skills and family relationships. The study is limited in generalizability. Although the study sample is representative of Chinese American older adults in a large metropolitan area, the findings are limited in generalizing nationally or internationally. In addition, the cross-sectional design definitely cannot establish the causal relationship between caregiving and psychological well-being. A reciprocal relationship could exist, that is, older adults with better mental and physical health are likely to care for their grandchild, which in turn, helps maintain or improve their evaluations of well-being. Other factors, such as physical and cognitive function, health behaviors, and social networks may explain the variance in psychological well-being among non-caregivers. Future research needs to evaluate the directionality of the longitudinal relationship between mental health and caregiving experiences after controlling for confounding factors. Another limitation is the lack of relevant information in defining caregiving. Caregiver status was defined based on weekly hours on childcare, but did not distinguish between coparenting caregivers and custodial caregivers, between involved caregivers and primary caregivers. Also missing is the length of caregiving, types of caring responsibilities, such as basic needs, personal care, medical care, and financial responsibility, and the number and characteristics of grandchildren being cared. In conclusion, grandchild care could be a burden, a blessing, or both, partly depending on older adults’ self-appraisal of their caregiving experience and partly on how they are treated in the family and in the community. Both informal support system and formal social and healthcare services are central to help older grandparents maintain or regain their psychological well-being in the face of caregiving challenges. Dr. Dong was supported by National Institute on Aging Grants R01AG042318, R01 MD006173, R01 CA163830, R34MH100443, R34MH100393, and RC4AG039085; a Paul B. Beeson Award in Aging; the Starr Foundation; the American Federation for Aging Research; the John A. Hartford Foundation; and the Atlantic Philanthropies. Table 1 Sample Descriptive and Comparisons between Grandparent Noncaregivers and Caregivers. Total sample1 Caregivers (n=818) Noncaregivers (n=1,547) P-value Psychological well-being  Depressive symptoms (range: 0–27), M ± SD 2.7 ± 4.1 1.9 ± 3.1 3.1 ± 4.5 <.001  Anxiety (range: 0–21), M ± SD 2.7 ± 3.3 2.2 ± 2.8 2.8 ± 3.5 <.001  Stress (range: 0–39), M ± SD 10.1 ± 6.6 8.9 ± 5.9 10.9 ± 6.7 <.001  Loneliness (range: 0–6), M ± SD 0.6 ± 1.2 0.4 ± 0.8 0.7 ± 1.3 <.001 Socio-demographics  Age (range: 59–105), M ± SD 72.8 ± 8.3 69.4 ± 6.3 75.0 ± 8.4 <.001  Female, n (%) 1,830 (58.0) 474 (58.0) 920 (59.5) .51  Education (range: 0–26), M ± SD 8.7 ± 5.1 8.6 ± 4.7 8.5 ± 5.2 .84  Income (range: 1–10), M ± SD 1.9 ± 1.1 1.8 ± 1.0 1.9 ± 1.1 <.001  Married, n (%) 2,236 (71.3) 665 (81.3) 1,035 (66.9) <.001  Number of children alive (range: 0–12), M ± SD 2.9 ± 1.5 2.8 ± 1.2 3.1 ± 1.6 <.001  Number of grandchildren (range: 0–15), M ± SD 4.5 ± 3.5 4.6 ± 2.7 5.2 ± 3.6 <.001  Household members (range: 0–10), M ± SD 1.9 ± 1.9 3.0 ± 2.2 1.5 ± 1.7 <.001  Years in the U.S. (range: 0–90), M ± SD 20.0 ± 13.2 16.5 (11.3) 21.2 ± 13.7 <.001  Self-rated health, n (%) <.01   Very good 87 (3.7) 27 (3.3) 60 (3.9)   Good 832 (35.2) 313 (38.3) 519 (33.6)   Fair 966 (40.9) 347 (42.4) 619 (40.0)   Poor 480 (20.3) 131 (16.0) 349 (22.6) Positive social support (range: 6–18), M ± SD 13.9 ± 3.0 12.8 ± 3.5 11.3 ± 3.8 <.001 Negative social support (range: 2–18), M ± SD 14.7 ± 3.6 15.3 ± 3.2 14.4 ± 3.7 <.001 Caregiver experience Caregiving time (range: 0–168), M ± SD _ 11.9 ± 24.9 _ _  Caregiving burden, n (%)   Never _ 664 (81.1) _ _   Little _ 58 (7.1) _ _   Sometimes/often/always _ 96 (11.8) _ _  Caregiving pressure, n(%)   Never _ 691 (84.5) _ _   Little _ 60 (7.3) _ _   Sometimes/often/always _ 67 (8.2) _ _  Perceived negative effect, n (%)   Never _ 657 (80.3) _ _   Little _ 76 (9.3) _ _   Sometimes/often/always _ 85 (10.4) _ _ 1 Sample sizes varied from 2,145 to 3,157 due to inapplicable question or missing values. Depression: 0–9 minor, 15–19 moderate, 20–17 severe Anxiety: 0–7 normal, 8–10 mild, 11–14 moderate, 15–21 severe Stress: 0–13 normal, 14–19 medium, 20–39 high Loneliness: 3–6 moderate/severe Table 2 Negative Binomial Regressions of Caregiving variables on Psychological Well-being. Depressive symptoms Anxiety Stress Loneliness Variables IRR (95% CI) P-value IRR (95% CI) P-value IRR (95% CI) P-value IRR (95% CI) P-value Noncaregiver 1.4 (1.2–1.6) <.001 1.2 (1.1–1.4) <.001 1.1 (1.1–1.2) <.001 1.6 (13–2.0) <.001 Caregiving time 1.0 (0.9–1.0) .16 1.0 (1.0–1.0) .48 1.0 (1.0–1.0) <.05 1.0 (1.0–1.0) .95 Caregiving pressure 1.4 (1.1–1.9) <.05 1.4 (1.1–1.9) <.01 1.4 (1.2–1.6) <.001 1.4 (0.8–2.2) .22 Caregiving burden 1.1 (1.0–1.1) <.05 1.1 (1.0–1.1) <.05 1.1 (1.0–1.1) <.001 1.1 (1.0–1.2) <.05 Perceived negative effect 1.5 (1.2–2.0) <.01 1.3 (1.0–1.6) .06 1.3 (1.2–1.5) <.001 1.7 (1.1–2.7) <.05 Note: Regression analyses controlled for age, gender, education, income, marital status, number of children alive, number of grandchildren, years living in the U.S., self-rated health, positive and negative supports. Conflict of Interest: The authors report no conflict of interest. Conflict of Interest Disclosures: Elements of Financial/Personal Conflicts All authors Author 2 Author 3 Author 4 Yes No Yes No Yes No Yes No Employment or Affiliation X Grants/Funds X Honoraria X Speaker Forum X Consultant X Stocks X Royalties X Expert Testimony X Board Member X Patents X Personal Relationship X Author Contributions: All authors contributed to this paper. Sponsor’s Role: None. 1 Burnette D Sun F Sun J 2013 A comparative review of grandparent care of children in the U.S. and China Ageing International 38 43 57 2 Xie X Xia Y 2011 Grandparenting in Chinese immigrant families Faculty Publications, Department of Child, Youth, and Family Studies Paper 82. http://digitalcommons.unl.edu/famconfacpub/82 3 Dong X Chang E Bergren S 2014a The burden of grandparenting among Chinese older adults in the Greater Chicago area – the PINE study AIMS Medical Science 1 125 140 4 Yee BWK Su J Kim SY Yancura L 2008 Asian American and Pacific Islander families Alvarez A Tewari N Asian American psychology: Current perspectives Mahwah, NJ Lawrence Erlbaum and Associates 5 Chen F Liu G Mair CA 2011 Intergenerational ties in context: Grandparents caring for grandchildren in China Social Forces 90 571 594 6 Zhou YR 2012 Space, time, and self: Rethinking aging in the context of immigration and transnationalism Clinical Key 26 232 242 7 Miyawaki CE 2015 A review of ethnicity, culture, and acculturation among Asian caregivers of older adults Sage Open 5 1 29 8 Lin X Bryant C Boldero J Dow B 2014 Older Chinese immigrants’ relationship with their children: A literature review The Gerontologist Advance Access published 10.1093/geront/gnu004 9 Kataoka-Yahiro MR 2010 Filipino American grandparent caregivers’ roles, acculturation, and perceived health status Journal of Cultural Diversity 17 24 33 20397571 10 Nagata DK Cheng WJY Tsai-Chae AH 2010 Chinese American grandmothering: A qualitative exploration Asian American Journal of Psychology 2 151 161 11 Lou VWQ 2011 Depressive symptoms of older adults in Hong Kong: The role of grandparent reward International Journal of Social Welfare 20 S135 S147 12 Ku LJE Stearns SC Van Houtven CH Lee SYD Dilworth-Anderson P Konrad TR 2013 Impact of caring for grandchildren on the health of grandparents in Taiwan Journals of Gerontology, Series B: Psychological Sciences and Social Sciences 68 6 1009 1021 13 Hayslip B Kaminski PL 2005 Grandparents raising their grandchildren: a review of the literature and suggestions for practice The Gerontologist 45 262 262 15799992 14 Chen F Liu G 2012 The health implications of grandparents caring for grandchildren in China The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences 67 1 99 112 15 Hayslip B Blumenthal H Garner A 2015 Social support and grandparent caregiver health: One-year longitudinal findings from grandparents raising their grandchildren Gerontol B Psych Sci Soc Sci, 2015 70 5 804 812 16 Burnette D 1999 Social relationships of Latino grandparent caregivers: A role theory perspective The Gerontologist 39 49 58 10028770 17 U.S. Census Bureau 2011 2010 census demographic profile Washington, DC Author 18 Conway F Jones S Speakes-Lewis A 2011 Emotional strain in caregiving among African American grandmothers raising their grandchildren Journal of Women and Aging 23 2 113 128 21534103 19 Arean P Ayalon L 2006 Assessment and treatment of depressed older adults in primary care Clinical Psychology: Science and Practice 12 321 335 20 Donnelly PK Kim KS 2008 The Patient Health Questionnaire (PHQ-9K) to screen for depressive disorders among immigrant Korean American elderly Journal of Cultural Diversity 15 24 29 19172976 21 Zigmond AS Snaith RP 1983 The hospital anxiety and depression scale Acta Psychiatr Scand 67 361 370 6880820 22 Lam CL Pan PC Chan AW Chan SY Munro C 1995 Can the Hospital Anxiety and Depression (HAD) Scale be used on Chinese elderly in general practice? Fam Pract 12 149 154 7589936 23 Leung CM Ho S Kan CS Hung CH Chen CN 1993 Evaluation of the Chinese version of the Hospital Anxiety and Depression Scale. A cross-cultural perspective Int J Psychosom 40 29 34 8070982 24 Cohen S Kamarch T Mermelstein R 1983 A global measure of perceived stress Journal of Health and Social Behavior 24 385 396 6668417 25 Hughes ME Waite LJ Hawkley LC Cacioppo JT 2004 A short scale for measuring loneliness in large surveys: Results from two population-based studies Research on Aging 26 655 672 18504506 26 Chandra NK Roy D 2012 A continuous version of the negative binomial distribution STATISTICA LXXII 1 81 92 27 Goodman C Silverstein M 2002 Grandmothers raising grandchildren: Family structure and well-being in culturally diverse families The Gerontologist 42 676 689 12351803 28 Tam VC-W Detzner DF 1998 Grandparents as a family resource in Chinese-American families: Perceptions of the middle generation McCubbin HI Thompson EA Thomposon AI Fromer JE Residency in Native American and Immigrant Families Thousand Oaks Sage 243 263

PMC005xxxxxx/PMC5118149.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8918110 2607 Eur J Neurosci Eur. J. Neurosci. The European journal of neuroscience 0953-816X 1460-9568 27657541 5118149 10.1111/ejn.13410 NIHMS818720 Article Optogenetic activation of MCH neurons increases NREM and REM sleep during the night in rats Blanco-Centurion Carlos 1 Liu Meng 1 Konadhode Roda P. 1 Zhang Xiaobing 2 Pelluru Dheeraj 1 van den Pol Anthony N. 2 Shiromani Priyattam J. 13 1 Department of Psychiatry & Behavioral Sciences, Medical University of South Carolina, SC, 29425 2 Department of Neurosurgery, Yale University, CT, 06510 3 Ralph H. Johnson Veterans Administration Medical Center, Charleston, SC, 29401 Corresponding author: Priyattam J. Shiromani, Ph.D., Ralph H. Johnson VA Medical Center, Medical University of South Carolina, Department of Psychiatry, 114 Doughty Street, MSC 404/STB 404, Charleston, SC 29425, shiroman@musc.edu, 843-789-6778 (office) 11 10 2016 16 10 2016 11 2016 01 11 2017 44 10 28462857 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Neurons containing melanin-concentrating hormone (MCH) are located in the hypothalamus. In mice optogenetic activation of the MCH neurons induces both NREM and REM sleep at night, the normal wake-active period for nocturnal rodents (Konadhode et al., 2013). Here we selectively activate these neurons in rats to test the validity of the sleep network hypothesis in another species. Channelrhodopsin-2 (ChR2) driven by the MCH promoter was selectively expressed by MCH neurons after injection of rAAV-MCHp-ChR2-EYFP into the hypothalamus of Long-Evans rats. An in vitro study confirmed that the optogenetic activation of MCH neurons faithfully triggered action potentials. In the second study, in Long-Evans rats, rAAV-MCH-ChR2, or the control vector, rAAV-MCH-EYFP, were delivered into the hypothalamus. Three weeks later baseline sleep was recorded for 48h without optogenetic stimulation (0 Hz). Subsequently, at the start of the lights-off cycle, the MCH neurons were stimulated at 5, 10, or 30Hz (1 mW at tip; 1 min on – 4 min off) for 24 h. Sleep was recorded during the 24h stimulation period. Optogenetic activation of MCH neurons increased both REM and NREM sleep at night, whereas during the day cycle only REM sleep was increased. Delta power, an indicator of sleep intensity, was also increased. In control rats without ChR2, optogenetic stimulation did not increase sleep or delta power. These results lend further support to the view that sleep-active MCH neurons contribute to drive sleep in mammals. Graphical abstract Neurons containing melanin concentrating hormone (MCH) or orexin are localized in the posterior hypothalamus. Activation of orexin neurons produces arousal, but the function of the MCH neurons is less clear. In mice, we determined that optogenetic activation of MCH neurons increases sleep at night. We now find that in wildtype rats activating these neurons also increases sleep and delta power. Such consistent results in two mammals underscore the role of the MCH neurons as drivers of sleep. Channelrhodopsin-2 REM sleep rat melanin concentrating hormone recombinant adenoassociated virus Introduction In the last hundred years successive groups of researchers have used new tools to identify neurons responsible for wake, non-rapid-eye movement (NREM) sleep and REM sleep [reviewed by (Konadhode et al., 2014)]. Initially, transections and electrolytic lesions helped locate the brain regions responsible for the different sleep-wake states (Jouvet, 1962). Subsequently, electrophysiology studies in freely-behaving animals identified neurons that were active only during a specific sleep-wake state (McCarley & Hobson, 1971; Chu & Bloom, 1973; McGinty & Harper, 1976; Siegel & McGinty, 1977; Aston-Jones & Bloom, 1981; Szymusiak & McGinty, 1986; Mileykovskiy et al., 2005; Hassani et al., 2009) indicating that activity of specific neurons regulated the sleep-wake states. The introduction of c-FOS as a functional neuroanatomical tool led to identifying the phenotype of neurons that were activated during sleep-wake states and also determined the connectivity between these neurons (Sherin et al., 1996; Basheer et al., 1997). The collective data from those studies has resulted in a neural network map that illustrates the various phenotypes of neurons regulating waking, NREM and REM sleep (Saper et al., 2010; Shiromani, 2011; Pelluru et al., 2013). Currently, optogenetics and pharmacogenetics are being used to directly test specific cell elements of this circuit. For instance, optogenetic activation of the arousal orexin neurons increased transitions to waking (Adamantidis et al., 2007) whereas inhibition resulted in sleep (Tsunematsu et al., 2011). Similar results were observed using the pharmacogenetic approach (Sasaki et al., 2011). Optogenetic activation of the noradrenergic locus coeruleus (LC) neurons also increased waking while photoinhibition decreased the length of the wake bouts (Carter et al., 2010). Pharmacogenetic activation of LC neurons ameliorated the excessive sleepiness in a mouse model of narcolepsy (Hasegawa et al., 2014). Optogenetics has also been used to drive neurons implicated in generating sleep. One phenotype of sleep-active neurons contains melanin concentrating hormone (MCH) (Hassani et al., 2009). MCH neurons are intermingled with the orexin neurons and project to many of the same targets as orexin neurons (Bittencourt et al., 1992; Steininger et al., 2004; Yoon & Lee, 2013). In wildtype mice we found that optogenetic activation of the MCH neurons increased sleep at a time when the mice would be normally awake. Subsequently, another group used MCH-cre mice and determined that acute stimulation during NREM sleep increased entries into REM sleep (Jego et al., 2013). Another group (Tsunematsu et al., 2014) genetically ablated the MCH neurons and found decreased amounts of NREM sleep without changing REM sleep. These studies in mice demonstrate that selective activation of MCH neurons increases sleep, and loss of these neurons decreases NREM sleep. Optogenetic studies need to be performed in other mammalian species that have both NREM and REM sleep, thereby validating the status of specific neurons within neural network models of sleep-wake regulation. To facilitate such studies, we linked the gene encoding channelrhodopsin-2 (ChR2) to the MCH promoter. In the present study, this virus was inserted into wildtype rats to test the hypothesis that activation of MCH neurons induces sleep not only in mice but also in another mammal. Materials and Methods Ethical statement All manipulations done to the animals adhered to the NIH Guide for the Care and Use of Laboratory Animals and were pre-approved by the Medical University of South Carolina (protocol 3355), Ralph H. Johnson VA (protocol 525, 584), and Yale University (protocol 2014-10117) Institutional Animal Care and Use Committee. Viral vectors The parent plasmid containing channelrhodopsin-2 gene (ChR2; H134R) was kindly donated by Dr. Karl Deisseroth (Stanford University/HHMI). The genes for ChR2 and enhanced yellow fluorescent protein (EYFP) were originally driven into neurons by the human synapsin promoter (hSyn-485 bp). We modified it by replacing the hSyn-485 with the MCH promoter sequence (462 bp). The MCH specific plasmid was then packaged by the Vector Core at the University of North Carolina (Chapel Hill, NC). The plasmid was packaged into recombinant adeno-associated viral vectors (rAAV; serotype 5) to a titer of 1×1010 particles per μL. The control vector containing no ChR2 gene (rAAV-MCH-EYFP) was constructed as previously described (Liu et al., 2011). After packaging, the MCH specific viral vectors were always stored at −85°C and thawed just before delivery into the brain. In vitro studies Optogenetic stimulation of the MCH neurons rAAV-MCH-ChR2-EYFP was injected into the lateral hypothalamus of five adult rats. Four to five weeks later, rats with selective ChR2 expression in lateral hypothalamic MCH neurons were deeply anesthetized with isoflurane and decapitated. Brains were immediately removed and immersed in ice-cold high-sucrose solution containing (in mM): 220 sucrose, 2.5 KCl, 6 MgCl2, 1 CaCl2, 1.23 NaH2PO4, 26 NaHCO3, and 10 glucose (gassed with 95% O2/5% CO2; 300–305 mOsm). Coronal brain slices (300 μm thick; vibratome) were transferred to an incubation chamber filled with artificial CSF (aCSF) solution containing (in mM) 124 NaCl, 2.5 KCl, 2 MgCl2, 2 CaCl2, 1.23 NaH2PO4, 26 NaHCO3, and 10 glucose (gassed with 95% O2/5% CO2; 300–305 mOsm) at room temperature (22 °C). Following a recovery period of 1–2 h, slices were transferred to a recording chamber mounted on a BX51WI upright microscope (Olympus, Tokyo, Japan). The slices were perfused with a continuous flow of gassed ACSF. The temperature of the recording solution was maintained at 33 ± 1 °C by a dual-channel heat controller (Warner Instruments, Hamden, CT). For comparison, we also recorded mouse MCH neurons from transgenic mice with GFP expressed in the MCH neurons, as described elsewhere (van den Pol et al., 2004; Zhang & van den Pol, 2012). MCH neurons with ChR2-EYFP expression were visualized using an infrared-differential interference contrast optical system combined with a monochrome CCD camera and monitor. Pipettes were pulled from thin-walled borosilicate glass capillary tubes (length 75 mm, O.D=1.5 mm, I.D.= 1.1 mm, World Precision Instruments, Sarasota, FL) using a P-97 micropipette puller (Sutter Instruments, Novato, CA). A pipette solution containing (in mM) 145 K-gluconate, 1 MgCl2, 10 HEPES, 1.1 EGTA, 2 Mg-ATP, 0.5 Na2-GTP, and 5 Na2-phosphocreatine (pH 7.3 with KOH; 290–295 mOsm) was used for the recording. Recording pipettes showed resistances ranging from 3 to 6 MΩ. An EPC-10 patch-clamp amplifier (HEKA Instruments, Bellmore, NY) and PatchMaster 2.20 software (HEKA Elektronik, Lambrecht/Pfalz, Germany) were used to acquire and analyze data. Neurons with series resistance greater than 20 MΩ and change of greater than 15% were excluded from the statistical analysis. Traces were processed using Igor Pro 6.36 (Wavemetrics, Lake Oswego, OR). The membrane potential for voltage-clamp recording was held at −60 mV. A blue light laser (470 nm, Laserglow Technologies, West Toronto, CA) set at 5–10 mWatts/mm2 was used to activate ChR2 in the MCH neurons. The blue light was delivered by a fiber optic (tip I.D.=100 μm, NA 0.22, Doric lenses, CA) to the recorded neuron. To test the photostimulation-evoked response, stimuli of 10 msec duration with fixed frequencies of 1, 5, 10, 20 and 30 Hz were used. Evoked response of transfected MCH neurons was recorded in voltage and current clamp conditions. In vivo studies Animals and groups Fourteen Long Evans adult rats (545.3 ± 41.3g; male) were used. Ten rats received rAAV-MCH-ChR2-EYFP while the other four rats received the control vector lacking ChR2 (rAAV-MCH-EYFP). All rats were housed singly in an isolated room, with controlled temperature (22 ± 2°C) and lighting (photoperiod =12:12 h; lights-off at 6 PM). Food and water were available ad-libitum. Gene transfection Under deep isofluorane anesthesia (1.5–2%) and aseptic conditions viral vectors were microinjected bilaterally into two adjacent rat brain sites targeting the entire population of MCH neurons (lateral zona incerta and perifornical area: AP= −2.8 mm, lateral= ±1.6 mm, vertical= −7.8 mm; medial zona incerta and dorsomedial hypothalamic area: AP= −3.1 mm, lateral= ± 0.8 mm, vertical= −7.8 mm). The coordinates are relative to bregma, lateral to the sagittal suture and vertical to the dura mater. For the lateral site the volume of injection was 2 μl and for the medial site it was 1 μl. Therefore, in both hemispheres a total of 6 μl was microinjected (virus load of 6×1010 particles per rat). The virus was delivered slowly over 20 min using a 10 μL Hamilton syringe. Following the microinjection, the injection needle (30 Ga) was left in place for an additional 5 min and then withdrawn slowly. Fiber optic and electrodes implant Two weeks after delivery of the viral vectors, the fiber optic probes and sleep recording electrodes were surgically implanted (under 2% isoflurane anesthesia). Two fiber optic probes (0.4 mm diameter), one in each hemisphere, were implanted at antero-posterior −2.9 mm, lateral −1.4 mm and vertical −7.0 mm. To record the electrocorticogram (ECoG) four miniature screws (Plastic One Inc.) were inserted to sit atop the frontal and occipital cerebral cortices. To record the skeletal muscle activity (EMG), two flexible wire electrodes (Plastic One Inc.) were bilaterally embedded into the nuchal muscles. The wires were inserted into a plastic socket, and along with the optic probes, affixed to the skull with dental cement. Optogenetic stimulation of MCH neurons Five weeks after the rAAV vector delivery rats were tethered to a light-weight cable connected to rotary swivels. The swivels allowed the rats to engage in complete freedom of behavior. Rats were adapted to tethering for four days and baseline sleep was recorded for 48 h. During this first sleep recording session there was no optogenetic stimulation and this initial session represents 0 Hz. Following the baseline recording, the rats were given optogenetic stimulation (blue LED, 473 nm wavelength; 1 mW intensity at tip; Doric Lenses, Québec, Canada). A programmable stimulator (Master-9, AMPI, Jerusalem, Israel) generated TTL pulses of 10 msec of duration for driving the photo stimuli. TTLs pulses were also recorded along with the ECoG and EMG activity. MCH neurons were activated by light at three frequencies: 5, 10, or 30 Hz (ON for 1 min and OFF for 4 min). The rats were stimulated for 24 h starting at lights-off (Zeitgeber hour 12). There was a 72 h interval between stimulation days and stimulation protocols were followed in a counterbalanced fashion. In 24 h MCH neurons were given 8.64 × 104 pulses at 5 Hz, 1.73 × 105 at 10 Hz and 5.18 × 105 at 30 Hz. MCH neurons were activated 1.0 % of total time in 24 h at 5 Hz, 2.0% at 10 Hz and 6.0% at 30 Hz. Sleep recording, scoring and spectral analysis The ECoG activity was recorded from the contralateral frontal-occipital electrodes. Using an analog polygraph (Grass Model 12), the ECoG activity was amplified 5000 times and band pass filtered between 0.3–100 Hz. To record the EMG activity, the muscle electrical activity was amplified between 5000–20,000 times and band pass filtered between 100–1 KHz. ECoG and EMG analog signals were then digitized and stored (at 128 Hz sampling rate) onto a computer hard drive by a data acquisition software (Vital Recorder, Kissei Comtec Co., Nagano, Japan). The acquisition software also stored and synchronized video streams of the rat’s behavior. Video was recorded with infrared CCD cameras that allowed recording of the animal’s behavior in the dark. The sleep records were scored in 12 sec epochs based on the Fast Fourier Transformation (FFT) algorithm of the ECoG activity for the delta bandwidth (0.5–4 Hz) along with the EMG integrated activity (SleepSign®, Kissei Comtec Co. Nagano, Japan). Each epoch was then manually scored as wake, non-REM sleep (NREM) or REM sleep (REMS). Video recordings assisted the identification of the sleep-wake states. SleepSign was also used to produce a FFT analysis of the ECoG during NREMS and REMS. Both delta (0.5–4 Hz) and theta (4–8 Hz) power were automatically determined by the software. ECoG FFT data was calculated in 3 h bins. ECoG FFT values for the 0 Hz day were normalized as percent of the 24 h average. Then power values during photostimulation days were expressed as percent of the 0 Hz power. Post mortem analysis of transfected rat brains At the end of the in-vivo experiments, rats were euthanized (5% isofluorane) and the brains perfused with 50 ml 0.9% saline and 150 ml 10% formalin-buffered solution. The brains were removed, left overnight in 10 % formalin PBS at 4 °C, and then transferred to a sucrose 30% PBS solution until equilibration. The brains were cut on a cryostat and 40 μm thick coronal sections obtained. The first and third sections, in a 1 in 5 series, were processed for immunolabeling of the MCH peptide. The brain sections were first washed to remove the cryoprotectant (PBS 0.01 M), then placed in normal donkey serum (2% in PBS+0.3% Triton-x). Sections were then incubated overnight at room temperature (RT) with the primary antibody (rabbit anti-MCH; 1:1000 dilution; Phoenix Pharmaceuticals, Cat # H070-47). Next day, after washing in PBS, the sections were incubated at RT for 1 h with donkey anti-rabbit IgG secondary antibody (Alexa Fluor® 568 conjugate; 1:500 dilutions, Life Technologies; Cat # A10042). Sections were then mounted onto gelatin coated glass slides, cover slipped, and stored at −20 °C. Hypothalamic tissue from one well in the animal depicted in figure 3 was incubated in goat anti-orexin antibody (1:500 dilution, Phoenix Pharmaceuticals, Cat #H003-30) to visualize orexin-immunoreactive neurons in conjunction with EYFP. To further test for specificity of the transgene construct a separate rat (off-site control) was given the rAAV-MCH-ChR2-EYFP into the caudate putamen (one side; 750 nl) where no MCH neurons are located. In this off-site control rat, a single injection (one side; 750nl) was also made in the zona incerta to confirm the efficacy of the vector. The rat was sacrificed 18 days after injection, and brain tissue was processed for visualization of the EYFP+ neurons. A confocal microscope (Nikon A1 + confocal microscope; Nikon Corporation, Tokyo, Japan) was used to visualize the MCH and EYFP positive neurons. The distribution of EYFP+ somata was plotted along with the location of the fiber optic tips. Each coronal section was scanned at low magnification (4×), dual channel (Channel 1:EYFP Laser line at 488 nm; Channel 2=Texas Red Laser line at 560 nm) and the acquired images exported as TIFF image files. The TIFF images were overlaid (Adobe Systems Inc.) and resized onto matching rat brain atlas digital plates. Areas showing strong EYFP positive signal in somata and proximal processes were traced and colored. Fiber optic tips indicated on the images by scar tissue were also traced. Brain sections from rats with extensive transfection and correct fiber optic placements were used to obtain a tally of the MCH neurons expressing EFYP. These sections were examined at high magnification (60×) by a person blinded to the type of treatment given to the rats. Single (MCH+ or EYFP+) and double (MCH+EYFP) labeled neurons were counted in five coronal brain sections separated at least 120 μm apart. Ten bilateral sampling areas per rat (147 × 196 μm) showing high density of EYFP positive somata were tallied. Cells were counted in images while scanning across the vertical plane (Z-axis in 1 μm increments). To assess colocalization between EYFP and MCH, Z-stacks were visualized both in orthogonal view in 2D and as XYZ-planes in 3D rendering. Only cells showing MCH labeling inside the cytoplasm were counted as specifically transfected. The tissue processed for visualization of orexin-immunoreactive neurons was similarly examined. Statistical analysis Sigma Stat software (Systat Software Inc., San Jose, CA) was used to statistically compare group means. Two way and one-way repeated measures ANOVA (General Linear Model) compared group means, followed by post-hoc analysis (Holm-Sidak). Independent and paired t-tests were also used as a-priori tests of hypotheses. Probability values equal or less than 0.05 were set for rejection of the null hypothesis. Results In vitro optogenetic activation of rat MCH neurons An in-vitro electrophysiological analysis with whole cell recording showed that rat MCH neurons that were ChR2-EYFP positive were also responsive to light. Similar to the majority of MCH neurons found within the mouse hypothalamus (van den Pol et al., 2004; Zhang & van den Pol, 2012), rat MCH neurons were generally silent with few if any spontaneous action potentials (Fig. 1A). The resting membrane potential of rat MCH neurons was almost identical to that of the mouse MCH neurons (rat=60.6±1.5 mV vs. mouse=59.4 ±1.1 mV n.s.; Fig. 1B). Blue light pulses (1–10 Hz; 10ms duration) evoked substantive inward currents with similar amplitudes and high fidelity responses. However, light pulses at frequencies faster than 20 Hz evoked relatively lower current amplitudes compared to the first pulse of the stimulation (Fig. 1C). In current-clamp configuration, light stimuli between 1–10 Hz generated action potentials with high fidelity. However, light pulses with a frequency faster than 20 Hz only evoked a single action potential following the initial pulse, and only small amplitude depolarizations evoked by the succeeding pulses (Fig. 1D). As we have demonstrated previously in this region of the hypothalamus, this spike frequency adaptation is a unique feature of MCH neurons, not shared by nearby hypocretin cells (van den Pol et al., 2004). In vivo studies Anatomical distribution of ChR2-EYFP+ somata In rats given rAAV-MCH-ChR2-EYFP the presence of the reporter gene, EYFP, served as proxy for the transfection of the ChR2 gene into MCH neurons. Of the ten rats given rAAV-MCH-ChR2-EYFP, six rats had extensive expression of EYFP. In the other four rats, few if any EYFP neurons were seen, perhaps because the cannula delivering the virus was clogged; these rats were therefore not included for further study. The gross anatomical distribution of the EYFP expression from the six rats with good ChR2 expression is summarized in figure 2. There were numerous EFYP+ somata within the zona incerta, the lateral hypothalamus and the perifornical area (Figure 3). The four WT rats given control vector (rAAV-MCH-EYFP) also had numerous EYFP+ somata. Tally of the MCH+EYFP+ neurons determined that 52.5 (± 2.0)% of MCH neurons were also EYFP+. The MCH promoter expressed the genes in MCH neurons, as shown by the strong and selective expression in which 97.3(± 0.97)% of EYFP+ were also MCH+ (Figure 3 and 4; left panels). In the tissue sections processed for visualization of orexin-ir neurons, none of the orexin+ neurons were also EYFP+ (Figure 4; right panel), which is consistent with our previously published report in mice (Konadhode et al., 2013). The supplementary figure shows the lack of EYFP+ neurons in the off-site (caudate-putamen) control rat. By contrast in the same rat, injection of the rAAV-MCH-ChR2-EYFP into the zona incerta robustly transduced EYFP into many neurons. This further supports the view that the viral gene vector employing the MCH gene promoter shows selective expression in MCH cells. Optogenetic activation of rat MCH neurons and sleep states Figure 5 summarizes the sleep data during the 12h night and day periods. During the night there was a significant difference between the experimental (ChR2) and control rats (no ChR2) in percent waking (F=8.7; df=1,29; p=0.018), NREM (F=6.47; df=1,29; p=0.035) and REM sleep (F=11.03; df=1,29; p=0.011). Within group post-hoc comparisons determined no effect of optogenetic stimulation in the control rats. Moreover, there was no difference between the groups at 0 Hz indicating comparable levels of wake, NREM and REM sleep at night (Holm-Sidak post-hoc test). However, in the rats given ChR2 10 Hz optogenetic stimulation during the 12h night cycle significantly decreased waking (p=0.001), and increased both NREM (p=0.001) and REM sleep (p=0.001) compared to 0 and 5Hz (Fig. 5). 5 Hz only increased REMS whereas 30 Hz did not have an effect (Fig. 5). These results indicate that in rats activation of MCH neurons at night induces both types of sleep, and that 10Hz is more effective compared to 5Hz. During the day cycle, there was no significant effect of stimulation on wake or NREM sleep. However, in rats with ChR2, activation of the sleep-active MCH neurons significantly increased REM sleep in all three frequencies compared to 0Hz (p=0.05) (Figure 5). In rats given the control virus (no ChR2) there were no changes in sleep in response to stimulation in waking, NREM or REM sleep (Figure 5). Thus, in rats stimulating MCH neurons during the day increased REM sleep, a finding consistent with data in MCH-Cre mice (Jego et al., 2013; Tsunematsu et al., 2014). To better understand the time course of the changes in sleep produced by optogenetic stimulation of MCH neurons, the data for 5 and 10Hz stimulation rates were arranged in 3h blocks (Figure 6). A one-way RMANOVA compared means separately for the night and day cycles. At night, 10Hz stimulation immediately increased total sleep time by increasing both NREM (p<0.05) and REM sleep (p<0.05). 5Hz stimulation did not increase NREM but increased REM sleep during the second half of the night cycle (p<0.05). Thus, 10Hz stimulation made the rats sleep more when they should normally be awake. During the day cycle 5Hz stimulation significantly increased REM sleep during all time points compared to 0Hz (p=0.01), while 10Hz increased it during the first 6h of the lights-on cycle compared to 0Hz (p=0.01). Next we examined the effect of MCH neuronal activation on sleep architecture (Figure 7). Only the effects of 10Hz stimulation in ChR2 rats was analyzed as this was most effective during both the night and day cycles. Data from the control rats was not analyzed as there were no changes in sleep in response to stimulation. Activation of rat MCH neurons significantly decreased the number of long wake bouts (>32 min) during both the night and day cycles and increased the number of short NREM and REMS bouts compared to 0Hz (p=0.05). Thus, 10Hz stimulation decreased the length of waking bouts and increased entry into both NREM and REM sleep. We also examined the effects of optogenetic activation of MCH neurons on the spectral ECoG power during sleep states. Figure 8 summarizes the changes observed in NREM sleep delta (0.5–4 Hz; top graph) and REM sleep theta power (4–8 Hz; bottom graph) in the rats given ChR2. With respect to delta power, only 10Hz stimulation data were analyzed since only this frequency significantly increased NREM sleep. At 0 Hz NREMS delta power waxed and waned across the 24 h cycle, a profile that is consistent with data in rats (Shiromani et al., 2000). 10Hz stimulation of the MCH neurons increased delta power, and during the daytime, NREMS delta power continued to progressively increase rather than wane. Activation of rat MCH neurons also increased theta power during REM sleep. Stimulation at 5 Hz resulted in higher theta power during the first 12 h of stimulation. At 10 Hz, REM sleep theta power nearly doubled for 18 h. The increase in theta power is also consistent with the increase in REM sleep time. Thus, activation of MCH neurons in rats not only increased both NREM and REMS, but also increased delta and theta power. Discussion In mammals a distributed network of neurons is implicated in generating wake, NREM and REM sleep. Because these neurons are intermingled with neurons serving other behaviors, optogenetics is used to selectively activate phenotype specific neurons. Previously (Konadhode et al., 2013), we determined that activation of the sleep-active neurons containing melanin concentrating hormone (MCH) in mice robustly induced sleep. We now demonstrate that in wildtype Long-Evans rats, similar to what is found in mice, optogenetic activation of MCH neurons induces both NREM and REM sleep at night when nocturnal rodents are primarily awake. Our two studies utilized the same viral vector, rAAV-MCH-ChR2(H134R)-EYFP, to insert the light-sensitive ChR2 gene into MCH neurons. In-vitro experiments determined that the MCH neurons in mice and rats were similarly activated by light. Both studies utilized the same experimental paradigm (one minute stimulation followed by 4 minutes of no stimulation, every 5 minutes) to stimulate the MCH neurons over 24h. In both studies, to better gauge the influence of the MCH neurons in inducing sleep, the optogenetic stimulation began at the start of the light-off period, which is when nocturnal rodents awaken in response to circadian signals. In our studies, activation of MCH neurons at night induced sleep by decreasing the length of wake bouts. Delta power, a measure of sleep intensity, was also increased in both rats and mice. Such consistent effects in two species that have both NREM and REM sleep validates the status of MCH neurons inducing sleep and suggests that MCH neurons may serve a common function in driving sleep in mammals. In the current study in rats, as in our previous study in mice (Konadhode et al., 2013), ChR2-EYFP was selectively colocalized in MCH neurons in the zona incerta, perifornical area and the lateral hypothalamus. In the present study, 97% of EYFP (proxy for ChR2) positive neurons were also MCH immunoreactive, indicating the selective nature of the gene expression. Specificity of the vector for MCH neurons was also confirmed by lack of colocalization between EYFP+ and orexin+, which is consistent with our previous report in mice (Konadhode et al., 2013). We used orexin neurons because these neurons are intermingled with MCH neurons. Further confirmation was established by an experiment showing the absence of transfection among neurons of the caudate putamen which lacks MCH neurons, and by the similarity of the unique electrophysiological characteristics of rat and mouse MCH neurons determined with whole cell recording. Similar rates of transfection of MCH neurons were observed in both species (52.5% in rats versus 53.4% in mice). Thus, in both our studies about half of the MCH neurons contained the light-sensitive ChR2. However, even if all of the MCH neurons contained ChR2, it is possible that light will not reach all of the light-sensitive MCH neurons, a potential limitation of the optogenetic approach. Nevertheless, there was a robust increase in both NREM and REM sleep at night with optogenetic activation of half of the MCH neurons. This indicates that only a subset of MCH neurons needs to be activated to increase sleep, underscoring the impact of MCH neurons in driving sleep. It is not known whether the subset of MCH neurons containing the light-sensitive opsin and activated by light recruits and activates the other MCH neurons. In the present study in rats, the sleep response was stronger than in our previous study in mice (Konadhode et al., 2013). For instance, in the present study both 5 and 10Hz were effective, whereas in our previous study only 10Hz was effective. This might be because the diameter of the fiber-optic probes in this study was twice as large as those used in our previous study (400μm versus 200μm). We selected the larger probes to reach the MCH neurons that are diffusely scattered along the hypothalamus. Thus, in the present study more MCH neurons were potentially activated by the broader beam of light, producing a stronger effect on sleep. Our two studies have also found that delta power, a marker of sleep intensity, is increased in response to activation of MCH neurons. An increase in delta power was also observed by another group who also activated the MCH neurons (Tsunematsu et al., 2014). In the present study, delta power and NREM sleep time increased upon stimulation at night. During the day 10Hz stimulation increased delta power but not NREM sleep time. We suggest that during the day NREM sleep time is at ceiling levels and cannot be increased further. However, delta power wanes across the day cycle and activating the MCH neurons prevented the decline. Two other groups have also examined the effect of optogenetic stimulation of MCH neurons on sleep (Jego et al., 2013; Tsunematsu et al., 2014). They used mice that were transgenic pMCH-cre (Jego et al., 2013) or the ChR2 was activated in MCH neurons via a tetracycline-controlled system (Tsunematsu et al., 2014). We used wildtype mice (Konadhode et al., 2013) or rats (the present study). They found an increase in REM sleep, while we determined that both NREM and REM sleep were increased in response to optogenetic activation of MCH neurons. One possibility for this difference is related to light-induced activation of specific clusters of MCH neurons. The MCH neurons are located in the zona incerta, lateral hypothalamus and perifornical area, with a smaller cluster located ventrally along the dorsal border of the ventral medial hypothalamus (Konadhode et al., 2014). The cluster of MCH neurons in the lateral and perifornical area innervate the pons (Hanriot et al., 2007; Torterolo et al., 2009; 2013) where REM sleep is generated. In the two studies where REM sleep was increased (Jego et al., 2013; Tsunematsu et al., 2014), the MCH neurons in the lateral and perifornical cluster may have shown a higher probability of activation, whereas in our studies we specifically targeted the probes to activate the MCH neurons in the zona incerta. In the present study, it is possible that because of the bigger probe light reached the lateral and perifornical clusters, which then induced REM sleep during the day. Thus, activating specific subsets of MCH neurons may exert a differential effect on NREM versus REM sleep, a perspective meriting further study. Although activation of MCH neurons promotes sleep, optogenetic inhibition has no effect (Jego et al., 2013; Tsunematsu et al., 2014). Interestingly, selective ablation of approximately 97% of MCH neurons via a cell-specific expression of diphtheria toxin decreases NREM sleep without affecting REM sleep (Tsunematsu et al., 2014). With 30% ablation of MCH neurons there is a decrease in NREM sleep intensity under basal conditions and in response to 4h total sleep deprivation (Varin et al., 2016). Together, the optogenetic and MCH neuron ablation studies underline the potential importance of MCH neurons in both NREM and REM sleep. Pharmacological studies also indicate that the MCH peptide, promotes both types of sleep. MCHR1 antagonists significantly decrease both NREM sleep and REM sleep (Ahnaou et al., 2008). Activation of the MCHR1 is the likely mechanism for the NREM sleep enhancing effect because MCHR1 null mice show significantly less NREM sleep(Ahnaou et al., 2011). Similar reduction of NREM sleep has been measured in the MCH null mice (Willie et al., 2008). In contrast, when MCH was infused ICV to rats, both types of sleep significantly increased (Verret et al., 2003). Local infusion of MCH into the locus coeruleus only increases REM sleep (Monti et al., 2014) suggesting that MCH neurons regulate NREM versus REM sleep depending on the projection site. In humans, MCH release was highest after sleep onset which corresponds to the light stages of NREM sleep (Blouin et al., 2013). In rats, we also measured higher MCH levels in the cerebrospinal fluid during the sleep phase (Pelluru et al., 2013). The MCH neurons are not active in waking, but increase their activity in NREM sleep (1–3Hz; average=0.5Hz), with peak activity in REM sleep (1–22Hz; average=1.1Hz) (Hassani et al., 2009). Thus, activating MCH neurons should promote sleep, which it does in both mice and rats. We and others have stimulated the MCH neurons at 5, 10, 20 and 30 Hz, rates that are higher compared to what may be found in natural NREM and REM sleep. However, as seen in figure 1, MCH neurons can be driven at 1–10Hz, but at 20 and 30Hz there is a decrement in amplitude and triggering of the action potential. We found that 10Hz induced both NREM and REM sleep, whereas 5, 10 and 30Hz induced only REM sleep. These stimulation frequencies are higher than normally recorded in MCH cells, but were necessary to offset the fact that only half the MCH are being stimulated, based on our histochemical corroboration. It seems unlikely that stimulating at 1Hz, the rate that is consistent with NREM and REM sleep, will induce more sleep, especially during the wake-active period because the faster stimulation rates may force the MCH neurons to release more of their contents quickly, thereby hastening the transition to sleep. The released transmitters are also likely to accumulate producing a cumulative effect in driving sleep. Indeed, in the present study, delta power, a very sensitive marker of accumulating endogenous somnogens, increased in response to the chronic stimulation. Optogenetics has been transformative in identifying the impact of specific neurons in behavior. Should the rate of optogenetic stimulation mimic the natural pattern of activity of the neurons? Although optogenetic mimicking of normal neuronal behavior may be ideal, for some behaviors, such as sleep, it may not be necessary to mirror the fine pattern of activity to induce the desired behavior. The cadence of the MCH neurons may be off in wake and on in sleep (loosely defined as binary). It may be sufficient to induce sleep to maintain this on-off cadence. In other words, turn-on the MCH neurons periodically to get a waking brain to fall asleep. Stimulation at 5Hz induced sleep, albeit weakly compared to 10Hz. It is possible that in rats stimulation at rates that may be closer to what occurs naturally may also induce sleep. How might MCH neurons induce sleep? During waking the activity of the MCH neurons is inhibited by the adjacent arousal orexin neurons (Apergis-Schoute et al., 2015). The MCH neurons are able to inhibit the combined activity of the arousal neurons and generate NREMS and REMS because MCH peptide inhibits the orexin neurons (Rao et al., 2008). However, GABA may also be colocalized in MCH neurons and similar to MCH, it may inhibit the arousal neurons (Elias et al., 2001; Del Cid-Pellitero & Jones, 2012); some MCH cells may also contain other fast transmitters (Chee et al, 2015). Another mechanism by which MCH stimulation may increase specifically NREM sleep is through indirect disinhibition of the reticular thalamic nucleus (RT). Recently it was found that optogenetic inhibition of hypothalamic GABA neurons projecting to the RT increase NREM sleep and EEG delta power ( Gutierrez-Herrera et al., 2016). If MCH neurons contact these hypothalamic GABA cells, the optogenetic activation of MCH could indirectly disinhibit the RT resulting in higher NREM sleep and EEG delta power. MCH neurons project throughout the brain, innervating neuronal populations implicated in waking and REMS (Cvetkovic et al., 2003; Cvetkovic et al., 2004; Hanriot et al., 2007; Sita et al., 2007; Croizier et al., 2010; Hong et al., 2011; Lima et al., 2013; Torterolo et al., 2013; Yoon & Lee, 2013). It is not known whether a single MCH neuron projects to multiple ascending and descending targets, but the MCH neurons located in the lateral cluster project primarily to the pons, where REMS generator neurons are located (Cvetkovic et al., 2004). Optogenetic mediated activation of MCH neurons may inhibit the local orexin neurons through release of inhibitory MCH, weakening the orexin drive onto downstream arousal neurons and attenuating the waking bout. It remains to be determined which projection(s) of the stimulated MCH neurons is responsible for the somnogenic effect. MCH neurons project to multiples sites, many of those include regions where the arousal neurons are located. Stimulation of specific projection(s) will be required to address this question. Nevertheless, the converging evidence from optogenetic, selective ablation of MCH neurons, and pharmacology studies underscores the inclusion of the MCH neurons in network models of sleep-wake regulation. That sleep was induced during the normal waking cycle suggests a potential role of MCH neurons in sleep disorders, such as insomnia, and also support the perspective that drugs targeting the MCH system may enhance sleep in certain sleep disorders. Supplementary Material Supp Fig S1 Supplementary Figure. Confocal scanning microscopy imaging of rat brain sections 18 days after microinjection of rAAV-ChR2-EYFP into the caudate-putamen (CPu; left panels) or into the zona incerta (ZI; right panels). At the CPu where there are no MCH neurons, there is no EYFP emission signal among neurons (top left panel). Non-neuronal auto fluorescence was observed under both 488nm and 552nm wavelength light but only in the area of cells damaged by the injection cannula. These autofluorescent cells may be traumatized or dying from direct injury or being scavenger cells infiltrating the damaged tissue. Activated macrophages are known to show strong autofluorescence. In contrast to the CPu, in the ZI numerous green fluorescent neurons along with its fibers are evident upon injection of the rAAV-ChR2-EYFP vector. These neurons were visible only when excited at 488nm wavelength light (top right panel) but not when excited at 552nm (middle right panel). Because EYFP emits light when excited at 488 nm wavelength, it confirms the specificity of the rAAV-ChR2-EYFP vector to transduce MCH neurons. In the ZI non-neuronal green and red autofluorescent cells with no processes are also evident only along the injection cannula track (top, middle and bottom right panels). We thank Wengxue Wang for assistance with the study and Dr. Amanda LaRue for use of the confocal microscope. This study was supported by Medical Research Service of the Department of Veterans Affairs (101 BX000798) and NIH grants 1K01AG041520 (to ML), NS052287, NS084477, NS079940 Abbreviations AP Anteroposterior CPu Caudate-Putamen ChR2 Channelrhodopsin2 ECoG Electrocorticography EMG Electromyography EYFP Enhanced Yellow Fluorescent Protein FFT Fast Fourier Transformation LC Locus Coeruleus MCH Melanin-Concentrating Hormone NREM no Rapid Eye Movement rAAV recombinant Adeno-Associated Virus REM Rapid Eye Movement ZI Zona incerta Figure 1 In vitro optogenetic activation of MCH neurons. A.- Representative traces showing membrane potentials recorded from a typical mouse MCH neuron (above) and a typical rat MCH neuron (bottom). Note how MCH neurons in both species are silent. B.- Bar graph showing similar resting membrane potential of MCH neurons from mice and rats. C.- Representative traces from rat MCH neurons showing the inward currents evoked by photostimulation of different frequencies under whole-cell voltage-clamp recording with cells held at −60 mV. D.- Representative traces showing the membrane depolarization and action potentials evoked by photostimulation of different frequencies under whole-cell current-clamp recording. Figure 2 Anatomical distribution maps of ChR2-EYFP+ expression in rats. ChR2-EYFP expression was evident across the dorsolateral tuberal hypothalamic area particularly within the zona incerta, perifornical, and lateral hypothalamus. ChR2-EYFP emission signal was less prominent in the medial aspect of hypothalamus. Overall EYFP emission signal spread roughly 1.4 mm along the anteroposterior axis and it was present in both hemispheres. Just above the EYFP emission signal, the blue rectangles indicate the locations of the optic fiber tips. Optic fiber locations were drawn based on the presence of its tracks. The alphanumeric codes on top represent the identification number of the rat whereas the numbers on the far left represent the anterior-posterior distance from bregma. Figure 3 rAAV-MCH-ChR2-EYFP vector successfully transfected MCH neurons. Top panel shows a panoramic view (4×) of the extent of transfection following a bilateral injection of rAAV-MCH-ChR2-EYFP. Confocal laser scanning microscopy indicated a robust expression of the reporter gene EYFP (green) across the lateral hypothalamus (LH), perifornical area (PeF) and the zona incerta (ZI). Middle magnification (20×) views of the PeF area (middle and bottom left panels). ChR2-EYFP signal is abundantly present in many somata but mostly its neuropil (middle left panel). Middle right panel shows that PeF contains many MCH-ir+ neurons. Bottom left panel illustrates the abundance of EYFP and MCH co-labeling. Bottom right panel was taken at 60× and it shows in great detail ChR2-EYFP expression associated to the plasma membrane of the neuron whereas MCH signal is located within the neuron’s cytoplasm. Cell counts were done at 60×. There was a selective expression of EYFP in MCH neurons (97.3± 0.97%), but 52.5 (± 2.0)% of MCH neurons were also EYFP+. Red arrows indicate examples of non-transfected MCH neurons whereas yellow arrows point toward multiple examples of positively transfected MCH neurons. Figure 4 EYFP colocalizes with MCH neurons but not with orexin (ORX) neurons. Confocal microscopy photomicrographs depict renderings of Z-stacks in XYZ planes. Left panel depicts colocalization of ChR2-EYFP in MCH-immunoreactive neurons, and right panel shows colocalization in orexin-ir neurons. Images were taken from sections from the same rat shown in figure 3 (WT38). Representative neurons are identified with arrowheads or arrows. White arrowheads identify double labeled somata (EYFP+MCH), white arrows identify either single MCH-ir (left panels) or ORX-ir (right panels) neurons, and green arrows identify single-labeled EYFP+ neurons. X axis=green line, Y axis=red line, Z=axis=blue line. Scale bars indicate the distance in microns. Figure 5 Effect of optogenetic stimulation of MCH neurons on percent (±SEM) of Wake, NREM sleep and REM sleep. The data summarizes the average percent of wake, NREM sleep and REM sleep during the 12 h period (night or day). Optogenetic stimulation started at lights-off and continued for 24 h (1 min on, 4 min off). *= significance versus 5, 10 or 30Hz within the ChR2 group (p=0.01). $= significance versus 0Hz ChR2 (p<0.02). #= significance versus no ChR2 (p=0.05) (Holm-Sidak post-hoc test after 2-RMANOVA). Figure 6 Time course of changes in wake, NREM sleep and REM sleep during 24 h of optogenetic stimulation. Stimulation started at night (lights-off). Grey horizontal bars at the bottom indicate the lights-off period. Data are 3 h percent (±SEM). *=significance versus 0 Hz (p=0.05; Holm-Sidak post-hoc test). Figure 7 Effects of 24 h of optogenetic activation of MCH neurons on sleep architecture. The first 12 h of stimulation occurred during the lights-off period whereas the second 12 h occurred during the lights-on period. To better visualize differences between treatments the number of waking bouts are expressed as log10 scale. NREMS and REMS bouts numbers are represented as linear scale. *=significance versus 0 Hz at p=0.01. Figure 8 Effect of optogenetic activation of rat MCH neurons on NREMS delta power and REMS theta power. ECoG power was determined only during NREM sleep (delta) or REM sleep (theta). To improve visualization hourly data was pooled in 3 h blocks for delta power and in 6 h for theta power. Gray horizontal bars denote the 12h lights-off, night period. *=significance versus 0 Hz (p=0.01). Conflict of Interests: No conflicts of interest, financial or otherwise are declared by the author(s). Adamantidis AR Zhang F Aravanis AM Deisseroth K de Lecea L 2007 Neural substrates of awakening probed with optogenetic control of hypocretin neurons Nature 450 420 424 17943086 Ahnaou A Dautzenberg FM Huysmans H Steckler T Drinkenburg WH 2011 Contribution of melanin-concentrating hormone (MCH1) receptor to thermoregulation and sleep stabilization: evidence from MCH1 (−/−) mice Behavioural brain research 218 42 50 21074567 Ahnaou A Drinkenburg WH Bouwknecht JA Alcazar J Steckler T Dautzenberg FM 2008 Blocking melanin-concentrating hormone MCH1 receptor affects rat sleep-wake architecture European journal of pharmacology 579 177 188 18062961 Apergis-Schoute J Iordanidou P Faure C Jego S Schone C Aitta-Aho T Adamantidis A Burdakov D 2015 Optogenetic evidence for inhibitory signaling from orexin to MCH neurons via local microcircuits The Journal of neuroscience: the official journal of the Society for Neuroscience 35 5435 5441 25855162 Aston-Jones G Bloom FE 1981 Nonrepinephrine-containing locus coeruleus neurons in behaving rats exhibit pronounced responses to non-noxious environmental stimuli J Neurosci 1 887 900 7346593 Basheer R Sherin JE Saper CB Morgan JI McCarley RW Shiromani PJ 1997 Effects of sleep on wake-induced c-fos expression The Journal of neuroscience: the official journal of the Society for Neuroscience 17 9746 9750 9391027 Bittencourt JC Presse F Arias C Peto C Vaughan J Nahon JL Vale W Sawchenko PE 1992 The melanin-concentrating hormone system of the rat brain: an immuno- and hybridization histochemical characterization J Comp Neurol 319 218 245 1522246 Blouin AM Fried I Wilson CL Staba RJ Behnke EJ Lam HA Maidment NT Karlsson KAE Lapierre JL Siegel JM 2013 Human hypocretin and melanin-concentrating hormone levels are linked to emotion and social interaction Nature communications 4 1547 Carter ME Yizhar O Chikahisa S Nguyen H Adamantidis A Nishino S Deisseroth K de Lecea L 2010 Tuning arousal with optogenetic modulation of locus coeruleus neurons Nature neuroscience 13 1526 1533 21037585 Chee MJ Arrigoni E Maratos-Flier E 2015 Melanin-concentrating hormone neurons release glutamate for feedforward inhibition of the lateral septum J Neurosci 35 3644 5 25716862 Chu N Bloom FE 1973 Norepinephrine-containing neurons: changes in spontaneous discharge patterns during sleeping and waking Science 179 908 910 4347167 Croizier S Franchi-Bernard G Colard C Poncet F La Roche A Risold PY 2010 A comparative analysis shows morphofunctional differences between the rat and mouse melanin-concentrating hormone systems PloS one 5 e15471 21103352 Cvetkovic V Brischoux F Griffond B Bernard G Jacquemard C Fellmann D Risold PY 2003 Evidence of melanin-concentrating hormone-containing neurons supplying both cortical and neuroendocrine projections Neuroscience 116 31 35 12535935 Cvetkovic V Brischoux F Jacquemard C Fellmann D Griffond B Risold PY 2004 Characterization of subpopulations of neurons producing melanin-concentrating hormone in the rat ventral diencephalon Journal of neurochemistry 91 911 919 15525345 Del Cid-Pellitero E Jones BE 2012 Immunohistochemical evidence for synaptic release of GABA from melanin-concentrating hormone containing varicosities in the locus coeruleus Neuroscience 223 269 276 22890079 Elias CF Lee CE Kelly JF Ahima RS Kuhar M Saper CB Elmquist JK 2001 Characterization of CART neurons in the rat and human hypothalamus The Journal of comparative neurology 432 1 19 11241374 Gutierrez-Herrera C Carus-Cadavieco M Jego S Ponomarenko A Korotkova T Adamantidis A Hypothalamic feedforward inhibition of thalamocortical network controls arousal and consciousness Nat neurosc 19 290 301 Hanriot L Camargo N Courau AC Leger L Luppi PH Peyron C 2007 Characterization of the melanin-concentrating hormone neurons activated during paradoxical sleep hypersomnia in rats The Journal of comparative neurology 505 147 157 17853446 Hasegawa E Yanagisawa M Sakurai T Mieda M 2014 Orexin neurons suppress narcolepsy via 2 distinct efferent pathways The Journal of clinical investigation 124 604 616 24382351 Hassani OK Lee MG Jones BE 2009 Melanin-concentrating hormone neurons discharge in a reciprocal manner to orexin neurons across the sleep-wake cycle Proceedings of the National Academy of Sciences of the United States of America 106 2418 2422 19188611 Hong EY Yoon YS Lee HS 2011 Differential distribution of melanin-concentrating hormone (MCH)- and hypocretin (Hcrt)-immunoreactive neurons projecting to the mesopontine cholinergic complex in the rat Brain research 1424 20 31 22015351 Jego S Glasgow SD Herrera CG Ekstrand M Reed SJ Boyce R Friedman J Burdakov D Adamantidis AR 2013 Optogenetic identification of a rapid eye movement sleep modulatory circuit in the hypothalamus Nature neuroscience Jouvet M 1962 Research on the neural structures and responsible mechanisms in different phases of physiological sleep Archives italiennes de biologie 100 125 206 14452612 Konadhode RR Pelluru D Blanco-Centurion C Zayachkivsky A Liu M Uhde T Glen WB Jr van den Pol AN Mulholland PJ Shiromani PJ 2013 Optogenetic stimulation of MCH neurons increases sleep The Journal of neuroscience: the official journal of the Society for Neuroscience 33 10257 10263 23785141 Konadhode RR Pelluru D Shiromani PJ 2014 Neurons containing orexin or melanin concentrating hormone reciprocally regulate wake and sleep Front Syst Neurosci 8 244 25620917 Lima FF Sita LV Oliveira AR Costa HC da Silva JM Mortara RA Haemmerle CA Xavier GF Canteras NS Bittencourt JC 2013 Hypothalamic melanin-concentrating hormone projections to the septo-hippocampal complex in the rat Journal of chemical neuroanatomy 47 1 14 23123956 Liu M Blanco-Centurion C Konadhode R Begum S Pelluru D Gerashchenko D Sakurai T Yanagisawa M van den Pol AN Shiromani PJ 2011 Orexin gene transfer into zona incerta neurons suppresses muscle paralysis in narcoleptic mice The Journal of neuroscience: the official journal of the Society for Neuroscience 31 6028 6040 21508228 McCarley RW Hobson JA 1971 Single neuron activity in cat gigantocellular tegmental field: selectivity of discharge in desynchronized sleep Science 174 1250 1252 5133450 McGinty DJ Harper RM 1976 Dorsal raphe neurons: depression of firing during sleep in cats Brain Res 101 569 575 1244990 Mileykovskiy BY Kiyashchenko LI Siegel JM 2005 Behavioral correlates of activity in identified hypocretin/orexin neurons Neuron 46 787 798 15924864 Monti JM Lagos P Jantos H Torterolo P 2014 Increased REM sleep after intra-locus coeruleus nucleus microinjection of melanin-concentrating hormone (MCH) in the rat Progress in neuro-psychopharmacology & biological psychiatry 56C 185 188 Pelluru D Konadhode R Shiromani PJ 2013 MCH neurons are the primary sleep-promoting group Sleep 36 1779 1781 24293750 Rao Y Lu M Ge F Marsh DJ Qian S Wang AH Picciotto MR Gao XB 2008 Regulation of synaptic efficacy in hypocretin/orexin-containing neurons by melanin concentrating hormone in the lateral hypothalamus The Journal of neuroscience: the official journal of the Society for Neuroscience 28 9101 9110 18784290 Saper CB Fuller PM Pedersen NP Lu J Scammell TE 2010 Sleep state switching Neuron 68 1023 1042 21172606 Sasaki K Suzuki M Mieda M Tsujino N Roth B Sakurai T 2011 Pharmacogenetic modulation of orexin neurons alters sleep/wakefulness states in mice PloS one 6 e20360 21647372 Sherin JE Shiromani PJ McCarley RW Saper CB 1996 Activation of ventrolateral preoptic neurons during sleep Science 271 216 219 8539624 Shiromani P Blanco-Centurion C 2011 Pontine Areas Inhibiting REM sleep Mallick BN Pandi-Perumal RS McCarley RW Morrison AR REM sleep: Regulation and function Cambridge University Press United Kingdom 280 284 Shiromani PJ Lu J Wagner D Thakkar J Greco MA Basheer R Thakkar M 2000 Compensatory sleep response to 12 h wakefulness in young and old rats American journal of physiology. Regulatory, integrative and comparative physiology 278 R125 133 Siegel JM McGinty DJ 1977 Pontine reticular formation neurons: relationship of discharge to motor activity Science 196 678 680 193185 Sita LV Elias CF Bittencourt JC 2007 Connectivity pattern suggests that incertohypothalamic area belongs to the medial hypothalamic system Neuroscience 148 949 969 17707116 Steininger TL Kilduff TS Behan M Benca RM Landry CF 2004 Comparison of hypocretin/orexin and melanin-concentrating hormone neurons and axonal projections in the embryonic and postnatal rat brain Journal of chemical neuroanatomy 27 165 181 15183202 Szymusiak R McGinty D 1986 Sleep-related neuronal discharge in the basal forebrain of cats Brain Res 370 82 92 3708324 Torterolo P Sampogna S Chase MH 2009 MCHergic projections to the nucleus pontis oralis participate in the control of active (REM) sleep Brain research 1268 76 87 19269278 Torterolo P Sampogna S Chase MH 2013 Hypocretinergic and non-hypocretinergic projections from the hypothalamus to the REM sleep executive area of the pons Brain research 1491 68 77 23122879 Tsunematsu T Kilduff TS Boyden ES Takahashi S Tominaga M Yamanaka A 2011 Acute optogenetic silencing of orexin/hypocretin neurons induces slow-wave sleep in mice The Journal of neuroscience: the official journal of the Society for Neuroscience 31 10529 10539 21775598 Tsunematsu T Ueno T Tabuchi S Inutsuka A Tanaka KF Hasuwa H Kilduff TS Terao A Yamanaka A 2014 Optogenetic Manipulation of Activity and Temporally Controlled Cell-Specific Ablation Reveal a Role for MCH Neurons in Sleep/Wake Regulation The Journal of neuroscience: the official journal of the Society for Neuroscience 34 6896 6909 24828644 van den Pol AN Acuna-Goycolea C Clark KR Ghosh PK 2004 Physiological properties of hypothalamic MCH neurons identified with selective expression of reporter gene after recombinant virus infection Neuron 42 635 652 15157424 Varin C Arthaud S Salvert D Gay N Libourel PA Luppi PH Leger L Fort P 2016 Sleep architecture and homeostasis in mice with partial ablation of melanin-concentrating hormone neurons Behavioural brain research 298 100 110 26529469 Verret L Goutagny R Fort P Cagnon L Salvert D Leger L Boissard R Salin P Peyron C Luppi PH 2003 A role of melanin-concentrating hormone producing neurons in the central regulation of paradoxical sleep BMC neuroscience 4 19 12964948 Willie JT Sinton CM Maratos-Flier E Yanagisawa M 2008 Abnormal response of melanin-concentrating hormone deficient mice to fasting: hyperactivity and rapid eye movement sleep suppression Neuroscience 156 819 829 18809470 Yoon YS Lee HS 2013 Projections from melanin-concentrating hormone (MCH) neurons to the dorsal raphe or the nuclear core of the locus coeruleus in the rat Brain research 1490 72 82 22967922 Zhang X van den Pol AN 2012 Thyrotropin-releasing hormone (TRH) inhibits melanin-concentrating hormone neurons: implications for TRH-mediated anorexic and arousal actions The Journal of neuroscience: the official journal of the Society for Neuroscience 32 3032 3043 22378876

PMC005xxxxxx/PMC5118153.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7605074 6087 Neuroscience Neuroscience Neuroscience 0306-4522 1873-7544 27746349 5118153 10.1016/j.neuroscience.2016.10.022 NIHMS823004 Article C3 transferase gene therapy for continuous conditional RhoA inhibition Gutekunst Claire-Anne PhD 1cguteku@emory.edu Tung Jack K. PhD 13jtung@emory.edu McDougal Margaret E. 1mmcdoug@emory.edu Gross Robert E. MD, PhD 123rgross@emory.edu 1 Department of Neurosurgery, Emory University School of Medicine, Atlanta, Georgia 2 Department of Neurology, Emory University School of Medicine, Atlanta, Georgia 3 Wallace H. Coulter Department of Biomedical Engineering, Emory University School of Medicine and Georgia Institute of Technology College of Engineering, Atlanta, Georgia Corresponding Authors: Claire-Anne Gutekunst, PhD, Emory University, Department of Neurosurgery, 101 Woodruff Circle, WMB, Suite 6000, Atlanta, GA 30322. cguteku@emory.edu. Robert E. Gross, MD, PhD, Emory University, Department of Neurosurgery, 101 Woodruff Circle, WMB, Suite 6000, Atlanta, GA 30322. rgross@emory.edu 15 10 2016 13 10 2016 17 12 2016 17 12 2017 339 308318 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Regrowth inhibitory molecules prevent axon regeneration in the adult mammalian central nervous system (CNS). RhoA, a small GTPase in the Rho family, is a key intracellular switch that mediates the effects of these extracellular regrowth inhibitors. The bacterial enzyme C3-ADP ribosyltransferase (C3) selectively and irreversibly inhibits the activation of RhoA and stimulates axon outgrowth and regeneration. However, effective intracellular delivery of the C3 protein in vivo is limited by poor cell permeability and a short duration of action. To address this, we have developed a gene therapy approach using viral vectors to introduce the C3 gene into neurons or neuronal progenitors. Our vectors deliver C3 in a cell-autonomous (endogenous) or a cell-nonautonomous (secretable/permeable) fashion and promote in vitro process outgrowth on inhibitory chondroitin sulfate proteoglycan substrate. Further conditional control of our vectors was achieved via the addition of a Tet-On system, which allows for transcriptional control with doxycycline administration. These vectors will be crucial tools for promoting continued axonal regeneration after CNS injuries or neurodegenerative diseases. Graphical abstract Introduction The Rho family of small GTPases comprises intracellular molecular switches that play critical roles in regulating diverse cellular processes from cell division and migration to axon outgrowth (Luo, 2000, Stankiewicz and Linseman, 2014). Three Rho GTPases – RhoA, Rac1 and Cdc42 – are central to the regulation of the actin and microtubule cytoskeleton involved in axon growth. In simplified terms, Rac1 regulates lamellipodia formation, Cdc42 regulates filipodia, and RhoA regulates axon retraction (stress fiber formation in non-neural cells). As such, RhoA is a pivotal switch in the axonal response to environmental cues that regulate axon extension versus retraction (Gross et al., 2007). The injured central nervous system (CNS) in the adult contains several types of molecules that inhibit the outgrowth and lead to retraction of axon growth cones, thus contributing to degeneration of fiber pathways and preventing regeneration of CNS pathways after various types of injury. Overcoming inhibitory molecules associated with myelin and the glial scar could greatly improve regeneration in the nervous system (McKerracher and Rosen, 2015). RhoA mediates the effects of diverse extracellular cues present after injury, including the myelin associated inhibitors (e.g. Nogo66), chondroitin sulfate proteoglycans (CSPGs), and some semaphorins that are commonly found in glial scars. Indeed, biochemical blockade of RhoA activity promotes axon growth and increased axon regeneration in the presence of these inhibitory molecules after CNS injury (Niederost et al., 2002, Fu et al., 2007). These promising effects of RhoA blockade are currently being evaluated in human clinical trials for the treatment of spinal cord injury (Fehlings et al., 2011). C3 transferase (C3) is a bacterial exoenzyme that specifically and irreversibly inhibits activation of RhoA by ADP ribosylation. Direct delivery of C3 to neurons has been shown to promote axon outgrowth (Niederost et al., 2002). However, C3 is not cell-permeable so modifications have been made to improve its entry into cells (Winton et al., 2002, Tan et al., 2007). In vivo inhibition of RhoA by direct injection of C3 promotes robust axonal regeneration in the CNS, as demonstrated in models of optic nerve crush (ONC) or spinal cord injuries (SCIs). C3 recombinant protein delivered directly to the injured optic nerve at the crush site allowed processes to extend beyond the lesion site, but was limited by the short period during which injured axon processes could take up the C3 reagent (Lehmann et al., 1999). A single application of a cell-permeable version of recombinant C3, C3-07, resulted in neuroprotection of RGCs for one week, as well as increased outgrowth of RGC axons across an ONC lesion (Bertrand et al., 2005). Additional injections resulted in improved survival and regeneration over a 2 week period over the single injection (Bertrand et al., 2007). Similarly, groups have documented axon regeneration by RhoA inhibition after SCIs. In rats, permeable C3 was delivered to a T7 dorsal -hemisection SCI model resulting in extensive axonal sprouting into the lesion site and scar. Subsequent SCI studies reconfirmed that a single injection of a cell permeable C3 (Cethrin) was detectable in cells 7 days later and blocked SCI – induced RhoA activation and apoptosis for that period (McKerracher and Higuchi, 2006). Further results following permeable C3 (Cethrin) injections into SCI have yet to be reported, but are the subject of a human clinical trial (Fehlings et al., 2011, McKerracher and Anderson, 2013). Although these modifications have increased the versatility of utilizing C3 for RhoA inhibition, these studies indicate that without a continuous source of cell-permeable C3, its cellular actions are limited to a duration of several days, which is likely insufficient for the regeneration of long axon pathways that are commonly damaged in neurodegenerative diseases. To address these limitations, we have generated viral vectors expressing C3 transferase to achieve specific, widespread, long term, and conditional RhoA inactivation. These novel vectors express either an endogenous C3 (eC3) or a secretable/permeable C3 (spC3) fused to the green fluorescent protein (GFP). Borrowing from the genetic nomenclature, we have called these C3 variants cell autonomous eC3 for expression within infected cells, and cell nonautonomous spC3, for an effect beyond the infected cells, respectively. We hypothesize that the latter should be able to affect a greater number of neurons than those infected with the cell autonomous approach. To temporally regulate and reduce any potential risks or side effects of C3 expression, we also developed expression vectors that are regulated by doxycycline (Szulc et al., 2006). As proof of principal, we have tested our transgenes in vitro and shown that C3 expression from both cell autonomous and cell non-autonomous vectors inhibit RhoA activation and promote neurite growth on inhibitory substrates. C3 expression in the rat striatum could also be conditionally controlled by doxycycline treatment with no significant adverse effects on striatal integrity. This novel toolbox of C3 vectors provides a versatile means of inhibiting RhoA in the nervous system and offers a promising translatable therapy for the regeneration of injured CNS pathways. EXPERIMENTAL PROCEDURES Plasmid construction The myc-tagged C3 gene was polymerase chain reaction (PCR) amplified from the pRK5-mycC3 vector generously provided by Dr. Alan Hall (Aktories et al., 1989) and inserted into the multiple cloning site of the pEGFP-N2 plasmid (Clontech, Mountain View, CA), in frame with the amino terminus of enhanced GFP (EGFP) to generate pEGFP-N2-C3 that expresses an endogenous form of C3GFP (eC3GFP). Next, a vector that expresses a secretable and permeable form of C3GFP was generated. Overlapping primers containing the DNA sequence for the synthetically modified trans-acting activator of transcription (TATκ) permeability peptide were PCR amplified and inserted at the amino terminus of C3GFP to create pEGFP-N2-TATκC3 (Tunnemann et al., 2006). The cassette containing TATκC3GFP was then PCR amplified and inserted into the multiple cloning site of the pSecTag2 Hygro A plasmid (Invitrogen) to add an immunoglobulin K (IgK) peptide signal sequence to the N-terminus of TATκC3GFP, thus allowing for the secretion of TATκC3GFP (spC3GFP) into the extracellular milieu. A control IgK-TATκGFP (spGFP) cassette was generated in the same fashion. The cassettes containing eGFP, eC3GFP, spGFP and spC3GFP were subsequently cloned into the FUGW lentiviral and the pAAV backbone plasmids for production of 2nd generation lentivirus and adeno-associated serotype 2 virus (AAV2), respectively. Cell culture Human embryonic kidney (HEK) 293T cells and NIH3T3 cells were both maintained in a humidified 5% CO2 atmosphere at 37°C in complete medium consisting of Dulbecco’s Modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 international units (IU)/ml penicillin, and 100 μg/ml streptomycin. Western Blot (WB) Transient transfections were carried out with the Lipofectamine 2000 reagent (Invitrogen). HEK293 cells were seeded in 6-well plates at 2×105 cells per well, grown for 24 hours (hr) and then incubated for 16hr with 1 ml of serum-free medium containing 2.5 μg of plasmid and 5 μl of Lipofectamine 2000 (Invitrogen, Carlsbad, CA). After incubation for 16hr, the medium was replaced by fresh medium containing 10% fetal calf serum for 24hr. Cell extracts were prepared in lysis buffer (Promega, Madison, WI) and protein concentration was determined using a bicinchoninic acid protein assay kit (ThermoFisher Scientific, Waltham, MA). Samples were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane. The membrane was blocked and incubated with either rabbit anti-C3 (generously provided by Dr. Lisa McKerracher, Bioaxone, Cambridge, MA (Winton et al., 2002)) or rabbit anti-GFP antibody (Clontech, Mountain View, CA) followed by a donkey anti-rabbit DyLight 800 secondary antibody (Pierce, ThermoFisher Scientific, Waltham, MA). Infrared (IR) detection was performed on an Odyssey IR scanner (Li-cor Biotechnology, Lincoln, NE). Rhotekin assay NIH3T3 cells were seeded in 10 cm dishes at 5×105 cells per plate and transfected as described above. NIH3T3 cells were incubated with permeable C3 (2 μg/ml for 4hr, Cytoskeleton Inc.), or dimethyl sulfoxide (DMSO). The next day, the effect of C3 expression on RhoA activation was measured using a pull down assay performed according to the manufacturer’s recommendations (Cytoskeleton Inc., Denver, CO). Briefly, cells were rinsed with ice-cold phosphate buffer solution (PBS), and then 250 μl of ice-cold 1X lysis/wash buffer was added. Lysates were cleared by centrifugation for 10 minutes. An aliquot of lysate was collected for measuring the total amount of Rho protein. The remaining lysate was added to Rhotekin agarose beads and incubated on ice for 45 minutes and washed three times. Beads and total Rho samples were then subjected to SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked and incubated with anti-Rho antibody followed by a DyLight 680 secondary antibody. IR detection was again performed on an Odyssey IR scanner (Li-cor Biotechnology, Lincoln, NE). Lentiviral vector production The FUGW and pLVUT-tTR-KRAB (pLVUT) backbones were obtained from Addgene (plasmids 14883 and 11651, respectively). FUGW is a 3rd generation lentiviral plasmid using the human Ubiquitin C (hUbC) promoter to drive enhanced GFP expression (Lois et al., 2002). pLVUT is a Tet-regulated (Tet-on) lentiviral vector for transgene expression under hUbC promoter control and a woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) (Szulc et al., 2006). Lentivirus were produced in-house based on methods described by Dr. Didier Trono (Barde et al., 2010). In brief, HEK293FT cells were grown in multiple 10 cm cell culture dishes to 90% confluency on the day of transfection. Each dish was then transfected with 7.5 μg Δ8.9 packaging vector, 3 μg VSVG envelope vector, and 10 μg of the transfer vector by calcium phosphate precipitation. Transfection efficiency was confirmed the following day by fluorescence microscopy and lentivirus was subsequently harvested for 2 days following transfection. The harvested lentivirus was purified through a 0.45 μM polyethersulfone (PES) filter and concentrated by ultracentrifugation with a 20% sucrose cushion to an approximate titer of 108–1010 infectious particles/ml. AAV viral vector production The AAV backbone with a chicken β-actin (CBA) promoter and WPRE was generously provided by Dr. Michael G. Kaplitt, Weill Cornell Medical Center, New York, NY. Stock of AAV serotype 2 expressing GFP (AAV-eGFP) was obtained from the University of North Carolina Vector Core (www.med.unc.edu/genetherapy/vectorcore). The laboratory of Dr. Michael G. Kaplitt prepared virus stocks for AAV-eC3GFP, AAV-spGFP and AAV-spC3GFP. Briefly, vector plasmids were packaged into AAV2 particles using a helper-free plasmid transfection system. The vectors were purified by heparin affinity chromatography and dialyzed against PBS. AAV titers were determined by quantitative PCR using primers to a fragment of the AAV backbone and adjusted to 1012 genomic particles per ml (Sondhi et al., 2012). Dot blot Dot blot assay was used to confirm the secretion of spGFP and spC3GFP. HEK293T cells were plated in 10 cm dishes and transfected as described. After several days in culture, cells were incubated overnight in unsupplemented DMEM. The next day, media was collected and concentrated using Amicon Ultra centrifugal filters (EDM Millipore, Billerica, MA). Concentrated samples were then bound to nitrocellulose membrane. The membrane was blocked and incubated with a rabbit polyclonal anti-GFP antibody (632460, Clontech, Mountain View, CA) followed by a donkey anti-rabbit DyLight 800 secondary antibody (Rockland, Limerick, PA). Again, IR detection was performed on an Odyssey IR scanner (Li-cor Biotechnology, Lincoln, NE). spC3 functional assay HEK293 cells were plated in a 6 well tray and transfected as described above with spC3GFP or DsRed2 expressing plasmids. Ten hours later, media was changed and DsRed2 expressing cells were passaged and transferred to a well containing either spC3GFP transfected cells or no cells. After 48 hr, the effect of spC3 on process outgrowth of the DsRed2 expressing cells was evaluated by fluorescence microscopy. CSPG spot assay Spots were prepared by placing six 2 μl droplets of Neurobasal media containing mixed CSPGs (50 μg/ml; CC117, EDM Millipore, Billerica, MA) and laminin (5 μg/ml; 23017-015, Invitrogen, Carlsbad, CA) onto poly-L-lysine (PLL) (20 μg/ml; P9155, Sigma-Aldrich, St Louis, MO) pretreated glass coverslips and allowed to dry completely. Rhodamine was added to the droplets at 1 μg/ml to help visualize the permissive/repulsive interface. Cortical neurons were dissociated from rat embryos at day 18 (E18) as described previously (Gutekunst et al., 2003). Neurons were plated on pre-coated coverslips and transduced with control or C3 expressing viruses. Five days post transduction cells were fixed with 4% paraformaldehyde and coverslipped with Vectashield mounting media (Vector Laboratories, Burlingame, CA) and visualized using a Leica DMREI fluorescent microscope equipped with a QImaging Retiga EXi camera and Simple PCI acquisition software. Expression of the GFP reporter was visible starting 3 days post infection. Twelve and eleven micrographs along the interface between PLL/CSPG were taken at random for eGFP and eC3GF infected cultures respectively, and the number of processes crossing the PLL/CSPG interface counted for each micrograph. For statistical analysis, the mean number of crosses were compared using a Student t-test independent sample, one tail, two sample equal variance with alpha (α)=0.05. Process outgrowth Mouse primary cortical neurons were dissociated and plated in 8-well slide chambers (3 chambers/viral vector) precoated with CSPGs (50 μg/ml; EDM Millipore, Billerica, MA) onto PLL (20 μg/ml) and infected with control or C3 expressing AAV2 at 1010 particles/ml. Three days post infection cells were fixed 4% paraformaldehyde and immunostained using mouse monoclonal anti-acetylated tubulin antibodies (T7451 at 1:500, Sigma-Aldrich, St Louis, MO). Cells were permeabilized in 0.01% Triton-X100 (Sigma-Aldrich, St Louis, MO) in PBS. Length of primary and secondary processes was assessed using free software ImageJ/NeuronJ (NIH) (Schneider et al., 2012). In this study a primary process is defined as directly extending from the cell body, while a secondary process is defined as a branch off the primary. Data are representative of a culture for which each vector was tested in triplicates and primary and secondary processes for 20, 19, and 31 neurons were used for analysis for eGFP, eC3GFP, and spC3GFP, respectively. For statistical analysis, a mixed linear model was used to estimate the means and variances (within- and -between neuron) for ln length (natural log) of the primary processes for each type of expression (eGFP, eC3GFP, and spC3GFP) using the SAS MIXED Procedure. A compound-symmetric variance-covariance form was assumed for the outcome and estimates of the standard errors of parameters were used to perform statistical tests and construct 95% confidence intervals (Diggle et al., 1994). The model-based means are unbiased with unbalanced and missing data, so long as the missing data are non-informative (missing at random). Since a significant overall difference was detected between the 3 types of expression, t-tests were used to compare the differences between the 3 model-based means for expression. The specific statistical tests were done within the framework of the mixed effects linear model. All statistical tests were 2-sided and unadjusted for multiple comparisons. A value of P < 0.05 indicated statistical significance for the overall expression effect. Length are reported as means along with 95% confidence intervals (CI). Similar analyses were performed for secondary processes and for longest length of primary and secondary processes. Immunocytochemistry HEK293T cells were prepared in 6-well plates (as described for WB) and infected with control or C3 expressing AAV vectors (final concentration 109 particles/ml). Three days post transduction, cells were fixed in 4% paraformaldehyde (PFA) in PBS for 15 min and processed for immunocytochemistry (ICC) using rabbit anti-C3 antibodies. Briefly, cells were permeabilized with 0.01% Triton X-100 in PBS for 5 min, (Sigma-Aldrich, St Louis, MO) in PBS for 5 min, blocked in 4% normal donkey serum (NDS, Jackson ImmunoResearch Laboratories, West Grove PA) for 30 min and incubated with rabbit anti-C3 antibodies (5 μg/ml) in 2% NDS in PBS overnight. After several washes, the cells were incubated in Alexa 594 donkey anti-rabbit secondary antibodies (Invitrogen) in PBS for 1hr. Cells were imaged using an inverted Leica DMIRE2 scope equipped with a QImaging Retiga EXi camera and Simple PCI acquisition software. The captured images were stored and processed using Adobe Photoshop software. In vivo viral vector injection Adult male Sprague Dawley rats (200–280gm, Charles River Laboratories) were housed in the Emory animal vivarium with a 12hr light/12hr dark cycle and ad libitum access to food and water. All procedures were conducted in accordance with approved guidelines from the Emory University Institutional Animal Care and Use Committee. Animals received bilateral injections of either LV-eGFP (n=3), LV-eC3GFp (n=3) or LV-spC3GFP (n=3) at 1010 viral particles/ml. Virus were injected (1 μl and 3μl on each side, respectively) using a pulled glass pipette connected to a Nanoject (Drummond Scientific, Bromall, PA) at 280 nl/min at the following stereotactic coordinates: at Bregma; ± 3 mm lateral to midline; 5.2 mm below dura. Starting 5 days post-injection, rats were randomly assigned to drink either sugar water or doxycycline supplemented sugar water (Szulc et al., 2006). For rat experiments, we added 0.2 g/l doxycycline (D3447, Sigma-Aldrich, St Louis, MO) and 5% sucrose into drinking water. The sugar was added to mask the bitterness of the doxycycline. Immunohistochemistry and double immunofluoresence After two weeks rats were deeply anesthetized with a lethal dose of Euthasol (130 mg/kg), injected intraperitoneally, and then perfused intracardially with 0.9% NaCl, followed by 4% PFA in 0.1 M PBS at pH 7.2 (PB) for 15 min at a rate of 20 ml per min. Brains were removed and cryoprotected in 30% sucrose at 4°C, sectioned in coronal plane at 50 μm thickness using a freezing microtome, collected and rinsed in 0.1 M PBS, pH 7.2. Immunohistochemistry was performed as described previously (Gutekunst et al., 1995, Gutekunst et al., 2010). Free-floating sections were incubated in 0.1% TritonX-100 and 3% hydrogen peroxide to eliminate endogenous peroxidase, rinsed in PBS, and preblocked in 4% normal goat serum (NGS) in PBS for 30 min at room temperature (RT). Sections were incubated in rabbit anti-GFP antibodies (2555, Cell Signaling Technology, Beverly, MA) in PBS containing 2% NGS at 4°C for 48hr, then rinsed and incubated for 1hr at RT in biotinylated anti-rabbit antibody (ABC Elite; Vector Laboratories, Burlingame, CA) in PBS containing 2% NGS. After several rinses in PBS, the sections were incubated in avidin-biotin complex (ABC Elite; Vector) for 90 min at 4°C. Immunoreactivity was visualized by incubation in 0.05% 3,3′-diaminobenzidine tetrahydrochloride (DAB; Sigma, St. Louis, MO) and 0.01% hydrogen peroxide in PBS, until a dark brown reaction product was evident (5–10 min). Sections were rinsed and mounted on glass slides, air dried and coverslipped. Sections were visualized using either a Nikon eclipse E400 microscope and images captured using a color digital camera (Nikon Instruments Inc, Melville, NY). Double immunofluorescence was used to co-localize C3 and KRAB as previously described (Gutekunst et al., 2012). Free-floating striatal sections were rinsed in PBS, blocked in 5% NDS and 0.1% Triton-X for 30 min at RT and rinsed in PBS. After rinses in PBS, sections were incubated overnight at 4°C in mouse anti-GFP and rabbit anti-KRAB (AB72609, Abcam, Cambridge, MA) in PBS containing 1% NDS. Sections were rinsed in PBS and incubated in Alexa Fluor 594 conjugated donkey anti-rabbit (Jackson Immunoresearch Laboratories, West grove, PA) and Alexa Fluor 488 conjugated donkey anti-mouse (Jackson Immunoresearch Laboratories, West grove, PA) secondary antibody in 1% NDS for 1hr at RT. Sections were rinsed with PBS, then mounted on glass slides with hard set Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Sections were visualized using either a Nikon eclipse E400 microscope equipped with 4 fluorescent cubes, a monochrome and color digital camera and Nikon BR software (Nikon Instruments Inc, Melville, NY). Immunofluorescence was also used to quantify brain response to C3 expression. Free floating sections were stained by immunofluorescence using mouse anti-glial fibrillary acidic protein (GFAP; G3893, Sigma-Aldrich, St. Louis, MO) or mouse anti-rat CD11b/c (OX42, 554859, Pharmingen, SanDiego, CA), followed by an Alexa Fluor 594 conjugated donkey antimouse (Jackson Immunoresearch Laboratories, West grove, PA) secondary antibody. Sections were mounted on glass slides with hard set Vectashield mounting medium (Vector Laboratories, Burlingame, CA). For each animal, an image was taken at the level of the injection site and GFAP and OX42 immunofluorescence intensity was assessed in 4 random windows per image using free software ImageJ/NeuronJ (NIH) (Schneider et al., 2012). For statistical analysis, mean intensity levels were compared between groups mixed model as described previously. RESULTS Development of an endogenous C3GFP (eC3GFP) cassette To generate a vector co-expressing C3 and GFP as a fusion protein (C3GFP), the C3 cDNA was PCR amplified and inserted into the cloning site of the pEGFP-N2 plasmid encoding an enhanced GFP (EGFP) variant that has been optimized for brighter fluorescence and higher expression in mammals (Di Polo et al., 1998, Harvey et al., 2009, Hellstrom et al., 2009). Transfected HEK293T cells expressed a band at ~ 60 kDa only in pEGFP-N2-C3 transfected cells, corresponding to the added molecular weights of EGFP and C3, confirming proper expression of the fusion protein (Figure 1A). The effectiveness of the C3GFP cassette in reducing RhoA activation was assessed with a Rhotekin RhoA activation assay. NIH3T3 cells, which are commonly used in Rho activation assays (Zheng et al., 2009), were treated with permeable C3/DMSO, DMSO alone, or transfected with either pEGFP-N2 or pEGFP-N2-C3. Cell lysates were incubated with beads coated with the Rho binding domain (RBD) of the Rho effector protein, Rhotekin, which shows high affinity for GTP-RhoA, and the Rhotekin- RBD/GTP-RhoA complexes bound to the beads were immunoprecipitated with RhoA antibodies; total RhoA (GTP- and GDP- bound) levels were assessed in the absence of immunoprecipitation. As expected, there were equal amounts of total RhoA across samples but decreased levels of active RhoA (RhoA-GTP) in C3 treated and pEGFP-N2-C3 transfected cells as compared to untreated cells and cells transfected with endogenous GFP control (pEGFP-N2) (Figure 1B). Functional cassettes were then moved into an AAV backbone and viral vectors were generated. C3 was only seen in cells transfected with pEGFP-N2-C3 (Figure 1C). Both GFP and C3GFP were present in the cytoplasm, filling the soma and processes consistent with endogenous and cytoplasmic protein expression (Figure 1C). eC3GFP promotes crossing at PLL/CSPG interface To confirm the functionality of C3 in the C3GFP fusion protein we used a CSPG spot assay and analyzed the effect of C3 expression on the growth of neuronal processes as they encounter an inhibitory substrate interface. CSPG is a key component of glial scars and a known potent inhibitor of axon regeneration (Fawcett and Asher, 1999, Sandvig et al., 2004, Silver and Miller, 2004, Wang et al., 2008). An interface between PLL and CSPG was created by placing CSPG/laminin-Rhodamine (CSPG-Rhod) or laminin-Rhodamine (Rhod) drops onto laminin coated glass coverslips. Rhodamine was used to visualize the interface and also served as a negative control. Primary dissociated rat E18 cortical neurons were plated on PLL/laminin with a CSPG drop in the center, and after 3 days infected with either eGFP or eC3GFP-expressing AAV. After 5 days most of the processes in cultures infected with eGFP-expressing vector turned at the PLL/CSPG interface, rarely crossing over to the CSPG side. In contrast, processes in cultures transduced with eC3GFP crossed the interface and extended onto the CSPG drop (Figure 2A). The mean number of crosses was 7 times greater in eC3GFP infected cultures compared to eGFP controls (Figure 2B) (Student t-test: t=3.882, df=21, p=0.004). Development of a secretable/permeable C3GFP (spC3GFP) cassette The functional eC3GFP cassette was then used to generate a vector that expresses a secreted and permeable form of C3GFP (spC3GFP). The synthetically modified trans-acting activator of transcription (TATκ) protein transduction domain from human immunodeficiency virus type 1 was used as a permeability tag (Tunnemann et al., 2006). TATκ is a commonly used transport peptide and has proven effective at transporting cargo to both the cytosol and the nucleus (Fawell et al., 1994, Vives et al., 1997, Schwarze et al., 1999, Flinterman et al., 2009). A secretable IgK signal peptide was then added to the amino terminus of TATκC3GFP using the pSecTag2 plasmid. C3 antibodies detected appropriately sized (~ 70 kDa) spC3GFP fusion proteins only in HEK293 cells transfected with pSecTag2-TATκ-C3GFP (Figure 3A). The IgK- TATκ-C3GFP cassette was subsequently moved into an AAV backbone; transfection of HEK293T cells with AAV-spC3GFP resulted in expression of spC3GFP, which appeared punctated as expected for secreted proteins (Figure 3B). Proper processing and secretion of spC3GFP was confirmed with a dot blot assay performed on media collected from transfected HEK293 cells (Figure 3C). GFP antibodies detected transgene expression only in the media of spC3GFP-transfected cells with no signal detected in the media of non-transfected cells or cells transfected with endogenous GFP or C3GFP (Figure 3C). The functional property of spC3GFP was also confirmed by culturing HEK293 cells expressing DsRed2 (as a marker) either alone or together with HEK293 cells that had been previously transfected to express spC3GFP. When co-cultured with spC3GFP expressing cells, DsRed-HEK293 cells developed long processes compared to when cultured alone (Figure 3D). The permeable property of the transgenes is again further characterized in a subsequent in vivo experiment (described below). eC3GFP and spC3GFP promote process outgrowth on CSPG substrate C3 transferase promotes growth of processes on inhibitory substrates. To confirm the functionality of endogenous and secretable/permeable C3GFP cassettes we assessed the growth of processes of cortical neurons growing on CSPG substrate. Primary cortical neurons were plated on CSPG substrate and infected with either eGFP-, eC3GFP- or spC3GFP-expressing AAV vectors. After 3 days, GFP was visible in neurons and cultures were immunostained with tubulin antibodies to visualize and measure primary and secondary processes (Figure 4). The average length of the primary processes was significantly increased in cultures expressing eC3GFP and spC3GFP compared to control cultures only expressing eGFP (p<0.05, t-tests). Whereas the mean length of primary processes in control cultures was 47.8 μm 95% CI (36.7, 58.8), the average length was 62.5 μm 95% CI (50, 75) and 96.4 μm 95% CI (79.8, 113.1) for eC3GFP and spC3GFp infected cultures respectively (Figure 4B). The effect of spC3GFP was significantly stronger than that of eC3GFP (p<0.005, t-test) and the average length of primary processes of spC3GFP cultures was twice as long than in the control cultures (p<0.001, t-test). When considering only the longest primary process for each neuron examined, eGFP and eC3GFP had similar longest primary processes (159 μm and 153 μm respectively), whereas the mean longest primary process of neurons in cultures infected with spC3GFP was significantly longer at 242.8 μm 95% CI (198.4, 287.1) (p<0.05, t-test) (Figure 4C). Unlike spC3GFP, eC3GFP also had a significant effect in increasing the average length of secondary processes compared to eGFP control (Figure 4B). The average length of secondary processes was 25.6 μm, 95% CI (21.1, 30) for eGFP, 21.4 μm, 95% CI (17.4, 25.3) for eC3GFP, compared to 31.5 μm, 95% CI (27.4, 35.6) for spC3GFP (p<0.05, t-test). The longest secondary processes were also significantly longer in spC3GFP infected cultures compared to control eGFP but not eC3GFP with an average length of 77.0 μm, 95% CI (59.5,94,5) (p<0.05, t-test), 46.1 μm, 95% CI (27.9,64.3) and 52.5 μm, 95% CI (28.8, 76.22) respectively (Figure 4C). In vivo expression of tet-on conditional eC3GFP and spC3GFP To validate in vivo expression, the eGFP, eC3GFP, and spC3GFP cassettes were moved into the tet-on lentiviral plasmid pLVUT for the production of inducible lentiviral vectors (LV). One of the advantages of using pLVUT is its drug-inducible design, which allows control of gene expression in mammalian cells (Szulc et al., 2006). In the absence of doxycycline, the Kruppel associated box (KRAB) transcriptional repressor binds to a tetracycline operator (tetO) and suppresses the expression of the transgene. Upon the addition of doxycycline, the suppression is released (Szulc et al., 2006). The incorporation of such controls in C3 gene therapy would ensure that C3 expression is turned on only during the desired time frame for regeneration. In addition, cells infected with pLVUT express KRAB constitutively (Groner et al., 2012). As such, KRAB can be used as a marker of expression and, for the purposes of this study, can be used to differentiate LV-spC3GFP-infected cells from those surrounding cells that have taken up the secreted and permeable C3GFP in the absence of cell-autonomous expression. Following injections of either LV-eGFP, LV-eC3GFP or LV-spC3GFP in the striatum, half of the rats received doxycycline-supplemented drinking water for 14 consecutive days. GFP expression was visible only in those animals that had received doxycycline-supplemented water (Figure 5A a–f). GFP immunohistochemistry confirmed GFP expression in cell bodies and processes and uncovered more GFP+ cells and processes than were visible by direct GFP visualization (Figure 5A j–l). More GFP was present in the neuropil of spC3GFP injected striatum (Figure 5A i) compared to eGFP and eC3GFP (Figure 5A g and h). GFP was found in all striatal neuronal cell types including DARPP-32+ medium spiny neurons and large ChAT+ cholinergic interneurons (data not shown). As expected with lentivirus, there was no detectable infection of glial cells. To further confirm the permeable aspect of secreted spC3GFP, striatal sections from rats injected with spC3GFP were stained for KRAB and GFP (Figure 5B). KRAB immunostaining is visible in the nuclei of many GFP+ cells, but in addition a number of cells positive for GFP but negative for KRAB (Figure 5B) were seen; the latter are most likely cells that are not infected but have taken up the secreted and permeable C3GFP in a cell-nonautonomous manner. The response of glial cells to C3 expression was also examined. Sections with comparable infection with GFP, eC3GFP or spC3GFP expression (confirmed by GFP ICC on adjacent sections) showed no difference in GFAP (reactive astrocytes) or OX42 (microglia) immunoreactivity (Figure 5C), early evidence of absence of inflammation. DISCUSSION In the present study we describe the development and validation of several viral vectors for long term and regulated expression of C3 transferase. We designed our first vector to express an endogenous form of full-length C3 transferase fused to GFP. There are only two other studies for which viral vectors have been developed to express C3 transferase. In both studies vectors were engineered to express C3 and GFP separated by an internal ribosomal entry site (IRES) (Fischer et al., 2004, Liu et al., 2005). In addition, for one of the vectors, a sequence encoding the first 10 amino acids of GAP-43 was added to the amino terminus of C3 to target its localization to the cell membrane (Fischer et al., 2004). Similar to our construct, these vectors express an endogenous form of full-length C3 transferase. However, in our design C3 and GFP are combined as a single fusion protein where GFP can be used as a direct reporter of not only infected cells but also the localization and potential site(s) of action of C3 transferase within the cell compartments. Conversely, with an IRES sequence, GFP can only report which cells have been transduced, and the precise location of C3 requires further manipulation such as immunohistochemistry. Consistent with previous studies, transfection with eC3GFP induced RhoA inhibition, which led to the reorganization of the cytoskeleton and process growth indicating that fusing GFP to C3 did not affect the activity of C3 transferase. Similarly, AAVeC3GFP was able to promote the growth of processes on an inhibitory substrate. This vector can be used as a gene therapy when aiming at inactivating RhoA within a specific cell population. The virus can be injected directly into the CNS or applied to tissue or cells prior to implantation. Although in the present form the construct utilizes the strong ubiquitous chicken β-actin promoter for fast expression with relatively high levels, more specific promoters and/or viral serotypes can be explored to obtain cell type specificity. One limitation of the eC3GFP and the two previously developed vectors is that the transgenes are expressed as cytoplasmic proteins, therefore limiting the effect of C3 to only cells that are transduced. To enable a more widespread distribution of C3, we designed our second vector to express a secretable/permeable form of full-length C3 transferase fused to GFP. In this approach, a cell non-autonomous spC3GFP cassette was engineered so that cells adjacent to those constitutively expressing C3 can also benefit from its effect. To our knowledge, this is the first development of a secretable/permeable C3 expression system. Similar to eC3GFP, spC3GFP promoted growth of processes on inhibitory substrate but showed an enhanced effect compared to eC3GFP, with 25% and 40% longer growth of primary and secondary processes, respectively. This novel vector can be used as a therapeutic approach when RhoA inhibition is needed in a large population of cells or when the cells needing treatment are too fragile to transduce or perhaps unable to express high levels of transgene. For example, instead of infecting degenerating neurons, a virus could be targeted at the healthy adjacent cells (i.e. different neuronal subtype or glia). Again, specific promoters and/or serotypes would be incorporated in the vector design to achieve specificity. It has been suggested that C3 may increase the glial inflammatory response post injury (Holtje et al., 2005, Hoffmann et al., 2008), an effect that may intensify with longer-term expression, which distinguishes our experiments from others with short-term direct C3 delivery. In this case, the spC3GFP vector design, which allows for a more widespread distribution of C3, could have deleterious side effects. With this in mind, we developed a third vector designed with a controllable system. Such an approach might maximize effectiveness vs. toxicity and can be used to identify the optimal interval for expression via dose and time–response curves with doxycycline. In addition, KRAB-mediated repression eliminates the leaky, nonspecific expression seen in other conditional systems based on transactivation (Szulc et al., 2006). Of note, this study found that the C3-expressing vectors did not appear to induce any glial response when injected directly into a healthy rat brain. Further studies need to be done to determine their effect in the context of an injury model. Virally mediated C3 expression can significantly reduce RhoA activation and represents a novel gene therapy approach that could be used alone or in conjunction with other therapies to promote axon regrowth in neurodegenerative disorders or injuries of the CNS. To validate the functionality of the eC3GFP and spC3GFP in vivo, the vectors are currently being tested in a rat optic nerve crush (ONC) model. The optic nerve offers several advantages for testing new approaches for CNS regeneration (Vidal-Sanz et al., 1987, Villegas-Perez et al., 1993, Berry et al., 1996, Fournier et al., 2003). Injured optic nerves were treated via intravitreous injection of AAV2-eC3GFP or AAV2-spC3GFP. Both vectors increased RGC survival and axon regeneration compared to controls up to 8 weeks post-injury (submitted). In summary, these new vectors are promising therapeutic approaches to regenerating and protecting neurons after CNS injury or neurodegenerative diseases. Depending on the need, they can be injected directly into the injured tissue to reverse the degenerative process or transduced to stem cells for cell replacement therapies. Further work is underway to test these vectors in other in vivo settings including implantation of ex-vivo infected neurons. We thank Elizabeth Jackson for her help with viral production and cell culture assays, Patty Murphy, Olga Laur, and Samantha Neumann for helping with vector construction, and Joseph Mertz for his help with viral injections and immunohistochemistry. We also thank Dr. Kirk Easley, MApStat, for help with statistical analysis. GRANT SUPPORT Supported by a National Institutes of Health grant 1R03NS091699-01 (CAG and REG) Supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR000454. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. (CAG and REG) Supported by the Emory University Research Committee (REG) Neurosurgery Research and Education Foundation via a Medical Student Summer Research Fellowship award (MEM) Viral Vector Core of the Emory Neuroscience NINDS Core Facilities grant, P30NS055077. The funding sources had no involvement in the study design, collection, analysis and interpretation, nor in the writing or the decision to submit this manuscript. Abbreviations AAV2 adeno-associated serotype 2 virus CNS central nervous system C3 C3 transferase CBA chicken β-actin CSPGs chondroitin sulfate proteoglycans DMEM Dulbecco’s Modified Eagle’s medium eC3 endogenous C3 EGFP enhanced GFP HEK Human embryonic kidney ICC immunocytochemistry ONC optic nerve crush SCIs spinal cord injuries KRAB Kruppel-associated box LV lentiviral vectors NGS normal goat serum NDS normal donkey serum PLL poly-L-lysine PCR polymerase chain reaction spC3 secretable/permeable C3 TATκ trans-acting activator of transcription Tet-on Tet-regulated hUbC Ubiquitin C WPRE woodchuck hepatitis virus posttranscriptional regulatory element Figure 1 Development of an endogenous C3GFP (eC3GFP) cassette A) Western blot of cell lysates from HEK293T cells transiently transfected with pEGFP-N2 and pEGPF-N2-C3 and stained with C3 antibodies. A band is visible in the pEGFP-N2-C3 transfected cells only. Molecular weight in KDa. B) Western-blot analysis of RhoA activity obtained from lysates of untreated NIH3T3 cells of cells pretreated as indicated. Active RhoA (aRhoA) fraction was isolated from freshly prepared NIH3T3 lysates using GST-Rhotekin-RBD beads following the manufacturer’s protocol (upper blot). The total RhoA (tRhoA) levels were determined in all lysates (lower blot). Xfect: transfected with. C) Micrographs of HEK293T cells infected with AAV2-eGFP or eC3GFP (green) as indicated and stained with C3 antibodies (red). Nuclei are stained with DAPI (blue). Scale bar: C) 10 μm. Figure 2 eC3GFP promotes crossing at PLL/CSPG interface A) Micrographs of neurons plated on coverslips showing the behavior of neurites at the PLL/CSPG interface. Rhodamine (red) was used to identify the CSPG drop. B) Graph of Average number of neurites crossing the PLL/CSPG interface + SEM. Scale bar A) 100 μm. *** p<0.05, Student t-test. Figure 3 Development of a secretable/permeable C3GFP (spC3GFP) cassette A) Western blot of cell lysates from HEK293T cells transiently transfected with pSecTag2 and pSecTag2- TATkC3GFP and stained with C3 antibodies. A band is visible in the pSecTag2-TATkC3GFP transfected cells only. Arrow head shows the molecular weight for pEGFP-N2-C3. Molecular weight in KDa. B) Micrographs of HEK293T cells infected with AAV2-spC3GFP (green) and stained with C3 antibodies (red). Nuclei are stained with DAPI (blue). C) Dot blots of concentrated media from HEK293T cells transiently transfected with pEGFP-N2 (eGFP), pEGPF-N2-C3 (eC3GFP) or pSecTag2-TATkC3GFP (spC3GFP) as indicated and stained with GFP antibodies. D) Micrographs of DsRed2-HEK293 cells growing alone (a–c) or in culture with spC3GFP expressing cells (d–f). DsRed2 fluorescence is shown in grey scale. Extensive processes (arrows) are seen in the cell in co-cultured. Scale bar: B) 10 μm, D) 100 μm, Figure 4 eC3GFP and spC3GFP promote process outgrowth on CSPG substrate A) Micrographs of neurons infected with AAV-eGFP, AAV-eC3GFP or AAV-spC3GFP plated on CSPG coated coverslips and stained with tubulin (red) for neurite length measurement analysis. B) Box and whisker plot showing length of primary and secondary processes with box quartiles (25–75%) and whiskers marking the 5–95% CI C) Box and whisker plot showing length of longest primary and secondary process with box quartiles and (25–75%) and whiskers marking the 5–95% CI. Scale bar: A) 100 μm. Statistics: * P<0.05, ** p<0.01, *** p<0.001, t-test. Figure 5 Conditional expression of eC3GFP and spC3GFP in rat striatum A) Micrographs of brain sections through the striatum of rats injected with LV-eGFP (a, d, g, j), LV-eC3GFP (b, e, h, k) or LV-spC3GFP (c, f, i, l) and subsequently given ad lib drinking sugar water unsupplemented (a–c) or supplemented (d–i) with doxycycline (Dox). Sections were stained for GFP (a–i) and DAB reaction product is seen in gray (a–f) or brown (g–i). At higher magnification (g–i), DAB reaction product is visible in neuronal cell bodies (g–i), neurite (g–i), and neuropil (i). GFP fluorescence can also be seen in unstained sections (j–l). DAPI (blue) is used as a nuclear marker. B) Micrographs of brain section of the striatum of an LV-spC3GFP injected rat immunostained for KRAB (red). GFP is visible in infected KRAB+ neurons (arrows) as well as non-infected KRAB− neurons (arrowheads). Panel a and b show composite images whereas panels c and d show spC3GFP and KRAB ICC respectively. C) Micrographs of brain section through the striatum of LV-spC3GFP injected rats receiving Dox or No Dox in their drinking water immunostained for GFAP or OX42 (red). Box and whisker plots showing intensity level of GFAP and OX2 with OX42 in the striatum of rats injected with LV-eGFP, LV-eC3GFP or LV-spC3GFP as indicated with box quartiles (25–75%) and whiskers marking the 5–95% CI. Scale bars: A) a–f: 0.3 mm, g–l: 20 μm; B) a: 20 μm, b–d: 10 μm. Author Contributions: Conception and design: Gutekunst, Gross Data collection: Gutekunst, Tung, McDougal Analysis and interpretation: Gutekunst, Tung, McDougal Manuscript Preparation: Gutekunst, Tung, McDougal, Gross This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Aktories K Braun U Rosener S Just I Hall A 1989 The rho gene product expressed in E. coli is a substrate of botulinum ADP-ribosyltransferase C3 Biochemical and biophysical research communications 158 209 213 2492192 Barde I Salmon P Trono D 2010 Production and titration of lentiviral vectors Curr Protoc Neurosci Chapter 4 Unit 4 21 Berry M Carlile J Hunter A 1996 Peripheral nerve explants grafted into the vitreous body of the eye promote the regeneration of retinal ganglion cell axons severed in the optic nerve J Neurocytol 25 147 170 8699196 Bertrand J Di Polo A McKerracher L 2007 Enhanced survival and regeneration of axotomized retinal neurons by repeated delivery of cell-permeable C3-like Rho antagonists Neurobiol Dis 25 65 72 17011202 Bertrand J Winton MJ Rodriguez-Hernandez N Campenot RB McKerracher L 2005 Application of Rho antagonist to neuronal cell bodies promotes neurite growth in compartmented cultures and regeneration of retinal ganglion cell axons in the optic nerve of adult rats J Neurosci 25 1113 1121 15689547 Di Polo A Aigner LJ Dunn RJ Bray GM Aguayo AJ 1998 Prolonged delivery of brain-derived neurotrophic factor by adenovirus-infected Muller cells temporarily rescues injured retinal ganglion cells Proceedings of the National Academy of Sciences of the United States of America 95 3978 3983 9520478 Diggle P Liang K-Y Zeger SL 1994 Analysis of longitudinal data Oxford, New York Clarendon Press; Oxford University Press Fawcett JW Asher RA 1999 The glial scar and central nervous system repair Brain Res Bull 49 377 391 10483914 Fawell S Seery J Daikh Y Moore C Chen LL Pepinsky B Barsoum J 1994 Tat-mediated delivery of heterologous proteins into cells Proceedings of the National Academy of Sciences of the United States of America 91 664 668 8290579 Fehlings MG Theodore N Harrop J Maurais G Kuntz C Shaffrey CI Kwon BK Chapman J Yee A Tighe A McKerracher L 2011 A phase I/IIa clinical trial of a recombinant Rho protein antagonist in acute spinal cord injury J Neurotrauma 28 787 796 21381984 Fischer D Petkova V Thanos S Benowitz LI 2004 Switching mature retinal ganglion cells to a robust growth state in vivo: gene expression and synergy with RhoA inactivation J Neurosci 24 8726 8740 15470139 Flinterman M Farzaneh F Habib N Malik F Gaken J Tavassoli M 2009 Delivery of therapeutic proteins as secretable TAT fusion products Molecular therapy: the journal of the American Society of Gene Therapy 17 334 342 19050698 Fournier AE Takizawa BT Strittmatter SM 2003 Rho kinase inhibition enhances axonal regeneration in the injured CNS J Neurosci 23 1416 1423 12598630 Fu Q Hue J Li S 2007 Nonsteroidal anti-inflammatory drugs promote axon regeneration via RhoA inhibition J Neurosci 27 4154 4164 17428993 Groner AC Tschopp P Challet L Dietrich JE Verp S Offner S Barde I Rodriguez I Hiiragi T Trono D 2012 The Kruppel-associated box repressor domain can induce reversible heterochromatization of a mouse locus in vivo J Biol Chem 287 25361 25369 22605343 Gross RE Mei Q Gutekunst CA Torre E 2007 The pivotal role of RhoA GTPase in the molecular signaling of axon growth inhibition after CNS injury and targeted therapeutic strategies Cell Transplant 16 245 262 17503736 Gutekunst CA Levey AI Heilman CJ Whaley WL Yi H Nash NR Rees HD Madden JJ Hersch SM 1995 Identification and localization of huntingtin in brain and human lymphoblastoid cell lines with anti-fusion protein antibodies Proceedings of the National Academy of Sciences of the United States of America 92 8710 8714 7568002 Gutekunst CA Stewart EN Franz CK English AW Gross RE 2012 PlexinA4 distribution in the adult rat spinal cord and dorsal root ganglia J Chem Neuroanat 44 1 13 22465808 Gutekunst CA Stewart EN Gross RE 2010 Immunohistochemical Distribution of PlexinA4 in the Adult Rat Central Nervous System Front Neuroanat 4 Gutekunst CA Torre ER Sheng Z Yi H Coleman SH Riedel IB Bujo H 2003 Stigmoid bodies contain type I receptor proteins SorLA/LR11 and sortilin: new perspectives on their function J Histochem Cytochem 51 841 852 12754295 Harvey AR Hellstrom M Rodger J 2009 Gene therapy and transplantation in the retinofugal pathway Prog Brain Res 175 151 161 19660654 Hellstrom M Ruitenberg MJ Pollett MA Ehlert EM Twisk J Verhaagen J Harvey AR 2009 Cellular tropism and transduction properties of seven adeno-associated viral vector serotypes in adult retina after intravitreal injection Gene Ther 16 521 532 19092858 Hoffmann A Hofmann F Just I Lehnardt S Hanisch UK Bruck W Kettenmann H Ahnert-Hilger G Holtje M 2008 Inhibition of Rho-dependent pathways by Clostridium botulinum C3 protein induces a proinflammatory profile in microglia Glia 56 1162 1175 18442097 Holtje M Hoffmann A Hofmann F Mucke C Grosse G Van Rooijen N Kettenmann H Just I Ahnert-Hilger G 2005 Role of Rho GTPase in astrocyte morphology and migratory response during in vitro wound healing Journal of neurochemistry 95 1237 1248 16150054 Lehmann M Fournier A Selles-Navarro I Dergham P Sebok A Leclerc N Tigyi G McKerracher L 1999 Inactivation of Rho signaling pathway promotes CNS axon regeneration J Neurosci 19 7537 7547 10460260 Liu X Hu Y Filla MS Gabelt BT Peters DM Brandt CR Kaufman PL 2005 The effect of C3 transgene expression on actin and cellular adhesions in cultured human trabecular meshwork cells and on outflow facility in organ cultured monkey eyes Mol Vis 11 1112 1121 16379023 Lois C Hong EJ Pease S Brown EJ Baltimore D 2002 Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors Science 295 868 872 11786607 Luo L 2000 Rho GTPases in neuronal morphogenesis Nat Rev Neurosci 1 173 180 11257905 McKerracher L Anderson KD 2013 Analysis of Recruitment and Outcomes in the Phase I/IIa Cethrin Clinical Trial for Acute Spinal Cord Injury J Neurotrauma McKerracher L Higuchi H 2006 Targeting Rho to stimulate repair after spinal cord injury J Neurotrauma 23 309 317 16629618 McKerracher L Rosen KM 2015 MAG, myelin and overcoming growth inhibition in the CNS Frontiers in molecular neuroscience 8 51 26441514 Niederost B Oertle T Fritsche J McKinney RA Bandtlow CE 2002 Nogo-A and myelin-associated glycoprotein mediate neurite growth inhibition by antagonistic regulation of RhoA and Rac1 J Neurosci 22 10368 10376 12451136 Sandvig A Berry M Barrett LB Butt A Logan A 2004 Myelin-, reactive glia-, and scar-derived CNS axon growth inhibitors: expression, receptor signaling, and correlation with axon regeneration Glia 46 225 251 15048847 Schneider CA Rasband WS Eliceiri KW 2012 NIH Image to ImageJ: 25 years of image analysis Nat Methods 9 671 675 22930834 Schwarze SR Ho A Vocero-Akbani A Dowdy SF 1999 In vivo protein transduction: delivery of a biologically active protein into the mouse Science 285 1569 1572 10477521 Silver J Miller JH 2004 Regeneration beyond the glial scar Nat Rev Neurosci 5 146 156 14735117 Sondhi D Johnson L Purpura K Monette S Souweidane MM Kaplitt MG Kosofsky B Yohay K Ballon D Dyke J Kaminksy SM Hackett NR Crystal RG 2012 Long-term expression and safety of administration of AAVrh.10hCLN2 to the brain of rats and nonhuman primates for the treatment of late infantile neuronal ceroid lipofuscinosis Hum Gene Ther Methods 23 324 335 23131032 Stankiewicz TR Linseman DA 2014 Rho family GTPases: key players in neuronal development, neuronal survival, and neurodegeneration Front Cell Neurosci 8 314 25339865 Szulc J Wiznerowicz M Sauvain MO Trono D Aebischer P 2006 A versatile tool for conditional gene expression and knockdown Nat Methods 3 109 116 16432520 Tan EY Law JW Wang CH Lee AY 2007 Development of a cell transducible RhoA inhibitor TAT-C3 transferase and its encapsulation in biocompatible microspheres to promote survival and enhance regeneration of severed neurons Pharm Res 24 2297 2308 17899323 Tunnemann G Martin RM Haupt S Patsch C Edenhofer F Cardoso MC 2006 Cargo-dependent mode of uptake and bioavailability of TAT-containing proteins and peptides in living cells FASEB journal: official publication of the Federation of American Societies for Experimental Biology 20 1775 1784 16940149 Vidal-Sanz M Bray GM Villegas-Perez MP Thanos S Aguayo AJ 1987 Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat J Neurosci 7 2894 2909 3625278 Villegas-Perez MP Vidal-Sanz M Rasminsky M Bray GM Aguayo AJ 1993 Rapid and protracted phases of retinal ganglion cell loss follow axotomy in the optic nerve of adult rats J Neurobiol 24 23 36 8419522 Vives E Brodin P Lebleu B 1997 A truncated HIV-1 Tat protein basic domain rapidly translocates through the plasma membrane and accumulates in the cell nucleus J Biol Chem 272 16010 16017 9188504 Wang H Katagiri Y McCann TE Unsworth E Goldsmith P Yu ZX Tan F Santiago L Mills EM Wang Y Symes AJ Geller HM 2008 Chondroitin-4-sulfation negatively regulates axonal guidance and growth J Cell Sci 121 3083 3091 18768934 Winton MJ Dubreuil CI Lasko D Leclerc N McKerracher L 2002 Characterization of new cell permeable C3-like proteins that inactivate Rho and stimulate neurite outgrowth on inhibitory substrates J Biol Chem 277 32820 32829 12091381 Zheng M Cierpicki T Momotani K Artamonov MV Derewenda U Bushweller JH Somlyo AV Derewenda ZS 2009 On the mechanism of autoinhibition of the RhoA-specific nucleotide exchange factor PDZRhoGEF BMC structural biology 9 36 19460155

PMC005xxxxxx/PMC5118174.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7605074 6087 Neuroscience Neuroscience Neuroscience 0306-4522 1873-7544 27746346 5118174 10.1016/j.neuroscience.2016.10.013 NIHMS827286 Article Elevated Levels of Calcitonin Gene-Related Peptide in Upper Spinal Cord Promotes Sensitization of Primary Trigeminal Nociceptive Neurons Cornelison Lauren E. MS Hawkins Jordan L. MS Durham Paul L. PhD Center for Biomedical and Life Sciences, Missouri State University, Springfield, MO USA Corresponding Author: Paul L. Durham, PhD, Distinguished Professor, Director, Center for Biomedical & Life Sciences, Missouri State University, Springfield, MO USA, Phone: 417-836-4869, Fax: 417-836-7602, pauldurham@missouristate.edu 5 11 2016 13 10 2016 17 12 2016 17 12 2017 339 491501 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Orofacial pain conditions including temporomandibular joint disorder and migraine are characterized by peripheral and central sensitization of trigeminal nociceptive neurons. Although calcitonin gene-related peptide (CGRP) is implicated in the development of central sensitization, the pathway by which elevated spinal cord CGRP levels promote peripheral sensitization of primary trigeminal nociceptive neurons is not well understood. The goal of this study was to investigate the role of CGRP in promoting bidirectional signaling within the trigeminal system to mediate sensitization of primary trigeminal ganglion nociceptive neurons. Adult male Sprague Dawley rats were injected in the upper spinal cord with CGRP or co-injected with the receptor antagonist CGRP8-37 or KT 5720, an inhibitor of protein kinase A (PKA). Nocifensive head withdrawal response to mechanical stimulation of trigeminal nerves was investigated using von Frey filaments. Expression of PKA, GFAP, and Iba1 in the spinal cord and P-ERK in the trigeminal ganglion was studied using immunohistochemistry. Some animals were co-injected intracisternally with CGRP and Fast Blue dye and trigeminal ganglion imaged using fluorescent microscopy. Intracisternal CGRP increased nocifensive responses to mechanical stimulation when compared to control levels. Co-injection of CGRP8-37 or KT 5720 with CGRP inhibited the nocifensive response. CGRP stimulated expression of PKA and GFAP in the spinal cord, and P-ERK in trigeminal ganglion neurons. Seven days post injection, Fast Blue was observed in trigeminal ganglion neurons and satellite glial cells. Our results demonstrate that elevated levels of CGRP in the upper spinal cord promote sensitization of primary trigeminal nociceptive neurons via a mechanism that involves activation of PKA centrally and P-ERK in trigeminal ganglion neurons. Our findings provide evidence of bidirectional signaling within the trigeminal system that can facilitate increased neuron-glia communication within the trigeminal ganglion associated with peripheral sensitization. calcitonin gene-related peptide nociception neuronal sensitization protein kinase A trigeminal ganglion glia Peripheral and central sensitization of trigeminal nociceptive neurons is associated with the pathology of prevalent and debilitating orofacial pain conditions including migraine and temporomandibular joint disorder (TMD). Migraine is a neurological disorder characterized by a severe headache, photophobia, phonophobia, and nausea that can persist up to 72 hours (Goadsby, 2005, Olesen et al., 2009). Similarly, TMD is a chronic painful condition that is characterized by persistent pain in the muscles and temporomandibular joint (TMJ) associated with mastication (Poveda Roda et al., 2007, Greenspan et al., 2011, Ohrbach et al., 2011, Furquim et al., 2015). The pathological pain and inflammation associated with migraine and TMD involves activation of trigeminal ganglion nerves, which provide sensory innervation of the head and face and relay nociceptive signals to the upper cervical spinal cord (Ballegaard et al., 2008, Bevilaqua Grossi et al., 2009, Dahan et al., 2015). Following peripheral activation of trigeminal nerves in response to tissue injury, the neuropeptide calcitonin gene-related peptide (CGRP) and other inflammatory mediators facilitate excitation of second order neurons and glial cells involved in the initiation and maintenance of central sensitization and persistent pain (Seybold, 2009, Sessle, 2011). Elevated CGRP levels in the spinal cord are implicated in the development of central sensitization by mediating changes in the expression of ion channels, receptors, and inflammatory genes in second order neurons and glial cells including astrocytes and microglia. Activation of astrocytes and microglia results in a prolonged inflammatory response that helps to sustain central sensitization and promotes a pathological pain state (Xie, 2008, Davies et al., 2010, Ikeda et al., 2012). CGRP is involved in the initiation and maintenance of central sensitization via activation of CGRP receptors that are localized on secondary neurons and glial cells within the spinal cord (Moreno et al., 2002, Marvizon et al., 2007, Lennerz et al., 2008). The CGRP receptor complex is comprised of three separate proteins: a G-protein-coupled receptor (GPCR), calcitonin receptor-like receptor (CLR), and an accessory protein, receptor activity-modifying protein 1 (RAMP1) that confers ligand-binding specificity (Benemei et al., 2007, Russell et al., 2014). Activation of CGRP receptors in neurons and glial cells causes an increase in intracellular levels of the secondary messenger cAMP that binds to and stimulates activation of protein kinase A (PKA). The signaling protein PKA induces expression of pro-inflammatory genes such as cytokines that are involved in sustaining a sensitized state of second order neurons (Staud, 2015). Elevated PKA levels in the cytosol are correlated with sensitization and activation of nociceptive neurons and glial cells via modulation of receptor expression and ion channel activity (Sun et al., 2004, Seybold, 2009). CGRP is also known to cause activation of the mitogen activated protein (MAP) kinases including p38, c-Jun kinase (JNK), and extracellular regulated kinase (ERK) in trigeminal neurons and glia (Thalakoti et al., 2007, Cady et al., 2011), that facilitate inflammation and nociception in the spinal cord (Ji et al., 2009). Similar to PKA, increased expression of these signaling proteins leads to a prolonged state of sensitization via modulation of ion channels, receptors, and transcription factors. In acute models of pain, peripheral sensitization, which results from the interaction of nociceptors with inflammatory substances released when tissue is damaged or inflamed (Sessle, 2011), has been shown to promote cellular changes that mediate central sensitization of second order neurons involved in pain transmission to the thalamus (Dodick and Silberstein, 2006, Sessle, 2011, Bernstein and Burstein, 2012). However, in prolonged pain states, central sensitization is maintained in the absence of evidence of peripheral tissue damage. In our study, we wanted to determine if elevated levels of CGRP in upper cervical spinal cord could promote bidirectional signaling and thus mediate peripheral sensitization of primary trigeminal neurons to mechanical stimulation. Findings from this study demonstrate that CGRP promotes sensitization of primary trigeminal nociceptive neurons via a mechanism involving PKA activation centrally, and is associated with increased levels of P-ERK and increased neuron-satellite glial cell coupling in the trigeminal ganglion. Furthermore, our results provide evidence of bidirectional signaling within the trigeminal system that may help explain how trigeminal nociceptive neurons become sensitized, as reported in chronic orofacial pain conditions, in the absence of any physical trauma. EXPERIMENTAL PROCEDURES Animals Sixty-seven Sprague-Dawley rats were used for this study. All animal studies were performed in accordance with the protocols approved by the Missouri State University Institutional Animal Care and Use Committee and were in agreement with guidelines set forth in the National Institutes of Health and the Animal Welfare Act of 2007. An effort was made to reduce the number of animals used in the study as well as to minimize suffering. Adult, male Sprague-Dawley rats (350-500 g) were obtained from Charles River Laboratories Inc. (Wilmington, MA) or purchased from Missouri State University (internal breeding colonies). All animals were housed in clean, plastic standard rat cages (VWR, West Chester, PA) in an animal holding room on a 12-hour light/dark cycle starting at 7 A.M. with ambient temperature maintained from 22-24 °C and access to food and water ad libitum. Animals were acclimated to the environment for a minimum of 1 week upon arrival prior to use. Reagents Stock solutions of CGRP or CGRP8-37 (American Peptide Company, Sunnyvale, CA) were prepared at a concentration of 1 mM in 0.9% saline solution (Fisher-Scientific, Fair Lawn, NJ). The PKA inhibitor KT 5720 (Tocris, Bristol, UK) was prepared at a stock concentration of 1 mM in DMSO (Sigma-Aldrich, St. Louis, MO). On the day of the experiment, an aliquot of 1 mM CGRP was thawed and diluted in 0.9% sterile saline to a concentration of 1 μM either alone or in solution with one of the two inhibitors. The inhibitors CGRP8-37 and KT 5720 were prepared in 0.9% saline solution with CGRP at concentrations of 5 μM and 500 nM, respectively. The retrograde labeling dye Fast Blue (Polysciences Inc., Warrington, PA) was diluted to a concentration of 4% in sterile phosphate buffered saline (PBS) containing 1 μM CGRP. Behavioral Testing All behavioral procedures were conducted between the hours of 7 A.M. and 11 A.M. Behavioral assessments were performed essentially as described in previously published studies in our laboratory using the Durham Animal Holder (Ugo Basile, Varese Lakes, Italy) (Garrett et al., 2012, Cady et al., 2014, Hawkins et al., 2015) on a total of 48 animals. Prior to testing, the rats were acclimated by guiding them into the holding device and secured in the holder for 5 minutes using a plastic blockade inserted behind the hindpaws. To minimize false responses during von Frey filament testing, a pipette tip was used to touch the animal’s head and face to acclimate the rats to having the cutaneous tissue over the masseter muscle touched with a filament. This was done for the three consecutive days prior to testing with von Frey filaments. During this acclimation period, if a rat appeared to be unwilling to go into the device or was continuously moving and shifting within the device, the animal was removed from the study. Following acclimations, nocifensive thresholds were determined in response to a series of calibrated von Frey filaments (North Coast Medical, Inc., Gilroy, CA; 60, 100, 180 grams) applied to the cutaneous tissue over the masseter muscle. A positive response was recorded when an animal visibly withdrew its head from a filament prior to it bending, while pressure was being applied. Each filament was applied 5 times on each side of the face, and the data are reported as the median number of responses obtained from 5 applications of each specific calibrated filament ± the interquartile range. The 100 g force was used for subsequent studies since the average number of positive head withdrawal responses to this force was less than 1 out of 5 for both right and left masseter muscles under basal conditions. Once baseline values were established, the animals were anesthetized by inhalation of 5% isoflurane. Animals were then injected intracisternally using a 26 ½ gauge needle (Becton Dickinson, Franklin Lakes, NJ) and a 50 μL Hamilton syringe (Hamilton Company, Reno, NV) at the midline between the occipital bone and the first cervical vertebrae (C1) with 20 μl of CGRP (1 μM) either alone, or co-injected with CGRP8-37 (5 μM) or KT 5720 (500 nM), a selective signaling inhibitor of PKA. Control animals were injected with saline alone or received no injection (naïve). Mechanical testing for nocifensive reactions at 2 h and days 1, 2, and 3 post injections was done using the same method as baseline values. Immunohistochemistry Twenty-three animals were used for immunohistochemical studies. To correlate behavioral responses to cellular changes in protein levels within the spinal cord, injections were performed as described above with CGRP alone or co-injected with KT 5720. The upper spinal cord (6 mm posterior to the obex) was removed at 2 h and at days 2 and 3 following injection and incubated in 4% paraformaldehyde at 4 °C for approximately 24 h. Tissues were then placed in 12.5% sucrose at 4 °C for approximately 1 h and then incubated in 25% sucrose for a minimum of 8 h. Following cryopreservation, tissues were stored at −20 °C. Tissues were embedded in Optimal Cutting Temperature compound (OCT; Sakura Finetek, Torrance, CA) and transverse sections 14 μm in thickness were taken between 4 and 5 mm caudal to the obex of the upper spinal cord, using a cryostat set at −24 °C. Sections from control and treated animals were placed on Superfrost Plus slides (Fisher Scientific, Pittsburg, PA) with the caudal side of the spinal cord in contact with the glass and stored at −20 °C. To determine changes in levels of the active form of ERK (P-ERK) in trigeminal ganglia, both ganglion were removed 2 h after intracisternal injection of CGRP and prepared for immunohistochemistry as described for the spinal cord tissue. Sections were blocked and permeabilized in a solution of 0.1% Triton X-100 in 5% donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA) for 20 min at room temperature. Following thorough rinsing in PBS, sections were incubated with primary antibodies for proteins of interest (Table 1) for either 3 h at room temperature or overnight in a humidified chamber at 4 °C. Sections were next incubated in solutions of secondary antibodies (Table 1) for one hour at room temperature, then mounted in Vectashield medium (H-1200) containing 4’,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA) to stain cell nuclei in preparation of viewing using fluorescent microscopy. As a control, some slides were incubated with only secondary antibodies to confirm specificity. Images were taken using a Zeiss Axiocam mRm camera mounted on a Zeiss Imager Z2 fluorescent microscope with an Apotome. Image acquisition was performed using Zeiss Zen 2012 software (Thornwood, NY). Retrograde labeling In initial studies, the fluorescent dye Fast Blue was injected into the intracisternal space between the occipital bone and C1 and its localization in the trigeminal ganglion observed using fluorescent microscopy. To directly investigate if CGRP could promote bidirectional signaling from terminals of primary sensory nerves fibers localized in the spinal cord, 50 μL of Fast Blue and 1 μM CGRP diluted in PBS were co-injected, while some animals were only injected with Fast Blue. Seven days later, the animals in both groups were euthanized via asphyxiation with carbon dioxide followed by decapitation. Trigeminal ganglia were removed and prepared for fluorescent microscopy as for tissue immunostaining. Briefly, longitudinal sections (14 μm) were prepared using a cryostat and mounted on Superfrost Plus slides with the caudal side in contact with the glass. Tissues were rehydrated with PBS and then mounted in Vectashield medium without DAPI to localize fluorescent staining from retrograde transport of the dye from the spinal cord. Statistical analysis Statistical analysis was performed essentially as described in previous published studies from our laboratory (Cady et al., 2014, Hawkins et al., 2015, Hawkins and Durham, 2016). For the mechanical stimulation studies, the data are reported as the median number of withdrawal responses ± the interquartile range to 100 g of force at each condition and time point. Subsequent analysis was then performed on data with n = 6 or greater for each experimental condition using a Friedman’s ANOVA to test for general statistical significance between time points for each group, followed by a Wilcoxon test to find changes within groups from basal, and a Kruskal Wallis followed by a Mann-Whitney U test for differences between groups at each time-point. Statistical significance was set at P < 0.05. For analysis of the immunohistological images of the spinal cord (n = at least 3 independent experiments per condition), relative levels of the proteins of interest were analyzed using NIH image J software. Fluorescent intensity was measured in ten rectangular regions in laminas I-III in the medullary horn. To normalize intensity measurements within each image, background intensity values were obtained from five non-overlapping regions in either the acellular area of the outer lamina as determined by DAPI, and average values subtracted from region of interest staining intensity values. All data are presented as mean fold-change from the average naïve value ± S.E.M. All immunohistochemical data were normally distributed as determined by a Shapiro-Wilk test. Analysis was performed for each separate time point using a one-way ANOVA with a Tukey’s post-hoc. For quantification of P-ERK expression in the trigeminal ganglion, the number of neuronal cells exhibiting nuclear localization of P-ERK was divided by the total number of visible neuronal nuclei as identified by NeuN staining in the overlapping V1/V2 region and distinct V3 region. Results are reported as the average percent ± S.E.M of neurons with P-ERK nuclear staining. Statistical significance was set at P < 0.05. RESULTS CGRP Mediates Nociceptive Response in Trigeminal Ganglion Neurons To evaluate the effect of increased CGRP levels in the upper spinal cord on primary nociceptor sensitivity, nocifensive responses were examined in response to mechanical stimulation of the cutaneous region over the masseter muscle. The number of withdrawals from the 100 g filament by animals was tested basally, 2 h post-injection, and on days 1, 2, and 3 post-injection (Fig. 1A). A Friedman test on all behavioral data indicated a significant difference between time-points for CGRP-injected animals (Χ2=15.921, n = 14, P ≤ 0.01), but not saline or naïve animals. Subsequent pairwise analysis showed no significant difference in basal responses between conditions. In contrast, significantly increased nocifensive responses were observed at 2 h (2.5 ± 3, Z = −2.9, n = 14, P ≤ 0.01) and on days 1 (2.5 ± 3, Z = −2.9, n = 14, P ≤ 0.05) and 2 (3.0 ± 4, Z = −3.0, n = 14, P ≤ 0.01) in animals injected with 1 μM CGRP when compared to basal values (0.0 ± 1). Nocifensive responses from animals injected with CGRP were also significantly different from saline controls at 2 h (0.5 ± 1.25, U = 32.5, nCGRP = 14, nsaline = 11, P ≤ 0.05) and on day 1 (1.0 ± 1.25, U = 35.0, n = 11, P ≤ 0.05) and day 2 post-injection (0.0 ± 2, U = 24.0, n = 11, P ≤ 0.01). Nocifensive responses in CGRP animals were also significantly higher than naïve control levels at 2 h (0.0 ± 1, U = 20.0, nCGRP = 14, nnaive = 10, P ≤ 0.01) and on day 1 (0.0 ± 1, U = 24.0, nCGRP = 14, nnaive = 10, P ≤ 0.01) and day 2 post-injection (0.0 ± 1, U = 27.0, nCGRP = 14, nnaive = 10, P ≤ 0.05). However, CGRP-induced responses were again similar to naïve and saline control values 3 days after injection. PKA Is Involved in CGRP-Mediated Nociception To investigate the signaling pathways involved in CGRP-mediated increases in nociception, animals were co-injected with a truncated version of CGRP (CGRP8-37) that is a known competitive CGRP receptor inhibitor or KT 5720, a selective inhibitor of the downstream signaling enzyme PKA. A Friedman test on behavioral data from animals receiving CGRP and CGRP8-37 showed no significance between time-points. Responses from animals that received CGRP8-37 in conjunction with CGRP were not significant from saline or naïve controls at 2 h, or days 1 and 2 post-injection. Nocifensive responses from these animals also failed to reach significance from basal levels at 2 h, or days 1 or 2 following injections. A Friedman test on behavioral data from animals receiving CGRP and KT 5720 indicated a significant difference between time-points (Χ2= 13.289, n = 7, P 0.01). Animals co-injected with CGRP and the PKA inhibitor KT 5720 exhibited significant increases in nocifensive responses to the 100 g filament at 2 h post-injection (2.0 ± 3.75) as compared to naïve control groups (U = 8.0, nnaive = 10, nKT5720 = 7, P ≤ 0.01), saline control groups (U)= 16.0, nsaline = 11, nKT5720 = 7, P ≤ 0.05), and basal readings (0.50 ± 1, Z = −2.2, n = 7, P ≤ 0.05) (Fig. 1B). In contrast, co-injected animals did not exhibit significantly increased nocifensive responses on day 1 (0.0 ± 2) or day 2 post-injection (0.0 ± 1) as compared to saline and naïve groups, as well as basal levels. Animals that received CGRP and KT 5720 reacted significantly less to the 100 g filament than those receiving CGRP alone at 1 d (U = 16.0, nCGRP = 14, nKT5720 = 7, P ≤ 0.5) and 2 d (U = 13.5, nCGRP = 14, nKT5720 = 7, P ≤ 0.05). Expression of PKA and GFAP but Not Iba1 Were Increased in Response to CGRP To identify the cell types influenced by CGRP, the expression of PKA, GFAP, and Iba1 were investigated using immunohistochemistry of upper spinal cord tissue. In naïve animals, low levels of the active form of PKA were detected in the upper spinal cord tissue (1.00 ± 0.06, n = 3; Fig. 2). General statistical significance was detected at 2 h (F(2, 6) = 14.4, P ≤ 0.01, η2 = 0.827) and 2 d (F(2, 6) = 17.3, P ≤ 0.05, η2 = 0.733) post-treatment. These changes were no longer statistically significant by 3 d post-treatment. The level of PKA staining was not significantly different from naïve controls in animals injected with sterile saline alone at any time point. At 2 h, CGRP-injected animals exhibited an increase in PKA immunostaining intensity (2.43 ± 0.11 fold, n = 3) in comparison to levels in naïve (P ≤ 0.01) and saline animals (P ≤ 0.01). PKA levels were still significantly elevated at 2 days post-injection (2.43 ± 0.15, n = 3) as compared to both naïve (P ≤ 0.05) and saline (P ≤ 0.05) animals. After 3 days, PKA levels in CGRP-injected animals were no longer significantly elevated when compared to naïve and saline control levels. Based on co-staining results, CGRP-mediated increase of PKA was observed in NeuN (neuronal nuclei marker) and GFAP stained cells, but not with Iba1. Thus, the stimulating effect of PKA involves changes in neurons and astrocytes. Levels of GFAP were compared between treatment conditions to evaluate changes in astrocyte activity in the spinal cord (Fig. 3A). Low levels of glial fibrillary acidic protein (GFAP) were observed in upper spinal cord tissues of naïve animals (1.00 ± 0.07, n = 3). Initial testing indicated a significant difference in GFAP levels in tissues taken 2 hours (F(2, 6) = 7.1, P ≤ 0.05, η2 = 0.702) 2 days (F(2, 6) = 17.3, P ≤ 0.01, η2 = 0.852) and 3 days post-injection (F(2, 6) = 9.3, P ≤ 0.05, η2 = 0.755). Post-hoc analysis showed that injection of sterile saline alone was not sufficient to elicit a change in expression of GFAP in astrocytes in the upper spinal cord at any time point as compared to naïve levels. Animals injected with CGRP were significantly different from naïve levels 2 h post-injection (1.88 ± 0.13, P ≤ 0.05), but did not show significant changes compared with saline controls at 2 h. In contrast, levels of GFAP in animals injected with CGRP were significantly higher from both naïve (P ≤ 0.01, n = 3) and saline controls (P ≤ 0.01, n = 3) 2 days after injection (3.03 ± 0.17). Three days post injection, spinal cord tissue from rats injected with CGRP still showed significantly different GFAP levels to naïve (1.93 ± 0.20, P ≤ 0.05) and saline (P ≤ 0.05) tissues. Relative levels of Iba1, a marker for active microglia, were assessed in the spinal cord to evaluate whether there were changes in microglia activity in response to CGRP (Fig. 3B). Statistically significant differences were not detected at any time point, indicating that neurons and astrocytes are primarily responsible for changes of sensitivity in the medullary horn. Expression of P-ERK in Trigeminal Ganglion Having shown that elevated levels of CGRP in the upper spinal cord led to an increased sensitivity of V3 trigeminal neurons that provide sensory innervation of the TMJ tissues, immunohistochemistry was utilized to study changes in P-ERK levels in primary neurons. While P-ERK immunostaining was observed primarily in the cytosol of neuronal cell bodies in naïve animals, the active form was more localized in the nuclei of trigeminal neurons in animals 2 h post CGRP injection (Fig. 4). P-ERK expression in neuronal nuclei was confirmed through tissue morphology and its colocalization with the NeuN expression. Nuclear localization of P-ERK expression was evaluated as a percentage of the total neuronal nuclei present in an image, normalized to number of visible neurons identified by DAPI staining. The staining pattern in the V3 region of naïve tissues (n = 3) showed an average of 65.3 ± 2.8 visible neuronal nuclei, with an average of 33.7 ± 0.72, or 51.9 ± 3.2%, of these expressing nuclear P-ERK. Animals injected with saline at 2 h post-injection (n = 3) contained an average of 64.3 ± 3.4 neurons present in pictures taken of the V3 region of the trigeminal ganglion, with 31.7 ± 2.3, or 49.2 ± 2.9%, of those neurons exhibiting nuclear P-ERK. Animals injected with CGRP at 2 h post-injection (n = 3) had an average of 57.7 ± 7.6 neurons present in pictures taken of the V3 region of the trigeminal ganglion. On average, 47.7 ± 5.6 of those neurons showed nuclear localization of P-ERK, making up 83.0 ± 1.4% of the total neurons per image. There was a statistically significant difference in nuclear P-ERK levels between groups (F(2, 6) = 39.8, P ≤ 0.001, η2 = 0.930). The difference in nuclear P-ERK between naïve and saline tissues was not significant. The percentage of nuclear P-ERK in the V3 region of CGRP tissues was significantly higher than that in naïve tissues (P 0.001) and saline tissues (P ≤ 0.001). In the V1/V2 region of the ganglion, naïve tissues (n = 3) had an average of 77.0 ± 14.6 neuronal nuclei visible in each tissue, with an average of 23.3 ± 5.25 of those neurons exhibiting nuclear P-ERK expression, comprising 30.2 ± 2.2% of the total neurons. Ganglia from animals that had been injected with saline alone (n = 3) had an average of 68.3 ± 7.3 neuronal nuclei visible in each tissue, with an average of 31.3 ± 4.8 neurons, or 45.9 ± 4.6% of total neurons exhibiting nuclear P-ERK expression. Tissues from animals injected with CGRP at 2 h post-injection (n = 3) were evaluated as having an average of 65.0 ± 7.7 neurons present in pictures taken of the V1V2 region of the trigeminal ganglion. On average, 52.3 ± 8.9 of those neurons showed nuclear localization of P-ERK, making up 79.1 ± 4.0% of the total neurons per tissue. A one-way ANOVA showed a statistically significant difference of nuclear P-ERK levels between groups (F(2, 6) = 34.3, P ≤ 0.001, η2 = 0.920). A Tukey post-hoc showed no significant difference in nuclear P-ERK between naïve and saline tissues. However, the percentage of nuclear P-ERK in the V1V2 region of CGRP tissues was significantly higher than that in naïve (P ≤ 0.001) or saline tissues (P ≤ 0.01). Evidence of Bidirectional Signaling within Trigeminal System The retrograde fluorescent tracer dye Fast Blue was observed in neuronal cell bodies located in the V3 region of the trigeminal ganglion seven days after dye injection into the upper spinal cord in unstimulated animals (Fig. 5). To test whether elevated CGRP levels in the spinal cord could promote increased neuron-glial signaling in the trigeminal ganglion, animals were co-injected with Fast Blue and 1 μM CGRP. Seven days post injection, the Fast Blue was observed in the cell body of trigeminal neurons and surrounding satellite glial cells throughout all three branches of the ganglion. In contrast, Fast Blue was not detected in Schwann cells, which are the other prevalent glial cell type present within the ganglion. DISCUSSION Elevated levels of CGRP are implicated in the development and maintenance of central sensitization and an enhanced pain state characterized by hyperalgesia and allodynia in diseases involving trigeminal nerves including TMD and migraine (Sun et al., 2003, Sun et al., 2004, Seybold, 2009, Sessle, 2011). However, the mechanisms by which elevated levels of CGRP in the upper spinal cord promote peripheral sensitization of primary nociceptive trigeminal neurons that provide sensory innervation to tissues in the head and face is not well understood. We found that administration of CGRP in the upper spinal cord resulted in a significant increase in the number of nocifensive head withdrawals in response to mechanical stimulation of the trigeminal nerve. This finding is in agreement with data from human studies in which both TMD and migraine patients report increased sensitivity to mechanical pressure applied to the head and face during an attack (Burstein et al., 2015, Dahan et al., 2015, Furquim et al., 2015). Under normal physiological conditions, the increase in pain sensitivity could provide a protective mechanism to minimize further damage to the tissue. As shown in our study, the enhanced nocifensive response was seen as early as 2 h post administration and this sensitized state was maintained for at least 48 h with resolution by 72 h post intracisternal CGRP injection. This finding is in agreement with the time course reported by migraine patients during an acute attack in which the severe pain and enhanced sensitivity is rarely sustained past 72 hours (Olesen et al., 2009). Our finding that elevated levels of CGRP within the upper spinal cord can lower the activation threshold of trigeminal primary sensory neurons to mechanical stimulation provide evidence to help explain the association of CGRP and the enhanced pain states reported by migraine and TMD patients. The physiological and cellular effects of CGRP are mediated via activation of the CGRP receptor, which is expressed on primary trigeminal neurons that synapse in the outer lamina of the spinal cord and the associated glial cells, astrocytes, and microglia (Wang et al., 2010, Hansen et al., 2016). To demonstrate the specificity of the CGRP-mediated response, we co-administered CGRP with the truncated CGRP molecule (CGRP8-37), which acts as a competitive inhibitor of the CGRP receptor (Chiba et al., 1989, Edvinsson et al., 2007). We found that CGRP8-37 could inhibit the stimulatory effect of CGRP on nociception at each of the time points. This finding provides evidence that blocking the CGRP receptor within the spinal cord is sufficient to suppress the initiation and maintenance of peripheral sensitization of trigeminal nociceptive neurons. Our result is in agreement with other studies that reported intrathecal administration of a CGRP receptor antagonist can block mechanically evoked nocifensive responses in tissues innervated by dorsal root ganglion sensory neurons (Sun et al., 2004, Adwanikar et al., 2007, Tzabazis et al., 2007, Hansen et al., 2016). Following binding of CGRP to its G-protein-coupled receptor, there is an increase in adenylate cyclase activity within those cells leading to an increase in the intracellular level of the secondary messenger cAMP (Brain and Cox, 2006). Elevated levels of cAMP then cause an increase in expression of the active form of the signaling kinase PKA via binding to an allosteric site in the protein. In contrast to blocking the CGRP receptor, we found that co-injection of the PKA inhibitor (KT 5720) with CGRP did not inhibit trigeminal sensitization to mechanical stimulation at the 2 h time point. However, blocking PKA activation did suppress mechanical sensitivity 24 and 48 h post CGRP injection. We can only speculate that the difference in temporal response may be due to CGRP eliciting an immediate increase in nuclear P-ERK in primary neurons, as shown in our study, independent of PKA activity in the central nervous system. The maintenance of the sensitization, however, is likely mediated by central PKA expression at least partially in astrocytes since CGRP stimulated GFAP expression, which is used as a biomarker of activated astrocytes. Taken together, our data support the notion that the initial increase in neuronal sensitivity in trigeminal nociceptors is due to cellular changes within the primary neurons while the more sustained sensitized state is attributable, at least in part, to activation of glial cells. Data from our behavioral studies provide evidence of the involvement of PKA activation in mediating the downstream stimulatory effects of CGRP. To determine if elevated CGRP levels in the spinal cord could stimulate increased expression of the active form of PKA in neurons and glial cells within the medullary horn, we used immunohistochemistry to directly study changes in PKA levels. The rationale for investigating PKA is further supported by evidence from other studies that PKA activation is associated with development of central sensitization (Levy and Strassman, 2002, Hu et al., 2003, Kohno et al., 2008). Based on colocalization of PKA with the proteins NeuN and GFAP, we found that PKA levels were significantly increased in cell bodies of second order neurons and astrocytes in response to CGRP at 2 hours post injection when compared to levels in naïve and saline treated animals. PKA levels remained significantly elevated at 48 and 72 h post injection. This finding is suggestive that, although a CGRP-mediated enhancement in nociceptive sensitivity had resolved by 72 h, the neurons and glia within the upper spinal cord retained a level of cellular sensitization. Our finding is in agreement with other studies that have shown that elevated levels of PKA within the lower spinal cord are involved in the initiation and maintenance of central sensitization (Aley and Levine, 1999, Hu et al., 2003, Hucho and Levine, 2007). The modulatory effects of PKA are thought to involve activation of pathways and transcription factors that regulate the expression and activity level of ion channels and receptors in nociceptive neurons and increase expression of pro-inflammatory molecules in both neurons and glial cells (Seybold, 2009). For example, elevated PKA levels in the spinal cord are implicated in the development of central sensitization by enhancing the activity of glutamate receptors that are expressed on second order nociceptive neurons (Aley and Levine, 1999, Hucho and Levine, 2007, Latremoliere and Woolf, 2009). Furthermore, activation of PKA intracellular signaling pathways have been shown to promote the initiation and prolonged state of sensitization and persistent pain via ion channel phosphorylation (Fitzgerald et al., 1999, Bhave et al., 2002, Han et al., 2005), and inducing pro-inflammatory cytokine genes containing CRE regulatory promoter sequences (Kawasaki et al., 2004). Further evidence for an important role of PKA in mediating nociception was provided by results demonstrating that blocking PKA signaling inhibits inflammation-induced hyperalgesic behaviors (Malmberg et al., 1997, Aley and Levine, 1999). In sum, our results provide evidence to further support the notion that PKA signaling plays a central role in mediating the stimulatory effects of CGRP, and thus is likely to be an important signaling pathway in promoting central sensitization associated with TMD and migraine. Sensitization of nociceptive neurons associated with the development of prolonged pain states is known to involve activation of spinal cord glial cells (Wieseler-Frank et al., 2004, Ren and Dubner, 2008, Gosselin et al., 2010). In support of this notion, we detected elevated immunoreactive levels of GFAP, which is a protein implicated in astrocyte activation, of CGRP injected animals. The observed increase in GFAP occurred at the 2 hour time point with levels greatest after 48 hours post injection, and remained significantly elevated even at 72 hours, a finding similar to our PKA results. In contrast, CGRP did not cause an increase in the expression of Iba1 in microglia when compared to control levels at any of the time points. Astrocytes can promote and sustain sensitization of peripheral and central neurons through the release of cytokines and other inflammatory molecules by increasing neuron-glial cell interactions in the spinal cord (Miller et al., 2009). Based on our findings, the stimulatory effects in response to intracisternal administration of CGRP appears to be mediated primarily by astrocytes with minimal contribution from microglial cells. To investigate a possible mechanism by which elevated CGRP levels in the spinal cord could lead to our observed increase in nocifensive head withdrawal response mediated by primary trigeminal nociceptive neurons, we determined changes in the level of the MAP kinase ERK in trigeminal neurons. Intracisternal CGRP caused a significant large increase in the nuclear localization of active, phosphorylated form of ERK (P-ERK) in the cell bodies of trigeminal neurons throughout the entire ganglion 2 hours post injection. In contrast in naïve control ganglion, P-ERK was mostly localized in the cytosol of neurons. Elevated P-ERK levels are reported to mediate a sensitized state of primary nociceptive neurons via modulating expression and activation levels of ion channels and membrane receptors (Cheng and Ji, 2008, Takeda et al., 2009). The importance of MAP kinases in the development of peripheral sensitization and an enhanced level of nociceptor sensitivity is supported by findings that selective inhibition of MAP kinase activity can suppress nociceptive cellular events (Milligan et al., 2003, Tsuda et al., 2004, Ji et al., 2009). Results from our study provide evidence that CGRP induces cellular changes of trigeminal ganglion neurons that correlate with the development of peripheral sensitization of primary nociceptive neurons. These data are in agreement with a previous study from our laboratory that demonstrated that nicotine, which promotes central sensitization by promoting an increase in neuron-glial signaling and cytokine production within the upper spinal cord, caused an significant elevation in P-ERK levels in primary trigeminal nociceptive neurons (Hawkins et al., 2015). Taken together, our findings provide evidence to support the notion of bidirectional signaling within the trigeminal system such that central sensitization can induce changes in trigeminal nociceptive neurons. To directly demonstrate that CGRP can promote retrograde signal transduction from the spinal cord to neuronal cell bodies located in the trigeminal ganglion, the retrograde dye Fast Blue was co-injected with CGRP in the upper spinal cord. Fast Blue is a fluorescent dye most commonly used as a retrograde neuronal tracer since it has been shown to be effectively transported retrogradely over long distances in various animal models (Casatti et al., 1999, Bossowska et al., 2009, Ivanusic, 2009). While we detected Fast Blue in the cell bodies of neurons throughout all regions of the trigeminal ganglion seven days post intracisternal injection in unstimulated animals, we observed the dye in both neuronal cell bodies and associated satellite glial cells in response to intracisternal CGRP. To our knowledge, these data for the first time provide direct evidence of bidirectional signaling from the cerebrospinal fluid to neuronal cell bodies within the trigeminal ganglion and coupling to satellite glial cells. The movement of the dye from neuronal cell bodies to the satellite glial cells likely involved the formation of gap junctions between these two cells. This type of neuron-glia coupling within the trigeminal ganglion has been observed following peripheral inflammation or in response to inflammation and nerve injury (Cherkas et al., 2004, Vit et al., 2008, Durham and Garrett, 2010, Villa et al., 2010). Importantly, increased signaling between neurons and glia with the ganglion is associated with development and maintenance of peripheral sensitization of nociceptive neurons. Our results support the idea that elevated levels of CGRP within the spinal cord can facilitate bidirectional signaling and sensitization of primary trigeminal neurons by mediating increased neuron-glial cell coupling in the trigeminal ganglion. In summary, findings from this study demonstrate that CGRP promotes peripheral sensitization of primary trigeminal nociceptive neurons to mechanical stimulation via a mechanism involving CGRP induction of PKA activity in neurons and glia and upregulation of GFAP in astrocytes in the upper spinal cord. Elevated levels of CGRP increased neuronal expression of P-ERK and promoted neuron-satellite glial cell coupling in the trigeminal ganglion. Thus, our results provide evidence to support the notion that CGRP-mediated central sensitization leads to an increase in trigeminal nociceptor sensitivity. Furthermore, we speculate that central to peripheral signaling as observed in our study may help to explain how peripheral nociceptors become sensitized, as reported in chronic orofacial pain conditions, even in the absence of any physical trauma or signs of inflammation. ACKNOWLEDGEMENTS We would like to thank Jennifer Cashler and Angela Goerndt for their assistance with the animals. This work was supported by the National Institutes of Health [DE024629]. Abbreviations CGRP calcitonin gene-related peptide TMJ temporomandibular joint PKA protein kinase A PBS phosphate buffered saline P-ERK phosphorylated extracellular signal-regulated kinase GFAP glial fibrillary acidic protein SEM standard error of the mean Iba1 ionized calcium-binding adapter molecule 1 Figure 1 A. Intracisternal injection of CGRP in upper cervical spinal cord increased nociceptive responses to mechanical stimulation of trigeminal neurons. The median number of nocifensive head withdrawals to the 100 g filament in naïve animals compared to animals basally and at 2 h, 1 day, 2 days, or 3 days post intracisternal injection of saline or CGRP is shown. B. The median number of nocifensive withdrawal responses to the 100 g filament was decreased in a time-dependent manner by inhibiting CGRP or PKA activity. Animals were injected intracisternally with CGRP or co-injected with CGRP and the truncated CGRP receptor antagonist peptide CGRP8-37 (CGRP + 8-37) or the selective PKA inhibitor KT 5720 (CGRP + KT 5720). Figure 2 CGRP injection increased expression of PKA in upper spinal cord compared to naïve and saline control. Representative images of sections from the upper spinal cord obtained from naïve (left), and saline treated (center) or CGRP treated (right) animals 2 days post injection are shown. All cell nuclei are identified by DAPI staining (top panel). The same sections were also stained for PKA or co-stained for PKA and the astrocyte biomarker GFAP. Enlarged images of the region of the medullary horn are shown. Scale bars = 200 μm. Figure 3 Intracisternal injection of CGRP increased expression of GFAP and transiently elevated Iba1 levels. Representative images of sections from the medullary horn of upper spinal cords obtained from naïve (left), and saline (center) or CGRP treated (right) animals 2 days post CGRP administration are shown. All cell nuclei identified by staining with DAPI are shown in the top panels, while immunostaining of the same tissue sections for GFAP (A) or Iba1 (B) are seen in the lower panels. Scale bars = 200 μm. Figure 4 Intracisternal CGRP injection increased expression of P-ERK in trigeminal ganglion neurons. Representative images of sections from the V1/V2 region of trigeminal ganglia obtained from naïve and CGRP treated animals are shown. All cell nuclei are identified by the nuclear dye DAPI (left panel), while the same tissue sections that were positive for P-ERK are seen in the second panel. Enlarged images of the region of the ganglion containing numerous neuronal cell bodies (white box) stained for neuronal protein NeuN (third panel) and the same region co-stained for P-ERK (far right panel) are shown. Scale bars = 100 μm. Figure 5 Evidence of bidirectional signaling within the trigeminal system from the upper spinal cord to trigeminal ganglion cells and increased neuron-glia coupling in response to CGRP. The fluorescent dye Fast Blue was localized primarily in the cell body of neurons 7 days after dye injection in unstimulated animals but the dye was seen in both neuronal cell bodies and satellite glial cells in animals co-injected with dye and CGRP. A white asterisk was used to indicate the cell body of several neurons, while arrows identify satellite glial cells containing the dye. Scale bars = 100 μm. Table 1 Antibodies and Incubation Conditions Used for Immunohistochemistry. Protein Dilution Company Incubation Time Incubation Temperature GFAP 1:5,000 Abcam 3 hours 20-22 °C Iba1 1:500 Abcam 3 hours 20-22 °C NeuN 1:1,000 Millipore 3 hours 20-22 °C P-ERK 1:500 Bioworld Overnight 20-22 °C PKA 1:500 Abcam 3 hours 4 °C Alexa 488 1:200 Life Technologies 1 hour 20-22 °C Alexa 567 1:200 Life Technologies 1 hour 20-22 °C Alexa 647 1:200 Life Technologies 1 hour 20-22 °C Intrathecal CGRP promotes sensitization of primary trigeminal nociceptive neurons Stimulatory effects of CGRP are mediated by PKA and involve astrocyte activation CGRP-dependent behavioral changes associate with increased P-ERK levels in ganglion Elevated CGRP levels in spinal cord promote neuron-glia communication in ganglion Our results provide direct evidence of bidirectional signaling within trigeminal system This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Conflict of interest: The authors of this paper do not have any conflicts of interest to report. REFERENCES Adwanikar H Ji G Li W Doods H Willis WD Neugebauer V Spinal CGRP1 receptors contribute to supraspinally organized pain behavior and pain-related sensitization of amygdala neurons Pain 2007 132 53 66 17335972 Aley KO Levine JD Role of protein kinase A in the maintenance of inflammatory pain J Neurosci 1999 19 2181 2186 10066271 Ballegaard V Thede-Schmidt-Hansen P Svensson P Jensen R Are headache and temporomandibular disorders related? A blinded study Cephalalgia: an international journal of headache 2008 28 832 841 18498400 Benemei S Nicoletti P Capone JA Geppetti P Pain pharmacology in migraine: focus on CGRP and CGRP receptors Neurol Sci 2007 28 Suppl 2 S89 93 17508187 Bernstein C Burstein R Sensitization of the trigeminovascular pathway: perspective and implications to migraine pathophysiology Journal of clinical neurology 2012 8 89 99 22787491 Bevilaqua Grossi D Lipton RB Bigal ME Temporomandibular disorders and migraine chronification Curr Pain Headache Rep 2009 13 314 318 19586596 Bhave G Zhu W Wang H Brasier DJ Oxford GS Gereau RWt cAMP-dependent protein kinase regulates desensitization of the capsaicin receptor (VR1) by direct phosphorylation Neuron 2002 35 721 731 12194871 Bossowska A Crayton R Radziszewski P Kmiec Z Majewski M Distribution and neurochemical characterization of sensory dorsal root ganglia neurons supplying porcine urinary bladder J Physiol Pharmacol 2009 60 Suppl 4 77 81 Brain SD Cox HM Neuropeptides and their receptors: innovative science providing novel therapeutic targets Br J Pharmacol 2006 147 Suppl 1 S202 211 16402106 Burstein R Noseda R Borsook D Migraine: multiple processes, complex pathophysiology J Neurosci 2015 35 6619 6629 25926442 Cady RJ Denson JE Sullivan LQ Durham PL Dual orexin receptor antagonist 12 inhibits expression of proteins in neurons and glia implicated in peripheral and central sensitization Neuroscience 2014 269 79 92 24685439 Cady RJ Glenn JR Smith KM Durham PL Calcitonin gene-related peptide promotes cellular changes in trigeminal neurons and glia implicated in peripheral and central sensitization Mol Pain 2011 7 94 22145886 Casatti CA Frigo L Bauer JA Origin of sensory and autonomic innervation of the rat temporomandibular joint: a retrograde axonal tracing study with the fluorescent dye fast blue J Dent Res 1999 78 776 783 10096453 Cheng JK Ji RR Intracellular signaling in primary sensory neurons and persistent pain Neurochem Res 2008 33 1970 1978 18427980 Cherkas PS Huang TY Pannicke T Tal M Reichenbach A Hanani M The effects of axotomy on neurons and satellite glial cells in mouse trigeminal ganglion Pain 2004 110 290 298 15275779 Chiba T Yamaguchi A Yamatani T Nakamura A Morishita T Inui T Fukase M Noda T Fujita T Calcitonin gene-related peptide receptor antagonist human CGRP-(8-37) Am J Physiol 1989 256 E331 335 2537579 Dahan H Shir Y Velly A Allison P Specific and number of comorbidities are associated with increased levels of temporomandibular pain intensity and duration J Headache Pain 2015 16 528 26002637 Davies AJ Kim YH Oh SB Painful Neuron-Microglia Interactions in the Trigeminal Sensory System TOPAINJ 2010 14 28 Dodick D Silberstein S Central sensitization theory of migraine: clinical implications Headache 2006 46 Suppl 4 S182 191 17078850 Durham PL Garrett FG Emerging importance of neuron-satellite glia interactions within trigeminal ganglia in craniofacial pain TOPAINJ 2010 3 3 13 Edvinsson L Nilsson E Jansen-Olesen I Inhibitory effect of BIBN4096BS, CGRP(8-37), a CGRP antibody and an RNA-Spiegelmer on CGRP induced vasodilatation in the perfused and non-perfused rat middle cerebral artery Br J Pharmacol 2007 150 633 640 17245362 Fitzgerald EM Okuse K Wood JN Dolphin AC Moss SJ cAMP-dependent phosphorylation of the tetrodotoxin-resistant voltage-dependent sodium channel SNS J Physiol 1999 516 Pt 2 433 446 10087343 Furquim BD Flamengui LM Conti PC TMD and chronic pain: a current view Dental Press J Orthod 2015 20 127 133 25741834 Garrett FG Hawkins JL Overmyer AE Hayden JB Durham PL Validation of a novel rat-holding device for studying heat- and mechanical-evoked trigeminal nocifensive behavioral responses J Orofac Pain 2012 26 337 344 23110274 Goadsby PJ Migraine pathophysiology Headache 2005 45 Suppl 1 S14 24 15833086 Gosselin RD Suter MR Ji RR Decosterd I Glial cells and chronic pain Neuroscientist 2010 16 519 531 20581331 Greenspan JD Slade GD Bair E Dubner R Fillingim RB Ohrbach R Knott C Mulkey F Rothwell R Maixner W Pain sensitivity risk factors for chronic TMD: descriptive data and empirically identified domains from the OPPERA case control study J Pain 2011 12 T61 74 22074753 Han JS Li W Neugebauer V Critical role of calcitonin gene-related peptide 1 receptors in the amygdala in synaptic plasticity and pain behavior J Neurosci 2005 25 10717 10728 16291945 Hansen RR Vacca V Pitcher T Clark AK Malcangio M Role of extracellular calcitonin gene-related peptide in spinal cord mechanisms of cancer-induced bone pain Pain 2016 157 666 676 26574822 Hawkins JL Denson JE Miley DR Durham PL Nicotine stimulates expression of proteins implicated in peripheral and central sensitization Neuroscience 2015 290 115 125 25637801 Hawkins JL Durham PL Prolonged Jaw Opening Promotes Nociception and Enhanced Cytokine Expression Journal of oral & facial pain and headache 2016 30 34 41 26817031 Hu HJ Glauner KS Gereau RWt ERK integrates PKA and PKC signaling in superficial dorsal horn neurons. I. Modulation of A-type K+ currents J Neurophysiol 2003 90 1671 1679 12750419 Hucho T Levine JD Signaling pathways in sensitization: toward a nociceptor cell biology Neuron 2007 55 365 376 17678851 Ikeda H Kiritoshi T Murase K Contribution of microglia and astrocytes to the central sensitization, inflammatory and neuropathic pain in the juvenile rat Mol Pain 2012 8 43 22703840 Ivanusic JJ Size, neurochemistry, and segmental distribution of sensory neurons innervating the rat tibia J Comp Neurol 2009 517 276 283 19757492 Ji RR Gereau RWt Malcangio M Strichartz GR MAP kinase and pain Brain Res Rev 2009 60 135 148 19150373 Kawasaki Y Kohno T Zhuang ZY Brenner GJ Wang H Van Der Meer C Befort K Woolf CJ Ji RR Ionotropic and metabotropic receptors, protein kinase A, protein kinase C, and Src contribute to C-fiber-induced ERK activation and cAMP response element-binding protein phosphorylation in dorsal horn neurons, leading to central sensitization J Neurosci 2004 24 8310 8321 15385614 Kohno T Wang H Amaya F Brenner GJ Cheng JK Ji RR Woolf CJ Bradykinin enhances AMPA and NMDA receptor activity in spinal cord dorsal horn neurons by activating multiple kinases to produce pain hypersensitivity J Neurosci 2008 28 4533 4540 18434532 Latremoliere A Woolf CJ Central sensitization: a generator of pain hypersensitivity by central neural plasticity J Pain 2009 10 895 926 19712899 Lennerz JK Ruhle V Ceppa EP Neuhuber WL Bunnett NW Grady EF Messlinger K Calcitonin receptor-like receptor (CLR), receptor activity-modifying protein 1 (RAMP1), and calcitonin gene-related peptide (CGRP) immunoreactivity in the rat trigeminovascular system: differences between peripheral and central CGRP receptor distribution J Comp Neurol 2008 507 1277 1299 18186028 Levy D Strassman AM Distinct sensitizing effects of the cAMP-PKA second messenger cascade on rat dural mechanonociceptors J Physiol 2002 538 483 493 11790814 Malmberg AB Brandon EP Idzerda RL Liu H McKnight GS Basbaum AI Diminished inflammation and nociceptive pain with preservation of neuropathic pain in mice with a targeted mutation of the type I regulatory subunit of cAMP-dependent protein kinase J Neurosci 1997 17 7462 7470 9295392 Marvizon JC Perez OA Song B Chen W Bunnett NW Grady EF Todd AJ Calcitonin receptor-like receptor and receptor activity modifying protein 1 in the rat dorsal horn: localization in glutamatergic presynaptic terminals containing opioids and adrenergic alpha2C receptors Neuroscience 2007 148 250 265 17614212 Miller RJ Jung H Bhangoo SK White FA Cytokine and chemokine regulation of sensory neuron function Handbook of experimental pharmacology 2009 417 449 19655114 Milligan ED Twining C Chacur M Biedenkapp J O’Connor K Poole S Tracey K Martin D Maier SF Watkins LR Spinal glia and proinflammatory cytokines mediate mirror-image neuropathic pain in rats J Neurosci 2003 23 1026 1040 12574433 Moreno MJ Terron JA Stanimirovic DB Doods H Hamel E Characterization of calcitonin gene-related peptide (CGRP) receptors and their receptor-activity-modifying proteins (RAMPs) in human brain microvascular and astroglial cells in culture Neuropharmacology 2002 42 270 280 11804624 Ohrbach R Fillingim RB Mulkey F Gonzalez Y Gordon S Gremillion H Lim PF Ribeiro-Dasilva M Greenspan JD Knott C Maixner W Slade G Clinical findings and pain symptoms as potential risk factors for chronic TMD: descriptive data and empirically identified domains from the OPPERA case-control study J Pain 2011 12 T27 45 22074750 Olesen J Burstein R Ashina M Tfelt-Hansen P Origin of pain in migraine: evidence for peripheral sensitisation The Lancet Neurology 2009 8 679 690 19539239 Poveda Roda R Bagan JV Diaz Fernandez JM Hernandez Bazan S Jimenez Soriano Y Review of temporomandibular joint pathology. Part I: classification, epidemiology and risk factors Medicina oral, patologia oral y cirugia bucal 2007 12 E292 298 Ren K Dubner R Neuron-glia crosstalk gets serious: role in pain hypersensitivity Curr Opin Anaesthesiol 2008 21 570 579 18784481 Russell FA King R Smillie SJ Kodji X Brain SD Calcitonin gene-related peptide: physiology and pathophysiology Physiol Rev 2014 94 1099 1142 25287861 Sessle BJ Peripheral and central mechanisms of orofacial inflammatory pain Int Rev Neurobiol 2011 97 179 206 21708311 Seybold VS The role of peptides in central sensitization Handbook of experimental pharmacology 2009 451 491 19655115 Staud R Cytokine and immune system abnormalities in fibromyalgia and other central sensitivity syndromes Curr Rheumatol Rev 2015 11 109 115 26088214 Sun RQ Lawand NB Willis WD The role of calcitonin gene-related peptide (CGRP) in the generation and maintenance of mechanical allodynia and hyperalgesia in rats after intradermal injection of capsaicin Pain 2003 104 201 208 12855330 Sun RQ Tu YJ Lawand NB Yan JY Lin Q Willis WD Calcitonin gene-related peptide receptor activation produces PKA- and PKC-dependent mechanical hyperalgesia and central sensitization J Neurophysiol 2004 92 2859 2866 15486424 Takeda M Takahashi M Matsumoto S Contribution of the activation of satellite glia in sensory ganglia to pathological pain Neurosci Biobehav Rev 2009 33 784 792 19167424 Thalakoti S Patil VV Damodaram S Vause CV Langford LE Freeman SE Durham PL Neuron-glia signaling in trigeminal ganglion: implications for migraine pathology Headache 2007 47 1008 1023 17635592 Tsuda M Mizokoshi A Shigemoto-Mogami Y Koizumi S Inoue K Activation of p38 mitogen-activated protein kinase in spinal hyperactive microglia contributes to pain hypersensitivity following peripheral nerve injury Glia 2004 45 89 95 14648549 Tzabazis AZ Pirc G Votta-Velis E Wilson SP Laurito CE Yeomans DC Antihyperalgesic effect of a recombinant herpes virus encoding antisense for calcitonin gene-related peptide Anesthesiology 2007 106 1196 1203 17525595 Villa G Ceruti S Zanardelli M Magni G Jasmin L Ohara PT Abbracchio MP Temporomandibular joint inflammation activates glial and immune cells in both the trigeminal ganglia and in the spinal trigeminal nucleus Mol Pain 2010 6 89 21143950 Vit JP Ohara PT Bhargava A Kelley K Jasmin L Silencing the Kir4.1 potassium channel subunit in satellite glial cells of the rat trigeminal ganglion results in pain-like behavior in the absence of nerve injury J Neurosci 2008 28 4161 4171 18417695 Wang Z Ma W Chabot JG Quirion R Calcitonin gene-related peptide as a regulator of neuronal CaMKII-CREB, microglial p38-NFkappaB and astroglial ERK-Stat1/3 cascades mediating the development of tolerance to morphine-induced analgesia Pain 2010 151 194 205 20691540 Wieseler-Frank J Maier SF Watkins LR Glial activation and pathological pain Neurochem Int 2004 45 389 395 15145553 Xie YF Glial involvement in trigeminal central sensitization Acta Pharmacol Sin 2008 29 641 645 18501109

PMC005xxxxxx/PMC5118180.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7605074 6087 Neuroscience Neuroscience Neuroscience 0306-4522 1873-7544 27717807 5118180 10.1016/j.neuroscience.2016.09.047 NIHMS820879 Article Sex differences in astrocyte and microglia responses immediately following middle cerebral artery occlusion in adult mice Morrison Helena W. PhD ab Filosa Jessica A. PhD jfilosa@augusta.edu a a Augusta University, 1120 15th St, Augusta, GA 30912 b Present address. University of Arizona, 1305 N. Martin Ave, P.O. Box 210203, Tucson, AZ 85721, hmorriso@email.arizona.edu 16 10 2016 4 10 2016 17 12 2016 17 12 2017 339 8599 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Epidemiological studies report that infarct size is decreased and stroke outcomes are improved in young females when compared to males. However, mechanistic insight is lacking. We posit that sex-specific differences in glial cell functions occurring immediately after ischemic stroke are a source of dichotomous outcomes. In this study we assessed astrocyte Ca2+ dynamics, aquaporin 4 (AQP4) polarity, S100β expression pattern, as well as, microglia morphology and phagocytic marker CD11b in male and female mice following 60 minutes of middle cerebral artery (MCA) occlusion. We reveal sex differences in the frequency of intracellular astrocyte Ca2+ elevations (F(1,86)=8.19, P=0.005) and microglia volume (F(1,40)=12.47, P=0.009) immediately following MCA occlusion in acute brain slices. Measured in fixed tissue, AQP4 polarity was disrupted (F(5,86)=3.30, P=0.009) and the area of non-S100β immunoreactivity increased in ipsilateral brain regions after 60 min of MCA occlusion (F(5,86)=4.72, P=0.007). However, astrocyte changes were robust in male mice when compared to females. Additional sex differences were discovered regarding microglia phagocytic receptor CD11b. In sham mice, constitutively high CD11b immunofluorescence was observed in females when compared to males (P=0.03). When compared to sham, only male mice exhibited an increase in CD11b immunoreactivity after MCA occlusion (P=0.006). We posit that a sex difference in the presence of constitutive CD11b has a role in determining male and female microglia phagocytic responses to ischemia. Taken together, these findings are critical to understanding potential sex differences in glial physiology as well as stroke pathobiology which are foundational for the development of future sex-specific stroke therapies. ischemic stroke calcium aquaporin 4 S100β microglia morphology CD11b The age-adjusted stroke mortality rate, which has been in decline since 2010, has been attributed to improved control of modifiable stroke risk factors such as hypertension, smoking, and dyslipidemia (Hoyert DL, 2012; Mozaffarian et al., 2015; Murphy et al., 2013). Yet In comparison, our understanding of a non-modifiable factors, such as sex differences, is lacking and therefore sex specific treatments are absent among stroke therapies. Emerging evidence taken from both epidemiological and pre-clinical stroke research supports the compelling assertion that sex has a dichotomous role in brain injury and functional outcomes after ischemic stroke (Go et al., 2014; Miller et al., 2011; Reeves et al., 2008). However, the diverging mechanisms of protection versus injury, which underlie this sexual dimorphism, have yet to be fully described and are not well understood (Banerjee et al., 2013; Ritzel et al., 2013). Brain ischemia, caused by stroke, results in immediate neuronal death and the inevitable release of cell contents which then further disrupts finely tuned neuronal functions and promotes neuroinflammation (Dirnagl et al., 1999; Dirnagl, 2012; Gelderblom et al., 2009). Structurally, astrocytes are well positioned to intervene because of their intimate association with other glial cells, neurons, and blood vessels (Filosa et al., 2015). Functionally, astrocytes have a vital role in maintaining, mediating and restoring neuronal function during physiologic and pathologic conditions (Araque et al., 1999; del Zoppo, 2010; Nedergaard et al., 2010). In this role, astrocytes are key in maintaining K+ homeostasis (Kofuji and Newman, 2004), removal of excess glutamate (Cheung et al., 2015), local blood flow regulation (Attwell et al., 2010; Kim et al., 2015), synaptogenesis and synaptic maintenance (Franke et al., 2012), among other functions. On the other hand, astrocyte dysfunction after ischemia may exacerbate brain injury via aberrant astrocyte Ca2+ signaling which can result in glutamate excitotoxicity (Nedergaard et al., 2010) or through the release of astrocyte-derived proteins and small molecular messengers such as ATP, TNFα, and S100β, which further increases neuroinflammation (Agulhon et al., 2012; Liu et al., 2011; Pascual et al., 2012). In addition, astrocyte aquaporin 4 (AQP4) is a water channel and key player in the development of brain edema. AQP4 is primarily localized to astrocyte endfeet rather than non-endfeet processes and this polarized location has an important role in brain fluid clearance (Gaberel et al., 2014; Iliff et al., 2013). AQP4 polarity is disrupted in the days following traumatic brain injury (Kress et al., 2014; Plog et al., 2015) and transient focal ischemia (Fukuda and Badaut, 2012; Vella et al., 2015) which may contribute to brain edema after injury. Importantly, brain edema is accompanied by increased astrocyte intracellular Ca2+ (Thrane et al., 2011). However, it is yet unknown if changes in astrocyte Ca2+ dynamics, the expression pattern of S100β and AQP4 polarity is altered immediately following middle cerebral artery (MCA) occlusion in adult male and female mice. Adding to the complexity of cell-to-cell interactions, microglia are phagocytes and, in conjunction with astrocytes, contribute to the inflammatory milieu after ischemic stroke. Microglia cells are highly sensitive to altered brain physiology because of their constant monitoring and immediate inflammatory response to pathophysiology (i.e. ischemia). These important functions are made possible by their ramified morphology (Morrison and Filosa, 2013), dynamic process movement (Davalos et al., 2005), consistent distribution throughout the brain (Lawson et al., 1990), and continuous input from neurons and astrocytes via small molecular messengers (i.e. fractalkine and S100β respectively) (Bianchi et al., 2010; Madry and Attwell, 2015). We previously reported that microglia morphology and phagocytic receptor CD11b are immediately changed by brain ischemia and therefore can be used as an indicator of an early microglia response to stroke (Morrison and Filosa, 2013). However, sex differences in microglia responses to ischemia are largely unknown. Microglia phagocytosis, initiated in part by CD11b-ligand interactions, is a double-edged sword; necessary for wound healing but also well documented in pre-clinical research as a source of secondary inflammatory injury after stroke (Brown and Neher, 2012; Lee et al., 2014). Understanding this balance in both males and females is relevant to future stroke therapies which are developed to promote brain recovery and yet must also account for inherent secondary injury. Unfortunately, few pre-clinical stroke-related studies have examined glial (astrocyte and microglia) responses to injury in male and female adult mouse models prior to 24 hours post-ischemia (Cordeau et al., 2008). As a result, our understandings of early glial events that may be central brain injury after ischemic stroke are only beginning to be revealed (Zheng et al., 2010; Zheng et al., 2013). Moreover, it is not clear if early astrocyte and microglia responses to ischemia are different between male and female mice. Such information is vital in establishing a timeline of sex differences after ischemic stroke and therefore vital to the discovery and implementation of potential sex-specific stroke treatment regimens. In this study, we determine sex differences in astrocyte and microglia responses to 60 min of MCA occlusion in adult male and female mice by assessing changes in astrocyte Ca2+ dynamics and microglia volume in acute brain slices. In addition, we illustrate sex differences in astrocyte S100β and AQP4 immunofluorescence patterns as well as microglia CD11b immunoreactivity after 60 min of MCA occlusion. 1.1 Experimental Procedures Animals Male CX3CR1GFP/ GFP mice with a C57BL6/J background (Jackson Laboratories, no. 005582) were bred to female C56BL6/J mice. Heterozygous male and female offspring CX3CR1GFP/+ weighing 20–25 g (6–8 weeks old) were used to record somatic astrocyte Ca2+ elevations and microglia volume after ischemia and in acute brain slices. For consistency, CX3CR1GFP/+ mice were used for additional immunohistochemistry (IHC) experiments. All animals were housed in climate controlled rooms with a 12-h light cycle. Female mice were not subjected to estrous cycle synchronization, but were allowed to cycle naturally. Female mice were sampled randomly to determine if ischemia induced astrocyte and microglia responses were present despite possible variations in sex hormone levels, an occurrence that more accurately models the human condition and an appropriate initial inquiry (McCarthy et al., 2012). All animal experiments were performed according to methods approved by and in compliance with the Augusta University Institutional Animal Care and Use Committee. Middle cerebral artery occlusion Focal cerebral ischemia to the right hemisphere was achieved using the filament method to occlude the right MCA for 60 min in male and female animals under isoflurane anesthesia delivered via a 20% oxygen/80% air mixture as previously described (Morrison and Filosa, 2013). After common, internal and external artery dissection, a heat blunted and silicone-coated filament (tip diameter 0.20–0.25 mm) was advanced to the ostea of the MCA through the external and internal common carotid artery. Cerebral ischemia to the right MCA region was verified by laser Doppler flowmetry (Perimed Periflux 5000, North Royalton, OH); animals were included in the study if ischemia resulted in a > 70% reduction in blood flow. Focal ischemia was maintained for 60 min. The sham procedure included all elements up to filament placement. All animals were euthanized without reperfusion and prior to brain tissue collection for acute brain slices or tissue processing for IHC methods. Acute brain slices and brain cell imaging Acute brain slices were acquired for ex vivo astrocyte and microglia imaging in a group of naïve animals (no surgery) or immediately after 60 min MCA occlusion. Brain tissue was extracted immediately after euthanasia in ice-cold artificial cerebral spinal fluid (aCSF) and immediately sliced into 280 µm coronal sections using a vibratome as previously described (Leica VT 1200S, Leica Microsystems, Wetzlar, Germany)(Morrison and Filosa, 2013). The composition of aCSF was (in nM): KCl 3, NaCl 120, MgCl2 1, NaHCO3 26, NaH2PO4 1.25, glucose 10, CaCl2 2 and 400 µM L-ascorbic acid; osmolarity was 300–305 mOsM and pH to 7.4 after 95% O2 and 5% CO2 equilibration. Brain hemisphere sections (no surgery, contralateral and ipsilateral) were incubated in aCSF containing the Ca2+ indicator Rhod-2AM (Molecular Probes, MP 01244, 5µM) and 2.5 µL 20% pluronic acid (Molecular Probes P3000MP) in a 95% O2/5% CO2 oxygenated chamber for 60 min just prior to imaging. Figure 1A summarizes the experimental protocol for Ca2+ imaging after 60 min MCA occlusion. Astrocyte Ca2+ imaging All images were acquired in cortical layers I–III using a 40× objective (Zeiss Achroplan 40×/0.8w) at a rate of 4 images per second. To test astrocyte viability, ATP (Sigma, A 9187, 500 µM) was bath applied at the end of each experiment; only astrocytes responsive to ATP, determined by a robust increase in intracellular Ca2+, were included in the data analysis. Live imaging was limited to a 3 hour window following incubation with the Ca2+ indicator. Calcium imaging was analyzed as previously described using Sparkan software [Dr. Adrian Bonev, University of Vermont (Filosa et al., 2004)]. Fluorescence intensity (F) was determined within 10 × 10 pixel squares placed over Rhod-2 loaded astrocytes with baseline fluorescence (F0) determined from 20 images showing no activity. Fractional fluorescence (F/F0) was calculated and peaks were automatically detected from oscillations crossing a set threshold value (>0.2 F/F0) and summarized as Ca2+ oscillations peak frequency (Hz) for each cell. We quantified the area under the curve (AUC) of each Ca2+ trace as an additional indicator of astrocyte Ca2+ which accounts for both peak frequency, amplitude and duration. Using the same 10 × 10 pixel region of interest, AUC was calculated as the integral over time for each cell exhibiting Ca2+ activity included in the study (Kim et al., 2015). All data was normalized to appropriate male and female naïve/no surgery group for statistical analysis. Microglia imaging CX3CR1GFP/+ microglia adjacent to astrocytes were imaged immediately following astrocyte Ca2+ imaging. We imaged microglia soma and associated processes by acquiring Z-stack confocal images every 1 µm for 30–35 µm to ensure that we included the same imaging plane used for astrocyte Ca2+ acquisition. IMARIS software (Bitplane) and filament building protocols were used to create representative 3D microglia models, necessary to quantify the volume of each microglia. Data were normalized to appropriate male and female naïve/no surgery group prior to two-way statistical analysis. Immunohistochemistry Immunohistochemistry (IHC) images were acquired in additional male and female groups of mice. After 60 min MCA occlusion or sham surgery, animals were euthanized, brain tissue removed, fixed for 24h in 4% paraformaldehyde and incubated in a 30% sucrose solution for 72h. Brain tissue was kept at −80°C until sectioning into coronal sections (Leica cryostat CM3050, 50µm) and stored at −20°C in a cryoprotectant solution until tissue processing for IHC staining. To identify astrocytes and microglia, free-floating brain sections were blocked in a 10% horse serum solution (0.01M PBS 0.05% Triton and 0.04% NaN3) for 1h followed by a 72h incubation with primary antibodies: rabbit anti-GFP 1:1,000 (Invitrogen A-6455), rat anti-Mac-1 1:500 (Chemicon MAB1387Z), rabbit anti-S100β 1:200 (Dako Z031101-2), and goat anti-AQP4 1:500 (Santa Cruz sc-9888). A 4-h incubation of 1:250 secondary primaries followed: donkey anti-rabbit Alexa 488, (711-546-152); donkey anti-rat CY3, (712-166-153); donkey anti-chicken Alexa 649 (702-605-155) and donkey anti-goat 594, (705-585-147, Jackson ImmunoResearch Laboratories). Our IHC protocol was for double staining of anti- AQP4 and anti-S100β in one set of male and female tissue and, in another set of tissue, double staining of anti-GFP and anti-CD11b. Male and female tissues from sham and ischemic groups were incubated together for consistent IHC staining. In addition, sections of ipsilateral proximal female brain tissue that was subjected to all aspects of our ICH protocol but without primary antibody. This secondary only antibody staining assessed unspecific binding that may have been prevalent in the infarcted area. All reactions were carried forward at room temperature; washes between incubations were done with 0.01M PBS for 15min. Slices were then mounted onto slides using Vectashield (Vector Laboratories, H-1000). Image acquisition and analysis A confocal microscope was used to acquire photomicrographs (30-µm Z-stack at 2-µm intervals, Zeiss 510, 40×/1.3 oil objective) of cortical layers I–III after IHC preparation in brain regions depicted in Figure 3A. Photomicrographs were stacked and split to obtain maximum intensity projections for all channels and saved as TIFF files prior to analysis. The location of astrocyte AQP4 and S100β was determined as well as microglia morphology and CD11b after 60 min MCA occlusion in male and female tissue. Astrocyte S100β analysis With injury, S100β is released in macro-molar concentrations from the astrocyte soma to the extracellular space. We determined changes in S100β location, from primarily somatic to non-somatic areas, after 60min MCA occlusion and tested if post-stroke changes were different between male and female mice. The distribution pattern of S100β was determined from photomicrographs with a three step analysis technique using Image J software (http://imagej.net). The S100β analysis technique was derived from previous studies examining changes in the area of AQP4 distribution [(Ren et al., 2013; Wang et al., 2012) described below]. For each photomicrograph, we first we determined the area of S100β positive immunofluorescence which included both somatic and non-somatic S100β immunoreactivity (Figure 3B low threshold). Second, we determined the area of S100β positive immunofluorescence that included primarily somatic S100β immunofluorescence (Figure 3B high threshold). Third, the area of non-somatic S100β was calculated: (area of somatic and non-somatic S100β immunoreactivity) − (area of somatic S100β immunoreactivity) = area of non-somatic immunoreactivity. An example of this process is illustrated in Figure 3B. Astrocyte AQP4 analysis In astrocytes, the water channel AQP4 is constitutively expressed and polarized to astrocyte endfeet rather than astrocyte soma or other processes. AQP4 polarity is disturbed with brain injury and this change in AQP4 distribution (from endfeet to non-endfeet locations) has been previously quantified using IHC photomicrographs and image analysis (Ren et al., 2013; Wang et al., 2012). We employ this method by first determining the area of AQP4 immunoreactivity at astrocyte endfeet as well as in non-endfeet astrocyte processes (Figure 4A low threshold). Second, we determined the area of AQP4 immunoreactivity in astrocyte endfeet (Figure 4A high threshold). AQP4 polarity was calculated: (area of endfeet and non-endfeet AQP4 immunoreactivity) − (area of endfeet AQP4 immunoreactivity) = un-polarized AQP4 immunoreactivity. An example of this process is illustrated in Figure 4A. Microglia analysis Microglia were analyzed for changes in morphology and phagocytic function using a computer-aided skeleton analysis method (microglia processes length and number of endpoints) and CD11b immunofluorescence, respectively, as previously described (Morrison and Filosa, 2013). For computer-aided morphological analysis, anti-GFP images were despeckled and then processed to create skeletonized images using Image J software. The Analyze Skeleton Plugin (Arganda-Carreras et al., 2010) was used to identify (tag) and microglia skeletons relevant to quantify ramification: slab voxels, junctions and endpoints. Figure 5B illustrates the conversion from despeckled to tagged image. We summarized the number of endpoints and averaged the length of all processes (slab voxels) from the Analyze Skeleton plugin data output. We then normalized all data by the number of somas/image to result in number of endpoints/cell and microglia process length/cell. Microglia CD11b total fluorescence intensity (TFI) was determined in male and female tissue as previously published. Briefly, microglia CD11b TFI was determined by multiplying the percent area of the image with positive immunoreactivity by the mean fluorescence intensity of CD11b immunoreactivity. Statistical Analysis Sex differences in control conditions (no surgery and sham groups) were determined using Student’s t-test. Data collected after 60min MCA occlusion were normalized to male or female no surgery or sham conditions and two-way ANOVA was then used to test for sex differences and brain regions affected by the MCA occlusion. Sidak’s multiple comparisons post-hoc test was used to test for specific differences between groups. All data are presented as the mean ± standard error of mean (SEM). GraphPad Prism 6 was used for all analyses. 1.2 Results Sex differences in astrocyte Ca2+ elevations after 60 min MCA occlusion We examined the effect of MCA occlusion on astrocyte Ca2+ dynamics in cortical brain regions immediately following 60 min MCA occlusion in adult male and female acute brain slices. A summarized protocol and brain regions (contralateral, ipsilateral distal and ipsilateral proximal) for astrocyte Ca2+ imaging are shown in Figure 1A. Representative astrocyte Ca2+ traces used to analyze AUC and peak frequency (Hz) are shown in Figure 1B. There were no sex differences in astrocyte Ca2+ AUC between male and female no surgery groups (Figure 1C, left). Data summarized by Figure 1C (right) was normalized to the no surgery group in each sex to determine changes in astrocyte AUC in brain regions affected by the MCA occlusion. There was no significant region or sex effect observed for Ca2+ AUC after 60min MCA occlusion (Two-way ANOVA: region F(3,72)=0.20, p = 0.9, sex: F(1,72)=2.639, p=0.11, interaction F(3,72)=0.50, p=0.68). We also summarized astrocyte Ca2+ peak frequency from astrocyte Ca2+ traces; there were no sex differences in astrocyte Ca2+ peak frequency between male and female no surgery groups (Figure 1D, left). Data summarized by Figure 1D (right) was normalized to the no surgery group in each sex to determine changes in the frequency of astrocyte Ca2+ elevations in brain regions affected by the MCA occlusion. A two-way ANOVA analysis reveals sex differences in Ca2+ peak frequency after ischemia but astrocyte Ca2+ peak frequency was not significantly different between brain regions (Figure 1D; region F(3,86)=0.25, p=0.86, sex: F(1,86)=8.19, p=0.005, interaction F(3,86)=1.75, p=0.16). Post-hoc testing of sex differences revealed that astrocyte Ca2+ peak frequency was increased in female ipsilateral tissue when compared to males. Sex differences in microglia volume resulting from 60 min MCA occlusion We imaged microglia from male and female CX3CR1GFP/+ mice subjected to 60 minutes MCA occlusion in mice in order to quantify changes in microglia volume. Microglia volume was calculated from 3D models generated using IMARIS (Bitplane) filament tracer protocols. A representative image of a microglia in each region (no surgery, contralateral, ipsilateral distal and ipsilateral proximal) and its corresponding IMARIS model used to determine volume is shown in Figure 2A. There were no sex differences in microglia volume between male and female no surgery groups (Figure 2B, left). Microglia volume data was normalized to male or female no surgery group in order to determine changes in volume affected by the MCA occlusion and summary data are shown in Figure 2B (right). Microglia volume was not significantly changed in brain regions after MCA occlusion but was different between male and female mouse groups after ischemia (two-way ANOVA: region F(3,40)=11.87, p<0.08, sex: F(1,40)=12.47, p=0.009, interaction F(3,4)=8.80, p=0.17). Post-testing of sex differences revealed no specific differences in microglia volume in the ipsilateral hemisphere after MCA occlusion. Sex differences in S100β expression pattern after 60 minutes of MCA occlusion The Ca2+ binding protein S100β is typically expressed in astrocyte soma, its release is a biomarker of brain injury (Dayon et al., 2011; Plog et al., 2015). Figure 3A illustrates the representative brain regions for data collection in proximity to the ischemic injury (contralateral, distal ipsilateral and ipsilateral proximal region); matching regions were acquired in sham tissue. The protocol used to measure non-somatic S100β expression, described in Methods, is illustrated in Figure 3B. Exemplar photomicrographs used for image analysis and data collection are shown in Figure 3C. Cropped images, shown below the full sized photomicrographs, illustrate the details of S100β immunofluorescence and changes to the expression pattern after MCA occlusion in male and female animals. We determined the change in S100β expression pattern after 60 min ischemic stroke in male and female mice using uncropped photomicrographs. Unspecific secondary binding of anti-rabbit 594 to the ipsilateral proximal region was not detectable. There were also no sex differences in the area of non-somatic S100β between male and female sham groups (Figure 3D, left). S100β data were normalized to the male or female sham group in order to determine changes in the area of S100β immunoreactivity after MCA occlusion according to sex and brain regions (Figure 3D, right). Significant effects were noted for both region and sex (two-way ANOVA: region: F(5,86)=4.72, p=0.007, sex: F(1,86)=10.25, p=0.0019, interaction F(5,86)=1.81, p=0.12). Post-hoc testing revealed that in the male, but not female, non-somatic S100β immunoreactivity was robustly increased after MCA occlusion in the proximal ipsilateral region when compared to sham and contralateral regions. Sex differences in AQP4 polarity after 60 minutes of MCA occlusion Changes in AQP4 expression are well documented in ischemic stroke (Ribeiro Mde et al., 2006; Vella et al., 2015; Wang et al., 2012). However, the expression pattern in male and female brain tissue has not been previously addressed. The protocol used to measure un-polarized AQP4 expression, described in Methods, is illustrated in Figure 4A. Examples of IHC-AQP4 images used for our image analysis protocol are shown in Figure 4B with cropped photomicrographs shown below to better illustrate changes in the AQP4 expression pattern. Unspecific secondary binding of anti-goat 647 to the ipsilateral proximal region was not detectable. In sham mice, sex differences exist in AQP4 polarity (Figure 4C, left). However, this change in AQP4 polarity was only observed in the proximal region (Student’s t-test, p = 0.05). AQP4 data were normalized to male or female sham group in order to determine changes in AQP4 polarity after MCA occlusion according to sex and brain regions (Figure 4C, right). Changes to AQP4 polarity was significantly changed in brain regions and according to sex after 60 min MCA occlusion (two-way ANOVA: region F(5,86)=3.30, p=0.009, sex: F(1,86)=17.21, p<0.0001, interaction F(5,86)=2.17, p=0.06). Post-hoc testing reveals significant changes in AQP4 polarity in ipsilateral brain regions after MCA occlusion in male and not female mice. Sex differences exist in AQP4 polarity in distal contralateral and ipsilateral brain regions. Sex differences in microglia morphology after 60 min MCA occlusion We investigated sex differences in microglia neuroinflammatory responses to ischemia in addition to astrocyte responses. We first quantified microglia morphology after 60 min MCA occlusion in male and female mice in sham, contralateral and ipsilateral (proximal and distal) regions as shown in Figure 3A. Cropped single cells from photomicrographs are included to better visualize microglia morphology details; data analysis was conducted on uncropped images. Microglia morphology (number of microglia process endpoints/cell and process length/cell) was determined from anti-GFP photomicrographs using a skeleton analysis method that was modified from our previous publication (Morrison and Filosa, 2013) and summarized in Figure 5B. As expected, unspecific secondary binding (anti rabbit 488) to the ipsilateral proximal region was not detected, however, faint GFP fluorescence of microglia in the CX3CR1GFP/+ mouse was present, not shown. There were no sex differences in microglia process length/cell or endpoints/cell between male and female sham groups (Figure 3D left and Figure 4D left, respectively). Microglia morphology data were normalized to male or female sham group in order to determine changes after MCA occlusion according to sex and brain regions; summary graphs of these normalized data are shown in Figure 5D and Figure 5E. Our analysis reveals that microglia process endpoints/cell were different after MCA occlusion according to region and sex (two-way ANOVA: region F(5,86)=4.74, p = 0.0007, sex: F(1,86)=18.29, p < 0.0001, interaction F(5,86)=4.74, p = 0.0007). However, a significant interaction effect, observed here, indicates that the two factors sex and ischemia are interdependent. On the other hand, microglia process length/cell after MCA occlusion was not altered by either sex or region (Figure 5D; two-way ANOVA: region F(5,86)=1.93, p=0.10, sex: F(1,86)=1.04, p=0.31, interaction F(5,86)=0.49, p=0.78). Sex differences in microglia CD11b in sham mice and after 60 min MCA occlusion In addition to morphology, we examined microglia function after 60 min MCA occlusion; microglia CD11b is an important receptor to microglia phagocytosis. Example images in all male and female tissue after sham and MCA occlusion are shown in Figure 6A. Cropped single cells from photomicrographs are included to better visualize CD11b immunofluorescence. The CD11b receptor is located on cell processes as well as, to a lesser extent, on the soma (Perego et al., 2011) which we show in Figure 6B (anti-GFP shows microglia (top) with matching anti-CD11b immunofluorescence below). Unspecific secondary binding of anti-rat CY3 secondary to the ipsilateral proximal region was not detectable. Typically, the intensity of CD11b immunoreactivity is increased in response to injury (Morrison and Filosa, 2013; Perego et al., 2011). However, we show that microglia CD11b immunofluorescence staining is prevalent in female sham tissue when compared to male sham tissue (Figure 6C; Student’s t-test: p = 0.03). In addition to these baseline sex differences, we also observed sex differences in microglia CD11b immunofluorescence after 60 min MCA occlusion. For this analysis, CD11b data was normalized to male or female sham group in order to determine changes after MCA occlusion according to sex and brain regions; summary graphs of these normalized data are shown in Figure 6D (two-way ANOVA: region F(5,86)=8.74, p=0.07, sex: F(1,86)=13.12, p=0.0001, interaction F(5,86)=7.44, p=0.11). Post-hoc testing reveal that changes to microglia CD11b is robust after 60 min of MCA occlusion in male proximal brain regions but absent in the female tissue. To summarize: 1) constitutive CD11b immunofluorescence is greater in females than males; 2) in female mice, constitutively high CD11b remains unchanged after 60 min MCA occlusion whereas, 3) in male mice, constitutively low CD11b is robustly increased in proximal ipsilateral regions after 60 min of MCA occlusion. 1.3 Discussion We provide evidence of sex differences in the frequency of astrocyte Ca2+ elevations and microglia volume after 60 min MCA occlusion which was measured in acute brain slices. Using IHC methods, we show that astrocytes and microglia have a robust immediate response to MCA occlusion that is sex dependent. Significant changes in the ipsilateral hemisphere immediately after MCA occlusion included: increased presence of non-somatic S100β; un-polarized AQP4; and higher microglia CD11b immunofluorescence in male but not female mice. In addition, the prevalence of microglia CD11b immunofluorescence at baseline in females, when compared to males, brings to light sex differences in constitutive microglia function which, we suggest, may have a role in determining male and female microglia responses to ischemia. Detrimental neuronal cortical spreading depolarizations and glutamate excitotoxicity, hypothesized to occur in the peri-infarct region following ischemia (Hinzman et al., 2015), stands as an early injurious event following stroke as well as a mechanism of delayed neuronal cell death (Dirnagl et al., 1999). Increased astrocyte Ca2+ elevations, mediated in part by neuronal release of glutamate (Kim et al., 1994), also contributes to edema (Thrane et al., 2011) and the neurotoxic events after ischemia (Ding et al., 2009; Hansson and Ronnback, 2003; Iwabuchi and Kawahara, 2009; Nedergaard and Dirnagl, 2005; Takano et al., 2009). We did not observe sex differences between our control (no surgery) groups and therefore baseline conditions did not have an effect on our post-ischemia analysis. Out data suggest that changes in the frequency of astrocyte Ca2+ elevations were small in the ipsilateral brain regions after 60 min of MCA occlusion in both male and female mice. Our findings differ from Ding and colleagues (2009) who revealed robust and sustained increases in astrocyte Ca2+ elevations measured in male mice using a photothrombotic stroke in vivo model (Ding et al., 2009). Methodological factors could account for observed differences between studies. For example, the bulk loading of Ca2+ indicators, a method used in this study, limits data collection to the cell soma and excludes detection of astrocyte Ca2+ responses in processes (Bazargani and Attwell, 2016). Therefore, our data may under-represent astrocyte Ca2+ dynamics after ischemic stroke (Srinivasan et al., 2015). Also possible, the focal ischemic injury from the filament method and MCA occlusion may not have provided sufficient cortical injury for a robust astrocyte Ca2+ response in the ipsilateral hemisphere. Methodologic constrains limited our ability to directly measure brain cell injury concurrent to ex vivo astrocyte imaging. In lieu, we measured microglia volume as an indirect indicator of cell injury. We show that microglia volume in control groups were similar and were not significantly changed after MCA occlusion in either male or female mice which suggests that brain injury was mild at this early time point. Similar to microglia volume, we did not observe sex differences in astrocyte Ca2+ elevations at baseline. However, we observed sex differences in astrocyte Ca2+ elevations after 60 min MCA occlusion. The frequency of astrocyte Ca2+ elevations in the ipsilateral hemisphere was increased in females when compared to males. It is possible that estradiol has a role in increasing astrocyte Ca2+ dynamics in female mice. In cell culture models, estradiol is shown to promote signaling mechanisms that increase intracellular Ca2+ concentrations and corresponding astrocyte Ca2+ dependent functions (Chaban et al., 2004; Kelly and Ronnekleiv, 2009). While not tested in our preparation, estradiol was reported to increase cytoplasmic Ca2+ release in cultured hypothalamic astrocytes collected from both males and females but not in females with estrogen receptor knockout (Kuo et al., 2010). It has also been shown that elevated intracellular Ca2+ stimulates astrocyte mitochondrial metabolism to maintain vital energy resources during the first few hours after ischemia. In this case, increased intracellular Ca2+ is associated with markedly improved post-stroke neurological outcomes (Zheng et al., 2010; Zheng et al., 2013). Taken together our data supports increased frequency of astrocyte Ca2+ elevation as a mechanism driving dichotomous stroke outcomes between males and females. However, it remains that our understanding of sex differences on constitutive or injury-evoked astrocyte Ca2+ dynamics in adult mice and resulting influence on brain function remains incomplete and is an area of study that warrants additional investigation. In addition to measuring astrocyte Ca2+ dynamics, we investigated the potential of sex differences in the location of astrocyte AQP4 and S100β, proteins suspected to play a role in brain injury after ischemic stroke (Benfenati et al., 2011; Donato et al., 2009; Kitchen et al., 2015; Vella et al., 2015). Under physiological conditions, AQP4 is highly polarized to astrocyte endfeet rather than non-endfeet processes and/or soma (Potokar et al., 2013). Our analysis reveals that AQP4 polarity is different between sham male and female mice, however, this finding was not consistent to both distal and proximal brain regions. Others have determined that estrogen increases AQP4 mRNA and protein quantities in adult mice (Shin et al., 2011). Following brain injury and ischemia, AQP4 expression is also increased (Fukuda and Badaut, 2012; Ribeiro Mde et al., 2006; Vella et al., 2015). Un-polarized AQP4 has been reported in proximity to the brain injury following more extended time points after stroke (Wang et al., 2012) and traumatic brain injury (Liu et al., 2015; Ren et al., 2013); a change in AQP4 polarity or absence of AQP4 impairs glymphatic clearance after injury (Iliff et al., 2014). Thus, we examined sex differences in AQP4 polarity (a distribution pattern) after 60 min MCA occlusion, rather than AQP4 quantity. Similar to others, we demonstrate that AQP4 is un-polarized in brain regions proximal to the ischemia induced by MCA occlusion in male mice. AQP4 also became un-polarized in female tissue, however, not as profoundly different as observations in male tissue. The lesser response noted in female proximal region is likely influenced by the normalization to sham, which was increased the female proximal region versus male. Therefore, our data best illustrate that the change from sham conditions was not as profound in female mice when compared to male. A change in AQP4 polarity just after 60 min of MCA occlusion is an indicator that AQP4 vesicle trafficking or insertion to the plasma membrane surface is disturbed due to cytotoxic events, cell swelling, and/or rearrangement of the cytoskeleton during gliosis (Potokar et al., 2013). Ensuring that AQP4 is correctly localized to endfeet is central to managing edema and ensuring efficient glymphatic clearance (Plog et al., 2015), both factors important to stroke recovery. S100β is an astrocyte Ca2+ binding protein and small molecular messenger present in the astrocyte cytoplasm. Astrocytes release S100β in macro-molar concentrations after brain injury to act as a damage-associated molecular pattern (DAMP) molecule to promote inflammatory responses and affect astrocyte-microglia communication (Bianchi et al., 2010; Donato et al., 2009; Zhang et al., 2011). As such, S100β is of increasing interest as a potential biomarker of brain injury and for its role in exacerbating inflammatory responses after stroke and traumatic brain injury (Dayon et al., 2011; Kiechle et al., 2014; Pham et al., 2010). The area of non-somatic S100β immunoreactivity was similar between male and female sham mice and therefore did not influence post-ischemia analysis. After ischemia, the area of non-somatic S100β is increased in males, with a lesser effect observed in female mice. In its role as a biomarker, Plog et al. (2015) provide evidence that S100β moves from brain parenchyma to the plasma via the glymphatic system and that the clearance of S100β to plasma is significantly impaired by un-polarized AQP4 distribution after traumatic brain injury or with the absence of AQP4 (APQ4-null mice). It will be vital to further clarify the role of sex differences in AQP4 polarity and glymphatic clearance of biomarkers such as S100β to the plasma. Such data is necessary to ensure the validity of interpreting peripheral concentrations of S100β as a biomarker of brain injury after stroke in males and females. In conjunction with astrocytes, microglia cells have a robust and immediate response to injury. In a previous study we characterized the early microglia morphologic and functional response to 60 min MCA occlusion in male mice (Morrison and Filosa, 2013). We extend these findings and now show sex differences in microglia morphology immediately following MCA occlusion, however these changes are subtle. There was no sex differences in baseline microglia ramification observed in sham animals. However, following MCA occlusion, there were sex differences in the number of microglia process endpoints per cell and no sex differences in process length/cell. We also reveal evidence of sex differences in microglia function. Microglia CD11b is an integrin receptor key to microglia phagocytosis of endogenous structures and debris after ischemia (Kettenmann et al., 2011; Wake et al., 2009). We show that CD11b immunofluorescence is constitutively high in female sham tissue compared to males and remains unchanged after ischemic stroke whereas, in males, constitutively low levels of CD11b immunofluorescence is increased after brain ischemia. We suggest that elevated CD11b expression pattern prior to ischemia, observed in female mice, enhances microglia capacity to phagocytize necrotic debris and cell contents that may otherwise exacerbate neurotoxicity and neuronal death after stroke. Additional studies are necessary to clarify the benefits of early microglia phagocytosis to improve stroke outcomes. We employ IHC methods in this study rather than other methods that quantify targeted antigens in tissue lysate; IHC is ideally suited to assess the location and distribution of targeted antigens. We and others have noted changes in AQP4 or S100β distribution, to become more diffuse, under experimental conditions or cell distribution (Dyck et al., 1993; Ren et al., 2013; Wang et al., 2012). However, IHC methods have inherent limitations and several factors must be considered when determining changes in AQP4 and S100β location and data interpretation (e.g. tissue integrity and associated diffusion of antibodies, unspecific immunodetection). To address this, we determined that there was negligible unspecific secondary immunoreactivity (for both astrocyte and microglia ICH protocols) in ipsilateral proximal tissue regions. In addition we provide evidence that after just 60 min of MCA occlusion (without reperfusion) cortical tissue integrity remians. We demonstrate that astrocytes remained active and viable and microglia volume was unchanged in the ipsilateral hemisphere after 60 min of MCA occlusion. In addition, using IHC, microglia morphology in the ipsilateral proximal tissue remained ramified and complex rather than amoeboid. Taken together, our observations support the notion that tissue integrity in cortical regions imaged in this study were not severely compromised (as seen after ischemia and 24hr of reperfusion) so as to introduce unspecific biding during IHC protocols that would confound data analysis and interpretation. We also acknowledge that sex differences in brain infarct size in this mouse model of ischemic stroke has been previously shown after ischemia and 24hr of reperfusion, a conventional time point to measure infarct size due to the methodological constraints of 2,3,5-Triphenyltetrazolium chloride (TTC) staining (Banerjee et al., 2013; Bederson et al., 1986; Isayama et al., 1991; Liszczak et al., 1984; Shin et al., 2011). In this study, infarct size was not measured after 60 min of MCA occlusion because of the early time point and methodological constraints. Sex differences in ischemic injury at this early time-point post stroke may exist and therefore may be a source of variability we observe in our data analysis. 1.4 Conclusions We present data of sex differences in astrocyte Ca2+ elevations as well as the astrocyte and microglia proteins known to have an important role in the evolution of brain injury after stroke. These findings are critical to contributions toward understanding sex differences not only in brain physiology but also stroke pathobiology. As we continue to elucidate origins of dichotomous stroke outcomes between the sexes, we move closer to the development of sex-specific and successful stroke therapies. Importantly, we address changes occurring at an early time point (prior to reperfusion) following ischemia with the goal to better understand and define the cellular targets involved in the evolution of ischemic stroke injury. This study received financial support from NIHLB (R01HL089067) to JAF, NINR (1F32NR013611) to HWM Abbreviations Ca2+ Calcium MCA Middle Cerebral Artery AQP4 Aquaporin 4 Figure 1 Astrocyte Ca2+ elevations in male and female mice after 60 min MCA occlusion A) Experimental protocol for ex vivo Ca2+ imaging of acute brain slices after 60 min MCA occlusion. B) Representative Ca2+ traces acquired from male and female mice: no surgery (aCSF), contralateral, distal ipsilateral and proximal ipsilateral brain regions. C) Summary data of area under the curve (AUC) from Ca2+ traces acquired from spontaneously active astrocytes in male and female brain slices without surgery (left) and after 60 min MCA occlusion (right). N (male/female animals): No surgery (11/10), contralateral (16/11), distal ipsilateral (7/9); and proximal ipsilateral (5/11). AUC in male and female no surgery groups (left, mean and SEM) are compared using unpaired Student’s t-test: p = 0.54. Astrocyte Ca2+ AUC after 60min MCA occlusion (right), are compared using two-way ANOVA (mean and SEM): region: p = 0.89, sex: p = 0.11, interaction: p = 0.68. D) Summary data of Ca2+ peak frequency from spontaneously active astrocytes from male and female brain slices without surgery (left) and after 60 min MCA occlusion (right). N (male/female animals): No surgery (12/10), contralateral (16/12), distal ipsilateral (10/10); and proximal ipsilateral (8/11). Astrocyte Ca2+ peak frequency in male and female no surgery groups (left, mean and SEM) are compared using unpaired Student’s t-test: p = 0.11. Peak frequency after 60min MCA occlusion (right) are compared using two-way ANOVA (mean and SEM): region: p = 0.16, sex: p = 0.005, interaction: p = 0.16 with Sidak multiple comparison analysis reported in figure. Figure 2 Sex differences in microglia volume after 60 min MCA occlusion A) Example of microglia photomicrographs from ex vivo imaging (left) and rendered IMARIS model (right) after 60 min MCA occlusion according to group: no surgery, contralateral and ipsilateral brain regions. Scale bar = 10µm. B) Summary data of microglia volume in male and female brain slices without surgery (left) and after 60 min MCA occlusion (right). N (male/female animals): No surgery (4/5), contralateral (11/6), distal ipsilateral (6/4) proximal ipsilateral (5/8). Microglia volume in male and female no surgery groups (left, data and SEM) are compared using unpaired Student’s t-test: p = 0.84. Microglia volume (mean and SEM) after 60min MCA occlusion (right) is compared using two-way ANOVA: region, p < 0.08, sex, p = 0.009, interaction: p = 0.17 with Sidak multiple comparison analysis reported in figure. Figure 3 Sex differences in S100β expression pattern after 60 min MCA occlusion A) Schematic drawing depicting regions imaged within cortical layers I–III in matching ipsilateral and contralateral brain sides with corresponding proximal and distal regions relative to the ischemic region. B) Example of high and low threshold binary images used to calculate non-somatic S100β. C) S100β immunostaining in male and female tissue after 60 MCA occlusion in sham, contralateral, and ipsilateral (distal and proximal) brain regions. Cropped photomicrographs corresponding to highlighted square shown in first row illustrate additional detail of somatic and non-somatic immunofluorescence. All data was collected using full sized photomicrographs (Scale bars = 10µm). D) Summary data of non-somatic S100β after sham procedure (left) and after 60 min MCA occlusion (right) in male and female mice. N (male/female animals): Distal: Sham (9/6), contralateral (9/8), ipsilateral (9/8); Proximal: Sham (9/6), contralateral (9/8), ipsilateral (9/8). Non-somatic S100β in male and female sham mice (mean and SEM) are compared using an unpaired Student’s t-test: distal p = 0.89; proximal, p = 0.81. Non-somatic S100β after 60min MCA occlusion in male and female mice (mean and SEM) are compared using two-way ANOVA: region, p= 0.0007, sex, p= 0.0019, interaction: p= 0.12 with Sidak multiple comparison analysis is reported in the figure. Figure 4 Sex differences in astrocyte AQP4 polarity after 60 min MCA occlusion A) Example of high and low threshold binary images used to calculate AQP4 polarity. B) AQP4 immunostaining in male and female tissue after 60 min MCA occlusion in sham, contralateral, and ipsilateral brain regions. Cropped photomicrographs corresponding to highlighted squares shown in first row depicting additional details of polarized and un-polarized AQP4 expression pattern. All data was collected using full sized photomicrographs (Scale bars = 10µm). C) Summary data of AQP4 polarity after sham procedure (left) and after 60 min MCA occlusion in male and female mice. N (male/female animals): Distal: Sham (9/6), contralateral (9/8), ipsilateral; Proximal: Sham (9/6), contralateral (9/8), ipsilateral. Un-polarized AQP4 in male and female sham mice (mean and SEM) are compared using an unpaired Student’s t-test: distal, p = 0.10; proximal p = 0.05. Un-polarized AQP4 after MCA occlusion in male and female mice (mean and SEM) are compared using two-way ANOVA: region p = 0.0089, sex: p < 0.0001, interaction: p = 0.064. Sidak multiple comparison analysis is reported in the figure. Figure 5 Microglia morphology in male and female mice after 60 min MCA occlusion A) Microglia GFP immunostaining in male and female tissue after sham or 60 min MCA occlusion in contralateral, and ipsilateral distal and proximal brain regions. Cropped photomicrographs (below) show processes morphology details of an individual microglia. Analysis was completed on full sized photomicrographs (Scale bar = 10µm). B) Example of photomicrographs showing the computer-aided microglia morphology analysis method using ImageJ: photomicrographs are despeckled, converted to a binary image, skeletonized and then tagged using skeleton analysis plugin for data collection. The analyze skeleton plugin identifies and tags skeletonized processes (orange) and endpoints (blue) http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:analyzeskeleton:start. C) Summary data of microglia process endpoints/cell after sham procedure (left) and after 60 min MCA occlusion in male and female mice: N (male/female animals): distal: sham (7/6), contralateral (9/9), ipsilateral (9/9); proximal: sham (7/6), contralateral (9/9), ipsilateral (9/9). Microglia endpoints/cell in male and female sham mice (mean and SEM) are compared using an unpaired Student’s t-test: distal, p = 0.44; proximal p = 0.74. Process length/cell after MCA occlusion in male and female mice (mean and SEM) are compared using two-way ANOVA: region p = 0.0001, sex: p < 0.0001, interaction: p = 0.0007 with Sidak multiple comparison analysis is reported in figure. D) Summary data of microglia process length/cell after sham procedure (left) and after 60 min MCA occlusion in male and female mice. N (male/female animals): distal: sham (7/6), contralateral (9/9), ipsilateral (9/9); proximal: sham (7/6), contralateral (9/9), ipsilateral (9/9). Microglia process length/cell in male and female sham mice (mean and SEM) are compared using an unpaired Student’s t-test: distal, p = 0.98; proximal p = 0.57. Process length/cell after MCA occlusion in male and female mice (mean and SEM) are compared using two-way ANOVA; region p=0.10, sex: p<0.31, interaction p=0.78. Figure 6 Microglia CD11b immunofluorescence in sham male and female mice and after 60 min MCA occlusion A) Microglia CD11b immunostaining in male and female tissue after sham or 60 min MCA occlusion in contralateral, and ipsilateral distal and proximal brain regions. Bottom row, cropped photomicrographs showing details for an individual microglia CD11b distribution. B) GFP and CD11b immunostaining from a single microglia. C) Summary data of microglia CD11b after sham procedure (left) and after 60 min MCA occlusion in male and female mice. N (male/female animals): Distal: Sham (8/6), contralateral (9/8), ipsilateral (9/8); Proximal: Sham (8/6), contralateral (9/8), ipsilateral (9/8). Summary data and analysis of microglia CD11b total fluorescence intensity (TFI)/cell in male and female sham tissue using Student’s t-test: distal, p = 0.07; proximal, p = 0.05. Microglia CD11b TFI/cell after 60min MCA occlusion (mean and SEM) are compared using two-way ANOVA: region p = 0.07, sex: p < 0.0001, interaction: p = 0.11 with Sidak multiple comparison analysis is reported in figure. Highlights Astrocyte Ca2+ dynamics are different in male and female mice after ischemia. Sex differences in astrocyte AQP4 polarity and non-somatic S100β after ischemia. Constitutive microglia CD11b is high in female sham cortex when compared to males. After ischemia, microglia CD11b immunoreactivity is not changed in female mice. Low baseline CD11b immunoreactivity is increased after ischemia in male mice. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Disclosure HWM contributed to the experimental design, execution of data collection, data analysis, data interpretation, and manuscript writing. JAF contributed to design of experiments, overall interpretation of data and manuscript editing. All authors read and approved the final manuscript. Agulhon C Sun MY Murphy T Myers T Lauderdale K Fiacco TA Calcium Signaling and Gliotransmission in Normal vs. Reactive Astrocytes Front Pharmacol 2012 3 139 22811669 Araque A Parpura V Sanzgiri RP Haydon PG Tripartite synapses: glia, the unacknowledged partner Trends Neurosci 1999 22 208 215 10322493 Arganda-Carreras I Fernandez-Gonzalez R Munoz-Barrutia A Ortiz-De-Solorzano C 3D reconstruction of histological sections: Application to mammary gland tissue Microsc Res Tech 2010 73 1019 1029 20232465 Attwell D Buchan AM Charpak S Lauritzen M Macvicar BA Newman EA Glial and neuronal control of brain blood flow Nature 2010 468 232 243 21068832 Banerjee A Wang J Bodhankar S Vandenbark AA Murphy SJ Offner H Phenotypic Changes in Immune Cell Subsets Reflect Increased Infarct Volume in Male vs. Female Mice Translational Stroke Research 2013 4 554 563 24187596 Bazargani N Attwell D Astrocyte calcium signaling: the third wave Nat Neurosci 2016 19 182 189 26814587 Bederson JB Pitts LH Germano SM Nishimura MC Davis RL Bartkowski HM Evaluation of 2,3,5-triphenyltetrazolium chloride as a stain for detection and quantification of experimental cerebral infarction in rats Stroke 1986 17 1304 1308 2433817 Benfenati V Caprini M Dovizio M Mylonakou MN Ferroni S Ottersen OP Amiry-Moghaddam M An aquaporin-4/transient receptor potential vanilloid 4 (AQP4/TRPV4) complex is essential for cell-volume control in astrocytes Proc Natl Acad Sci 2011 108 2563 2568 21262839 Bianchi R Giambanco I Donato R S100B/RAGE-dependent activation of microglia via NF-kappaB and AP-1 Co-regulation of COX-2 expression by S100B, IL-1beta and TNF-alpha Neurobiol Aging 2010 31 665 677 18599158 Brown GC Neher JJ Eaten alive! Cell death by primary phagocytosis: 'phagoptosis' In Trends Biochem Sci 2012 37 325 332 Chaban VV Lakhter AJ Micevych P A membrane estrogen receptor mediates intracellular calcium release in astrocytes Endocrinology 2004 145 3788 3795 15131017 Cheung G Sibille J Zapata J Rouach N Activity-Dependent Plasticity of Astroglial Potassium and Glutamate Clearance Neural Plast 2015 2015 109106 26346563 Cordeau P Jr Lalancette-Hebert M Weng YC Kriz J Live imaging of neuroinflammation reveals sex and estrogen effects on astrocyte response to ischemic injury Stroke 2008 39 935 942 18258827 Davalos D Grutzendler J Yang G Kim JV Zuo Y Jung S Littman DR Dustin ML Gan WB ATP mediates rapid microglial response to local brain injury in vivo In Nat Neurosci Vol 2005 8 752 758 Dayon L Turck N Garci-Berrocoso T Walter N Burkhard PR Vilalta A Sahuquillo J Montaner J Sanchez JC Brain extracellular fluid protein changes in acute stroke patients J Proteome Res 2011 10 1043 1051 21142207 del Zoppo GJ The neurovascular unit in the setting of stroke J Intern Med 2010 267 156 171 20175864 Ding S Wang T Cui W Haydon PG Photothrombosis ischemia stimulates a sustained astrocytic Ca2+ signaling in vivo Glia 2009 57 767 776 18985731 Dirnagl U Iadecola C Moskowitz MA Pathobiology of ischaemic stroke: an integrated view Trends in neurosciences 1999 22 391 397 10441299 Dirnagl U Pathobiology of injury after stroke: the neurovascular unit and beyond Ann N Y Acad Sci 2012 1268 21 25 22994217 Donato R Sorci G Riuzzi F Arcuri C Bianchi R Brozzi F Tubaro C Giambanco I S100B's double life: intracellular regulator and extracellular signal Biochim Biophys Acta 2009 1793 1008 1022 19110011 Dyck RH Van Eldik LJ Cynader MS Immunohistochemical localization of the S-100 beta protein in postnatal cat visual cortex: spatial and temporal patterns of expression in cortical and subcortical glia Brain Res Dev Brain Res 1993 72 181 192 8485842 Filosa JA Bonev AD Nelson MT Calcium dynamics in cortical astrocytes and arterioles during neurovascular coupling In Circ Res 2004 95 e73 e81 Filosa JA Morrison HW Iddings JA Du W Kim KJ Beyond neurovascular coupling, role of astrocytes in the regulation of vascular tone Neuroscience 2015 323 96 109 25843438 Franke H Verkhratsky A Burnstock G Illes P Pathophysiology of astroglial purinergic signalling Purinergic Signal 2012 8 629 657 22544529 Fukuda AM Badaut J Aquaporin 4: a player in cerebral edema and neuroinflammation J Neuroinflammation 2012 9 279 23270503 Gaberel T Gakuba C Goulay R Martinez De Lizarrondo S Hanouz JL Emery E Touze E Vivien D Gauberti M Impaired glymphatic perfusion after strokes revealed by contrast-enhanced MRI: a new target for fibrinolysis? Stroke 2014 45 3092 3096 25190438 Gelderblom M Leypoldt F Steinbach K Behrens D Choe CU Siler DA Arumugam TV Orthey E Gerloff C Tolosa E Magnus T Temporal and spatial dynamics of cerebral immune cell accumulation in stroke Stroke 2009 40 1849 1857 19265055 Go AS Mozaffarian D Roger VL Benjamin EJ Berry JD Blaha MJ Dai S Ford ES Fox CS Franco S Fullerton HJ Gillespie C Hailpern SM Heit JA Howard VJ Huffman MD Judd SE Kissela BM Kittner SJ Lackland DT Lichtman JH Lisabeth LD Mackey RH Magid DJ Marcus GM Marelli A Matchar DB McGuire DK Mohler ER 3rd Moy CS Mussolino ME Neumar RW Nichol G Pandey DK Paynter NP Reeves MJ Sorlie PD Stein J Towfighi A Turan TN Virani SS Wong ND Woo D Turner MB Executive summary: heart disease and stroke statistics--2014 update: a report from the American Heart Association Circulation 2014 129 399 410 24446411 Hansson E Ronnback L Glial neuronal signaling in the central nervous system FASEB J Vol 2003 17 341 348 Hinzman JM DiNapoli VA Mahoney EJ Gerhardt GA Hartings JA Spreading depolarizations mediate excitotoxicity in the development of acute cortical lesions Exp Neurol 2015 267 243 253 25819105 Hoyert DL XJ Deaths: Preliminary data for 2011 2012 61 Hyattsville, MD National vital statistics reports National Center for Health Statistics ed.^eds Iliff JJ Wang M Zeppenfeld DM Venkataraman A Plog BA Liao Y Deane R Nedergaard M Cerebral arterial pulsation drives paravascular CSF-interstitial fluid exchange in the murine brain J Neurosci 2013 33 18190 18199 24227727 Iliff JJ Chen MJ Plog BA Zeppenfeld DM Soltero M Yang L Singh I Deane R Nedergaard M Impairment of glymphatic pathway function promotes tau pathology after traumatic brain injury J Neurosci 2014 34 16180 16193 25471560 Isayama K Pitts LH Nishimura MC Evaluation of 2,3,5-triphenyltetrazolium chloride staining to delineate rat brain infarcts Stroke 1991 22 1394 1398 1721247 Iwabuchi S Kawahara K Oxygen-glucose deprivation-induced enhancement of extracellular ATP-P2Y purinoceptors signaling for the propagation of astrocytic calcium waves Biosystems 2009 96 35 43 19063934 Kelly MJ Ronnekleiv OK Control of CNS neuronal excitability by estrogens via membrane-initiated signaling Mol Cell Endocrinol 2009 308 17 25 19549588 Kettenmann H Hanisch UK Noda M Verkhratsky A Physiology of microglia Physiol Rev 2011 91 461 553 21527731 Kiechle K Bazarian JJ Merchant-Borna K Stoecklein V Rozen E Blyth B Huang JH Dayawansa S Kanz K Biberthaler P Subject-specific increases in serum S-100B distinguish sports-related concussion from sports-related exertion PLoS One 2014 9 e84977 24416325 Kim KJ Iddings JA Stern JE Blanco VM Croom D Kirov SA Filosa JA Astrocyte contributions to flow/pressure-evoked parenchymal arteriole vasoconstriction J Neurosci 2015 35 8245 8257 26019339 Kim WT Rioult MG Cornell-Bell AH Glutamate-induced calcium signaling in astrocytes Glia 1994 11 173 184 7927645 Kitchen P Day RE Taylor LH Salman MM Bill RM Conner MT Conner AC Identification and Molecular Mechanisms of the Rapid Tonicity-induced Relocalization of the Aquaporin 4 Channel J Biol Chem 2015 290 16873 16881 26013827 Kofuji P Newman EA Potassium buffering in the central nervous system Neuroscience 2004 129 1045 1056 15561419 Kress BT Iliff JJ Xia M Wang M Wei HS Zeppenfeld D Xie L Kang H Xu Q Liew JA Plog BA Ding F Deane R Nedergaard M Impairment of paravascular clearance pathways in the aging brain Ann Neurol 2014 76 845 861 25204284 Kuo J Hamid N Bondar G Dewing P Clarkson J Micevych P Sex differences in hypothalamic astrocyte response to estradiol stimulation Biol Sex Differ 2010 1 7 21208471 Lawson LJ Perry VH Dri P Gordon S Heterogeneity in the distribution and morphology of microglia in the normal adult mouse brain Neuroscience 1990 39 151 170 2089275 Lee Y Lee SR Choi SS Yeo HG Chang KT Lee HJ Therapeutically targeting neuroinflammation and microglia after acute ischemic stroke Biomed Res Int 2014 2014 297241 25089266 Liszczak TM Hedley-Whyte ET Adams JF Han DH Kolluri VS Vacanti FX Heros RC Zervas NT Limitations of tetrazolium salts in delineating infarcted brain Acta Neuropathol 1984 65 150 157 6084391 Liu H Qiu G Zhuo F Yu W Sun S Li F Yang M Lost Polarization of Aquaporin4 and Dystroglycan in the Core Lesion after Traumatic Brain Injury Suggests Functional Divergence in Evolution Biomed Res Int 2015 2015 471631 26583111 Liu W Tang Y Feng J Cross talk between activation of microglia and astrocytes in pathological conditions in the central nervous system Life Sci 2011 89 141 146 21684291 Madry C Attwell D Receptors, ion channels, and signaling mechanisms underlying microglial dynamics J Biol Chem 2015 290 12443 12450 25855789 McCarthy MM Arnold AP Ball GF Blaustein JD De Vries GJ Sex differences in the brain: the not so inconvenient truth J Neurosci 2012 32 2241 2247 22396398 Miller VM Kaplan JR Schork NJ Ouyang P Berga SL Wenger NK Shaw LJ Webb RC Mallampalli M Steiner M Taylor DA Merz CN Reckelhoff JF Strategies and methods to study sex differences in cardiovascular structure and function: a guide for basic scientists Biol Sex Differ 2011 2 14 22152231 Morrison HW Filosa JA A quantitative spatiotemporal analysis of microglia morphology during ischemic stroke and reperfusion J Neuroinflammation 2013 10 4 23311642 Mozaffarian D Benjamin EJ Go AS Arnett DK Blaha MJ Cushman M de Ferranti S Despres JP Fullerton HJ Howard VJ Huffman MD Judd SE Kissela BM Lackland DT Lichtman JH Lisabeth LD Liu S Mackey RH Matchar DB McGuire DK Mohler ER 3rd Moy CS Muntner P Mussolino ME Nasir K Neumar RW Nichol G Palaniappan L Pandey DK Reeves MJ Rodriguez CJ Sorlie PD Stein J Towfighi A Turan TN Virani SS Willey JZ Woo D Yeh RW Turner MB Heart disease and stroke statistics-2015 update: a report from the American Heart Association Circulation 2015 131 e29 e322 25520374 Murphy S Xu J Kochanek K Deaths: Final data for 2010 National vital statistics reports 2013 61 Hyattsville, MD National Center for Health Statistics ed.^eds Nedergaard M Dirnagl U Role of glial cells in cerebral ischemia Glia 2005 50 281 286 15846807 Nedergaard M Rodriguez JJ Verkhratsky A Glial calcium and diseases of the nervous system In Cell Calcium 2010 47 140 149 Pascual O Ben Achour S Rostaing P Triller A Bessis A Microglia activation triggers astrocyte-mediated modulation of excitatory neurotransmission In Proc Natl Acad Sci 2012 109 197 205 Perego C Fumagalli S De Simoni MG Temporal pattern of expression and colocalization of microglia/macrophage phenotype markers following brain ischemic injury in mice J Neuroinflammation 2011 8 174 22152337 Pham N Fazio V Cucullo L Teng Q Biberthaler P Bazarian JJ Janigro D Extracranial sources of S100B do not affect serum levels PLoS One 2010 5 Plog BA Dashnaw ML Hitomi E Peng W Liao Y Lou N Deane R Nedergaard M Biomarkers of traumatic injury are transported from brain to blood via the glymphatic system J Neurosci 2015 35 518 526 25589747 Potokar M Stenovec M Jorgacevski J Holen T Kreft M Ottersen OP Zorec R Regulation of AQP4 surface expression via vesicle mobility in astrocytes Glia 2013 61 917 928 23505074 Reeves MJ Bushnell CD Howard G Gargano JW Duncan PW Lynch G Khatiwoda A Lisabeth L Sex differences in stroke: epidemiology, clinical presentation, medical care, and outcomes Lancet Neurol 2008 7 915 926 18722812 Ren Z Iliff JJ Yang L Yang J Chen X Chen MJ Giese RN Wang B Shi X Nedergaard M 'Hit & Run' model of closed-skull traumatic brain injury (TBI) reveals complex patterns of post-traumatic AQP4 dysregulation J Cereb Blood Flow Metab 2013 33 834 845 23443171 Ribeiro Mde C Hirt L Bogousslavsky J Regli L Badaut J Time course of aquaporin expression after transient focal cerebral ischemia in mice J Neurosci Res 2006 83 1231 1240 16511868 Ritzel RM Capozzi LA McCullough LD Sex, stroke, and inflammation: the potential for estrogen-mediated immunoprotection in stroke Horm Behav 2013 63 238 253 22561337 Shin JA Choi JH Choi YH Park EM Conserved aquaporin 4 levels associated with reduction of brain edema are mediated by estrogen in the ischemic brain after experimental stroke Biochim Biophys Acta 2011 1812 1154 1163 21641993 Srinivasan R Huang BS Venugopal S Johnston AD Chai H Zeng H Golshani P Khakh BS Ca(2+) signaling in astrocytes from Ip3r2(−/−) mice in brain slices and during startle responses in vivo Nat Neurosci 2015 18 708 717 25894291 Takano T Oberheim N Cotrina ML Nedergaard M Astrocytes and ischemic injury In Stroke 2009 40 United States S8 S12 ed.^eds Thrane AS Rappold PM Fujita T Torres A Bekar LK Takano T Peng W Wang F Rangroo Thrane V Enger R Haj-Yasein NN Skare O Holen T Klungland A Ottersen OP Nedergaard M Nagelhus EA Critical role of aquaporin-4 (AQP4) in astrocytic Ca2+ signaling events elicited by cerebral edema Proc Natl Acad Sci 2011 108 846 851 21187412 Vella J Zammit C Di Giovanni G Muscat R Valentino M The central role of aquaporins in the pathophysiology of ischemic stroke Front Cell Neurosci 2015 9 108 25904843 Wake H Moorhouse AJ Jinno S Kohsaka S Nabekura J Resting microglia directly monitor the functional state of synapses in vivo and determine the fate of ischemic terminals In J Neurosci 2009 29 3974 3980 Wang M Iliff JJ Liao Y Chen MJ Shinseki MS Venkataraman A Cheung J Wang W Nedergaard M Cognitive deficits and delayed neuronal loss in a mouse model of multiple microinfarcts J Neurosci 2012 32 17948 17960 23238711 Zhang L Liu W Alizadeh D Zhao D Farrukh O Lin J Badie SA Badie B S100B attenuates microglia activation in gliomas: possible role of STAT3 pathway Glia 2011 59 486 498 21264954 Zheng W Watts LT Holstein DM Prajapati SI Keller C Grass EH Walter CA Lechleiter JD Purinergic receptor stimulation reduces cytotoxic edema and brain infarcts in mouse induced by photothrombosis by energizing glial mitochondria PLoS One 2010 5 e14401 21203502 Zheng W Talley Watts L Holstein DM Wewer J Lechleiter JD P2Y1R-initiated, IP3R-dependent stimulation of astrocyte mitochondrial metabolism reduces and partially reverses ischemic neuronal damage in mouse J Cereb Blood Flow Metab 2013 33 600 611 23321785

PMC005xxxxxx/PMC5118182.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8902541 3967 Head Neck Head Neck Head & neck 1043-3074 1097-0347 27265898 5118182 10.1002/hed.24519 NIHMS812089 Article Effect of surgical intervention on circulating tumor cells in patients with squamous cell carcinoma of the head and neck using a negative enrichment technology Jatana Kris R. MD 12* Balasubramanian Priya PhD 3 McMullen Kyle P. MD 2 Lang Jas C. PhD 2 Teknos Theodoros N. MD 2 Chalmers Jeffrey J. PhD 3 1 Department of Pediatric Otolaryngology – Head and Neck Surgery, Nationwide Children's Hospital, Columbus, Ohio 2 Department of Otolaryngology – Head and Neck Surgery, Wexner Medical Center at Ohio State University, Columbus, Ohio 3 William G. Lowrie Department of Chemical and Biomolecular Engineering, Ohio State University, Columbus, Ohio. * Corresponding author: K. R. Jatana, Department of Otolaryngology – Head and Neck Surgery, Nationwide Children's Hospital, 555 South 18th Street, Suite 2A, Columbus, Ohio 43205. Kris.Jatana@osumc.edu 1 9 2016 5 6 2016 12 2016 01 12 2016 38 12 17991803 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background The purpose of this study was to investigate the impact of surgical intervention on detection of circulating tumor cells (CTCs) in patients with squamous cell carcinoma of the head and neck (SCCHN.) Methods We utilized a negative depletion technique to identify cytokeratin (CK)-positive CTCs. The numbers of CTCs immediately before and after surgical resection were compared. Results Seventy-six blood samples from 38 patients with SCCHN were examined. Seventy-nine percent of the patients had CTCs detected before and after surgery. A total of 7.89% had no CTCs before surgery, yet had CTCs identified after surgery. Overall, 60.5% of patients had an increased number of CTCs/mL after surgery with a mean increase of 6.63-fold. A statistically significant increase in CTCs was seen after surgery (p = .02). Conclusion The timing of sample collection in patients with SCCHN who have surgical intervention can potentially impact the number of CTCs identified. circulating tumor cell squamous cell carcinoma surgery immunomagnetic separation INTRODUCTION Cancer of the upper aerodigestive system is estimated to have 59,340 cases in the United States in 2015 and comprise 3.5% of all cancers.1 Squamous cell carcinoma of the head and neck (SCCHN) makes up approximately 95% of these tumors. Despite medical advances, the overall 5-year survival rate of approximately 50% has not changed significantly over the last several decades.2–4 Studies in other solid cancers, such as breast,5 prostate,6 lung,7,8 and colon cancers,9 have shown that the presence of circulating tumor cells (CTCs) that disseminates from the primary tumor has prognostic significance, and this dissemination may be an important step in hematogenous metastasis. We have demonstrated that CTCs can be detected in the peripheral blood of patients with SCCHN and the presence of CTCs correlated with reduced short-term disease-free survival.10 Before our studies, Partridge et al11 used a similar negative depletion technology and found that detection of disseminated tumor cells in patients with SCCHN correlated with a higher risk of local/distant recurrence as well as reduced survival.11 In inoperable SCCHN, the presence of CTCs has been linked to regional lymph nodal staging of 2b or higher.12 To date, the presence of lymph node metastasis in SCCHN is the most predictive prognostic factor that impacts survival.13,14 Beyond basic enumeration of CTCs in a patient's blood, it has also been reported that CTCs may appear in the blood stream because of mechanical manipulation of the tumor during surgical intervention,15–18 fine-needle aspirations,19 during colonoscopy,20 and endorectal ultrasound.21 The release of these cells is a potential concern on multiple levels, the least of which is influencing traditional CTC counts. As far as we know, CTC counts before and after open surgical resection in the operating room has not been performed in SCCHN. Our previously published SCCHN studies have used blood samples taken after surgical intervention. The head and neck region is a highly vascularized tissue. A released cell, such as a CTC, potentially could enter circulation after physical manipulation during surgery, through lymphatics or veins that return blood to the superior vena cava to the right atrium and subsequently the ventricle of the heart. From the right side of the heart, this blood circulates to the lungs for oxygenation through a capillary network and then returns to the heart for pumping into arterial circulation where distal end organs again contain a capillary network before entering venous circulation and returning to the heart. A common assumption of CTC is that they are relatively large compared to normal peripheral blood cells; several methodologies to identify CTCs are based on size alone.22,23 With head and neck surgery, the first small vasculature that such a proposed released cell would encounter would be the capillary network in the lungs. It should be noted, however, as we have reported previously, that the relative size of the putative CTCs in our previous studies are similar in size to normal peripheral blood cells. Such cells may continue to repeatedly circulate around the body. MATERIALS AND METHODS This study was approved by the Institutional Review Board. Informed consent was obtained from patients who were undergoing surgical intervention at the Arthur G. James Cancer Hospital and Solove Research Institute and Comprehensive Cancer Center at The Ohio State University. Inclusion criteria were adults >18 years with histologically confirmed SCCHN undergoing an open surgical excision of the primary tumor. Exclusion criteria included patients with known metastases and any patient who received a blood transfusion intraoperatively. Peripheral venous blood samples were procured from 38 patients (total of 76 samples.) Five to 10 mL were collected in green-top BD Vacutainer tubes immediately before and after surgery while in the operating room. Each venous sample was collected from an extremity (arm or leg) where the intravenous fluids were not being administered and the initial 2 to 3 mL was drawn off and discarded to prevent contamination. A fresh syringe was then used to obtain the sample. The samples were stored at 4°C and were processed within 24 hours. Operators were completely blinded to the time point of the samples as well as the clinical and correlative information during the sample processing and CTC enumeration steps. The CTC enrichment process has been previously discussed in detail.24 Briefly, the red blood cells were lysed using a lysis buffer and the remaining leukocytes were immunomagnetically labeled using anti-CD45 tetrameric antibody complex and dextran-coated magnetic nanoparticles (cat #18259; Stem Cell Tech, Vancouver, BC). The magnetically labeled cells were then passed through our optimized immunomagnetic separation system, to deplete a majority of the leukocytes. An aliquot of the enriched sample containing the CTCs was used for CTC enumeration. Cell suspension containing CTCs was stored in 10% neutral buffered formalin until further processing. Cells were cytospun onto a microscopic slide using a Shandon Cytospin instrument at 1800 rpm for 5 minutes. The cytospun slides were hydrated in phosphate-buffered saline before staining with anti-cytokeratin fluorescein isothiocyanate (FITC) antibody (1:10, Miltenyi Biotech, clone:CK3-6H5, 130-080-101) targeting cytokeratins 8, 18, and 19 for 30 minutes at 37°C before mounting with mounting media containing 4′,6′-diamidino-2-phenylindole (DAPI; Vector Laboratories, H-1500). Slides were observed under a Nikon 80i epifluorescent microscope equipped with band pass filters for diamidino-phenylindole (DAPI) and FITC emission, and cells that were positive for both FITC and DAPI and had high nuclear to cytoplasmic ratios were counted as CTCs. Images of the slides were taken using a Zeiss LSM510 confocal microscope. The number of CTCs per mL of peripheral blood was calculated. Blood samples were drawn from healthy volunteers or patients with benign conditions and processed in the same manner as a clinical sample. Statistical analyses were performed using GraphPad Prism version 6 (San Diego, CA). The D'Agostino and Pearson omnibus normality test was performed to test if the CTC counts before or after surgery were normally distributed. Wilcoxon matched-pair sign rank test was used to determine if there was a statistically significant difference between CTC numbers before and after surgery. Any p value ≤ .05 was considered statistically significant. RESULTS A total of 38 patients with SCCHN (mean age, 64.2 years; range, 51–83 years) who all underwent open surgical excision in the operating room, had their venous blood examined for CTCs. Men comprised 63% and women comprised 37% of the patients. The sites were 62% in the oral cavity, 15% in the oropharynx, and 23% in the larynx. Overall staging included: 14% stage 1; 20% stage 2; 26% stage 3; and 40% stage 4. Figures 1 and 2 display examples of enriched patient samples from a patient in which no change in putative CTC concentration was observed, and a second patient in which a large increase in CTC concentration was observed. Overall, as shown in Table 1, CTCs were detected in 79.0% of patients before and after surgery. It was noted that 7.89% of patients had detected CTCs after surgical intervention when none were detected in the before sample. An additional 10.5% of patients had CTCs detected in sample before surgery, but not in the after surgery sample. The D'Agostino & Pearson omnibus normality test showed that the data were not normally distributed (p value < .0001 for both presurgery and postsurgery). Because the data were not normally distributed, a nonparametric Wilcoxon matched-pair sign rank test was performed to test for statistical significance. There was a significant difference in CTC counts using matched pairs analysis as a result of surgical intervention (2-tailed p value = .0224). In particular, CTCs measured soon after surgery were significantly higher than CTCs measured before surgery (1-tailed p value = .0112). There was no correlation to age, sex, or overall stage of cancer. There was no correlation between CTCs and positive margins, disruption of tumor or aerodigestive tract, or length of surgery. It was found that 60.5% of patients had an increased number of CTCs after surgical intervention (see Figure 3). The range of CTCs in the peripheral blood before/after surgical intervention from oral cavity, oropharyngeal, and laryngeal sites is shown in Figure 4. None of the healthy control blood samples had similar, brightly stained, cytokeratin-positive cells, or cells with any morphology resembling CTCs. DISCUSSION There has been limited data on the impact of an intervention on SCCHN with regard to CTCs. Ramani et al25 looked at the presence of cytokeratin (CK)-19 by reverse transcriptase-polymerase chain reaction before and after incisional biopsy of oral cavity squamous cell carcinoma. They found that, in the 10 patients studied, there was no dissemination of cancer cells after incisional biopsy in these patients.25 A prior study by Kusukawa et al26 demonstrated the detection of CK-19 reverse transcriptase-polymerase chain reaction in this same tumor site, and reported that dissemination of cancer cells into circulation does occur after an incisional biopsy. Although both of these studies looked at the presence of mRNA alone, neither had visual conformation of CTCs. Studies by Ohtake et al27 have shown that an incision has caused increased lymph node metastasis in dimethylbenz[a]anthracene-induced tongue cancer in the hamster model; the incidence of regional lymph node metastasis increased by 65.9% after repeated incisions of the lesion. It is not clear what the clinical relevance of such findings is at this time. It has been suggested that CTCs may be destroyed in the microvascular capillary networks by the host immune surveillance system, including macrophages and natural killer cells.28 We have identified CTCs in the blood from the distal extremity venous system with patients with SCCHN undergoing open surgical resection +/− cervical lymphadenectomy, meaning that these CTCs have gone through a minimum of 2 capillary networks (likely circulating numerous times) before detection. The open surgical procedures ranged from approximately 90 minutes to 6 hours duration. As shown in Figures 1 and 2, clumps of CTCs are seen postsurgery, but these aggregates likely would form intravascularly after passing through capillary circulation. The mechanisms by which this happens is not yet elucidated, but the CTCs in SCCHN are likely of similar size to circulating leukocytes. Some have reported that CTCs may be only transiently detectable, and that it is a dynamic and periodic event.29 There are several unanswered questions with regard to timing of blood sample collection when any tumor-related intervention is performed, the location of venous blood sample collection with regard to the tumor location, and what clinical relevance exists. Each solid cancer should be individually investigated as the specific protocols of sample collection with invasive procedures may be critical and influence CTC detection. This is important as most published studies with CTCs in all solid cancers have not addressed this issue in the literature. Even within a specific type of cancer, like SCCHN, the tumor location may also influence identification of CTCs, as shown in Figure 4. Another important concept to consider with respect to the concentration of these rare cells is the intravascular fluid balance in patients under general anesthesia during these surgical procedures. There is always an estimated blood loss that occurs. None of the patients in this study received any blood transfusion; however, all surgical patients had a positive fluid balance, indicating they received additional intravenous fluid (directly into a distal vein, unrelated to the location of the blood sample taken for this study), which increases blood volume several fold more than the estimated blood loss, resulting in hemodilution. Because of dilution of the blood, the concentration of CTCs/mL blood may be underestimated. We have previously demonstrated some of the cells we are reporting as putative CTC in this study are also epithelial cell adhesion molecule negative, cytokeratin and vimentin positive, and can be further positive for epidermal growth factor receptor or CD44.30 At this point, it is unclear what, if any, clinical significance can be associated with these changes in concentrations of putative CTC after surgery. Epithelial to mesenchymal transition may occur as cancer cells acquire a more aggressive migratory phenotype.31–34 It is possible that these surgically released CTCs are inherently different than CTCs that enter the intravascular space naturally. None of the 7.89% of patients who had CTCs detected postoperatively, not preoperatively, developed locoregional disease recurrence or distant disease recurrence. Further characterization of these CTCs identified before and after surgery, with multimarker analysis in combination with long-term clinical follow-up may yield a greater understanding of the role of surgically released CTCs in SCCHN. CONCLUSIONS A statistically significant increase in the CTC counts immediately after surgical intervention, while still in the operating room, was seen in patients with SCCHN. The timing of blood sample collection for such solid cancers that undergo surgical intervention, such as SCCHN, can potentially impact the number of CTCs identified. Although a prognostic blood test for CTCs could have important treatment and surveillance implications, the viability and clinical significance of potentially surgically released CTCs in SCCHN is still not known; it is possible that these cells have a different phenotype than CTCs that enter the intravascular space by other mechanisms. Contract grant sponsor: This work was supported by the following agencies: the National Science Foundation (BES-0124897); the National Cancer Institute (R01 CA97391-01A1); the State of Ohio Third Frontier Program (ODOD 26140000:TECH 07-001); the National Cancer Institute CCC Core Grant (P30 CA16058), and the National Science Foundation (EEC 0425626). FIGURE 1 Representative patient with squamous cell carcinoma of the head and neck (SCCHN) sample with no change in circulating tumor cells/mL peripheral blood post-surgery (a) compared to pre-surgery (b), as determined by dual color immunofluorescent staining: green = cytokeratin; blue = diamidino-phenylindole. FIGURE 2 Representative patient with squamous cell carcinoma of the head and neck (SCCHN) sample demonstrating a 2.5-fold change in circulating tumor cells/mL peripheral blood post-surgery (a) compared to pre-surgery (b), as determined by dual color immunofluorescent staining: green = cytokeratin; blue = diamidino-phenylindole. FIGURE 3 Ladder plots displaying the number and degree of circulating tumor cell increase (left) and decrease (right) in pre-surgical and post-surgical specimens. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com] FIGURE 4 Changes in circulating tumor cells (CTCs)/mL of peripheral blood before and after surgical intervention by general site of tumor (L = larynx; OC = oral cavity; OP, oropharynx). Each data point, minimum and maximum, is shown for these sites. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com] TABLE 1 Correlation between circulating tumor cell detection and the time of distal venous blood draw (presurgery or postsurgery) in 38 patients undergoing surgical intervention for squamous cell carcinoma of the head and neck. Before surgical intervention After surgical intervention Overall frequency Negative Negative 1/38 (2.6%) Negative Positive 3/38 (7.89%) Positive Negative 4/38 (10.5%) Positive Positive 30 (79.0%) 34/38 (89.5%) positive 33/38 (86.8%) positive REFERENCES 1 Siegel RL Miller KD Jemal A Cancer statistics, 2015. CA Cancer J Clin 2015 65 5 29 25559415 2 Gath HJ Brakenhoff RH Minimal residual disease in head and neck cancer. Cancer Metastasis Rev 1999 18 109 126 10505550 3 Cohen EE Lingen MW Vokes EE The expanding role of systemic therapy in head and neck cancer. J Clin Oncol 2004 22 1743 1752 15117998 4 Cooper JS Pajak TF Forastiere AA Postoperative concurrent radio-therapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 2004 350 1937 1944 15128893 5 Braun S Vogl FD Naume B A pooled analysis of bone marrow micrometastasis in breast cancer. N Engl J Med 2005 353 793 802 16120859 6 Köllermann J Weikert S Schostak M Prognostic significance of disseminated tumor cells in the bone marrow of prostate cancer patients treated with neoadjuvant hormone treatment. J Clin Oncol 2008 26 4928 4933 18794550 7 Kasimir-Bauer S Schleucher N Weber R Neumann R Seeber S Evaluation of different markers in non-small cell lung cancer: prognostic value of clinical staging, tumour cell detection and tumour marker analysis for tumour progression and overall survival. Oncol Rep 2003 10 475 482 12579292 8 Pantel K Izbicki J Passlick B Frequency and prognostic significance of isolated tumour cells in bone marrow of patients with non-small-cell lung cancer without overt metastases. Lancet 1996 347 649 653 8596379 9 Leinung S Würl P Schönfelder A Weiss CL Röder I Schönfelder M Detection of cytokeratin-positive cells in bone marrow in breast cancer and colorectal carcinoma in comparison with other factors of prognosis. J Hematother Stem Cell Res 2000 9 905 911 11177604 10 Jatana KR Balasubramanian P Lang JC Significance of circulating tumor cells in patients with squamous cell carcinoma of the head and neck: initial results. Arch Otolaryngol Head Neck Surg 2010 136 1274 1279 21173379 11 Partridge M Brakenhoff R Phillips E Detection of rare disseminated tumor cells identifies head and neck cancer patients at risk of treatment failure. Clin Cancer Res 2003 9 5287 5294 14614011 12 Hristozova T Konschak R Stromberger C The presence of circulating tumor cells (CTCs) correlates with lymph node metastasis in nonresectable squamous cell carcinoma of the head and neck region (SCCHN). Ann Oncol 2011 22 1878 1885 21525401 13 Jones AS Roland NJ Field JK Phillips DE The level of cervical lymph node metastases: their prognostic relevance and relationship with head and neck squamous carcinoma primary sites. Clin Otolaryngol Allied Sci 1994 19 63 69 8174305 14 Leemans CR Tiwari R Nauta JJ van der Waal I Snow GB Regional lymph node involvement and its significance in the development of distant metastases in head and neck carcinoma. Cancer 1993 71 452 456 8422638 15 Sawabata N Okumura M Utsumi T Circulating tumor cells in peripheral blood caused by surgical manipulation of non-small-cell lung cancer: pilot study using an immunocytology method. Gen Thorac Cardiovasc Surg 2007 55 189 192 17554991 16 Healy WL Pfeifer BA Kurtz SR Evaluation of autologous shed blood for autotransfusion after orthopaedic surgery. Clin Orthop Relat Res 1994 299 53 59 17 Hansen E Wolff N Knuechel R Ruschoff J Hofstaedter F Taeger K Tumor cells in blood shed from the surgical field. Arch Surg 1995 130 387 393 7710337 18 Papavasiliou P Fisher T Kuhn J Nemunaitis J Lamont J Circulating tumor cells in patients undergoing surgery for hepatic metastases from colorectal cancer. Proc (Bayl Univ Med Cent) 2010 23 11 14 20157496 19 Hu XC Chow LW Fine needle aspiration may shed breast cells into peripheral blood as determined by RT-PCR. Oncology 2000 59 217 222 11053989 20 Koch M Kienle P Sauer P Hematogenous tumor cell dissemination during colonoscopy for colorectal cancer. Surg Endosc 2004 18 587 591 14735340 21 Koch M Antolovic D Kienle P Increased detection rate and potential prognostic impact of disseminated tumor cells in patients undergoing endorectal ultrasound for rectal cancer. Int J Colorectal Dis 2007 22 359 365 16758164 22 Vona G Sabile A Louha M Isolation by size of epithelial tumor cells: a new method for the immunomorphological and molecular characterization of circulating tumor cells. Am J Pathol 2000 156 57 63 10623654 23 De Giorgi V Pinzani P Salvianti F Application of a filtration- and isolation-by-size technique for the detection of circulating tumor cells in cutaneous melanoma. J Invest Dermatol 2010 130 2440 2447 20535130 24 Yang L Lang JC Balasubramanian P Optimization of an enrichment process for circulating tumor cells from the blood of head and neck cancer patients through depletion of normal cells. Biotechnol Bioeng 2009 102 521 534 18726961 25 Ramani P Thomas G Ahmed S Use of Rt-PCR in detecting disseminated cancer cells after incisional biopsy among oral squamous cell carcinoma patients. J Cancer Res Ther 2005 1 92 97 17998634 26 Kusukawa J Suefuji Y Ryu F Noguchi R Iwamoto O Kameyama T Dissemination of cancer cells into circulation occurs by incisional biopsy of oral squamous cell carcinoma. J Oral Pathol Med 2000 29 303 307 10947245 27 Ohtake K Shingaki S Nakajima T Effects of incision and irradiation on regional lymph node metastasis in carcinoma of the hamster tongue. Oral Surg Oral Med Oral Pathol 1990 70 62 69 2115153 28 Hanna N The role of natural killer cells in the control of tumor growth and metastasis. Biochim Biophys Acta 1985 780 213 226 3896313 29 Romsdahl MD Chu EW Hume R Smith RR The time of metastasis and release of circulating tumor cells as determined in an experimental system. Cancer 1961 14 883 888 13743092 30 Balasubramanian P Lang JC Jatana KR Multiparameter analysis, including EMT markers, on negatively enriched blood samples from patients with squamous cell carcinoma of the head and neck. PLoS One 2012 7 e42048 22844540 31 Berx G Rasp e E Christofori G Thiery JP Sleeman JP Pre-EMTing metastasis? Recapitulation of morphogenetic processes in cancer. Clin Exp Metastasis 2007 24 587 597 17978854 32 Chaffer CL Thompson EW Williams ED Mesenchymal to epithelial transition in development and disease. Cells Tissues Organs 2007 185 7 19 17587803 33 Mani SA Guo W Liao MJ The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 2008 133 704 715 18485877 34 Sarriö D Rodriguez-Pinilla SM Hardisson D Cano A Moreno-Bueno G Palacios J Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype. Cancer Res 2008 68 989 997 18281472

PMC005xxxxxx/PMC5118191.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101129617 30409 Genes Brain Behav Genes Brain Behav. Genes, brain, and behavior 1601-1848 1601-183X 27489246 5118191 10.1111/gbb.12314 NIHMS807658 Article Functional conservation of MBD proteins: MeCP2 and Drosophila MBD proteins alter sleep Gupta Tarun 1 Morgan Hannah R. 2 Bailey Jessica A. 2 Certel Sarah J. 12* 1 Neuroscience Graduate Program, The University of Montana, Missoula, MT 2 Division of Biological Sciences and Center for Structural and Functional Neuroscience, The University of Montana, Missoula, MT 59812 * Corresponding author: sarah.certel@umontana.edu 12 8 2016 6 9 2016 11 2016 01 11 2017 15 8 757774 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Proteins containing a methyl-CpG-binding domain (MBD) bind 5mC and convert the methylation pattern information into appropriate functional cellular states. The correct readout of epigenetic marks is of particular importance in the nervous system where abnormal expression or compromised MBD protein function, can lead to disease and developmental disorders. Recent evidence indicates the genome of Drosophila melanogaster is methylated and two MBD proteins, dMBD2/3 and dMBD-R2, are present. Are Drosophila MBD proteins required for neuronal function, and as MBD-containing proteins have diverged and evolved, does the MBD domain retain the molecular properties required for conserved cellular function across species? To address these questions, we expressed the human MBD-containing protein, hMeCP2, in distinct amine neurons and quantified functional changes in sleep circuitry output using a high throughput assay in Drosophila. hMeCP2 expression resulted in phase-specific sleep loss and sleep fragmentation with the hMeCP2-mediated sleep deficits requiring an intact MBD-domain. Reducing endogenous dMBD2/3 and dMBD-R2 levels also generated sleep fragmentation, with an increase in sleep occurring upon dMBD-R2 reduction. To examine if hMeCP2 and dMBD-R2 are targeting common neuronal functions, we reduced dMBD-R2 levels in combination with hMeCP2 expression and observed a complete rescue of sleep deficits. Furthermore, chromosomal binding experiments indicate MBD-R2 and MeCP2 associate on shared genomic loci. Our results provide the first demonstration that Drosophila MBD-containing family members are required for neuronal function and suggest the MBD domain retains considerable functional conservation at the whole organism level across species. methyl-CpG Binding Protein 2 (MeCP2) MBD proteins Drosophila sleep octopamine methylation Introduction Gene expression and even more fundamentally, DNA architecture, is controlled by chemical modifications on histone proteins and DNA. In plants, vertebrates and more recently Drosophila, the chemical mark is an added methyl group at position 5 of cytosine (5mC) (Capuano et al., 2014, Gehring, 2013, Schubeler, 2015, Takayama et al., 2014, Varriale, 2014, Zilberman, 2008). Most methyl-CpG binding domain (MBD)-containing proteins bind methylated DNA and function to translate the chemical modification into appropriate cellular states (Bogdanovic & Veenstra, 2009, Fatemi & Wade, 2006, Sasai & Defossez, 2009). By interacting with diverse partners, MBD-containing proteins regulate the differentiation and function of a cell by maintaining or altering chromatin structure, interpreting genomic imprinting, gene-specific transcriptional activation/repression and controlling RNA splicing (Chahrour & Zoghbi, 2007, Lyst & Bird, 2015, Samaco & Neul, 2011). Due to this wide array of nuclear functions, MBD-containing proteins and in particular, the MBD family member, methyl-CpG-binding protein 2 (MeCP2), have been described as genome-wide modulators of gene expression and cellular differentiation (Cohen et al., 2011, Della Ragione et al., 2012, Skene et al., 2010, Yasui et al., 2013). Alterations in MeCP2 levels, either through loss-of-function mutations or gene duplication, results in the postnatal neurodevelopmental disorders, Rett Syndrome (RTT) and MeCP2 duplication syndrome. MeCP2 dysregulation is also an important component of neuropsychiatric and neurological disorders ranging from Alzheimer’s and Huntington’s to depression and drug addiction (Ausio et al., 2014, Hutchinson et al., 2012, Lv et al., 2013, Ramocki et al., 2009, Zimmermann et al., 2015). One prevalent phenotype among children with alterations in MeCP2 function and a common feature of neurodegenerative disease and neuropsychiatric disorders is sleep abnormalities (Angriman et al., 2015, Kakkar & Dahiya, 2015, Mccarthy & Welsh, 2012, Musiek et al., 2015). Such sleep impairments include delays in the onset of sleep, alterations in total sleep, and frequent wakings resulting in fragmented sleep (Cortesi et al., 2010, Nomura, 2005, Piazza et al., 1990, Souders et al., 2009, Young et al., 2007). Recently, it has become increasingly clear that epigenetic factors play fundamental roles in transcriptional and post-transcriptional regulation within the circadian clock network (Liu & Chung, 2015, Qureshi & Mehler, 2014). For example, mouse studies indicate day length changes alter promoter DNA methylation within the SCN and in humans methylation levels display a 24-hr rhythmicity (Angriman et al., 2015, Kakkar & Dahiya, 2015) (Azzi et al., 2014). In Drosophila, many oscillatory transcripts including several non-coding RNAs have been identified, (Hughes et al., 2012), in mice two miRNAs, miR-134 and miR-132, are implicated in circadian regulation, MeCP2 regulates miR-134 processing in the brain, and finally the levels of MeCP2 are regulated during the circadian cycle (Alvarez-Saavedra et al., 2011, Cheng et al., 2014, Gao et al., 2010) Therefore, as sleep is a relevant behavior at the molecular and phenotypic level, we used the well-characterized sleep paradigm in Drosophila to ask if MBD-containing protein family members from different species retain the functional conservation required for neuronal output. Sleep and arousal are regulated by multiple neurotransmitters including octopamine, dopamine, γ-aminobutyric acid (GABA), and serotonin (5HT) through different but interacting circuits (Cirelli, 2009, Crocker & Sehgal, 2010, Potdar & Sheeba, 2013). Due to the defined role of octopamine (OA, the invertebrate homolog of mammalian noradrenaline) in promoting wakefulness, we manipulated the OA system through a series of experiments. First, we expressed human MeCP2 (hMeCP2, the e2 isoform) and examined the cell-type specific effects on OA neuron function. Human MeCP2 was selected due to the extensive characterization of its MBD domain, the availability of a series of hMeCP2 alleles capable of conditional expression in Drosophila, and the opportunity to examine if MBD domain function is conserved from fly to human. We found that hMeCP2 expression in Drosophila leads to phase-specific sleep deficits, increased sleep latency, and sleep fragmentation. To separate the role of disrupted amine production versus disrupted neuron function, we expressed hMeCP2 in OA neurons that lacked OA and established that a portion of the nighttime sleep reduction is dependent on amine function. Next by using multiple MeCP2 alleles including the RTT-causing allele, MeCP2R106W, we determined the MBD-domain is required for the MeCP2-mediated sleep deficits. Second, as the Drosophila genome contains two proteins with extended homologies to vertebrate MBD family members and the recent confirmation of cytosine methylation in Drosophila, we asked if reducing dMBD2/3 and dMBD-R2 could also change the function of OA neurons. As with hMeCP2 expression, sleep fragmentation occurred upon dMBD2/3 and dMBD-R2 manipulation. If OA neuron function is altered due to the targeting of similar or the same genomic areas by hMeCP2 and the endogenous MBD proteins, then reducing dMBD2/3 or dMBD-R2 in conjunction with hMeCP2 expression should decrease or eliminate the hMeCP2-mediated sleep deficits. Our results indicate the phase-specific sleep deficits that occur due to hMeCP2 are rescued with a concomitant reduction in dMBD-R2 and by labeling 3rd instar larval polytene chromosomes, we found that hMeCP2 and dMBD-R2 accumulate together at distinct chromosomal bands. Finally, increasing the expression of dMBD2/3 in OA neurons decreased phase-specific sleep levels in the same manner as hMeCP2 expression. Taken together, our results provide the first demonstration that a reduction in Drosophila MBD-containing proteins can alter neuron output and indicate conservation in the cell-specific functions of epigenetic translators. Materials and methods Drosophila Stocks Canton-S, UAS-Red Stinger (BL 8545, BL 8546), UAS-mCD8:GFP (BL 5130), UAS-MBD-R2-IR (BL 30481), UAS-dMBD2/3-IR (BL 35347), UAS-EP1112 (BL16987), and UAS-EP04582 (BL15753) were obtained from the Bloomington Stock Center (Bloomington, IN). Tested EP lines obtained from the Kyoto Stock Center include DGRC #’s 201-693, 204-320, and 206-139. The UAS-MeCP2, UAS-MeCP2R294X, UAS-MeCP2R106W, and UAS-MeCP2Δ166 lines were generously provided by Juan Botas (Cukier et al., 2008). dTdc2-Gal4 was obtained from Jay Hirsh (Cole et al., 2005), th-Gal4 was provided by Sirge Birman (Friggi-Grelin et al., 2003), and trh-Gal4 was a gift from Olga Alekseenko (Alekseyenko et al., 2010). Husbandry All fly stocks were maintained in a temperature (25 °C) and humidity-controlled (~50%) environment on a standard cornmeal based medium (agar, cornmeal, sugar, yeast extract, Triton-X). During development and post-eclosion, all flies were entrained to standard 12hr-12hr light:dark (L:D) conditions under 1400 ± 200 lx fluorescent light intensity. Transgenic control males were generated by crossing Canton S females with males from the respective UAS- or gal4-lines. Before experimentation, male pupae were isolated and aged individually in 16X100mm borosilicate glass tubes containing standard food medium described above. Behavioral Analysis For activity and sleep monitoring, 2–3 day old socially naive males were transferred to 65×5mm glass tubes with 15mm food on one end and a cotton plug on the other. Flies were transferred under CO2 anesthesia and allowed 24-hr to recuperate and acclimatize to new housing conditions before data collection. The locomotor activity counts were recorded for both control and experimental males using Drosophila Activity Monitoring (DAM) system (Trikinetics, Waltham, MA) for a period of 10 consecutive days at 1-min bin acquisition mode. Count data for the first and the last day were removed to minimize mechanical noise. Data from 8 consecutive days was analyzed further using Counting Macro 5.19.5 (CM) program generously provided by R. Allada (Northwestern University, Evanston, IL). Various indices of sleep including temporal organization, duration and latency of sleep and the number and length of sleep bouts were analyzed as described previously (Pfeiffenberger et al., 2010). Sleep was defined as complete inactivity for a period of 5 consecutive minutes (Shaw et al., 2000). Graphs were generated with Graphpad Prism and Adobe Illustrator CS5. Immunohistochemistry and Imaging Adult male brains were dissected and fixed in 4% paraformaldehyde (Electron Microscopy Sciences) for 40 minutes and labeled as described previously (Certel et al., 2010). The following primary antibodies were used: rabbit anti-MeCP2 (1:30, Cell Signaling Technologies), mouse anti-MeCP2 (1:500, Abcam), rat anti-CD8 (1:100, Molecular Probes), monoclonal rabbit anti-GFP (1:200, Molecular Probes), mouse nc82 (1:100) and anti-MBD-R2 (1:200) (Prestel et al., 2010). Secondary antibodies include Alexa Fluor 488-conjugated donkey anti-mouse, Alexa Fluor 594-conjugated goat anti-rabbit, Alexa Fluor 647-conjugated donkey anti-mouse, Alexa Fluor 488-conjugated goat anti-rat cross-adsorbed antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA). Brain samples were mounted in a drop of Vectashield™ (Vector Laboratories Inc, Burlingame, CA) and Images were collected on an Olympus Fluoview FV1000 laser scanning confocal mounted on an inverted IX81 microscope and processed with Image-J 1.33 (NIH) and Adobe Photoshop (Adobe, CA). Polytene Chromosome Immunofluorescence For Drosophila polytene chromosomal preparation and immunofluorescence, third instar larvae raised at raised at 25°C and dissected in 0.1% Triton X-100 solution in phosphate buffer saline (PBS). Salivary glands were placed in 250μm of solution 2 (3.7% paraformaldehyde, 1% Triton X-100 in PBS) for 30–45 seconds. Solution 2 was replaced with solution 3 (3.7% paraformaldehyde, 50% acetic acid) for another 2 minutes. Salivary glands were pipetted along with 20μl of solution 3 on siliconised glass cover slips and picked up onto a poly-L-lysine coated slide (Sigma), tapped to aid chromosomal spreading and frozen in liquid nitrogen. Cover slips were removed and slides were processed for IF as described previously (Capelson et al., 2010). Mouse α-MeCP2 was used at 1:100 and rabbit anti-dMBDR2 at 1:200 (a gift from Dr. Peter Becker). Secondary antibodies include Alexa Fluor 594-conjugated goat anti-rabbit and Alexa Fluor 647-conjugated donkey anti-mouse for spectral non-overlap with DAPI (1μg/ml) which was used as a DNA counterstain. Polytene samples were mounted in a drop of Vectashield™ and imaged as described previously. Images were processed for background subtraction and contrast enhancement with contrast-limited adaptive histogram equalization (CLAHE) in ImageJ. Theoretical PSF (point spread function) was calculated for images used for colocalization analysis followed by an iterative 2D deconvolution for each channel (macro code and algorithm parameters are available upon request). Pearson’s correlation coefficient (PCC) and Manders colocalization coefficient (MCC) were estimated and then PCC was statistically evaluated against randomized images using Costes’ randomization methods (Costes et al., 2004). Percentile based thresholding was applied to segment polytene chromosomes from the background for MCC calculations within the JaCoP plugin for ImageJ. RT-qPCR dMBD2/3 and dMBD-R2 expression levels were measured quantitatively by RT-qPCR. Heads from socially naive 3–5 day old adult males from control and experimental (n-syb-Gal4;UAS-dMBD2/3-IR, n-syb-Gal4;UAS-dMBD-R2-IR groups were extracted under CO2 anesthesia and frozen immediately in sets of three in 1.5-ml Eppendorf tubes kept in dry ice. Total RNA from each pool (~35 heads / pool) was isolated by Tri-Reagent, (Molecular Research Center, Cincinnati, OH). RNA samples were DNase treated and reverse transcribed as described previously (Hess-Homeier et al, 2014). qPCR reactions were carried out in quadruplicate for each gene and genotype on an Agilent Stratagene Mx3005P platform using following thermal protocol: 95°C – 10min; 40 X (95°C – 30sec; 53°C – 1min; 72°C – 1min) followed by 0.5°C stepwise increment from 65°C to 95°C. Cdc2c (cyclin-dependent kinase 2) reference gene was used for data normalization. Expression levels were calculated using the ΔCT method. dMBD-R2 expression was quantified from the total head RNA using the following primer pair: F: 5′-GGCCAGTTTGGATATAGCATCCC-3′ and R: 5′-GCACGATAACAGTGGGTTTCTGG-3′. For dMBD2/3, the primers used were: F: 5′-AGAAGCGACTGGAACGACTACG-3′ and R: 5′-CGGTCTGTTCGTTGACATTGGG-3′. For cdc2c reference gene, pre-designed exon-spanning primer pair PP1255 was used from the FlyPrimerBank: F: 5′-CGAGGGCACCTACGGTATAGT-3′ and R: 5′-CGCCTTCTAGCCGAATCTTTTTG-3′. Using the same experimental protocol, total RNA was recovered from UAS-AP-1muEP1112(BL#16987) controls and tdc2-gal4;UAS-AP-1muEP1112 experimental males. The tdc2-gal4 driver was used as n-syb-Gal4;UAS-AP-1muEP1112 males had subpar viability. dMBD2/3 expression was quantified using F: 5′-ACTGCCCAAGACCATAC-3′ and R: 5′-TGTCGTCCTCCGAAATG-3′. RPL32 expression was used as a reference in the EP line qPCR reactions and was amplified using the following primer set, F: 5′-ATGCTAAGCTGTCGCACAAATG-3′ AND R: 5′-GTTCGATCCGTAACCGATGT-3′. qPCR reactions were carried out in quadruplicate for each gene and genotype on an Agilent Stratagene Mx3005P platform using following thermal protocol: 95°C – 10min; 40 X (95°C – 30sec; 55°C – 1min; 72°C – 1min) followed by 0.5°C stepwise increment from 65°C to 95°C. HPLC For HPLC analysis, brains from socially naive 3–5-day old adult males from control and experimental groups were dissected in ice-cold PBS (137 mM NaCl/2.7 mM KCl/10 mM Na2HPO4/1.8 mM KH2PO4, pH 7.4) and frozen immediately in sets of three in 1.5-ml Eppendorf tubes at −20°C. To measure OA levels from the central brain, the photoreceptors were removed in all dissections. Each pool (n=15) of brains were homogenized in 150μL of ice-cold 0.05M perchloric acid containing 30 ng/mL DBA and chilled on ice before analysis. Immediately before analysis, the samples were centrifuged at 14,100g for 20 min at 4ºC. The supernatant was removed and 50μL injected into the HPLC. Amine levels were measured with an ESA CoulArray Model 5600A HPLC with electrochemical detection equipped with a C18 column (Varian), and a 200μl loop (Rheodyne). The flow rate was set at 0.8 ml/min. The mobile phase was composed of 10% acetonitrile (Fisher, HPLC grade), 14.18g monochloroacetic acid, 4.80g NaOH (pH adjusted to 3.0–3.5 with glacial acetic acid), and 0.301g sodium octyl sulfate (SOS) in 1000mL of sterile, polished water and filtered with 0.2μm filter. The electrodes were set at −50, 0, 100, 200, 300, 400, 500, 600, 700, 800, 900, and 920 mV. OA was detected in the 600-mV channel. Retention times and concentrations of the amines were determined by comparison to a standard composed of 80, 160, 320, 800, and 1200pg of octopamine hydrochloride in 0.1 M perchloric acid containing 30ng/mL DBA. The data from three groups of pooled males (n=15 in each pool) were averaged. Peaks were identified based on elution times. Statistical Analysis One-way ANOVA with Holm-Sidak’s multiple comparisons test was used to evaluate effects of genotype on various sleep parameters in three or more groups. Multiplicity-adjusted p-values are obtained for each pairwise comparison and only the most conservative/numerically higher values are reported here. Data was examined for Gaussian distribution and homogeneity of variance using D’Augustino Pearson omnibus normality test and Brown-Forsythe test respectively. Data were log-transformed or central limit theorem was assumed for datasets with n>30 in case of violations of assumptions of normality. Otherwise, non-parametric Kruskal-Wallis with Dunn’s post-hoc test was used. Generalized ESD test (Rosner, 1993) was used to examine outliers. Results are presented as either mean±s.e.m. or mean±c.i., as indicated in the text. The consolidation index is defined as a weighted average bout length where each sleep bout was weighted according to its duration in minutes. To calculate the consolidation index, we summed the square of all sleep bout lengths (in minutes) and divided by the total amount of sleep according to (Pitman et al., 2006). This calculation reduces the influence of transient awakenings during the sleep phase. The empirical cumulative distribution (CDF) for sleep bouts was plotted using the ecdf function in MATLAB (The MathWorks, Natick, MA). Ordinary two-way Multivariate ANOVA (MANOVA) was carried out in SPSS23 using the general linear model (GLM) procedure to explore interactions between the effect of hMeCP2 and dMBDs on linear composite of various measures of sleep. Multivariate outliers were detected for all sleep parameters based on a chi-square distribution using Mahalanobis distance (MD). Cases with MD>18.47 (critical χ2 value assessed at p < .001, df = 4) were identified as outliers and removed. Box-Cox transformed dependent variables (i.e. total sleep, waking activity, consolidation index, and number of sleep bouts) were auto-scaled for the purposes of scale standardization and univariate outliers were identified using +3.0 z-score criterion. Multi-collinearity was checked against the variance inflation factor (VIF; threshold=5). As our dataset contained an unbalanced design (unequal sample size across groups), and violated the assumption of homogeneity of covariance matrices, Pillais’ trace criterion (which is most robust to such violations) was reported. These results were cross-validated by employing a non-parametric or permutation MANOVA (NPMANOVA / PERMANOVA) in PASTv3.09 (Hammer et al., 2001) which is insensitive to such violations (Anderson, 2001). Homology modeling The SWISS-MODEL template library (SMTL version 2015-04-15, PDB release 2015-04-17) was searched with Blast (Altschul et al., 1997) and HHBlits (Remmert et al., 2012) for evolutionary related structures matching the target MBD amino acid sequence for both MBD-R2 and MBD2/3. The templates with the highest quality predicted from features of the target-template alignment were then selected for model building. Models were built based on the target-template alignment using Modeller (Sali & Blundell, 1993) within the UCSF Chimera package (Pettersen et al., 2004). The model quality/reliability was assessed using the z-DOPE (Shen & Sali, 2006) and GA341 (Melo et al., 2002) scoring functions through ModEval Model Evaluation Server (http://modbase.compbio.ucsf.edu/evaluation/). RESULTS MeCP2 expression in octopamine neurons results in fragmented and reduced sleep Examining sleep output in fruit flies provides an ideal paradigm for investigating the role of MBD proteins in neuronal function for several reasons. First, numerous behavioral parameters can be quantified in a large cohort of genetically identical control and experimental populations (Bellen et al., 2010, Venken & Bellen, 2014). Second, behavioral output can be measured at the single minute level, which provides a formidable temporal resolution of function, and finally this functional output is responsive to changing environmental stimuli thus requiring an active readout of the neuronal nuclear state. To determine if MeCP2 expression in distinct amine neurons can alter sleep-wake circuitry function, we used the Gal4-UAS gene expression system and previously generated UAS-hMeCP2 transgenic lines (Cukier et al., 2008). As norepinephrine and OA regulate sleep levels by promoting wakefulness (Crocker & Sehgal, 2008, Mitchell & Weinshenker, 2010, Robbins, 1997), we expressed hMeCP2 (the MeCP2e2 isoform) in OA/tyramine (TA) neurons via the tyrosine decarboxylase2 (tdc2)-gal4 driver (Cole et al., 2005) (Fig. 1a-a′) and quantified sleep-wake patterns, sleep onset, duration, and the quality of sleep over a 10-day period using a standard automated high-throughput activity monitoring system (Ho & Sehgal, 2005) (Drosophila Activity Monitor, Trikinetics, Waltham, MA). Groups of experimental and control adults were assayed concurrently. Adult males expressing hMeCP2 in OA neurons exhibited specific deficits in sleep quantity and quality including a significant reduction in total sleep as compared to transgenic controls (tdc2-gal4/+, and UAS-hMeCP2/+) and the nuclear protein expression control (tdc2-Gal4;UAS-dsRed) (Fig. 1b). This sleep reduction occurred during day and night time frames (Zeitgeber hours ZT04-10 and ZT14.5-22) (Fig. 1c, d). A comparison across the 10-day assay time and replicate groups indicating the consistency of the hour-specific deficit is provided in Fig. S1. We found that the sleep of hMeCP2-expressing males was fragmented as measured by an increase in the average number of sleep bouts (Fig. 1e) and a significant decrease in the consolidation index (C.I.), a weighted measure of average bout length (Fig. 1f, see Materials and Methods). This difficulty in maintaining sleep was also evident by plotting sleep bout data using the empirical cumulative distribution function (ECDF). The ECDF demonstrates that longer consolidated bouts of sleep are replaced with a greater proportion of short sleep bouts in experimental males but not controls (Fig. 1g). Experimental males also displayed a significant reduction in the latency to initiate sleep (Fig. 1h), suggesting the need for recovery after sleep loss and homeostatic relevance of the observed sleep deficits. This sleep loss and hMeCP2 expression in OA neurons in general did not shorten the lifespan of experimental vs. control males rather median lifespan is significantly increased (Fig. S2). In addition to controlling for nuclear protein expression, we further verified the sleep defects observed in tdc2-gal4;UAS-hMeCP2 adults are not due to general changes by asking if different sleep deficits occur as a result of hMeCP2 expression in serotonin neurons (Fig. S3). Males expressing hMeCP2 in 5HT neurons via the tryptophan hydroxylase (trh)-Gal4 line (Alekseyenko et al., 2010) did exhibit sleep loss similar to hMeCP2 effects in OA neurons during specific nighttime hours (ZT19-22.5; Fig. S3). However, the nighttime sleep deficits caused by hMeCP2 expression in 5HT neurons were not accompanied by sleep fragmentation changes (Fig. S3). The conserved nighttime sleep reduction suggests that hMeCP2 expression may alter a specific aspect of neuronal function that is shared by neurons that express different neurotransmitters, yet other sleep impairments are cell-specific. OA is required for a subset of hMeCP2-mediated sleep deficits Alterations in the duration of sleep and a reduction in the latency to initiate sleep are phenotypes observed in flies lacking OA (Crocker & Sehgal, 2008). Therefore, one possible explanation for this particular sleep deficit is that the expression of genes necessary for OA synthesis has been altered. To address this question, we quantified OA levels extracted from the heads of control and experimental males using High Performance Liquid Chromatography (HPLC). Heads were removed during the period of decreased daytime sleep, ZT04-10, to determine if OA levels changed during the sleep reduction time periods. OA concentrations per head did not differ between control (tdc2-gal4/+; and UAS-hMeCP2/+) and experimental (tdc2-gal4;UAS-hMeCP2) males (Fig. 2a). Although we cannot rule out the possibility of OA level differences in specific neurons contributing to sleep deficits, these results demonstrate that a global reduction in OA production does not occur as a result of hMeCP2 expression. Although hMeCP2 expression in OA neurons does not alter OA production, it is possible that the observed sleep deficits require OA function. To test this possibility, we expressed hMeCP2 in flies that completely lack OA due to a null mutation in tyramine-β-hydroxylase (TβhnM18), the rate-limiting enzyme in OA biosynthesis (Monastirioti et al., 1996). Although the total amount of sleep by OA null males expressing hMeCP2 was not different from controls (Fig. 2c), TβhnM18 tdc2-gal4;;UAS-hMeCP2 males exhibited daytime hourly specificity in sleep reduction like wildtype males expressing hMeCP2 (Fig. 2b,d). Strikingly however, the nighttime sleep deficit (ZT14-17.5) quantified in Figure 1 is completely rescued in hMeCP2-expressing males that lack OA (Fig. 2d). This result suggests OA is required to translate the hMeCP2-mediated neuronal defects into a reduction in nighttime sleep during specific hours. Not all hMeCP2-mediated sleep deficits rely on OA neurotransmitter function, as the consolidation index and sleep bout number alterations (Fig. 2e, f) were similar between hMeCP2-expressing males lacking OA or hMeCP2-expressing males with OA. In contrast to the rescued dark phase sleep reduction, the daytime sleep deficits observed during ZT04-10 in tdc2-Gal4;UAS-hMeCP2 adults persisted in males that lack OA (Fig. 2d). A possible explanation for any sleep reduction is a concomitant increase in activity. As Tβh converts tyramine (TA) to OA, the absence of this enzyme results in an accumulation of TA [3,4]. To determine if the periods of sleep reduction observed in males lacking OA are due to elevated TA-induced increases in locomotion rather than hMeCP2 expression (Hardie et al., 2007, Monastirioti, 1999), we quantified activity levels. Changes in waking activity were not observed in the absence of OA (Fig. 2g). Finally, hMeCP2 expression in the nucleus of octopamine neurons may provide some protection against the OA deficient circuit alterations as the increase in sleep observed in OA null males is returned to control levels in the same males now expressing hMeCP2 (TβhnM18 tdc2-gal4;;UAS-hMeCP2) (Fig. 2c, dark gray vs. yellow column). The C-terminal region of hMeCP2 is not sufficient to generate sleep deficits in OA neurons One approach to understanding the potential targets of multi-domain containing proteins is to link protein domain(s) with a corresponding phenotype, therefore we investigated which conserved domains are essential in generating the observed sleep impairments by expressing hMeCP2 alleles that lack the CTD and separately the MBD (Cukier et al., 2008). Due to the relatively sparse distribution of 5mC methylation in Drosophila, we first postulated hMeCP2 exerts its affects through methylation-independent interactions mediated by the C-terminal transcriptional repression domain (TRD) and the C-terminal domain (CTD). The TRD functions as a recruitment center for several transcriptional and epigenetic regulators including components of the transcription repression machinery Sin3a, HDAC1, and HDAC2 (Ghosh et al., 2010, Nan et al., 1998), while the CTD (residues 295 to 486) contains one or more chromatin binding regions (Ausio et al., 2014, Roloff et al., 2003). Together the TRD and CTD domains have been implicated in nucleosomal clustering, array compaction and oligomerization, and gene repression (Nikitina et al., 2007). To remove the C-terminus, we expressed the early truncating mutation encoded by the hMeCP2R294X allele which is found in ~5–6% of RTT patients (Laccone et al., 2001, Wan et al., 1999). In the resulting R294X protein, the TRD is partially truncated and the CTD is completely removed (Fig. 3a) (Wan et al., 1999). The Gal4-driven protein expression of UAS-hMeCP2R294X was previously verified by western blot analysis (Cukier et al., 2008). If the sleep deficits observed in males expressing hMeCP2 in OA neurons were mediated through the C-terminus, we would predict sleep would be normal in males expressing hMeCP2R294X. However, removing TRD and CTD function, did not eliminate the daytime sleep reduction observed in tdc2-gal4;UAS-hMeCP2 males, and only a partial recovery in the nighttime sleep deficits occurred (ZT14.5-22, Figure 3b,c). Males expressing R294X exhibited a decrease in the latency to initiate sleep (Fig. 3d) and changes in sleep architecture (Fig. 3e–g) in a manner similar in males expressing full-length hMeCP2. Specifically, the number of sleep bouts and weighted average bout lengths exhibited by tdc2-gal4;UAS-hMeCP2R294X males remained significantly different than controls (Fig. 3e, f). These results indicate that the hMeCP2-induced changes that drive sleep alterations in the OA neuronal population do not occur primarily through the CTD and TRD domains. The N-terminus and MBD domain are necessary for MeCP2-induced alterations in sleep architecture We next asked if the majority of the sleep deficits observed in tdc2-gal4;UAS-hMeCP2 males are due to the conserved MBD domain. To test this question, we used the UAS-hMeCP2Δ166 line to express a truncated hMeCP2 allele that lacks the N-terminal and MBD domain (Cukier et al., 2008) (Fig. 4a, b). We found the sleep deficits caused by hMeCP2 expression including the amount of sleep, latency to sleep, sleep bout number, and sleep bout length were absent in tdc2-gal4;UAS-hMeCP2Δ166 males (Fig. 4c–h). This lack of sleep defects could be explained if the Δ166 protein was not expressed, however hMeCP2 Δ166 accumulates in the nucleus of tdc2-gal4;UAS-hMeCP2 Δ166 adult brains as demonstrated by immunohistochemistry (Fig. 4b). In addition, previous studies determined hMeCP2 Δ166 is present along polytene chromosomes, is phosphorylated at amino acid S423, and is able to cause Drosophila neuronal morphology and dendritic defects (Cukier et al., 2008, Vonhoff et al., 2012). To determine if the MBD domain itself is required for the MeCP2-induced changes in sleep output, we expressed the severe RTT-causing missense hMeCP2R106W allele. MeCP2-induced alterations in sleep output are dependent on the MBD domain The R106W mutation in the MBD domain impacts the MeCP2 protein by severely disrupting its ability to bind methylated DNA (Chapleau et al., 2009, Kudo et al., 2001), thus potentially altering target gene repression and chromatin condensation. Males expressing hMeCP2R106W in OA neurons (tdc2-gal4;UAS-hMeCP2R106W), completely lack the sleep deficits including all sleep reductions and fragmentation phenotypes caused by wildtype hMeCP2 function (Fig. 5a–e). These results demonstrate that an intact MBD domain is necessary to cause the hMeCP2-mediated changes in sleep behavior. Furthermore, if the hMeCP2-induced changes were a result of non-specific methylation-independent cellular effects in OA neurons, we would expect the sleep deficits to remain as was observed in a previous study describing R106W-induced neuron morphology and motor performance defects (Cukier et al., 2008). However, our results indicate methylation-dependent mechanisms may play a key role in hMeCP2-induced changes in OA neuron output. Recent experiments examining hMeCP2-induced motorneuron dendritic defects also reported an absence of morphology changes upon R106W expression (Vonhoff et al., 2012). Males with reduced dMBD-R2 levels in OA neurons displayed increased sleep levels At this point, our results describe specific hMeCP2-induced sleep deficits and establish the MBD of MeCP2 is a critical component. We next simply asked if endogenous MBD-containing proteins are required for amine neuron function and sleep-wake circuitry output. At least two diverse proteins in Drosophila belong to the MBD family: a) dMBD-R2 and b) dMBD2/3 (Fig. 6a) (Hendrich & Tweedie, 2003, Roder et al., 2000). dMBD2/3 is a small protein consisting strikingly of three MBD domains (Fig. 6b) in contrast; dMBD-R2 contains a THAP, TUDOR, and PHD-type Zinc finger in addition to the MBD domain (Fig. 6c). dMBD2/3 and the dMBD2/3Δ splice variant associate with the nucleosome remodeling and deacetylase (NuRD) complex (Marhold et al., 2004a), repress transcription in in vitro assays (Ballestar & Wolffe, 2001), and dMBD2/3Δ preferentially recognizes mCpG-containing DNA through its MBD (Roder et al., 2000). In addition, the expression of both dMBD2/3 and MBD2/3Δ is developmentally regulated and is present in adult tissues suggesting selective roles in transcriptional regulation (Marhold et al., 2004a, Marhold et al., 2004b). Unlike dMBD2/3, it has not been determined if MBD-R2 binds 5mC, however, dMBD-R2 is a part of the multi-subunit chromatin remodeling NSL (non-specific lethal) complex, which regulates gene expression at genome wide levels (Roder et al., 2000). The human MeCP2 MBD contains 8 known DNA binding sites, half of which are lysine residues (K107, K109, R111, K119, D121, K130, R133 and E137; Conserved domain database: 238690). At least five of these eight DNA-binding sites are present in the Drosophila dMBD-R2 protein (R111, K119, D121, K130, R133) and four in dMBD2/3 (R111, K119, D121, K130). These conserved sites and their location in reference to the hMeCP2 residue positions are depicted in Fig. 6a. In addition, a predicted homology model suggests similarity between specific secondary structural features among the MBD domains of dMBD-R2, dMBD2/3 MBD domains and hMeCP2 (Fig. 6d, f). Therefore, we asked if reducing dMBD2/3 or dMBD-R2 levels using RNA interference could alter the function of neurons as measured by changes in the sleep network. To measure the RNAi effect on transcript levels, quantitative reverse transcription PCR (RT-qPCR) was performed on RNA extracted from the heads of n-syb-Gal4;UAS-dMBD2/3-IR (IR=inverted repeats) and n-syb-Gal4;UAS-dMBD-R2-IR adults. Transcript levels were reduced by 26.84% (Fig. 6e and 36.79% respectively (Fig. 6f). When dMBD-R2 and dMBD2/3 levels were reduced in OA neurons by separately expressing the UAS-dMBD-R2-IR and UAS-dMBD2/3-IR lines under control of the tdc2-gal4 driver, we found that sleep fragmentation and sleep deficits occurred in both tdc2-Gal4;UAS-dMBD-R2-IR and tdc2-Gal4;UAS-dMBD2/3-IR males. Sleep fragmentation was quantified by the increase in sleep bout number along with a decrease in the consolidation index (Figs. 7a–d). Males with reduced dMBD-R2 levels in OA neurons exhibited an increase in the amount of total sleep over a 24 hr period (Fig. 7a,b) with phase ZT1-3 (Fig. 7a, arrow), contributing significantly to the overall increase (tdc2-gal4;UAS-dMBD-R2-IR, 27.7 ± 0.12 vs. tdc2-gal4/+, 13.1 +/− 0.4, UAS-dMBD-R2-IR/+, 19.4 ± 0.5, F=105.2, p=6.66E-08 (i.e. p<0.0001; Welch F test in the case of unequal variances)). This increase in total sleep exhibited by dMBD-R2 deficient adults was not due to diminished locomotor activity as the experimental males were more active during waking periods than controls (Fig. 7c). Males with reduced dMBD-R2 levels also displayed changes in sleep architecture as sleep bout number increased and the weighted average bout length (consolidation index) decreased (Fig. 7d,e). Reducing dMBD-R2 rescues hMeCP2-mediated phase-specific sleep deficits The observation that total sleep increased with a reduction in dMBD-R2 levels is the opposite of the reduction in sleep due to hMeCP2 expression. This result led us to speculate that as both are able function as potential gene repressors, a reduction in dMBD-R2 could lead to an increase in specific gene expression whereas hMeCP2 may be reducing specific gene expression. If hMeCP2 and dMBD-R2 are functioning at similar gene loci or genomic regions, then we predict a reduction in dMBD-R2 levels in combination with hMeCP2 should rescue the phase-specific sleep loss. We tested this hypothesis by generating tdc2-gal4;UAS-hMeCP2/UAS-dMBD-R2-IR adults and found the hMeCP2-mediated night and daytime sleep deficits are indeed restored to control levels when dMBD-R2 levels are reduced (Fig. 7f). These results indicate reducing the levels of a Drosophila MBD protein can generate the same phenotype as the expression of hMeCP2 and suggests possible functional conservation. In support of this hypothesis, we tested whether the effect of relative dMBD levels on sleep output varies in the presence or absence of hMeCP2 by performing a two-way multivariate analysis of variance (MANOVA). A significant interaction (dMBD-R2 × hMeCP2) effect was observed between relative dMBD-R2 and hMeCP2 expression on combined measures of sleep (F(3, 190) = 28.192, p < 0.0001; V = 0.308; Obs. Power = 1.00, Fig. 7g,h). To examine on a genomic level if hMeCP2 and MBD-R2 are able to associate together at chromosomal locations, we expressed hMeCP2 in polytene salivary gland chromosomes using the 48B10-gal4 driver. Isolated larval polytene chromosomes from 48B10-gal4;UAS-hMeCP2 larvae were labeled with MBD-R2 and MeCP2 antibodies. As expected, dMBD-R2 localizes to many sites on polytene chromosomes due to its role as a general facilitator of transcription and as a component of the non-specific-lethal and male-specific-lethal complexes (Pascual-Garcia et al., 2014, Prestel et al., 2010). However, hMeCP2 and dMBD-R2 are detected together at a number of chromosomal sites (Fig. 7i–k, arrows, n=6) suggesting the possibility of common gene loci or chromatin organization targets. As a whole, our results indicate the conserved MBD domain even among disparate MBD-containing proteins, such as hMeCP2 and dMBD-R2, is capable of conferring shared neuronal phenotypes and shared genomic binding sites. Increased MBD2/3 expression recapitulates the hMeCP2-mediated sleep deficits To determine if increasing the levels of endogenous MBD-containing proteins alters sleep parameters in the same manner as hMeCP2, we searched for Enhancer and Promoter (EP)-element inserted lines upstream of dMBD-R2 or dMBD2/3. When crossed to a Gal4 driver, the UAS-containing EP lines can induce the overexpression of the gene of interest (Rorth, 1996). Five EP stocks were tested (see Materials and Methods). Two lines, P{EP}MBD2/3EY0482 and P{EP}AP1muEP1112 located upstream and within dMBD2/3 (Fig. 8a) altered sleep output in the same manner. An increase in MBD2/3 transcript levels were verified in tdc2-gal4;UAS-AP1muEP1112 adults (Fig. 8a). To compare with specific sleep phenotypes observed upon hMEeCP2 expression, we quantified the effects of dMBD2/3 overexpression on sleep during the two specific day and night phases (see Fig. 1d). tdc2-gal4;UAS-AP1muEP1112 males exhibited a reduction in sleep during day and night time frames (ZT04-10 and ZT15-17.5) (Fig. 8b) as well as sleep fragmentation as measured by an increase in the average number of sleep bouts (Fig. 8c). Likewise, tdc2-gal4;UAS-MBD2/3EY0482 males displayed sleep reductions during the same day and night phases (Fig. 8d). Next we asked if sleep alterations occur in dMBD2/3 deficient males. In contrast to hMeCP2 and dMBD2/3 overexpression, tdc2-gal4;UAS-dMBD2/3-IR males did not display a decrease in sleep during specific day and night phases (Fig. 8e and see Fig. 1d), however reducing dMBD2/3 levels in OA neurons did result in sleep fragmentation (Fig. 8f) and an increase in locomotor activity (data not shown). Finally, we generated tdc2-gal4;UAS-hMeCP2/UAS-MBD2/3-IR males and observed a significant interaction (dMBD2/3 × hMeCP2) effect between relative dMBD2/3 and hMeCP2 expression on sleep fragmentation using Pillais’ trace with 0.05 significance criterion (F(3, 194) = 30.665, p < 0.0001; V = 0.322; Obs. Power = 1.00, Fig. 8h,i). Discussion In this study, we tested the hypothesis that MBD-containing proteins retain considerable functional conservation by measuring neuronal output through an automated, reproducible sleep assay. Sleep impairments are a major feature in a substantial number of neurodegenerative and neuropsychiatric disorders [37]. However more fundamentally, this data can be viewed as a relevant behavioral representation of circuit dysfunction in general, which is a common theme in neurodevelopmental syndromes including RTT (Cortesi et al., 2010, Shepherd & Katz, 2011). A powerful advantage of using Drosophila sleep to analyze the functional differentiation of circuits and neurons is the ability to measure behavior at the single minute level. This formidable temporal resolution in combination with amine neuron-specific manipulation allowed us to analyze the functional consequences of wildtype hMeCP2, domain-specific mutations, and Drosophila MBD proteins with powerful phenotypic resolution. Our results demonstrate that adults expressing hMeCP2 in OA neurons sleep less, however this sleep loss is not a general or random phenomena but rather occurs during specific day and nighttime intervals. In a similar manner, hMeCP2 expression in 5-HT neurons also results in a loss of nighttime sleep however, with the fine temporal resolution, we can identify sleep loss intervals that are both unique and overlapping when compared to hMeCP2 expression in OA neurons. Finally, in a previous study we determined that hMeCP2 expression in astrocytes non-cell-autonomously alters the sleep network only during distinct nighttime hours (Hess-Homeier et al., 2014). How might hMeCP2 expression in amine neurons reduce sleep amounts and sleep quality? At the DNA level, MeCP2 binds to the promoters of genes involved in amine synthesis including dopa decarboxylase (Urdinguio et al., 2008) and MeCP2 levels themselves are under circadian cycle control (Martinez De Paz et al., 2015). Previous studies have demonstrated that a loss of OA promotes sleep (Crocker & Sehgal, 2008) and our HPLC studies indicate OA brain levels are not reduced upon hMeCP2 expression. However, it is possible that the MeCP2-induced reduction in nighttime sleep is mediated through an increase in OA signaling. This hypothesis is consistent with previous observations including an overexpression of Tdc2 or genetically activating OA neurons significantly decreases nighttime but not daytime sleep (Crocker & Sehgal, 2008). In addition, components of the arousal circuitry respond to OA wake-promoting signals including the large-lateral ventral neurons (l-LNvs) neurons (Crocker et al., 2010). When hyper-excited, l-LNv neurons result in a reduction in sleep quality and amount and express octopamine receptors (Kula-Eversole et al., 2010, Shang et al., 2008). In our experiments, MeCP2 expression could potentially increase OA neuron activity by modulating presynaptic function either through changes in levels of OA biosynthetic enzymes, or components of OA transport and release, or conserved RNA-binding proteins such as Lark, which regulate neuronal excitability in the circadian system (Ishimoto et al., 2012). Finally, our results notably demonstrate a loss of OA function can completely rescue the most severe hMeCP2-mediated reduction in nighttime sleep found within ZT14-17.5 (Fig. 2c). As MBD family members have a highly similar DNA-binding surface that shows high affinity for methylated DNA, a key question is whether individual proteins bind differentially to distinct regions within the genome. Variations in the affinity for binding methylated targets include double-stranded vs. single-stranded, sequence dependent vs. sequence independent, and CpG vs. non-CpG (CpH; H=A/C/T) methylation (Baubec et al., 2013, Fatemi & Wade, 2006, Guo et al., 2014). Recently, a role for MeCP2 binding to CpH sites and regulating the expression of genes enriched for neuronal function has been described (Chen et al., 2015). Non-CpG methylation has been reported in vertebrate neurons (Fatemi & Wade, 2006, Guo et al., 2014, Pinney, 2014) and in Drosophila; the methylation present in the adult genome is enriched on non-CpG motifs, particularly CpT and CpA dinucleotides (Boffelli et al., 2014, Capuano et al., 2014, Takayama et al., 2014). In our sleep experiments, MeCP2 may be functioning to translate endogenous CpH methylation into changes in gene expression. This idea is especially compelling as we demonstrated that an intact MBD-binding domain is required for all hMeCP2-induced sleep deficits (Fig. 5). Furthermore, males with reduced levels of dMBD2/3, which binds methylated DNA, exhibited overlapping sleep quality deficits (Fig. 7, SF4). In this context, Drosophila may provide an ideal in vivo system to examine the functional consequences of CpH-mediated MBD protein interactions as future studies can address the significance of CpH methylation at candidate genes that control circadian rhythm and aspects of sleep. In conclusion, epigenetically modifying chromatin structure in response to different stimuli may be a key mechanism in generating shifts in gene expression not only at successive stages of neuron development but successive stages of neuron function. Such functional changes may include responses to pheromones (predators or conspecifics), odors (food resources), or light (sleep) all critical aspects of reproduction and survival in any organism. In this study we examined the consequences of a hypomorphic reduction of endogenous MBD proteins in a relevant neuronal subpopulation to provide a whole organism readout of changes in neuron function that should be interpretable at the chromatin level in future studies due to ever-increasing advances in circadian rhythm and sleep gene identification. Our results provide the first demonstration that Drosophila MBD proteins are required for neuron function and that MBD-containing proteins indicate conservation in the cell-specific functions of epigenetic translators. Supplementary Material Supp Fig S1 Supplemental Figure 1: hMeCP2 expression in OA neurons reduces the sleep of adult males: Comparison across days and replicates (a) The sleep of control (tdc2-gal4/+, and UAS-hMeCP2/+) males and experimental tdc2-gal4;UAS-hMeCP2 adults was recorded and averaged across a 10-day time period. The data shown are subsets of controls and reflect the results from batch 1 of the experimental males. (b) The average sleep recorded per batch of experimental males does not differ. Supp Fig S2 Supplemental Figure 2: Males expressing hMeCP2 in OA neurons live longer than controls A Kaplan-Meier survival curve with the dotted boundaries around the curves representing standard error (SE). The survival distribution of experimental males, tdc2-gal4;UAS-hMeCP2 males is statistically different than controls (standard log-rank test, P≪0.0001). Supp Fig S3 Supplemental Figure 3: Adults expressing hMeCP2 in 5HT neurons exhibit a nighttime sleep reduction (a) hMeCP2 nuclear expression (green) in 5HT neurons from a trh-gal4; UAS-MeCP2/+ male brain. (b–h) The quality and amount of sleep in individual adult males averaged over an 8-day period from control and experimental groups. (b) The total amount of sleep per 24-hr day is not significantly changed in experimental males as compared to UAS-hMeCP2/+ controls (Padj=0.2051). (c) Eduction graph displaying the average amount of sleep per 30 minute bin (daytime/light phase: white bar; nighttime/dark phase: black bar, shaded grey) in control and experimental males. trh-gal4/+; UAS-MeCP2/+ males displayed a reduction in sleep during Zeitgeber hours ZT19-22.5 (arrow). These deficits are quantified in (d) P=0.0011, Mann Whitney test. (e–h) Sleep fragmentation in males expressing MeCP2 in 5HT neurons. (e) The daytime consolidation index is significantly reduced in experimental vs. control males (Padj<0.0001). The nighttime consolidation index is not altered (Padj=0.7262). (f) The average number of daytime sleep bouts is increased in experimental males vs. controls (Padj<0.0001), without alterations in the average number of nighttime sleep bouts (Padj=0.8316). (g) Daytime, but not nighttime, waking activity is increased in experimental males vs. controls (Padj<0.0001). (h) The empirical cumulative distribution function demonstrates experimental males exhibit a greater proportion of short sleep bouts as compared to controls. Data are shown as means ± standard error of the mean (SEM). Unless noted otherwise, results were analyzed by one-way ANOVA with Holm-Sidak’s multiple comparison test. The authors thank Juan Botas, Jay Hirsh, Sirge Birman, Olga Alekseenko, Edward Kravitz and the Bloomington Stock Center for fly stocks. We are grateful to Conor Jacobs for initial sleep studies, Peter Becker for the MBD-R2 antibody, Jennene Lyda for assistance with HPLC analysis, and members of the Certel lab for helpful discussions. We also thank the University of Montana Molecular Histology and Fluorescence Imaging Core and Lou Herrit for technical expertise. The nc82 antibody was developed by Eric Buchner obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the Department of Biology, University of Iowa (Iowa City, IA). The National Institutes of Health Center of Biomedical Research Excellence grant P20RR015583 and the University of Montana Research Grant Program to SJC supported this work. Figure 1 hMeCP2 expression in OA neurons reduces the sleep of adult males (a-a″) hMeCP2 expression (red) in OA neurons from an adult tdc2-gal4/UAS-mCD8:gfp; UAS-hMeCP2/+ male (anti-GFP, green; mAb nc82, labels neuropil regions, blue). (b–h) Sleep profiles of individual adult males averaged over 8 days from control and experimental groups. Controls: tdc2-gal4/+ (white), UAS-hMeCP2/+ (light grey), tdc2-gal4/+; UAS-dsRed/+ (dark grey) and experimental: tdc2-gal4/+; UAS-hMeCP2/+ (red). (b) Total sleep per 24-hr day is reduced in experimental males as compared to controls (Padj=0.0013; one-way ANOVA with Holm-Sidak’s multiple comparison test). (c) Eduction graph displaying 30 minute bins of averaged sleep (daytime/light phase: white bar; nighttime/dark phase: black bar, shaded grey). tdc2-gal4/+; UAS-hMeCP2/+ males displayed a reduction in the average amount of sleep during both day and night (arrows) as compared to controls. These deficits are quantified in (d) for Zeitgeber hours ZT04-10, (P<0.0001; two-tailed Mann Whitney test) and ZT14.5-22, (P<0.0001; two-tailed Mann Whitney test). (e–g) Sleep fragmentation in males expressing MeCP2 expression in OA neurons. As compared to controls, the average number of sleep bouts per day (e) is increased (Padj<0.0001) and weighted average bout length measured by the consolidation index (f) is reduced significantly in experimental males (Padj<0.0001). (g) The empirical cumulative distribution function (ECDF) demonstrating experimental males exhibit a greater proportion of short sleep bouts as compared to controls. (h) Latency to initiate sleep (the delay in minutes from the lights OFF to the time to the first sleep bout) is significantly reduced in tdc2-gal4/+; UAS-hMeCP2/+ males as compared to controls (Padj=0.0009; one-way ANOVA with Holm-Sidak’s multiple comparison test). Data are shown as means ± standard error of the mean (SEM). Figure 2 The loss of OA rescues a subset of hMeCP2-induced sleep deficits (a) HPLC quantification of OA levels in whole brain extracts of 3–5 day old adult males collected during ZT04-10. OA levels between control and experimental groups did not differ. (b–f) Sleep profiles of individual adult males averaged over an 8-day period from control and experimental groups. Controls: tdc2-gal4/+ (white bar), UAS-hMeCP2/+ (light grey), tβhnM18 tdc2-gal4 (dark grey) and experimental: tdc2-gal4; UAS-hMeCP2 (red), tβhnM18 tdc2-gal4; UAS-hMeCP2 (yellow). (b) Eduction graph displaying average amount of sleep per 30 minute bin (daytime/light phase: white bar; nighttime/dark phase: black bar) in control and experimental males. hMeCP2-induced sleep deficits (red line) are restored to control levels in tβhnM18 tdc2-gal4; UAS-hMeCP2 males during ZT14-17.5 (yellow line, arrow). (c) Total sleep increased in the OA deficient control males (tβhnM18 tdc2-gal4, black column) as compared to transgenic controls (white/gray columns, Padj = 0.0070). Expression of hMeCP2 in OA deficient males (tβhnM18 tdc2-gal4; UAS-hMeCP2, black vs. yellow columns) returned total sleep values to wildtype levels (Padj = 0.6563; one-way ANOVA with Holm-Sidak’s multiple comparison). (d) The reduction in sleep during ZT04-10 remained in OA deficient males expressing hMeCP2. The sleep reduction during ZT14-17.5 was completely rescued in the absence of OA (multiplicity adjusted P-value for pooled controls vs. tβhnM18 tdc2-gal4; UAS-hMeCP2 experimental males; P= 0.8447). (e–f) Sleep fragmentation remains in hMeCP2-expressing OA deficient males. The consolidation index (e) is reduced significantly in both experimental groups (Padj = 0.1658) and the average number of sleep bouts is increased (f) (Padj = 0.2409). (g) No difference was observed in the waking activity between OA deficient controls (tβhnM18 tdc2-gal4) and experimental males (tβhnM18 tdc2-gal4; UAS-hMeCP2/+; Padj = 0.6325). Figure 3 hMeCP2-induced sleep deficits remain in males expressing the R294X allele (a) Schematic depicting the structural domains MeCP2 and the loss of domains due to the R294X mutation. (b–h) The sleep profiles of control and experimental adult males averaged over an 8-day period. (b) Eduction graph displaying the average amount of sleep per 30 minute bin (daytime/light phase: white bar; nighttime/dark phase: black bar, shaded grey). Average sleep during Zeitgeber hours ZT04-10 and ZT14.5-22 are quantified in (c). Males expressing the R294X allele displayed a similar reduction in the average amount of sleep during ZT04-10 as males expressing the full-length allele (Padj=0.0103). During ZT14.5-22, the average sleep deficit in males expressing R294X allele remains reduced as compared to controls (P<0.0001). This 294X-induced sleep reduction is partially recovered in comparison to hMeCP2-expressing males (P<0.0001). (d) Males expressing full-length or R294X alleles exhibited a reduction in the latency to initiate sleep as compared to controls (Padj=0.0001). (e–g) Sleep fragmentation in males expressing the full-length MeCP2 and R294X alleles in OA neurons. (e) The average number of sleep bouts increases to a lesser extent in R294X males as compared to males expressing full-length MeCP2 (Padj<0.0001) however the increase in sleep bouts of tdc2-gal4;UAS-hMeCP2294X is significantly higher than controls (P<0.0001). (f) The consolidation index was reduced significantly in both full-length and R294X males as compared to controls (Padj<0.0001). (g) Experimental males exhibited a greater proportion of short sleep bouts as calculated by the empirical cumulative distribution function. Data are shown as means ± standard error of the mean (SEM). Unless noted otherwise, one-way ANOVA with Holm-Sidak’s multiple comparison test was used. Figure 4 Sleep fragmentation and sleep deficits are rescued in males expressing the hMeCP2Δ166 allele in OA neurons (a) Schematic diagram depicting MeCP2 structure and the loss of domains due to the Δ166 truncation. (b) hMeCP2Δ166 (green) is expressed in adult OA neurons via the tdc2-gal4 driver (tdc2-gal4; UAS-MeCP2Δ166). (c–h) The sleep profiles of control and experimental adult males averaged over an 8-day period. (c) The latency to initiate sleep is not significantly reduced in males expressing hMeCP2Δ166 as compared to controls (Padj=0.2611). (d) Eduction graph displaying average amounts of sleep per 30-minute bin in control and experimental males. The overall sleep profile and average sleep during Zeitgeber hours ZT04-10 and ZT14.5-22 is completely rescued in males expressing hMeCP2Δ166. (e) The average amount of sleep does not differ between controls and males expressing hMeCP2Δ166: ZT04-10, (Padj=0.514), and ZT14.5-22, (P=0.7853). (e–h) Sleep is not fragmented in males expressing hMeCP2Δ166 in OA neurons. (f) The average number of sleep bouts is not significantly different in tdc2-gal4; UAS-MeCP2Δ166 vs. the tdc2-gal4 and UAS-MeCP2 control (Padj=0.2923). (g) The consolidation index does not differ between males expressing hMeCP2Δ166 and controls (Padj=0.1308). (h) The empirical cumulative distribution function demonstrates experimental males exhibit a greater proportion of short sleep bouts as compared to controls. Data are shown as means ± standard error of the mean (SEM). The one-way ANOVA with Holm-Sidak’s multiple comparison test was used. Figure 5 Expression of the MeCP2 R106W allele does not confer sleep deficits or fragmentation (a–e) Sleep patterns averaged over a period of 8 days from control and experimental males. (a) Eduction graph displaying average amount of sleep per 30-min bin. The sleep patterns and sleep quality of males expressing hMeCP2R106W in OA neurons are the same as controls. (b) The average sleep during Zeitgeber hours ZT04-10 and ZT14.5-22 does not differ between males expressing R106W and controls: ZT04-10, Padj=0.7406, and ZT14.5-22, P=0.0974. (c–e) Sleep fragmentation does not occur in males expressing R106W. (C) The average number of sleep bouts in males expressing R106W is not significantly different from controls (Padj=0.8849). (d) The consolidation index does not differ from the R106W-expressing experimental males and controls (Padj=0.9843). (e) Experimental males exhibited a greater proportion of short sleep bouts as calculated by the empirical cumulative distribution function. Data are shown as means ± standard error of the mean (SEM). The one-way ANOVA with Holm-Sidak’s multiple comparison test was used. Fig 6 Alignment and conservation of MBD-containing proteins (a) The structural domains of hMeCP2 with domain-specific multiple sequence alignment of select MBD-family proteins in human (h) and Drosophila (d). Identical sequences are highlighted in various shades of blue depending on the degree of conservation across groups. The histogram (yellow) represents conserved physico-chemical properties for each column of the alignment. Higher scores (max=10) for non-identical columns indicate amino acid substitutions that belong to the same physico-chemical class (Livingstone & Barton, 1993). (b) A schematic diagram depicting the size and conserved domains of dMBD-2/3. (c) Schematic representation of dMBD-R2 showing the conserved structural domains. (d) A structural model of the dMBD2/3 MBD domain (Template: MBD3 (pdb: 2mb7), sequence identity = 40.9%, GA341 score = 0.955, z-DOPE score = −0.234. (e) For semi-quantitative RT-PCR experiments, RNA from the heads of adults expressing dMBD2/3-IR in OA neurons (n-syb-Gal4-gal4;UAS-dMBD2/3-IR, blue column), and controls (n-syb-gal4-Gal4/+, white column; UAS-dMBD2/3-IR/+, gray column). dMBD-2/3 transcript levels were significantly reduced in n-syb-Gal4-gal4;UAS-dMBD2/3-IR adults as compared to age-matched control adults (Ordinary one way ANOVA, Padj=0.0026). Reactions were performed in quadruplicate. Rpl32 expression was used as the reference control to normalize expression between treatment groups (error bars indicate s.e.m.). (f) A structural model of the dMBD-R2 MBD domain (Template: MeCP2 (pdb: 3c2i), sequence identity = 34%, GA341 score = 0.931, z-DOPE score = −0.213). (g) RNA from the heads of adults expressing dMBD-R2-IR (n-syb-Gal4-gal4;UAS-dMBD-R2-IR, blue column), and controls (n-syb-gal4-Gal4/+, white column; UAS-dMBD-R2-IR/+, gray column) were used for semi-quantitative RT-PCR experiments. dMBD-R2 transcript levels were significantly reduced in n-syb-Gal4-gal4;UAS-dMBD-R2-IR adults as compared to age-matched control adults (Ordinary one way ANOVA, Padj=0.0045). Reactions were performed in quadruplicate. Cdc2 expression was used as the reference control to normalize expression between treatment groups. Fig 7 Concomitant reduction of dMBD-R2 and hMeCP2 expression rescues hMeCP2-mediated sleep deficits (a–e) Sleep quality and quantity exhibited by individual males averaged over an 8-day period from control and experimental groups. (a) Eduction graph displaying 30 minute bins of averaged sleep between males expressing dMBD-R2-IR in OA neurons (tdc2-gal4;UAS-dMBD-R2-IR) and controls (daytime: white bar; nighttime: black bar, shaded grey). Total sleep over a 24 hr period is increased (see quantification in panel e) with the ZT1-3 interval (arrow) of increased sleep noted. (b) MBD-R2-deficient males displayed an increase in total sleep as compared to controls (Padj<0.0001). (c) Waking activity is increased in experimental males (tdc2-gal4;UAS-dMBD-R2-IR) vs. controls (Padj<0.0001). (c) Sleep fragmentation as measured by an increase in the number of sleep bouts (Padj<0.0) and a decrease in the consolidation index (e) occurred in tdc2-gal4;UAS-dMBD-R2-IR males as compared to controls (Padj=0.001Data are shown as means ± standard error of the mean (SEM). (f) Eduction graph displaying 30 minute bins of averaged sleep between males expressing hMeCP2 in OA neurons, males expressing hMeCP2 and dMBD-R2-IR (tdc2-gal4;UAS-hMeCP2/UAS-dMBD-R2-IR) and controls (daytime: white bar; nighttime: black bar, shaded grey). The phase-specific sleep reductions quantified in tdc2-gal4;UAS-hMeCP2 males (red square line) have been rescued to control levels with the reduction in dMBD-R2 levels (arrows). (g–h) The effect of relative dMBD-R2 expression on sleep output in the presence or absence of hMeCP2, was quantified using a two-way multivariate analysis of variance (MANOVA). Using Pillais’ trace and 0.05 criterion for significance, a significant interaction (dMBD-R2 × hMeCP2) effect was observed between relative dMBD-R2 expression and hMeCP2 gain of function on combined measures of sleep (F(3, 190) = 28.192, p < 0.0001; V = 0.308; Obs. Power = 1.00). This interaction effect explained 32.2% of multivariate variance of sleep composite in dMBD2/3-deficient males and 30.8% of multivariate variance in dMBDR2-deficient males (V = partial η2). This interaction effect explained 32.2% of multivariate variance of sleep composite in dMBD2/3-deficient males and 30.8% of multivariate variance in dMBDR2-deficient males (V = partial η2). (i–k) Polytene chromosomes from 48B10-gal4 UAS-hMeCP2 3rd instar larvae. Both dMBDR2 (red) and hMeCP2 (green) display extensive chromosomal binding. Co-immunofluorescence is observed at selected bands (arrowheads, PCC: r = 0.508; MCC1: 0.64, MCC2: 0.694 ; Costes’ randomization test: P-value=100%). Individual channels in panels (e–f) correspond to the white ROI. Fig 8 Overexpression of dMBD2/3 in OA neurons causes phase-specific sleep decreases similar to hMeCP2 expression (a) Locations of the overexpression EP lines, AP-1mu[EP1112] (red arrowhead) and MBD2/3[EP04582] (orange arrowhead) are depicted in relation to the MBD2/3 loci as well as the approximate location of the targeted UAS-driven inverted repeat sequences (blue). RNA from the heads of tdc2-Gal4-gal4;UAS-AP-1muEP1112adults (red column) and the UAS-AP-1muEP1112/+ (gray column) were used for semi-quantitative RT-PCR experiments. dMBD2/3 transcript levels were significantly increased in tdc2-Gal4-gal4;UAS-AP-1muEP1112 adults as compared to age-matched control adults (unpaired t test, P=0.0103). Reactions were performed in quadruplicate. RPL32 expression was used as the reference control to normalize expression between treatment groups. (b) tdc2-gal4/;UAS-AP-1muEP1112 males (n=46) displayed a reduction in the average amount of sleep during Zeitgeber hours ZT04-10, (p<0.0001; Kruskal Wallis test) and ZT15-17.5, (p<0.0001) as compared to controls, tdc2-gal4/+ (n=28) and UAS-AP-1muEP1112/+ (n=20). (c) Sleep fragmentation as measured by the average number of sleep bouts per day in tdc2-gal4/;UAS-AP-1muEP1112 males is increased as compared to controls (Padj<0.0001, Kruskal-Wallis with Dunn’s post-hoc test). (d) A reduction in the average amount of sleep during ZT04-10, (p<0.0001; Kruskal Wallis test) and ZT15-17.5, (p<0.0001) decreased in tdc2-gal4/; UAS-MBD2/3EP04582 males (n=23, orange column) as compared to controls, tdc2-gal4/+ (n=28) and UAS-MBD2/3EP04582/+ (n=57). (e) The average amount of sleep is not altered between controls (tdc2-gal4/+, n=43, white column; UAS-dMBD2/3-IR/+, n=33, gray column) and males with reduced dMBD2/3 levels in OA neurons, tdc2-gal4;UAS-dMBD2/3-IR, n=31, blue column) during ZT14.5-22 (P=0.1374). During ZT04-9 the average sleep of tdc2-gal4;UAS-dMBD2/3-IR males does not differ from the UAS-dMBD2/3-IR/+ control (P=0.4801). The two control groups are statistically different (P<0.0001). (f) The average number of sleep bouts per 24-hr period is increased in tdc2-gal4; UAS-dMBD2/3-IR males as compared to controls (Padj=0.0041). (g) The consolidation index is significantly reduced in dMBD2/3-deficient males as compared to controls (Padj=0.0032). (h, i) The effect of relative dMBD2/3 expression on sleep output in the presence or absence of hMeCP2, was quantified using a two-way multivariate analysis of variance (MANOVA). Using Pillais’ trace and 0.05 criterion for significance, a significant interaction (dMBD2/3 × hMeCP2) effect was observed between relative dMBD2/3 expression and hMeCP2 gain of function on combined measures of sleep (F(3, 194) = 30.665, p < 0.0001; V = 0.322; Obs. Power = 1.00). Alekseyenko OV Lee C Kravitz EA 2010 Targeted manipulation of serotonergic neurotransmission affects the escalation of aggression in adult male Drosophila melanogaster PLoS ONE 5 e10806 20520823 Altschul SF Madden TL Schaffer AA Zhang J Zhang Z Miller W Lipman DJ 1997 Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acids Res 25 3389 3402 9254694 Alvarez-Saavedra M Antoun G Yanagiya A Oliva-Hernandez R Cornejo-Palma D Perez-Iratxeta C Sonenberg N Cheng HY 2011 miRNA-132 orchestrates chromatin remodeling and translational control of the circadian clock Hum Mol Genet 20 731 751 21118894 Anderson MJ 2001 A new method for non-parametric multivariate analysis of variance Austral ecology 26 1 32 46 Angriman M Caravale B Novelli L Ferri R Bruni O 2015 Sleep in Children with Neurodevelopmental Disabilities Neuropediatrics Ausio J de Paz AM Esteller M 2014 MeCP2: the long trip from a chromatin protein to neurological disorders Trends Mol Med 20 487 498 24766768 Azzi A Dallmann R Casserly A Rehrauer H Patrignani A Maier B Kramer A Brown SA 2014 Circadian behavior is light-reprogrammed by plastic DNA methylation Nat Neurosci 17 377 382 24531307 Ballestar E Wolffe AP 2001 Methyl-CpG-binding proteins. Targeting specific gene repression Eur J Biochem 268 1 6 11121095 Baubec T Ivanek R Lienert F Schubeler D 2013 Methylation-dependent and -independent genomic targeting principles of the MBD protein family Cell 153 480 492 23582333 Bellen HJ Tong C Tsuda H 2010 100 years of Drosophila research and its impact on vertebrate neuroscience: a history lesson for the future Nat Rev Neurosci 11 514 522 20383202 Boffelli D Takayama S Martin DI 2014 Now you see it: genome methylation makes a comeback in Drosophila Bioessays 36 1138 1144 25220261 Bogdanovic O Veenstra GJ 2009 DNA methylation and methyl-CpG binding proteins: developmental requirements and function Chromosoma 118 549 565 19506892 Capelson M Liang Y Schulte R Mair W Wagner U Hetzer MW 2010 Chromatin-bound nuclear pore components regulate gene expression in higher eukaryotes Cell 140 372 383 20144761 Capuano F Mulleder M Kok R Blom HJ Ralser M 2014 Cytosine DNA methylation is found in Drosophila melanogaster but absent in Saccharomyces cerevisiae, Schizosaccharomyces pombe, and other yeast species Anal Chem 86 3697 3702 24640988 Certel SJ Leung A Lin CY Perez P Chiang AS Kravitz EA 2010 Octopamine neuromodulatory effects on a social behavior decision-making network in Drosophila males PLoS ONE 5 Chahrour M Zoghbi HY 2007 The story of Rett syndrome: from clinic to neurobiology Neuron 56 422 437 17988628 Chapleau CA Calfa GD Lane MC Albertson AJ Larimore JL Kudo S Armstrong DL Percy AK Pozzo-Miller L 2009 Dendritic spine pathologies in hippocampal pyramidal neurons from Rett syndrome brain and after expression of Rett-associated MECP2 mutations Neurobiol Dis 35 219 233 19442733 Chen L Chen K Lavery LA Baker SA Shaw CA Li W Zoghbi HY 2015 MeCP2 binds to non-CG methylated DNA as neurons mature, influencing transcription and the timing of onset for Rett syndrome Proc Natl Acad Sci U S A 112 5509 5514 25870282 Cheng TL Wang Z Liao Q Zhu Y Zhou WH Xu W Qiu Z 2014 MeCP2 suppresses nuclear microRNA processing and dendritic growth by regulating the DGCR8/Drosha complex Developmental cell 28 547 560 24636259 Cirelli C 2009 The genetic and molecular regulation of sleep: from fruit flies to humans Nat Rev Neurosci 10 549 560 19617891 Cohen S Gabel HW Hemberg M Hutchinson AN Sadacca LA Ebert DH Harmin DA Greenberg RS Verdine VK Zhou Z Wetsel WC West AE Greenberg ME 2011 Genome-wide activity-dependent MeCP2 phosphorylation regulates nervous system development and function Neuron 72 72 85 21982370 Cole SH Carney GE McClung CA Willard SS Taylor BJ Hirsh J 2005 Two functional but noncomplementing Drosophila tyrosine decarboxylase genes: distinct roles for neural tyramine and octopamine in female fertility J Biol Chem 280 14948 14955 15691831 Cortesi F Giannotti F Ivanenko A Johnson K 2010 Sleep in children with autistic spectrum disorder Sleep Med 11 659 664 20605110 Costes SV Daelemans D Cho EH Dobbin Z Pavlakis G Lockett S 2004 Automatic and quantitative measurement of protein-protein colocalization in live cells Biophysical journal 86 3993 4003 15189895 Crocker A Sehgal A 2008 Octopamine regulates sleep in Drosophila through protein kinase A-dependent mechanisms J Neurosci 28 9377 9385 18799671 Crocker A Sehgal A 2010 Genetic analysis of sleep Genes Dev 24 1220 1235 20551171 Crocker A Shahidullah M Levitan IB Sehgal A 2010 Identification of a neural circuit that underlies the effects of octopamine on sleep:wake behavior Neuron 65 670 681 20223202 Cukier HN Perez AM Collins AL Zhou Z Zoghbi HY Botas J 2008 Genetic modifiers of MeCP2 function in Drosophila PLoS Genet 4 e1000179 18773074 Della Ragione F Filosa S Scalabri F D’Esposito M 2012 MeCP2 as a genome-wide modulator: the renewal of an old story Front Genet 3 181 22973303 Fatemi M Wade PA 2006 MBD family proteins: reading the epigenetic code J Cell Sci 119 3033 3037 16868031 Friggi-Grelin F Iche M Birman S 2003 Tissue-specific developmental requirements of Drosophila tyrosine hydroxylase isoforms Genesis 35 260 269 12717737 Gao J Wang WY Mao YW Graff J Guan JS Pan L Mak G Kim D Su SC Tsai LH 2010 A novel pathway regulates memory and plasticity via SIRT1 and miR-134 Nature 466 1105 1109 20622856 Gehring M 2013 Genomic imprinting: insights from plants Annu Rev Genet 47 187 208 24016190 Ghosh RP Nikitina T Horowitz-Scherer RA Gierasch LM Uversky VN Hite K Hansen JC Woodcock CL 2010 Unique physical properties and interactions of the domains of methylated DNA binding protein 2 Biochemistry 49 4395 4410 20405910 Guo JU Su Y Shin JH Shin J Li H Xie B Zhong C Hu S Le T Fan G Zhu H Chang Q Gao Y Ming GL Song H 2014 Distribution, recognition and regulation of non-CpG methylation in the adult mammalian brain Nat Neurosci 17 215 222 24362762 Hammer Ø Harper DAT Ryan PD 2001 PAST: Paleontological statistics software package for education and data analysis Palaeontologia Electronica 4 1 9p Hardie SL Zhang JX Hirsh J 2007 Trace amines differentially regulate adult locomotor activity, cocaine sensitivity, and female fertility in Drosophila melanogaster Dev Neurobiol 67 1396 1405 17638385 Hendrich B Tweedie S 2003 The methyl-CpG binding domain and the evolving role of DNA methylation in animals Trends Genet 19 269 277 12711219 Hess-Homeier DL Fan CY Gupta T Chiang AS Certel SJ 2014 Astrocyte-specific regulation of hMeCP2 expression in Drosophila Biol Open Ho KS Sehgal A 2005 Drosophila melanogaster: an insect model for fundamental studies of sleep Methods Enzymol 393 772 793 15817324 Hughes ME Grant GR Paquin C Qian J Nitabach MN 2012 Deep sequencing the circadian and diurnal transcriptome of Drosophila brain Genome Res 22 1266 1281 22472103 Hutchinson AN Deng JV Cohen S West AE 2012 Phosphorylation of MeCP2 at Ser421 contributes to chronic antidepressant action J Neurosci 32 14355 14363 23055506 Ishimoto H Lark A Kitamoto T 2012 Factors that Differentially Affect Daytime and Nighttime Sleep in Drosophila melanogaster Front Neurol 3 24 22375135 Kakkar AK Dahiya N 2015 Management of Parkinsons disease: Current and future pharmacotherapy Eur J Pharmacol 750 74 81 25637088 Kudo S Nomura Y Segawa M Fujita N Nakao M Dragich J Schanen C Tamura M 2001 Functional analyses of MeCP2 mutations associated with Rett syndrome using transient expression systems Brain Dev 23 Suppl 1 S165 173 11738866 Kula-Eversole E Nagoshi E Shang Y Rodriguez J Allada R Rosbash M 2010 Surprising gene expression patterns within and between PDF-containing circadian neurons in Drosophila Proc Natl Acad Sci U S A 107 13497 13502 20624977 Laccone F Huppke P Hanefeld F Meins M 2001 Mutation spectrum in patients with Rett syndrome in the German population: Evidence of hot spot regions Hum Mutat 17 183 190 11241840 Liu C Chung M 2015 Genetics and epigenetics of circadian rhythms and their potential roles in neuropsychiatric disorders Neuroscience bulletin 31 141 159 25652815 Livingstone CD Barton GJ 1993 Protein sequence alignments: a strategy for the hierarchical analysis of residue conservation Comput Appl Biosci 9 745 756 8143162 Lv J Xin Y Zhou W Qiu Z 2013 The epigenetic switches for neural development and psychiatric disorders J Genet Genomics 40 339 346 23876774 Lyst MJ Bird A 2015 Rett syndrome: a complex disorder with simple roots Nat Rev Genet 16 261 275 25732612 Marhold J Brehm A Kramer K 2004a The Drosophila methyl-DNA binding protein MBD2/3 interacts with the NuRD complex via p55 and MI-2 BMC Mol Biol 5 20 15516265 Marhold J Kramer K Kremmer E Lyko F 2004b The Drosophila MBD2/3 protein mediates interactions between the MI-2 chromatin complex and CpT/A-methylated DNA Development 131 6033 6039 15537686 Martinez de Paz A Vicente Sanchez-Mut J Samitier-Marti M Petazzi P Saez M Szczesna K Huertas D Esteller M Ausio J 2015 Circadian Cycle-Dependent MeCP2 and Brain Chromatin Changes PLoS One 10 e0123693 25875630 McCarthy MJ Welsh DK 2012 Cellular circadian clocks in mood disorders J Biol Rhythms 27 339 352 23010657 Melo F Sanchez R Sali A 2002 Statistical potentials for fold assessment Protein Sci 11 430 448 11790853 Mitchell HA Weinshenker D 2010 Good night and good luck: norepinephrine in sleep pharmacology Biochem Pharmacol 79 801 809 19833104 Monastirioti M 1999 Biogenic amine systems in the fruit fly Drosophila melanogaster Microsc Res Tech 45 106 121 10332728 Monastirioti M Linn CE Jr White K 1996 Characterization of Drosophila tyramine beta-hydroxylase gene and isolation of mutant flies lacking octopamine J Neurosci 16 3900 3911 8656284 Musiek ES Xiong DD Holtzman DM 2015 Sleep, circadian rhythms, and the pathogenesis of Alzheimer disease Exp Mol Med 47 e148 25766617 Nan X Ng HH Johnson CA Laherty CD Turner BM Eisenman RN Bird A 1998 Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex Nature 393 386 389 9620804 Nikitina T Shi X Ghosh RP Horowitz-Scherer RA Hansen JC Woodcock CL 2007 Multiple modes of interaction between the methylated DNA binding protein MeCP2 and chromatin Mol Cell Biol 27 864 877 17101771 Nomura Y 2005 Early behavior characteristics and sleep disturbance in Rett syndrome Brain Dev 27 Suppl 1 S35 S42 16182496 Pascual-Garcia P Jeong J Capelson M 2014 Nucleoporin Nup98 associates with Trx/MLL and NSL histone-modifying complexes and regulates Hox gene expression Cell Rep 9 433 442 25310983 Pettersen EF Goddard TD Huang CC Couch GS Greenblatt DM Meng EC Ferrin TE 2004 UCSF Chimera--a visualization system for exploratory research and analysis J Comput Chem 25 1605 1612 15264254 Pfeiffenberger C Lear BC Keegan KP Allada R 2010 Processing sleep data created with the Drosophila Activity Monitoring (DAM) System Cold Spring Harb Protoc 2010 pdb prot5520 Piazza CC Fisher W Kiesewetter K Bowman L Moser H 1990 Aberrant sleep patterns in children with the Rett syndrome Brain Dev 12 488 493 2288379 Pinney SE 2014 Mammalian Non-CpG Methylation: Stem Cells and Beyond Biology (Basel) 3 739 751 25393317 Pitman JL McGill JJ Keegan KP Allada R 2006 A dynamic role for the mushroom bodies in promoting sleep in Drosophila Nature 441 753 756 16760979 Potdar S Sheeba V 2013 Lessons from sleeping flies: insights from Drosophila melanogaster on the neuronal circuitry and importance of sleep J Neurogenet 27 23 42 23701413 Prestel M Feller C Straub T Mitlohner H Becker PB 2010 The activation potential of MOF is constrained for dosage compensation Mol Cell 38 815 826 20620953 Qureshi IA Mehler MF 2014 Epigenetics of sleep and chronobiology Current neurology and neuroscience reports 14 432 24477387 Ramocki MB Peters SU Tavyev YJ Zhang F Carvalho CM Schaaf CP Richman R Fang P Glaze DG Lupski JR Zoghbi HY 2009 Autism and other neuropsychiatric symptoms are prevalent in individuals with MeCP2 duplication syndrome Ann Neurol 66 771 782 20035514 Remmert M Biegert A Hauser A Soding J 2012 HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment Nat Methods 9 173 175 Robbins TW 1997 Arousal systems and attentional processes Biol Psychol 45 57 71 9083644 Roder K Hung MS Lee TL Lin TY Xiao H Isobe KI Juang JL Shen CJ 2000 Transcriptional repression by Drosophila methyl-CpG-binding proteins Mol Cell Biol 20 7401 7409 10982856 Roloff TC Ropers HH Nuber UA 2003 Comparative study of methyl-CpG-binding domain proteins BMC Genomics 4 1 12529184 Rorth P 1996 A modular misexpression screen in Drosophila detecting tissue-specific phenotypes Proc Natl Acad Sci U S A 93 12418 12422 8901596 Sali A Blundell TL 1993 Comparative protein modelling by satisfaction of spatial restraints J Mol Biol 234 779 815 8254673 Samaco RC Neul JL 2011 Complexities of Rett syndrome and MeCP2 J Neurosci 31 7951 7959 21632916 Sasai N Defossez PA 2009 Many paths to one goal? The proteins that recognize methylated DNA in eukaryotes Int J Dev Biol 53 323 334 19412889 Schubeler D 2015 Function and information content of DNA methylation Nature 517 321 326 25592537 Shang Y Griffith LC Rosbash M 2008 Light-arousal and circadian photoreception circuits intersect at the large PDF cells of the Drosophila brain Proc Natl Acad Sci U S A 105 19587 19594 19060186 Shaw PJ Cirelli C Greenspan RJ Tononi G 2000 Correlates of sleep and waking in Drosophila melanogaster Science 287 1834 1837 10710313 Shen MY Sali A 2006 Statistical potential for assessment and prediction of protein structures Protein Sci 15 2507 2524 17075131 Shepherd GM Katz DM 2011 Synaptic microcircuit dysfunction in genetic models of neurodevelopmental disorders: focus on Mecp2 and Met Curr Opin Neurobiol 21 827 833 21733672 Skene PJ Illingworth RS Webb S Kerr AR James KD Turner DJ Andrews R Bird AP 2010 Neuronal MeCP2 is expressed at near histone-octamer levels and globally alters the chromatin state Mol Cell 37 457 468 20188665 Souders MC Mason TB Valladares O Bucan M Levy SE Mandell DS Weaver TE Pinto-Martin J 2009 Sleep behaviors and sleep quality in children with autism spectrum disorders Sleep 32 1566 1578 20041592 Takayama S Dhahbi J Roberts A Mao G Heo SJ Pachter L Martin DI Boffelli D 2014 Genome methylation in D. melanogaster is found at specific short motifs and is independent of DNMT2 activity Genome Res 24 821 830 24558263 Urdinguio RG Lopez-Serra L Lopez-Nieva P Alaminos M Diaz-Uriarte R Fernandez AF Esteller M 2008 Mecp2-null mice provide new neuronal targets for Rett syndrome PLoS One 3 e3669 18989361 Varriale A 2014 DNA methylation, epigenetics, and evolution in vertebrates: facts and challenges Int J Evol Biol 2014 475981 24551476 Venken KJ Bellen HJ 2014 Chemical mutagens, transposons, and transgenes to interrogate gene function in Drosophila melanogaster Methods 68 15 28 24583113 Vonhoff F Williams A Ryglewski S Duch C 2012 Drosophila as a model for MECP2 gain of function in neurons PLoS One 7 e31835 22363746 Wan M Lee SS Zhang X Houwink-Manville I Song HR Amir RE Budden S Naidu S Pereira JL Lo IF Zoghbi HY Schanen NC Francke U 1999 Rett syndrome and beyond: recurrent spontaneous and familial MECP2 mutations at CpG hotspots Am J Hum Genet 65 1520 1529 10577905 Yasui DH Xu H Dunaway KW Lasalle JM Jin LW Maezawa I 2013 MeCP2 modulates gene expression pathways in astrocytes Mol Autism 4 3 23351786 Young D Nagarajan L de Klerk N Jacoby P Ellaway C Leonard H 2007 Sleep problems in Rett syndrome Brain Dev 29 609 616 17531413 Zilberman D 2008 The evolving functions of DNA methylation Curr Opin Plant Biol 11 554 559 18774331 Zimmermann CA Hoffmann A Raabe F Spengler D 2015 Role of mecp2 in experience-dependent epigenetic programming Genes (Basel) 6 60 86 25756305

PMC005xxxxxx/PMC5119462.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101190261 31733 Nat Rev Microbiol Nat. Rev. Microbiol. Nature reviews. Microbiology 1740-1526 1740-1534 27211789 5119462 10.1038/nrmicro.2016.46 NIHMS829529 Article Reassortment in segmented RNA viruses: mechanisms and outcomes McDonald Sarah M. 12 Nelson Martha I. 3 Turner Paul E. 4 Patton John T. 5 1 Virginia Tech Carilion School of Medicine and Research Institute, 2 Riverside Circle, Roanoke, Virginia 24016, USA 2 Department of Biomedical Sciences and Pathobiology, Virginia–Maryland College of Veterinary Medicine, 205 Duck Pond Drive, Blacksburg, Virginia 24061, USA 3 Fogarty International Center, National Institutes of Health, 31 Center Drive, MSC 2220, Bethesda, Maryland 20892, USA 4 Department of Ecology and Evolutionary Biology, Yale University, Osborn Memorial Labs, 165 Prospect Street, P. O. Box 208106, New Haven, Connecticut 06520, USA 5 Virginia–Maryland College of Veterinary Medicine, University of Maryland, 8075 Greenmead Drive, College Park, Maryland 20742, USA Correspondence to S.M.M. mcdonaldsa@vtc.vt.edu 12 11 2016 23 5 2016 7 2016 22 11 2016 14 7 448460 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Segmented RNA viruses are widespread in nature and include important human, animal and plant pathogens, such as influenza viruses and rotaviruses. Although the origin of RNA virus genome segmentation remains elusive, a major consequence of this genome structure is the capacity for reassortment to occur during co-infection, whereby segments are exchanged among different viral strains. Therefore, reassortment can create viral progeny that contain genes that are derived from more than one parent, potentially conferring important fitness advantages or disadvantages to the progeny virus. However, for segmented RNA viruses that package their multiple genome segments into a single virion particle, reassortment also requires genetic compatibility between parental strains, which occurs in the form of conserved packaging signals, and the maintenance of RNA and protein interactions. In this Review, we discuss recent studies that examined the mechanisms and outcomes of reassortment for three well-studied viral families — Cystoviridae, Orthomyxoviridae and Reoviridae — and discuss how these findings provide new perspectives on the replication and evolution of segmented RNA viruses. Viruses that maintain their genomes as several distinct RNA molecules are called segmented RNA viruses. They are ubiquitous in nature, infecting a wide variety of animals, plants and bacteria. To date, 11 different segmented RNA virus families have been described in the literature: Arenaviridae, Birnaviridae, Bromoviridae, Bunyaviridae, Chrysoviridae, Closteroviridae, Cystoviridae, Orthomyxoviridae, Partitiviridae, Picobirnaviridae and Reoviridae (TABLE 1). These families and their type species can differ considerably in their virion structure, genome architecture, replication cycle, host tropism and pathology. For example, some segmented RNA virus species, such as influenza A virus and rotavirus A, are associated with substantial human disease and serious economic burdens to society1–3. Other segmented RNA virus species affect humans indirectly through the infection of domestic animals and crops (for example, bluetongue virus, infectious bursal disease virus and tomato spotted wilt virus)4–6. However, a shared feature of all segmented RNA viruses is their capacity to exchange genome segments in toto during co-infection through a process called reassortment. Specifically, when two or more viruses infect a single host cell, they can package each other’s genome segments into a nascent virion, thereby producing hybrid progeny (FIG. 1a). For multipartite viruses in the Bromoviridae, Chrysoviridae, Partitiviridae and Picobirnaviridae families, which incorporate their genome segments into several independent virus particles (TABLE 1), reassortment is stochastic and creates virions that have a random mix of genome segments from each parent. By contrast, for viruses that package their genome segments into a single virion, such as species in the Cystoviridae, Orthomyxoviridae and Reoviridae families, reassortment generally results in segment replacement, such that one co-infecting virus incorporates the genome segment (or segments) of another co-infecting virus in place of its own. In this case, genetic exchange requires the conservation of intricate assortment signals and preservation of the RNA–RNA and/or RNA–protein interactions that mediate genome packaging. For this reason, strain-specific differences in the sequences or structures of homologous RNAs and/or in the packaging proteins of co-infecting parent viruses can severely restrict the generation of reassortant progeny during co-infection. Moreover, for reassortants to selectively emerge at appreciable levels in the viral population, they must have a genomic composition that confers at least some modest advantage to viral fitness. Conceptually, reassortment shares some features with sexual reproduction in eukaryotes, whereby chromosomes are segregated during meiosis and combined in various ways during gamete fusion (FIG. 1b). Sexual reproduction is argued to be evolutionarily advantageous to eukaryotes, in part, because it purges deleterious mutations and increases population-level genetic diversity, which is a prerequisite to evolution by natural selection7–9. Thus, by analogy to sexual reproduction, one theory posits that the reassortment capacity of segmented RNA viruses contributes to the maintenance of this genome structure10–13. However, rapidly evolving viral populations have more opportunities to remove unfit mutations than eukaryotes, and reassortment is clearly not necessary for the evolutionary success of the numerous non-segmented viruses. Therefore, it is possible that reassortment is a by-product of genome segmentation, rather than a key evolutionary driver of such a genome structure. Indeed, the evolution of genome segmentation may have been driven by other advantages that are conferred by this arrangement, such as the control of gene expression14, increased coding potential of the genome15 and enhanced stability of such virus particles16. Regardless of the original drivers of segmentation, the capacity of important human pathogens such as influenza A viruses and rotavirus A strains to reassort has important implications for their ongoing evolution and impact on global health. The mechanisms and outcomes of reassortment can differ from recombination, which is another form of genetic exchange that occurs readily for some non-segmented RNA viruses, particularly those with positive-sense RNA ((+)RNA) genomes17 (FIG. 1c). During recombination, the viral polymerase begins copying the RNA template of one parental strain, and it then switches templates mid-synthesis to use that of a different parental strain. Therefore, the result of recombination is the generation of chimeric RNA molecules that contain regions of nucleotide sequence derived from each parent. Unlike reassortment, during which entire genes (or sets of genes) are exchanged by the swapping of segments, recombination can occur nearly anywhere in the RNA genome, even in the middle of a gene. Therefore, recombination can result in the formation of non-functional chimeric fusion proteins, whereas reassortment cannot17. In other words, reassortment is a mechanism that maintains the ORF of a gene and, consequently, maintains protein integrity, whereas recombination typically introduces changes in ORFs and their encoded proteins. Recombination has rarely been reported for segmented RNA viruses18–21, and some of the detected recombination events may be the result of sequencing artefacts22. The lack of robust recombination among segmented RNA viruses is likely to be a reflection of their biology, specifically related to the manner in which their polymerases transcribe and replicate the genome segments in the absence of template switching. A detailed discussion of the mechanism, origins and evolutionary consequences of recombination in RNA viruses compared with reassortment is provided in REF. 23. In this Review, we focus on the mechanisms and outcomes of reassortment for non-multipartite, segmented RNA viruses in the well-studied Cystoviridae, Orthomyxoviridae and Reoviridae families. In particular, we describe the strategies that are used by these viruses to ensure efficient incorporation of their genome segments into nascent virions, and we discuss how genetic incompatibilities during segment assortment and packaging can directly restrict the generation of reassortants during co-infection. We also highlight recent experimental and comparative genomic studies that elucidate the possible selection pressures that promote or temper the emergence of reassortant viruses in the population. Our goal is to provide new perspectives on the replication and evolution of segmented RNA viruses, which may in turn stimulate the development of measures for the prevention of disease. Genome segment assortment and packaging Cystoviridae The Cystoviridae family is composed of segmented double-stranded RNA (dsRNA) viruses that infect Gram-negative bacteria20. The type species for this family is Pseudomonas phage φ6 (hereafter referred to as φ6), an extensively researched bacteriophage that primarily replicates within the various pathovars of the plant pathogen Pseudomonas syringae. Since its discovery in the early 1970s24, φ6 has been used as a tractable model system to test evolutionary hypotheses within controlled laboratory settings and to uncover mechanisms of virus biology. Additional Cystoviridae family members have been found at various geographical locations around the world, which suggests that these viruses are widespread in nature25–27. The φ6 virion consists of an outer lipid envelope surrounding a nucleocapsid shell and an icosahedral procapsid core28. Within the core reside three dsRNA genome segments, totalling >13 kb in length and encoding 13 viral proteins25 (FIG. 2a). The segments each contain several ORFs that are flanked by 5′ and 3′ UTRs, and they are named according to their sizes: small (S; 2.9 kb), medium (M; 4.1 kb) and large (L; 6.4 kb). During the replication cycle of φ6, dsRNA genome segments are transcribed into (+)RNA molecules by viral polymerases29. In addition to acting as templates for protein synthesis, these (+)RNAs are the form of the φ6 genome that is incorporated into nascent particles25 (FIG. 2b). Using an in vitro packaging system, it was shown that φ6 (+)RNAs are inserted individually and sequentially into a pre-formed procapsid core through an entry portal at one five-fold icosahedral axis30,31 (FIG. 2b). The empty core initially displays only the binding site for the S (+)RNA segment, leading to its recruitment and packaging. Thereafter, a conformational change occurs in the core that reveals a binding site for the M (+)RNA segment32. Again, only after packaging of the M segment is the binding site for the L (+)RNA segment revealed. It was also demonstrated through in vitro assays that the cis-acting RNA sequence and structural elements that are crucial for packaging are located in the 5′ UTRs33 (FIG. 2a). A 5′-terminal 18 bp sequence is shared among the S, M and L segments and enables φ6 to distinguish between viral RNAs and host RNAs. The gene-specific packaging signals that differentiate S, M and L segments during packaging are located ~200 bp downstream of the 18 bp conserved sequence. Following encapsidation of all three φ6 (+)RNAs, the procapsid core expands, thereby triggering the core-associated viral polymerases to convert the (+)RNAs into dsRNA genome segments through a single round of negative-sense RNA ((−) RNA) synthesis20. Additional virion morphogenesis, including the acquisition of an outer envelope, leads to the production of fully infectious φ6 particles. It is predicted, albeit not experimentally demonstrated, that the vast majority of nascent φ6 virions that are produced during the viral life cycle contain all three genome segments. This prediction is based on the observations that (+)RNA packaging is sequential and inter-segmentally dependent, and that the three φ6 genome segments are present in equimolar amounts at the viral population level. Nevertheless, it has been shown that φ6 packaging can be drastically manipulated in vitro, yielding particles with more or fewer than three genome segments or with rearranged genome segments34–36. For example, one study created a φ6 mutant that did not package the S segment owing to an amino acid substitution in one of its core proteins35. This mutant still efficiently packaged and replicated the M and L segments, thereby propagating a virus that contains two segments in a non-lytic carrier state in the bacterial host. Furthermore, it was shown that the entire φ6 genome can be concatenated into a single RNA molecule and still produce a viable mutant with only moderate replication defects36. The observation that non-segmented variants of φ6 can be created in the laboratory but do not emerge at detectable levels in nature suggests that genome segmentation provides a fitness advantage. Orthomyxoviridae The Orthomyxoviridae family of segmented (−)RNA viruses consists of six different genera, three of which (Influenzavirus A, Influenzavirus B and Influenzavirus C) cause respiratory disease in humans37. Of these three genera, Influenzavirus A (consisting of a single species, influenza A virus) imparts the largest medical and economic burdens; seasonal epidemics of strains of influenza A virus account for 27,000–55,000 deaths each year in the United States alone, with an annual cost of US$87.1 billion to the healthcare system2,38. Influenza A viruses can also cause pandemics, the most severe of which occurred in 1918–1919 and is estimated to have killed 20–50 million people globally39. In addition to infecting humans, influenza A viruses are endemic in several other animal species, including pigs, dogs, horses, bats and birds, which provide natural reservoirs for viral evolution40. Influenza A viruses exist as pleomorphic, enveloped virion particles, each encasing a genome of eight (−)RNA molecules (FIG. 2c). The individual genome segments of an influenza A virus range in size from 0.9–2.3 kb in length, and the total genome length is approximately 13.5 kb (REF. 41). Each (−)RNA contains 13 ORFs, in the antisense orientation, which are flanked by 3′ and 5′ UTRs. Altogether, an influenza A virus encodes at least 13 proteins in its eight genome segments. The termini of the viral (−)RNA consist of highly conserved 12–13 bp sequences that can partially anneal with each other in cis so that the molecules fold over and form a corkscrew shape (FIG. 2c). Multiple copies of the viral nucleocapsid protein (NP) bind to the length of each (−)RNA, and a heterotrimeric polymerase complex is attached to the end where the 3′ and 5′ termini connect. Thus, the eight influenza A virus genome segments are packaged into virions as eight distinct ribonucleoprotein (RNP) complexes42,43 (FIG. 2c). The manner in which influenza A viruses package each of their eight genome segments has not yet been fully resolved. However, the available data are most consistent with the notion that this is a selective, non-random process that it is mediated by interactions between the (−)RNA molecules themselves44,45 (FIG. 2d). Individual RNPs are assembled in the nucleus, and they must then translocate to the plasma membrane, where they are incorporated into a budding enveloped virion. Using fluorescence in situ hybridization, it was shown that the RNPs are exported from the nucleus as subcomplexes, which further assort into a supramolecular complex that contains all eight RNPs while trafficking to budding sites46–48. Additional studies support the idea that the subcomplexes consist of specific pairs of (−)RNAs that directly engage each other and that the supramolecular complex is formed through an elaborate interaction network between the (−)RNAs of the subcomplexes49–52. Although an in vitro packaging system is lacking for influenza A viruses, studies of defective-interfering RNAs have shed light on which regions of the viral genome are crucial for the assortment process. Specifically, defective-interfering RNAs have been engineered to contain large deletions in the central ORFs but to maintain the extreme ORF termini as well as the 3′ and 5′ UTRs of the genome segments, and such RNAs are capable of competing with full-length segments for packaging53. This indicates that the segment-specific assortment signals are located within ~300 bp from the termini of the (−)RNA molecules (FIG. 2c). Furthermore, several studies have used reverse genetics approaches to engineer viruses that encapsidate reporter genes, thereby defining those nucleotides that are crucial for the packaging of each (−)RNA segment into a virion44. However, how these sequence elements are recognized in the context of the RNP is unclear. One possibility is that some regions of the (−)RNA termini lack NP, enabling them to adopt local secondary or tertiary structures and to mediate RNA–RNA interactions. It was originally proposed that the packaging efficiency for influenza A viruses was very high and that most nascent virions contained a full complement of all eight RNPs44,45. This theory was supported by structural analysis of individual virions using thin-section electron microscopy and electron tomography43,52,54. However, influenza A virions can be engineered in the laboratory to contain more or fewer than eight genome segments, which suggests that some level of inefficiency is tolerated55,56. Furthermore, additional studies have demonstrated that when cells are infected at a low multiplicity, most fail to express at least one of the viral proteins57, providing evidence for a model of influenza A virus packaging that is less than perfect. This result suggests that the gene encoding the protein that failed to be expressed was defective or missing altogether in these semi-infectious particles. Moreover, the efficiency of segment packaging was found to vary between virus strains and to be influenced by mutations in specific viral proteins57,58. Finally, semi-infectious particles are estimated to outnumber complete particles by 6/1 (REF. 58), and they readily participate in reassortment events during co-infection with complete particles59. Thus, further studies aimed at elucidating the effect of semi-infectious particles on the long-term evolution of influenza A viruses are warranted. Reoviridae The Reoviridae family of segmented dsRNA viruses includes several clinically and economically important human, animal and plant pathogens, such as rotaviruses, bluetongue viruses and rice dwarf viruses1,4,60. Rotaviruses are well-studied members of the Reoviridae family because they cause life-threatening gastroenteritis in infants and young children. Before the worldwide introduction of two vaccines in 2006, strains of the rotavirus A species were estimated to have killed ~450,000 children each year1,61. Strains of rotavirus A also infect numerous mammalian and avian species, including pigs, cows, horses, rabbits, cats, dogs, mice and birds, which are reservoirs for viral evolution. Ongoing epidemiological surveillance data also show that strains from the divergent species rotavirus B and rotavirus C are important causes of morbidity and mortality in pigs and cows62–65, and that they may be underappreciated causes of disease in humans66–69. The rotavirus A virion is a non-enveloped, triple-layered particle that encloses a dsRNA genome of 11 segments70. The viral genome segments range in size from 0.5–3.3 kb, and the genome as a whole totals 23.0 kb (FIG. 2e). The segments are each organized as a central ORF flanked by 5′ and 3′ UTRs. In general, each gene is monocistronic, encoding a single viral protein. An additional ORF has been described in one segment for some rotavirus strains71, enabling the expression of up to 12 proteins in total. The assortment and packaging process of rotaviruses is very poorly understood because the field lacks both in vitro packaging assays and efficient reverse genetics methods. Nevertheless, the available data suggest that this process shares aspects of both φ6 and influenza A virus assortment and packaging72 (FIG. 2f). For example, as for φ6, the dsRNA genome segments of rotavirus A are transcribed by viral polymerases into (+)RNA molecules, which are the form of the genome that is assorted and packaged into nascent virion particles73–76. However, unlike the φ6 genome, the rotavirus A (+)RNAs are not inserted one by one into a pre-formed core. Instead, it is hypothesized that rotavirus A genome assortment occurs in a manner that is similar to influenza A virus genome assortment. In particular, it is thought that the 11 distinct (+)RNAs engage each other through cis-acting RNA elements to form a supramolecular complex that is encapsidated by the core shell protein during early virion assembly (FIG. 2f). The 5′ and 3′ UTR sequences differ for each of the 11 (+)RNAs, but these sequences are highly conserved among homologous gene segments from different strains of rotavirus A. The segment-specific packaging signals for rotavirus A (+)RNAs are predicted to reside within these 5′ and 3′ termini. In silico analyses of nucleotide sequences from strains of rotavirus A have identified several putative RNA structural elements in these terminal regions that may represent assortment signals77,78. For strains of the bluetongue virus and mammalian orthoreovirus species, two other Reoviridae family members, some packaging signals have been identified with the help of in vitro assembly and reverse genetics; they involve the 5′ and 3′ UTRs, and include some coding sequences79–83. Therefore, the location of the packaging signals for rotavirus A strains and other Reoviridae family members may be similar to the location of the influenza A virus signals (FIG. 2f). During or immediately following their packaging into a core assembly intermediate, rotavirus A (+)RNAs are converted into dsRNAs by viral polymerases70. The nascent core assembly intermediate then undergoes additional morphogenesis to become an infectious triple-layered, non-enveloped particle. The efficiency of genome packaging is poorly understood for the Reoviridae family, members of which have 9, 10, 11 or 12 dsRNA genome segments. The observation that no family members have 13 or more segments may be a reflection of the packaging process and the icosahedral capsid architecture. Specifically, each of the 12 fivefold axes of the inner core shell is predicted to have space for only one dedicated polymerase complex and one associated genome segment84. However, it is unclear why some Reoviridae family members package fewer than 12 segments, thereby leaving one or more fivefold vertices unoccupied. For example, strains of rotavirus A have 11 genome segments, which are present in equimolar amounts at the population level. Moreover, variants of rotavirus A with partially duplicated genome segments and/or foreign sequence insertions have been isolated or engineered85–87. This suggests that these viruses can accommodate extra nucleic acid and, theoretically, that they may be able to package additional segments. That being said, there have been no reports of strains of rotavirus A that contain an extra copy of a genome segment, nor have there been reports of variants that lack one or more genome segments. This particular observation is intriguing, given that some strains do not express the accessory protein NSP1 owing to spontaneous deletions or mutations in the NSP1-coding genome segment88,89. For these strains, the defective NSP1-coding genome segment is still efficiently packaged into nascent virions, which suggests that the (+)RNA molecule itself is crucial for viral replication, perhaps during assortment. Therefore, we hypothesize that rotaviruses and other members of the Reoviridae family use an all-or-none packaging mechanism similar to that of φ6. More specifically, we predict that the full complement of (+) RNA segments must be incorporated into a core assembly intermediate for genome replication to take place. In this model, each packaged (+)RNA would have a dedicated polymerase that acts only on the specific associated segment but that functions in concert with the ten other polymerases to simultaneously replicate the dsRNA genome. Future investigations into the details of rotavirus assortment, packaging and genome replication are warranted, but such investigations may require the development of robust in vitro assays. Generation of reassortants during co-infection Given the exquisite selective packaging mechanisms for Cystoviridae, Orthomyxoviridae and Reoviridae family members, it is no surprise that successful reassortment between two parental strains during co-infection requires a high degree of genetic compatibility. More specifically, the capacity of one parental strain to package the genome segment of another requires the maintenance of intricate RNA–RNA and/or protein–RNA interactions. Indeed, there has been no description of reassortment occurring between segmented RNA viruses that belong to different families (for example, an Orthomyxoviridae member and a Reoviridae member); these viruses are simply too divergent to participate in genetic exchange. Even for more closely related viruses within the same genus, subtle differences in viral RNAs and proteins can temper the efficiency with which reassortants are generated during co-infection. It is likely that molecular failures at the level of segment assortment and packaging are a major reason for why the frequency of reassortants is lower than expected following experimental co-infections in the laboratory setting. For influenza viruses, the compatibility of packaging signals in the form of conserved RNA–RNA interactions is a primary determinant that dictates the reassortment potential for any two co-infecting parental strains44 (FIG. 3a). A remarkable example of this was provided by the demonstration that the reassortment restriction between influenza A viruses and influenza B viruses can be overcome, at least for the genome segment encoding haemagglutinin (HA), simply by using reverse genetics to swap the packaging signals90. However, it is important to note that studying the molecular determinants of reassortment restriction using reverse genetics does not fully recapitulate restrictions during co-infection because such studies do not take into account the important role of competition among homologous segments. In support of this idea, it was shown that although reverse genetics can create all possible reassortants between an avian and a human influenza A virus strain, such hybrid progeny are not readily produced during co-infection91. The reason for this discrepancy is related to the fact that the human influenza A virus RNAs interact suboptimally with those from avian strains (FIG. 3a). In other words, low-affinity interactions between the non-cognate RNAs (that is, those that are derived from different parental viruses) are readily outcompeted by the optimal, higher-affinity interactions between cognate RNAs (that is, those that are derived from the same parental virus). In addition to influencing the overall frequency of reassortants for influenza A viruses, subtle differences in RNA–RNA interactions during assortment and packaging also affect the constellation of genome segments in any resulting hybrid progeny. In fact, it has long been observed that some segments are preferentially packaged together such that the genotypes of influenza A virus reassortants are not random. For example, the segment-specific RNA–RNA interactions that occur during assortment have been shown to differ from strain to strain, suggesting that only reassortants that co-package interacting segments would maintain the supramolecular network and produce viable progeny49. Furthermore, in the absence of segment mismatch, influenza A viruses reassort with high frequency, demonstrating that there are few extrinsic barriers to exchange92. The genetic limitations on the capacity to create reassortants during co-infection may be less stringent for the Cystoviridae family than for the Orthomyxoviridae family. In fact, isolates from the Cystoviridae family with a high level of sequence divergence are able to reassort with φ6 in the laboratory setting and in nature26,93. However, in vitro packaging assays suggest that there may be some direct restrictions to genetic exchange. For example, it was shown that Pseudomonas phage φ13 (hereafter referred to as φ13) can efficiently package the φ6 M (+)RNA segment, even though this segment carries a packaging signal very divergent from the φ13 packaging signal94. By contrast, φ6 was incapable of packaging the M (+)RNA segment of φ13 unless the φ6 packaging sequences were appended to the molecule94. This result for these members of the Cystoviridae family is similar to the reports for influenza A viruses and suggests that even if genetic exchange can occur, not all combinations of genome segments are tolerated, and restrictions to reassortment may be strain specific. Similar to the restriction on reassortment between influenza A viruses and influenza B viruses, strains of rotavirus A are incapable of reassorting with strains of other rotavirus species following experimental co-infection of cells or animals. However, there seem to be restrictions that prevent successful genetic exchange even in strains of rotavirus A, which are closely related, as the frequency of reassortants in a given population of progeny is usually much lower than the frequency predicted based on chance alone95. Similar observations have been made for other members of the Reoviridae family, including mammalian orthoreoviruses96,97 and bluetongue viruses98. However, for all members of the family Reoviridae, it remains to be tested whether reassortants are simply not generated during co-infections, or whether they are generated but do not emerge in the population because they are less fit than their parental strains (see below). An interesting aspect of the replication cycles of members of the Reoviridae and Cystoviridae families, and a factor that may influence the generation of hybrid progeny during co-infection, is that genome replication (that is, dsRNA synthesis) occurs following segment assortment and packaging. Thus, for a reassortant progeny to be generated, the viral polymerase of one strain must be capable of replicating the packaged (+)RNAs of a different parental strain. For rotaviruses, the polymerase recognizes the (+)RNA template by a sequence-specific interaction with seven nucleotides that are located at the 3′ end of the molecule99. Rotavirus A, rotavirus C, rotavirus D and rotavirus F strains have a similar seven-nucleotide sequence (UGUGACC or UGUGGCU), which differs substantially from that of rotavirus B, rotavirus G and rotavirus H strains (AAAACCC, AAGACCC or UAUACCC)100. Therefore, the polymerases of rotavirus A, C, D and F strains would not be able to efficiently bind to and replicate the (+)RNA templates of rotavirus B, G and H strains, and vice versa (FIG. 3b). Similar strain-determined template specificities were found for the polymerases of Cystoviridae members φ6, φ13 and Pseudomonas phage φ8 (REF. 101). In light of this, suboptimal protein–RNA interactions during assortment, packaging and replication are expected to influence the generation of reassortants for viruses in the Cystoviridae and Reoviridae families. Emergence of reassortants in nature The increased capacity for whole-genome sequencing has facilitated new approaches that have revealed the importance of reassortment in the emergence of viruses with novel phenotypes (FIG. 4), including those that are associated with outbreaks. Large-scale comparative genomics studies of Cystoviridae, Orthomyxoviridae and Reoviridae family members in various hosts have detected numerous reassortants in viral populations26,102–114. For example, reassortment can lead to the creation of more-fit variants that outcompete previously circulating strains, and such cases are extremely well documented for influenza A viruses. Several influenza A viruses endemic in swine or birds have been successfully transmitted to humans, and in many cases, reassortment has been instrumental in the major evolutionary transition that is required for this transmission to humans. This is illustrated by the 1957 ‘Asian’ and 1968 ‘Hong Kong’ pandemics, which were both associated with reassortant viruses comprising both human and avian virus genome segments. Similarly, the 2009 pandemic resulted from a reassortment event between highly divergent North American swine viruses and Eurasian swine viruses. In all three pandemics, the reassortment event resulted in novel human viruses that carried divergent genes encoding HA and neuraminidase (NA) derived from the animal viruses; these human viruses express HA and NA antigens that are not well recognized by human adaptive immune responses (FIG. 4a). Reassortment among co-circulating human strains of the same HA–NA subtype is also important for the evolution and emergence of seasonal strains of influenza A virus115, including those that are antigenically novel106, those with enhanced transmissibility105 and those that are resistant to antiviral drugs116. Although reassortment can provide fitness advantages to the progeny virus if that progeny acquires a beneficial allele, reassortment can alternatively confer fitness costs if it uncouples a set of alleles that operate best when kept together (FIG. 4b). For example, reassortment has the potential to unlink RNAs or their encoded proteins that interact functionally during the viral replication cycle. As a consequence, a reassortant might exhibit suboptimal RNA–RNA, protein–RNA and/or protein–protein interactions during its de novo replication cycle, thereby making it less able to propagate and spread (that is, less fit) than the non-reassortant parental strains. The observation that reassortment can lead to attenuated viruses with reduced replicative fitness in this way has fostered the development of vaccine strains for influenza A viruses and rotavirus A (BOX 1). Box 1 Reassortant viruses as live-attenuated vaccine strains The capacity of influenza viruses and rotaviruses to generate functional new variants through reassortment has been harnessed to produce highly effective vaccines that stimulate immune responses without causing disease. The vaccines contain reassortants generated in the laboratory that combine the immunogenic surface proteins from field strains within the genetic backbones of specific laboratory-adapted ‘master donor’ strains that exhibit desired properties, such as high titre growth or attenuation. For example, in the Unites States, the seasonal influenza immunization programme is anchored by two types of vaccines, an inactivated influenza vaccine (IIV) and a live-attenuated influenza vaccine (LAIV; called FluMist)128,129. Both vaccines are quadrivalent formulations that consist of two strains of influenza A virus (H3N2 and H1N1) and two strains of influenza B virus (Yamagata and Victoria lineages). During vaccine production, 6/2 reassortants are generated; these contain the 6 internal genome segments from the laboratory-adapted master donor strain and the 2 haemagglutinin (HA)-encoding and neuraminidase (NA)-encoding gene segments from the field isolates. These vaccines are modified bi-annually on the basis of genetic and antigenic analyses of the dominant circulating global strains. However, the extensive lead time that is required to produce and evaluate candidate vaccine strains occasionally results in mismatches between vaccine strains and field strains, which results in reduced vaccine effectiveness. In the future, the use of reverse genetics to directly engineer reassortant vaccine candidates may shorten this lead time and reduce mismatches. Moreover, the directed introduction of growth-restricting mutations into field isolates through reverse genetics may bypass the need to create reassortants and could enable the rapid production of new live-attenuated vaccines that more closely match circulating strains. For human strains of rotavirus A, two live-attenuated vaccines are widely used globally: the monovalent Rotarix vaccine (GlaxoSmithKline) and the pentavalent RotaTeq vaccine (Merck). The RotaTeq vaccine is composed of five bovine–human reassortant strains (10/1 or 9/2) that contain 9 or 10 internal bovine rotavirus genes (from strain WC3), the human virus VP7 coding genes with rotavirus G1, G2, G3 and G4 genotype specificities, and the human virus VP4 coding gene with a strain P[8] genotype specificity130. The attenuated phenotype that is conferred by the reassortant gene constellation of the RotaTeq vaccine strains enables them to induce intestinal mucosal immunity without causing disease. Interestingly, although rotaviruses and influenza A viruses are both segmented viruses that use reassortment to advance their evolution and diversity, the pace of antigenic change among circulating influenza viruses has necessitated frequent adjustments of the IIV and LAIV vaccine formulations, whereas the rotavirus reassortant vaccine has remained effective for nearly 10 years without change61,128. Importantly, in some cases, the failure to detect reassortants following experimental co-infection might reflect the poor fitness of hybrid progeny caused by mismatched alleles, rather than restrictions on the actual generation of the reassortant during co-infection. For example, for influenza A viruses, uncoupling of the three polymerase-coding genes (PA, PB1 and PB2) by reassortment can lead to the formation of viruses with a diminished capacity for RNA synthesis117–119. Essentially, the polymerase proteins of some non-cognate strains are not able to effectively interact to form a functional enzyme complex (proteins of human viruses do not interact with proteins of avian viruses, for instance). Similar observations have been made for rotavirus replicase complex proteins, whereby the subunits of rotavirus A and rotavirus C strains cannot functionally substitute for each other120–122. Furthermore, comparative genomics studies also support the notion that inter-segmental RNA or protein co-adaptation tempers reassortment among co-circulating strains. For example, a mutual information-based algorithm was used to define amino acid residues that co-varied in multi-sequence alignments of proteins from rotavirus A strains112. The data revealed a vast network of interconnected amino acids in various viral proteins, some of which are not known to physically interact with each other. Thus, reassortment may also be limited by the selective constraints that are placed on functionally co-adapted, albeit non-interacting, proteins. However, it is also important to mention that less-fit reassortants with mismatched-allele constellations can acquire corrective mutations that restore interaction interfaces between non-cognate (that is, not co-adapted) proteins (FIG. 4c). In fact, it has been shown that low-fitness influenza A virus reassortants can accumulate fitness-restoring mutations in functionally interacting proteins if the reassortants are serially passaged in the laboratory117,123–126. There is also increasing evidence to support the notion that reassortment events cause a temporary increase in the rate of amino acid changes for influenza A viruses as the viral proteins adapt to a new genetic environment127. To date, there have been no studies that address the role of co-adaptive changes influencing RNA–RNA interactions and, in turn, reassortment for segmented RNA viruses, but this remains an important area for future investigation. These observations for influenza A viruses and strains of rotavirus A regarding allele combinations are in contrast to those for φ6 and members of the Cystoviridae family in general. As mentioned above, even members of the Cystoviridae family with a high level of sequence divergence were found to undergo frequent reassortment in nature26. It is interesting to speculate that perhaps members of the Cystoviridae family are not subjected to the same purifying selection pressures that are imposed on influenza A viruses and strains of rotavirus A following reassortment, maybe simply because of the way the genes are organized within the segments. In particular, the genes that encode interacting proteins of φ6 are usually located on the same segment (for example, all procapsid proteins are encoded by the L segment); therefore, reassortment would not uncouple functionally interacting alleles. Thus, φ6 represents an ideal experimental system for further investigation of the genetic linkages among genome segments for members of the Cystoviridae family and the effect of those linkages on the frequency of reassortment. Outlook In summary, segmented RNA viruses include some of the most important human, animal and plant pathogens in recent history. The biological mechanism that originally produced these viruses from non-segmented precursors remains unknown (BOX 2), but one of the most apparent consequences of this genome structure is the generation of novel reassortant progeny during co-infection. Reverse genetics and other experimental advances have increased our ability to investigate the molecular constraints on reassortment under controlled experimental conditions. Moreover, recent advances in genome-sequencing technologies have furthered our understanding of segmented RNA virus diversity and have shed light on the frequency of reassortants in natural populations. Importantly, for influenza A viruses and rotaviruses, these discoveries have shown how reassortment contributes to viral pathogenic and zoonotic potentials, and have enabled the generation of live-attenuated reassortant vaccines. However, some key outstanding questions remain unanswered and should be the focus of future research endeavours. How do influenza viruses and rotaviruses selectively package their genome segments? What are the relative contributions of failed RNA–RNA, protein–RNA and protein–protein interactions to reassortment restriction between any given strains? Are there virus-extrinsic barriers to reassortment within the infected host or within the environment? What are the biological and temporal dynamics that are required to achieve robust reassortment? What is the contribution of semi-infectious particles or defective-interfering particles to viral replication, reassortment and evolution? Do some individual genome segments evolve biased packaging so that they are over-represented in the reassortant progeny (that is, are there ‘selfish genes’)? The answers to these questions are expected not only to inform disease prevention and control strategies, but also to shed light on our basic understanding of organismal evolution. Box 2 Possible origins of RNA virus genome segmentation It is unknown how an ancestral non-segmented RNA virus underwent genome segmentation in the first place, but different theories have been proposed to explain the origin of segmented RNA viruses. One theory posits that genome segmentation may have arisen following the accidental merging of RNA genomes from two different viruses (see the figure, part a). This theory is supported by a recent analysis of Jingmen tick virus (a taxonomically unclassified segmented RNA virus), as two of the four positive-sense RNA ((+)RNA) genome segments are genetically related to those of flaviviruses, whereas the other two are completely unique, which suggests that they were acquired independently from a still-unidentified parental ancestor131. It is possible that the acquisition of a novel RNA virus genome provided the Jingmen tick virus ancestor an evolutionary advantage over parental strains that lacked such extra genome segments. Alternatively, genome segmentation could have arisen as a downstream consequence of diploidy or polyploidy, whereby the precursor non-segmented RNA virus may have randomly packaged more than one copy of its genome into a nascent virion (see the figure, part b). Diploidy and polyploidy are argued to be evolutionarily advantageous in complex organisms because they buffer against the effects of deleterious mutations. Accumulation of mutations over time may have enabled the ‘duplicate’ genome to encode new proteins, evolving into a new genome segment132. Indeed, diploidy and polyploidy have been documented for measles viruses and Ebola viruses, which normally have single-stranded negative-sense RNA ((−)RNA) genomes133–135. Moreover, diploidy is a hallmark of the RNA genome structure of several retroviruses, including HIV. As diploidy and polyploidy require that there are few restrictions on the amount of nucleic acid that can be packaged into a virus particle, such genomes may have evolved more easily for enveloped viruses than for non-enveloped viruses with stringent capsid sizes. S.M.M. receives financial support from the Virginia Tech Carilion School of Medicine and Research Institute and the US National Institutes of Health (NIH; grants R01AI116815, R21AI113402 and R21AI119588). P.E.T. receives financial support from the US National Science Foundation BEACON Center for Study of Evolution in Action and the NIH (grant R01AI09164601). J.T.P. is supported by funding from the University of Maryland, College Park, USA. Segmented RNA viruses Viruses in which the genome consists of more than one RNA molecule (that is, segments). The genome segments can be packaged within a single virion particle or into separate particles Type species A representative viral strain that is studied to understand the biology of an entire viral genus or family Reassortment A process of genetic exchange whereby two or more parental viruses co-infect a single host cell and exchange genome segments. The outcome is the formation of hybrid viral progeny with genome segments derived from multiple parental strains Assortment The mechanism by which a segmented virus packages one of each genome segment into a virion particle Viral fitness The capacity of an individual virus to generate infectious progeny, relative to other virus genotypes in the population Pathovars Bacterial strains with the same or similar characteristics In vitro packaging system A simplified experimental system in which viral genome segments are incorporated into a virion particle; this occurs in a test tube and outside the context of an infected host cell Defective-interfering RNAs Spontaneously generated mutant RNA molecules that usually contain large gene deletions but maintain sequences that are crucial for their replication and packaging. These RNAs reduce the fitness of full-length viruses during cellular co-infection HA–NA subtype A binomial system of classification for influenza A viruses that is based on the neutralizing antibody response to the virion structural proteins haemagglutinin (HA) and neuraminidase (NA) Diploidy or polyploidy, In virology when an individual virus encapsidates two (diploidy) or more (polyploidy) copies of the genome into a single virus particle Figure 1 Reassortment, sexual reproduction and recombination a | Reassortment in non-multipartite RNA viruses. Two virus particles are shown, each with a full complement of three viral genome segments. Following reassortment, hybrid progeny can be formed that contain segments derived from both parents. b | Sexual reproduction. Two parent gamete cells are shown, each with a haploid genome of three chromosomal segments. Following sex between the two parents, a hybrid diploid progeny is produced that contains one copy of each chromosome from each parent. c | Recombination in non-segmented, single-stranded RNA viruses. Following recombination between two virus particles, chimeric genomes are produced that have regions derived from each parent. Figure 2 Pseudomonas phage φ6, influenza A virus and rotavirus genome organization and assortment a | The Pseudomonas phage φ6 genome consists of three double-stranded RNA (dsRNA) segments: small (S), medium (M) and large (L). Blue indicates ORFs, and grey represents intergenic regions; lines at the 5′ and 3′ termini represent UTRs. Sequences that are known to be important for the selective packaging of φ6 single-stranded positive-sense RNA ((+)RNA) replication intermediates are shown in red. b | A model of φ6 genome segment assortment and packaging. φ6 (+)RNAs are packaged sequentially. Initially, the procapsid has a binding site only for the S (+)RNA segment, enabling it to be inserted. Following the packaging of the S segment, a binding site for the M (+)RNA segment is revealed in the procapsid, enabling that segment to be inserted. Finally, a binding site for the L (+)RNA segment is revealed, the segment is inserted, and the entire complement of φ6 (+)RNAs are encapsidated. Following packaging of all three (+)RNA segments, the procapsid core expands, which triggers the conversion of the three (+)RNAs into double-stranded RNA (dsRNA) genome segments by viral polymerases. c | The influenza A virus genome comprises eight negative-sense RNA ((−)RNA) segments. A representative segment is shown as a linear (−)RNA molecule (top) and as a ribonucleoprotein (RNP; bottom), in which the (−)RNA is bound by a heterotrimeric polymerase complex and nucleocapsid protein (NP). The ORF, UTRs and sequences that are important for selective genome packaging are coloured as in part a. d | A model of genome segment assortment and packaging in influenza A viruses. Eight influenza A virus RNPs are synthesized in the nucleus and individually exported into the cytosol, where they pair up with each other. While en route to the plasma membrane, the eight RNPs form a supramolecular complex that is encapsidated by a lipid envelope during budding to form the virion. e | The genome of rotavirus A is composed of 11 dsRNA segments, one of which is shown as a (+)RNA precursor in linear form (top) and folded into a putative panhandle shape (bottom). The ORF, UTRs and sequences that are important for selective packaging are coloured as in part a and part b. A polymerase–capping enzyme complex is thought to be bound to the 3′-terminal UGUGACC sequence. A putative stem–loop structure may act as an assortment and/or packaging signal. f | A model of genome segment assortment and packaging in rotaviruses. The 11 (+)RNAs, each with a bound polymerase–capping enzyme complex, are thought to pair up and eventually form a supramolecular complex that is encapsidated by a forming virion particle. During or immediately after encapsidation, the (+)RNAs are converted into dsRNA genome segments by their dedicated polymerase. The polymerases function while tethered to the viral capsid (not shown). Figure 3 Direct restrictions on the generation of reassortants a | Incompatibility of RNA–RNA interactions. Two influenza A virus genome segments are shown as ribonucleoproteins (RNPs), each derived from a parent strain (strain A is shown in red and strain B is shown in blue). If the packaging signals are compatible (left), the RNA molecules can interact, which leads to co-packaging and reassortment. However, if the packaging signals are not compatible, the RNA molecules will interact suboptimally, thereby preventing their co-packaging. b | Incompatibility of protein–RNA interactions. A rotavirus A positive-sense RNA ((+)RNA) molecule from one strain may be recognized only by the polymerase from that same strain. If the polymerase in the virion is from a different strain and is unable to recognize the (+)RNA molecule, replication does not occur, thus restricting the generation of reassortants. Figure 4 Fitness consequences of reassortment a | Increase in viral fitness. Following reassortment, hybrid progeny can be formed that contain segments derived from both parents. In some cases, the new allelic combination confers phenotypic changes to the reassortant. For example, reassortment can produce an antigenically novel variant that is not recognized by the host immune system. This more-fit reassortant emerges in the host, whereas the less-fit parental strains are eliminated. b | Decrease in viral fitness. In some cases, reassortment can uncouple essential cognate protein sets that interact optimally when kept together. If non-cognate proteins do not interact, the reassortant would be less fit than parental strains and would therefore be eliminated from the population. c | Post-reassortment adaptations. A less-fit reassortant can accumulate mutations that restore the interaction interface between the non-cognate proteins. Such post-reassortment adaptive changes will enable the reassortant to regain fitness and emerge. Table 1 Segmented RNA virus families: genome organization, type species and hosts Family Genome organization Packaging Type species of genera within family Hosts Arenaviridae 2 (−)RNA or ambisense* molecules Single virion Lymphocytic choriomeningitis mammarenavirus Animals Birnaviridae 2 dsRNA molecules Single virion Infectious bursal disease virus Animals Bromoviridae 3 (+)RNA molecules Multipartite Brome mosaic virus Plants Bunyaviridae 3 (−)RNA or ambisense molecules Single virion Rift Valley fever virus, Tomato spotted wilt virus Animals and plants Chrysoviridae 4 dsRNA molecules Multipartite Penicillium chrysogenum virus Fungi Closteroviridae 2 (+)RNA molecules Multipartite Lettuce infectious yellows virus Plants Cystoviridae ‡ 3 dsRNA molecules Single virion Pseudomonas phage φ6, Pseudomonas phage φ10, Pseudomonas phage φ13 Bacteria Orthomyxoviridae ‡ 6–8 (−)RNA molecules Single virion Influenza A virus, Influenza B virus Animals Partitiviridae 2 dsRNA molecules Multipartite White clover cryptic virus 1 Plants, fungi and protozoa Picobirnaviridae 2 dsRNA molecules Multipartite Human picobirnavirus Animals Reoviridae ‡ 8–12 dsRNA molecules Single virion Rotavirus A, Bluetongue virus Animals and plants dsRNA, double-stranded RNA; (+)RNA: positive-sense RNA; (−)RNA, negative-sense RNA. * Ambisense refers to an RNA molecule that is positive-sense in some regions and negative-sense in other regions. ‡ This Review focuses on these three families. Competing interests statement The authors declare no competing interests. 1 Tate JE 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis Lancet Infect Dis 12 136 141 2012 22030330 2 Klepser ME Socioeconomic impact of seasonal (epidemic) influenza and the role of over-the-counter medicines Drugs 74 1467 1479 2014 25150045 3 GBD 2013 Mortality and Causes of Death Collaborators Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013 Lancet 385 117 171 2015 25530442 4 Maclachlan NJ Mayo CE Potential strategies for control of bluetongue, a globally emerging, Culicoides-transmitted viral disease of ruminant livestock and wildlife Antiviral Res 99 79 90 2013 23664958 5 Mahgoub HA Bailey M Kaiser P An overview of infectious bursal disease Arch Virol 157 2047 2057 2012 22707044 6 Hanssen IM Lapidot M Thomma BP Emerging viral diseases of tomato crops Mol Plant Microbe Interact 23 539 548 2010 20367462 7 Bernstein H Genetic damage, mutation, and the evolution of sex Science 229 1277 1281 1985 3898363 8 Bernstein ME Does variation in the testosterone level of the seminal plasma affect the primary sex ratio? J Theor Biol 126 377 378 1987 3657237 9 Felsenstein J The evolutionary advantage of recombination Genetics 78 737 756 1974 4448362 10 Chao L Evolution of sex in RNA viruses J Theor Biol 133 99 112 1988 This seminal article describes how segment reassortment in RNA viruses is conceptually similar to sexual reproduction in eukaryotes and theorizes about the role of such genetic exchange mechanisms during viral evolution 3226144 11 Chao L Levels of selection, evolution of sex in RNA viruses, and the origin of life J Theor Biol 153 229 246 1991 1787738 12 Chao L Evolution of sex in RNA viruses Trends Ecol Evol 7 147 151 1992 21235989 13 Turner PE Searching for the advantages of virus sex Orig Life Evol Biosph 33 95 108 2003 12967275 14 Nee S On the evolution of sex in RNA viruses J Theor Biol 138 407 412 1989 2593681 15 Takeda M Generation of measles virus with a segmented RNA genome J Virol 80 4242 4248 2006 16611883 16 Ojosnegros S Viral genome segmentation can result from a trade-off between genetic content and particle stability PLoS Genet 7 e1001344 2011 21437265 17 Lai MM Genetic recombination in RNA viruses Curr Top Microbiol Immunol 176 21 32 1992 1600753 18 Boni MF No evidence for intra-segment recombination of 2009 H1N1 influenza virus in swine Gene 494 242 245 2012 22226809 19 Boni MF Homologous recombination is very rare or absent in human influenza A virus J Virol 82 4807 4811 2008 18353939 20 Mindich L Packaging, replication and recombination of the segmented genome of bacteriophage Φ6 and its relatives Virus Res 101 83 92 2004 15010219 21 Onodera S Sun Y Mindich L Reverse genetics and recombination in Φ8, a dsRNA bacteriophage Virology 286 113 118 2001 11448164 22 Woods RJ Intrasegmental recombination does not contribute to the long-term evolution of group A rotavirus Infect Genet Evol 32 354 360 2015 25847696 23 Simon-Loriere E Holmes EC Why do RNA viruses recombine? Nat Rev Microbiol 9 617 626 2011 This review article compares the mechanism of recombination in non-segmented RNA viruses to that of reassortment in segmented RNA viruses. The authors also describe theories regarding the evolutionary significance of genetic exchange among viruses 21725337 24 Vidaver AK Koski RK Van Etten JL Bacteriophage Φ6: a lipid-containing virus of Pseudomonas phaseolicola J Virol 11 799 805 1973 16789137 25 Mindich L Precise packaging of the three genomic segments of the double-stranded-RNA bacteriophage Φ6 Microbiol Mol Biol Rev 63 149 160 1999 10066834 26 Silander OK Widespread genetic exchange among terrestrial bacteriophages Proc Natl Acad Sci USA 102 19009 19014 2005 This first comparative genomics study of members of the Cystoviridae family in nature reveals that reassortment occurs frequently and between highly divergent viral strains. The results of this study suggest that few genetic restrictions to reassortment exist in this family 16365305 27 O’Keefe KJ Geographic differences in sexual reassortment in RNA phage Evolution 64 3010 3023 2010 20500219 28 Jäälinoja HT Huiskonen JT Butcher SJ Electron cryomicroscopy comparison of the architectures of the enveloped bacteriophages Φ6 and Φ8 Structure 15 157 167 2007 17292834 29 Emori Y Iba H Okada Y Transcriptional regulation of three double-stranded RNA segments of bacteriophage Φ6 in vitro J Virol 46 196 203 1983 6827650 30 Gottlieb P In vitro packaging of the bacteriophage Φ6 ssRNA genomic precursors Virology 181 589 594 1991 2014638 31 Gottlieb P In vitro replication, packaging, and transcription of the segmented double-stranded RNA genome of bacteriophage Φ6: studies with procapsids assembled from plasmid-encoded proteins J Bacteriol 172 5774 5782 1990 2211512 32 Huiskonen JT Structure of the bacteriophage Φ6 nucleocapsid suggests a mechanism for sequential RNA packaging Structure 14 1039 1048 2006 16765897 33 Gottlieb P Qiao X Strassman J Frilander M Mindich L Identification of the packaging regions within the genomic RNA segments of bacteriophage Φ6 Virology 200 42 47 1994 This study uses an in vitro system to map the location of packaging signals within the φ6 RNA segments 8128636 34 Onodera S Construction of a transducing virus from double-stranded RNA bacteriophage Φ6: establishment of carrier states in host cells J Virol 66 190 196 1992 1727482 35 Onodera S Isolation of a mutant that changes genomic packaging specificity in Φ6 Virology 252 438 442 1998 9878623 36 Onodera S Directed changes in the number of double-stranded RNA genomic segments in bacteriophage Φ6 Proc Natl Acad Sci USA 95 3920 3924 1998 This report describes how the three segments of the φ6 genome can be engineered as a single concatenated genome segment, providing insights into the replication advantages of genome segmentation for Cystoviridae family members 9520468 37 Palase P Shaw M Fields Virology Howley PM Knipe DM 1647 1690 Lippincott Williams & Wilkins 2007 38 Dushoff J Mortality due to influenza in the United States — an annualized regression approach using multiple-cause mortality data Am J Epidemiol 163 181 187 2006 16319291 39 Johnson NP Mueller J Updating the accounts: global mortality of the 1918–1920 “Spanish” influenza pandemic Bull Hist Med 76 105 115 2002 11875246 40 Parrish CR Murcia PR Holmes EC Influenza virus reservoirs and intermediate hosts: dogs, horses, and new possibilities for influenza virus exposure of humans J Virol 89 2990 2994 2015 25540375 41 Gultyaev AP Fouchier RA Olsthoorn RC Influenza virus RNA structure: unique and common features Int Rev Immunol 29 533 556 2010 20923332 42 Zheng W Tao YJ Structure and assembly of the influenza A virus ribonucleoprotein complex FEBS Lett 587 1206 1214 2013 23499938 43 Noda T Kawaoka Y Structure of influenza virus ribonucleoprotein complexes and their packaging into virions Rev Med Virol 20 380 391 2010 20853340 44 Gerber M Selective packaging of the influenza A genome and consequences for genetic reassortment Trends Microbiol 22 446 455 2014 This review article summarizes the genetic, biochemical and structural data supporting the current model of genome segment assortment and packaging in influenza A viruses 24798745 45 Hutchinson EC Fodor E Transport of the influenza virus genome from nucleus to nucleus Viruses 5 2424 2446 2013 24104053 46 Chou YY One influenza virus particle packages eight unique viral RNAs as shown by FISH analysis Proc Natl Acad Sci USA 109 9101 9106 2012 22547828 47 Chou YY Colocalization of different influenza viral RNA segments in the cytoplasm before viral budding as shown by single-molecule sensitivity FISH analysis PLoS Pathog 9 e1003358 2013 23671419 48 Lakdawala SS Influenza A virus assembly intermediates fuse in the cytoplasm PLoS Pathog 10 e1003971 2014 This investigation uses multicolour single-molecule fluorescent in situ hybridization to show that influenza A virus RNAs assort in the cytoplasm while en route to the plasma membrane 24603687 49 Gavazzi C An in vitro network of intermolecular interactions between viral RNA segments of an avian H5N2 influenza A virus: comparison with a human H3N2 virus Nucleic Acids Res 41 1241 1254 2013 23221636 50 Gavazzi C A functional sequence-specific interaction between influenza A virus genomic RNA segments Proc Natl Acad Sci USA 110 16604 16609 2013 This paper describes in vitro experiments that identify direct intermolecular interactions between influenza A virus genomic RNA segments, shedding light on the mechanism of assortment 24067651 51 Fournier E Interaction network linking the human H3N2 influenza A virus genomic RNA segments Vaccine 30 7359 7367 2012 23063835 52 Fournier E A supramolecular assembly formed by influenza A virus genomic RNA segments Nucleic Acids Res 40 2197 2209 2012 22075989 53 Hutchinson EC Genome packaging in influenza A virus J Gen Virol 91 313 328 2010 19955561 54 Sugita Y Configuration of viral ribonucleoprotein complexes within the influenza A virion J Virol 87 12879 12884 2013 This study uses electron microscopy to analyse the fine structure of purified RNPs and their configuration within influenza A virions, thereby revealing novel aspects of assortment and packaging 24067952 55 Enami M An influenza virus containing nine different RNA segments Virology 185 291 298 1991 1833874 56 Gao Q A nine-segment influenza a virus carrying subtype H1 and H3 hemagglutinins J Virol 84 8062 8071 2010 20519387 57 Brooke CB Most influenza A virions fail to express at least one essential viral protein J Virol 87 3155 3162 2013 This article reports experiments showing that the vast majority of influenza A virus particles do not express at least one viral protein. This implies that the genome segment packaging efficiency for influenza A viruses may be lower than previously predicted 23283949 58 Brooke CB Influenza A virus nucleoprotein selectively decreases neuraminidase gene-segment packaging while enhancing viral fitness and transmissibility Proc Natl Acad Sci USA 111 16854 16859 2014 25385602 59 Fonville JM Influenza virus reassortment is enhanced by semi-infectious particles but can be suppressed by defective interfering particles PLoS Pathog 11 e1005204 2015 This work shows that influenza A virus particles that lack a full complement of functional genome segments (that is, semi-infectious particles) can reassort with infectious particles 26440404 60 Otuka A Migration of rice planthoppers and their vectored re-emerging and novel rice viruses in East Asia Front Microbiol 4 309 2013 24312081 61 Tate JE Parashar UD Rotavirus vaccines in routine use Clin Infect Dis 59 1291 1301 2014 25048849 62 Marthaler D Rapid detection and high occurrence of porcine rotavirus A, B, and C by RT-qPCR in diagnostic samples J Virol Methods 209 30 34 2014 25194889 63 Moutelikova R Prodelalova J Dufkova L Prevalence study and phylogenetic analysis of group C porcine rotavirus in the Czech Republic revealed a high level of VP6 gene heterogeneity within porcine cluster I1 Arch Virol 159 1163 1167 2014 24212886 64 Amimo JO Vlasova AN Saif LJ Prevalence and genetic heterogeneity of porcine group C rotaviruses in nursing and weaned piglets in Ohio, USA and identification of a potential new VP4 genotype Vet Microbiol 164 27 38 2013 23428382 65 Park SI Genetically diverse group C rotaviruses cause sporadic infection in Korean calves J Vet Med Sci 73 479 482 2011 21099189 66 Lobo PD Phylogenetic analysis of human group C rotavirus in hospitalized children with gastroenteritis in Belem, Brazil J Med Virol 88 728 733 2016 26369400 67 Luchs A do Carmo Sampaio Tavares Timenetsky M Phylogenetic analysis of human group C rotavirus circulating in Brazil reveals a potential unique NSP4 genetic variant and high similarity with Asian strains Mol Genet Genom 290 969 986 2015 68 El-Senousy WM Ragab AM Handak EM Prevalence of rotaviruses groups A and C in Egyptian children and aquatic environment Food Environ Virol 7 132 141 2016 69 Lahon A Walimbe AM Chitambar SD Full genome analysis of group B rotaviruses from western India: genetic relatedness and evolution J Gen Virol 93 2252 2266 2012 22815276 70 Trask SD McDonald SM Patton JT Structural insights into the coupling of virion assembly and rotavirus replication Nat Rev Microbiol 10 165 177 2012 22266782 71 Rainsford EW McCrae MA Characterization of the NSP6 protein product of rotavirus gene 11 Virus Res 130 193 201 2007 17658646 72 McDonald SM Patton JT Assortment and packaging of the segmented rotavirus genome Trends Microbiol 19 136 144 2011 21195621 73 Lawton JA Estes MK Prasad BV Mechanism of genome transcription in segmented dsRNA viruses Adv Virus Res 55 185 229 2000 11050943 74 Gallegos CO Patton JT Characterization of rotavirus replication intermediates: a model for the assembly of single-shelled particles Virology 172 616 627 1989 This seminal investigation analyses the protein composition and activity of rotavirus replication–assembly intermediates and suggests a model for genome segment assortment 2552662 75 Patton JT Gallegos CO Structure and protein composition of the rotavirus replicase particle Virology 166 358 365 1988 2845649 76 Patton JT Gallegos CO Rotavirus RNA replication: single-stranded RNA extends from the replicase particle J Gen Virol 71 1087 1094 1990 2161046 77 Suzuki Y A candidate packaging signal of human rotavirus differentiating Wa-like and DS-1-like genomic constellations Microbiol Immunol 59 567 571 2015 26224654 78 Li W Genomic analysis of codon, sequence and structural conservation with selective biochemical-structure mapping reveals highly conserved and dynamic structures in rotavirus RNAs with potential cis-acting functions Nucleic Acids Res 38 7718 7735 2010 20671030 79 Burkhardt C Structural constraints in the packaging of bluetongue virus genomic segments J Gen Virol 95 2240 2250 2014 24980574 80 Sung PY Roy P Sequential packaging of RNA genomic segments during the assembly of bluetongue virus Nucleic Acids Res 42 13824 13838 2014 25428366 81 Roner MR Steele BG Localizing the reovirus packaging signals using an engineered m1 and s2 ssRNA Virology 358 89 97 2007 This paper describes reverse genetics experiments that enabled the localization of the packaging signals for two viral genome segments of Reoviridae family members 16987539 82 Roner MR Bassett K Roehr J Identification of the 5′ sequences required for incorporation of an engineered ssRNA into the reovirus genome Virology 329 348 360 2004 15518814 83 Roner MR Joklik WK Reovirus reverse genetics: incorporation of the CAT gene into the reovirus genome Proc Natl Acad Sci USA 98 8036 8041 2001 11427706 84 Zhang X In situ structures of the segmented genome and RNA polymerase complex inside a dsRNA virus Nature 527 531 534 2015 26503045 85 Desselberger U Genome rearrangements of rotaviruses Arch Virol Suppl 12 37 51 1996 9015100 86 Troupin C Rearranged genomic RNA segments offer a new approach to the reverse genetics of rotaviruses J Virol 84 6711 6719 2010 20427539 87 Navarro A Trask SD Patton JT Generation of genetically stable recombinant rotaviruses containing novel genome rearrangements and heterologous sequences by reverse genetics J Virol 87 6211 6220 2013 This study uses reverse genetics to show that additional RNA, including heterologous gene sequences, can be packaged into rotavirus virions 23536662 88 Tian Y Genomic concatemerization/deletion in rotaviruses: a new mechanism for generating rapid genetic change of potential epidemiological importance J Virol 67 6625 6632 1993 8411365 89 Taniguchi K Kojima K Urasawa S Nondefective rotavirus mutants with an NSP1 gene which has a deletion of 500 nucleotides, including a cysteine-rich zinc finger motif-encoding region (nucleotides 156 to 248), or which has a nonsense codon at nucleotides 153 to 155 J Virol 70 4125 4130 1996 This report describes a naturally occurring rotavirus mutant that does not express a functional NSP1 protein but that efficiently incorporates the NSP1-coding gene. The implications of this study are that the NSP1-coding RNA molecule itself is essential for assortment and packaging of the other ten genome segments 8648754 90 Baker SF Influenza A and B virus intertypic reassortment through compatible viral packaging signals J Virol 88 10778 10791 2014 This article presents experiments showing that an influenza A virus can package an influenza B virus genome segment, provided that the segment contains cognate influenza A virus packaging signals 25008914 91 Essere B Critical role of segment-specific packaging signals in genetic reassortment of influenza A viruses Proc Natl Acad Sci USA 110 E3840 E3848 2013 This investigation shows the importance of compatible packaging signals for efficient reassortment to occur among divergent influenza A virus strains 24043788 92 Marshall N Influenza virus reassortment occurs with high frequency in the absence of segment mismatch PLoS Pathog 9 e1003421 2013 This paper describes experiments that quantify, for the first time, the efficiency of reassortment between two nearly genetically identical strains of influenza A virus, indicating that reassortment is restricted mainly by genetic incompatibility of strains 23785286 93 Diaz-Munoz SL Electrophoretic mobility confirms reassortment bias among geographic isolates of segmented RNA phages BMC Evol Biol 13 206 2013 24059872 94 Qiao X Qiao J Onodera S Mindich L Characterization of Φ13, a bacteriophage related to Φ6 and containing three dsRNA genomic segments Virology 275 218 224 2000 11017801 95 Ramig RF Genetics of the rotaviruses Annu Rev Microbiol 51 225 255 1997 9343350 96 Wenske EA Genetic reassortment of mammalian reoviruses in mice J Virol 56 613 616 1985 4057359 97 Nibert ML Margraf RL Coombs KM Nonrandom segregation of parental alleles in reovirus reassortants J Virol 70 7295 7300 1996 This work documents the frequency of reassortment between two strains of the Reoviridae family following experimental co-infection, and provides evidence for genetic linkages among segments 8794386 98 Ramig RF Analysis of reassortment and superinfection during mixed infection of Vero cells with bluetongue virus serotypes 10 and 17 J Gen Virol 70 2595 2603 1989 2552005 99 Lu X Mechanism for coordinated RNA packaging and genome replication by rotavirus polymerase VP1 Structure 16 1678 1688 2008 This article reports the high-resolution X-ray crystal structure of the rotavirus polymerase in the presence and absence of RNA template, informing an understanding of genome segment packaging and replication 19000820 100 Ogden KM Johne R Patton JT Rotavirus RNA polymerases resolve into two phylogenetically distinct classes that differ in their mechanism of template recognition Virology 431 50 57 2012 22687427 101 Yang H Makeyev EV Bamford DH Comparison of polymerase subunits from double-stranded RNA bacteriophages J Virol 75 11088 11095 2001 11602748 102 Nelson MI Evolution of novel reassortant A/H3N2 influenza viruses in North American swine and humans, 2009–2011 J Virol 86 8872 8878 2012 22696653 103 Dugan VG The evolutionary genetics and emergence of avian influenza viruses in wild birds PLoS Pathog 4 e1000076 2008 18516303 104 Lindstrom SE Cox NJ Klimov A Genetic analysis of human H2N2 and early H3N2 influenza viruses, 1957–1972: evidence for genetic divergence and multiple reassortment events Virology 328 101 119 2004 15380362 105 Nelson MI Multiple reassortment events in the evolutionary history of H1N1 influenza A virus since 1918 PLoS Pathog 4 e1000012 2008 18463694 106 Rambaut A The genomic and epidemiological dynamics of human influenza A virus Nature 453 615 619 2008 18418375 107 Holmes EC Whole-genome analysis of human influenza A virus reveals multiple persistent lineages and reassortment among recent H3N2 viruses PLoS Biol 3 e300 2005 This first large-scale comparative genomics study of influenza A viruses shows that multiple co-circulating lineages exist and documents reassortment events 16026181 108 Ghedin E Large-scale sequencing of human influenza reveals the dynamic nature of viral genome evolution Nature 437 1162 1166 2005 16208317 109 Westgeest KB Genome-wide analysis of reassortment and evolution of human influenza A (H3N2) viruses circulating between 1968 and 2011 J Virol 88 2844 2857 2014 24371052 110 McDonald SM Evolutionary dynamics of human rotaviruses: balancing reassortment with preferred genome constellations PLoS Pathog 5 1000634 2009 This first large-scale comparative genomics study of human rotaviruses shows that multiple co-circulating lineages exist and documents reassortment events 111 McDonald SM Diversity and relationships of cocirculating modern human rotaviruses revealed using large-scale comparative genomics J Virol 86 9148 9162 2012 22696651 112 Zhang S Analysis of human rotaviruses from a single location over an 18-year time span suggests that protein coadaption influences gene constellations J Virol 88 9842 9863 2014 This comparative genomics study of human rotavirus identifies persistent lineages and suggests that co-adaptation of functionally interacting viral proteins may restrict reassortment over time 24942570 113 Dennis AF Molecular epidemiology of contemporary G2P[4] human rotaviruses cocirculating in a single U.S. community: footprints of a globally transitioning genotype J Virol 88 3789 3801 2014 24429371 114 Nomikou K Widespread reassortment shapes the evolution and epidemiology of bluetongue virus following European invasion PLoS Pathog 11 e1005056 2015 26252219 115 Nelson MI Phylogenetic analysis reveals the global migration of seasonal influenza A viruses PLoS Pathog 3 1220 1228 2007 17941707 116 Simonsen L The genesis and spread of reassortment human influenza A/H3N2 viruses conferring adamantane resistance Mol Biol Evol 24 1811 1820 2007 17522084 117 Li C Compatibility among polymerase subunit proteins is a restricting factor in reassortment between equine H7N7 and human H3N2 influenza viruses J Virol 82 11880 11888 2008 This is the first investigation to reveal how incompatibilities among polymerase subunits can restrict reassortment among divergent influenza A viruses 18815312 118 Hara K Co-incorporation of the PB2 and PA polymerase subunits from human H3N2 influenza virus is a critical determinant of the replication of reassortant ribonucleoprotein complexes J Gen Virol 94 2406 2416 2013 23939981 119 Octaviani CP Goto H Kawaoka Y Reassortment between seasonal H1N1 and pandemic (H1N1) 2009 influenza viruses is restricted by limited compatibility among polymerase subunits J Virol 85 8449 8452 2011 21680507 120 McDonald SM Shared and group-specific features of the rotavirus RNA polymerase reveal potential determinants of gene reassortment restriction J Virol 83 6135 6148 2009 This is the first report to demonstrate how incompatibilities between the rotavirus polymerase and core shell protein may restrict reassortment among divergent strains 19357162 121 McDonald SM Patton JT Rotavirus VP2 core shell regions critical for viral polymerase activation J Virol 85 3095 3105 2011 21248043 122 Taraporewala ZF Structure-function analysis of rotavirus NSP2 octamer by using a novel complementation system J Virol 80 7984 7994 2006 16873255 123 Ilyushina NA Postreassortment changes in a model system: HA–NA adjustment in an H3N2 avian–human reassortant influenza virus Arch Virol 150 1327 1338 2005 15789269 124 Kaverin NV Postreassortment changes in influenza A virus hemagglutinin restoring HA–NA functional match Virology 244 315 321 1998 This article reports that compensatory mutations can correct mismatched protein interactions that were due to reassortment for influenza A viruses in cell culture experiments 9601502 125 Kaverin NV Intergenic HA–NA interactions in influenza A virus: postreassortment substitutions of charged amino acid in the hemagglutinin of different subtypes Virus Res 66 123 129 2000 10725545 126 Rudneva IA Effect of gene constellation and postreassortment amino acid change on the phenotypic features of H5 influenza virus reassortants Arch Virol 152 1139 1145 2007 17294090 127 Neverov AD Intrasubtype reassortments cause adaptive amino acid replacements in H3N2 influenza genes PLoS Genet 10 e1004037 2014 This study shows that influenza A virus reassortment in nature causes a temporary increase in the rate of amino acid changes as the viral proteins adapt to a new genetic environment 24415946 128 Krammer F Palese P Steel J Advances in universal influenza virus vaccine design and antibody mediated therapies based on conserved regions of the hemagglutinin Curr Top Microbiol Immunol 386 301 321 2015 25007847 129 Jin H Subbarao K Live attenuated influenza vaccine Curr Top Microbiol Immunol 386 181 204 2015 25059893 130 Chandran A Santosham M RotaTeq: a three-dose oral pentavalent reassortant rotavirus vaccine Expert Rev Vaccines 7 1475 1480 2008 19053204 131 Qin XC A tick-borne segmented RNA virus contains genome segments derived from unsegmented viral ancestors Proc Natl Acad Sci USA 11 6744 6749 2014 132 Andersson DI Jerlstrom-Hultqvist J Nasvall J Evolution of new functions de novo and from preexisting genes Cold Spring Harb Perspect Biol 7 a017996 2015 26032716 133 Shirogane Y Watanabe S Yanagi Y Cooperation between different RNA virus genomes produces a new phenotype Nat Commun 3 1235 2012 23212364 134 Rager M Polyploid measles virus with hexameric genome length EMBO J 21 2364 2372 2002 12006489 135 Beniac DR The organisation of Ebola virus reveals a capacity for extensive, modular polyploidy PLoS ONE 7 e29608 2012 22247782

PMC005xxxxxx/PMC5119465.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101282906 33137 ACS Chem Biol ACS Chem. Biol. ACS chemical biology 1554-8929 1554-8937 27310134 5119465 10.1021/acschembio.6b00398 NIHMS824491 Article New Aspercryptins, Lipopeptide Natural Products, Revealed by HDAC Inhibition in Aspergillus nidulans Henke Matthew T. † Soukup Alexandra A. || Goering Anthony W. † McClure Ryan A. ‡ Thomson Regan J. ‡ Keller Nancy P. ||⊥# Kelleher Neil L. *†‡§ † Department of Molecular Biosciences, Northwestern University, Evanston, Illinois 60208, United States ‡ Department of Chemistry, Northwestern University, Evanston, Illinois 60208, United States § Feinberg School of Medicine, Northwestern University, Evanston, Illinois 60208, United States || Department of Genetics, University of Wisconsin, Madison, Wisconsin 53706, United States ⊥ Department of Bacteriology, University of Wisconsin, Madison, Wisconsin 53706, United States # Department of Medical Microbiology and Immunology, University of Wisconsin, Madison, Wisconsin 53706, United States * Corresponding Author: n-kelleher@northwestern.edu 22 10 2016 24 6 2016 19 8 2016 19 8 2017 11 8 21172123 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Unlocking the biochemical stores of fungi is key for developing future pharmaceuticals. Through reduced expression of a critical histone deacetylase in Aspergillus nidulans, increases of up to 100-fold were observed in the levels of 15 new aspercryptins, recently described lipopeptides with two non-canonical amino acids derived from octanoic and dodecanoic acids. In addition to two NMR-verified structures, MS/MS networking helped uncover an additional 13 aspercryptins. The aspercryptins break the conventional structural orientation of lipopeptides and appear “backward” when compared to known compounds of this class. We have also confirmed the 14-gene aspercryptin biosynthetic gene cluster, which encodes two fatty acid synthases and several enzymes to convert saturated octanoic and dodecanoic acid to α-amino acids. Graphical Abstract For decades the search for new natural products by screening for bioactivity has been hampered by rediscovering the same compounds.1 To avoid the rediscovery of natural products, some laboratories have decoupled discovery from screening for bioactivity by measuring accurate mass as the primary, high-throughput screen for the expression of new natural products.2 This new approach to molecular discovery has been applied to bacteria3,4 but less so in fungi, which offer an enormous biosynthetic potential.5 The genome of the mold Aspergillus nidulans contains more than 50 gene clusters annotated to be involved in the biosynthesis of natural products. Yet just over 20 of these biosynthetic gene clusters (BGCs) have associated natural products.6 Several strategies have been developed to better harness the biosynthetic repertoire of fungi.7 One such strategy involves the inhibition of histone deacetylase activity (HDACi) and has shown some promise.8–10 HDACi increases global histone acetylation levels and can increase transcription of otherwise repressed natural product BGCs. Our previous work used quantitative mass spectrometry-based analyses to compare the extracellular metabolomes of A. nidulans before and after both chemical and genetic HDACi.11 We found that HDACi both up-regulated and down-regulated the expression of many compounds. Among the up-regulated compounds were several not known to be produced by A. nidulans, such as the fellutamides.11 Continuing those efforts here, we have solved the structures of two new lipopeptides, aspercryptin A1 and A2, that are up-regulated by up to 90-fold using the HDACi-based strategy. The aspercryptins are six-amino-acid peptides containing two noncanonical α-amino acids derived from saturated C8- and C12-fatty acids and a C-terminal alcohol and are related to two previous aspercryptins published by Chiang et al. in 2016.12 Unlike the overwhelming majority of lipopeptides in the literature, the aspercryptins appear “backward” and have their lipid tail at the C-terminus. We also map the BGC responsible for the aspercryptins by genetic disruption and Northern blotting. It features a NRPS backbone gene encoding a terminal reductase, atypical activation domains for unusual “fatty amino acid” substrates, and enzymes putatively involved in fatty amino acid biogenesis. The aspercryptins join a small group of lipopeptide natural products where incorporation of the lipid moiety occurs at sites other than the N-terminus. Comparative Metabolomics of HDAC-Deficient A. nidulans Reveals the Aspercryptins In previous work, we generated a mutant of A. nidulans with constitutively lower levels of the HDAC RpdA (AN4493).11 Using differential metabolomics on this strain (rpdAKD, RAAS58.4) compared with wildtype fungus (Figure 1, left), we detected many signals that were up-regulated in the rpdAKD mutant (Figure 1, middle panels). Dereplication of these metabolites based on their accurate masses allowed us to cull the list to only those that have not yet been characterized (Figure 1). Several dozen of the putatively new metabolites displayed a high degree of similarity in their MS/MS fragmentation spectra, suggesting structural and therefore biosynthetic relatedness (Figure 2). For the members of this compound family, all were up-regulated over a wide range from 2- to 130-fold in rpdAKD vs wildtype (data for one compound are shown in Figure 1, third panel). The structures of two of these compounds, aspercryptin A1 (m/z 758.5386 [M + H]+, C37H71N7O9, −0.02 ppm) and aspercryptin A2 (m/z 742.5440 [M + H]+, C37H71N7O8, 0.45 ppm; Figure 1, right panel) were determined completely by MS with stable isotope feeding and extensive NMR (Supporting Information Table S1). The Aspercryptins Are “Backward” Lipopeptides The aspercryptins are linear lipopeptides built from six amino acids and appear to be the first example of peptide natural products with two lipid groups. They also seem to be the only known lipopeptides with a lipid tail at the C-terminus; thus they appear “backward” to other lipopeptides. This is despite the overwhelming literature precedent for lipopeptides having an N-terminal lipid group. Aspercryptin A1 and A2 differ from each other only at the N-terminal residue—serine for aspercryptin A1 and alanine for aspercryptin A2. The most striking feature of the aspercryptins is that of the six amino acids, two are highly unusual and nonproteogenic: 2-amino-octanoic acid and 2-amino-dodecanol. Stereochemical assignment of the aspercryptins was determined for each residue by derivatization with Marfey’s reagent and comparison with standards using HPLC-MS (Supporting Information Figure S1). For aspercryptin A1, the first two residues (serine and threonine) are epimerized to D-serine and D-allo-threonine while the remaining four residues are the L-isomers. The stereocenters of aspercryptin A2 are the same as those in aspercryptin A1; however, there is also an epimer of aspercryptin A2 such that the alanine is the L-isomer, thus we name this analogue epi-aspercryptin A2 (Supporting Information Figure S1). Several of these monomers (threonine, isoleucine, and serine) were also confirmed by metabolic feeding of stable isotope analogues (Supporting Information Figure S2). Furthermore, stable-isotope labeling shows that when fed d3-serine, labeled aspercryptin A1 only shows incorporation of two deuterons (Supporting Information Figure S2d). This further supports the epimerization of this residue by loss of the deuteron on the α carbon. MS/MS Networking Reveals a Large Family of Aspercryptins We observed that the MS/MS spectra of many other metabolites up-regulated by rpdAKD had striking similarity to those of aspercryptin A1 and A2. We then turned to MS/MS networking to visualize the relatedness of these metabolites in the data set. The ability for MS/MS networking to cluster biosynthetically related compounds is now established.13 When performing network analysis of the wildtype and rpdAKD extracts, we saw several examples of biosynthetically related natural products clustering together as expected; for example, the emericellamides group tightly together based on MS/MS spectral similarity (Figure 2a, circled). One cluster of metabolites consistently upregulated in the rpdAKD extracts (Figure 2a, boxed) contained aspercryptins A1 and A2 (Figure 2b). Using the MS/MS fragmentation patterns of aspercryptin A1 and A2 (Supporting Information Figures S3 and S4) and the recently published aspercryptins B1 and B312 as anchor points, we detected and have proposed putative structures for an additional 13 aspercryptins (Figure 2b, yellow circles; Table 1). These new aspercryptins fall into subfamilies, and we were readily able to quantify their abundance increases in the rpdAKD mutant (Supporting Information Figure S5). The putative structures for 8 of the additional 13 are shown in Figure 3. A subset was chosen for clarity in the main text (for a complete version of this chart, see Supporting Information Figure S6). The variations in the structures of the aspercryptins are standard for what is seen from NRPS pathways: incorporation of alanine instead of serine, valine for isoleucine, and a C10 fatty amino alcohol instead of the C12 version. Taking the structure of aspercryptin A1 as the “canonical” sequence of the aspercryptins, we have made a hierarchical nomenclature system, where the letter represents the status of the N-terminus, A for a free amine, B for cichorine capped, C for acetyl, and D for propionyl, and where the number represents “variants” of the base amino acid sequence, 1 for canonical, 2 for serine to alanine, etc. Isolation and full NMR structure determination for each of the aspercryptins are beyond the scope of this work. While the de novo structure elucidation of natural products from analysis of MS and MS/MS data alone is not possible, such analyses can readily be used for the detection and putative characterization of highly similar structural analogues. Though the stereochemistry of the aspercryptins proposed by MS/MS analysis is likely the same as those that have been experimentally determined, we have refrained from assuming that this is indeed the case (Figure 3). Confirmation of the Aspercryptin Biosynthetic Gene Cluster Linking natural products to their gene clusters is key to annotating the biosynthetic potential of phylogenetically diverse fungi. The genome of A. nidulans contains a 14-gene BGC12 (AN7884 to AN7872, Figure 4a and Supporting Information Table S3) that was recently shown to produce aspercryptins B1 and B3.12 This BGC contains a six-module NRPS (AN7884) and two fatty acid synthase (FAS) subunits (AN7880 and AN7873). To experimentally link all of the aspercryptins to this BGC, we deleted the NRPS gene, AN7884, and observed no detectable levels of any of the aspercryptins (Figure 4b). We also generated an overexpression mutant of the nearby transcription factor, AN7872, which led to increased transcript levels for many of the genes predicted to be in the cluster. This mirrored the overexpression of these genes in the rpdAKD strain (Supporting Information Figure S7) and confirmed the predicted cluster boundaries. Further solidifying its role as the transcription factor for the cluster, we also observed a >2-fold increase in the levels of aspercryptin A1 upon overexpression of AN7872 (Figure 4c). Additionally, deletion of the FAS gene AN7880 abolished production of the aspercryptins. Levels of the aspercryptins could be rescued partially by supplementing the media with octanoic and dodecanoic acids, which are likely the direct products of the FAS enzymes (Figure 4c). The NRPS, the transcription factor, and the intervening genes have been named atnA (NRPS) through atnN (transcription factor). Upon exploring the phylogenetic distribution of the atn BGC in the Aspergillus genome repository (http://www.aspergillusgenome.org), we found the atn BGC present in six other Aspergilli and to be most conserved in A. versicolor (NRPS genes 87% identical, Supporting Information Figure S8). The BGC borders were confirmed by transcript mapping using Northern blots of all genes in the cluster (Supporting Information Figure S7). Proposed Biosynthesis for the Aspercryptins Based on annotations of the BGC and the above deletion experiments, we propose the following for aspercryptin biosynthesis (Supporting Information Figure S9). The first step is the generation of the fatty acid precursors, octanoic and dodecanoic acids, by the FAS subunits AtnF and AtnM (AN7880 and AN7873). A. nidulans has three other pairs of FAS genes. One, fasA and fasB, is required for fatty acids involved in primary metabolism; deletion of either fasA or fasB is lethal.14 The other two pairs of FAS genes make precursor fatty acids for natural products—pkiB and pkiC for the polyketides made by the PKS pkiA (AN3386)15 and stcJ and stcK for sterigmatocystin biosynthesis.14 The fatty acid precursors are perhaps the most interesting aspect of the system and are likely transformed into the corresponding α-amino fatty acids in three steps. First, they are hydroxylated by the cytochrome P450 AtnE (AN7881), then oxidized to the corresponding α-keto acids by the NAD(P)-dependent oxidoreductase AtnD (AN11028), and finally converted to the α-amino fatty acids by the PLP-dependent aminotransferases AtnH or AtnJ (AN7878 and AN7876). Similar pathways to convert a fatty acid or polyketide precursor to the corresponding α-amino acid have been proposed for the cyclosporins and apicidins/HC-toxin, and experimentally supported to varying degrees.16–18 The only experimental evidence to support this transformation pathway from these systems comes from the deletion of a branched chain aminotransferase which eliminates HC-toxin biosynthesis.19 A recent publication by Chiang et al. also proposes this pathway for production of the fatty amino acids and shows that deletion of these genes abolished levels of aspercryptin B1 (except in the case of the aminotransferases, AtnH and AtnJ, which could reasonably compensate for each other).12 Unlike the other FAS genes in the A. nidulans genome, the aspercryptin FAS elements are “interrupted” by the amino-transferase genes that we suggest are involved in generation of the α-amino fatty acids (Supporting Information Figure S10). Perhaps this chromosomal organization evolved to decrease the likelihood that the gene cassette necessary for the fatty amino acid biosynthesis will be separated by chromosomal reorganization. If they were not disrupted by the aminotransferases, the expression of the FAS genes could conceivably result in unproductive synthesis of free medium-chain fatty acids that could disrupt membrane stability. In fact, the addition of dodecanoic acid above 10 μM for rescue experiments resulted in little to no growth. Once made, we propose that the α-amino fatty acids, 2-amino-octanoic and 2-amino-dodecanoic acids, are recognized, activated, and covalently tethered to the NRPS AtnA by its fourth and sixth adenylation domains (Supporting Information Figure S9). For typical lipopeptides, lipid moieties are added to the N-terminus by an initial terminal condensation domain of the NRPS.20 Incorporation of the lipid group as an α-amino fatty acid by an adenylation domain has been proposed for members of the apicidin/HC-toxin family and demonstrated in the biosynthesis of cyclosporin, where the NRPS has been shown to adenylate and covalently bind to the polyketide-derived α-amino acid (4R)-4-[(E)-2-butenyl]-4-methyl-L-threonine (bmt).21,22 Experiments are ongoing to directly establish the activity of adenylation domains 4 and 6 of AtnA in recognizing and activating 2-amino-octanoic and 2-amino-dodecanoic acids, respectively. To date, only a handful of α-amino fatty acids have been observed in cyclic lipopeptides (Supporting Information Figure S11). Intriguingly all of these are derived from fungi, with the exception of the piperazimycins.23 Perhaps this strategy for lipid incorporation into natural products is a fungal innovation that spread into Streptomycetes. In general, using the NRPSpredictor2 (modeled from bacterial NRPS) did not confidently predict the amino acid substrates recognized by each of the six adenylation domains of AtnA (Supporting Information Table S4). While Chiang et al. propose cichorine-serine as the monomer activated by the first adenylation domain of AtnA for the biosynthesis of aspercryptin B1,12 we believe it to be serine, and that the N-terminal amine of mature aspercryptin A1 subsequently reacts during the biosynthesis of cichorine24 to form aspercryptin B1. Additionally, despite AtnA having only one epimerase domain in the threonine module, the first two amino acids of aspercryptin A1 are D-serine and D-allo-threonine. This suggests that serine is either loaded directly as D-serine or that the epimerase domain in the threonine module epimerizes both L-serine and L-threonine. Iterative epimerase domains are not without precedent.25 Because we observed that the alanine residue of aspercryptin A2 exists as a mixture of L- and D-enantiomers, we predict that the epimerase domain acts iteratively. Chronologically, L-serine (for aspercryptin A1) or L-alanine (for aspercryptin A2) is activated by the first adenylation domain of AtnA, and while serine is fully epimerized to the D-enantiomer by the epimerase domain, alanine is only partially converted to the D-enantiomer. The complete conversion of serine and threonine and partial conversion of alanine suggests that the side chain hydroxyl of serine and threonine may aid in governing substrate recognition by the epimerase domain. The final step in the biosynthesis of the aspercryptins is the reduction of the C-terminus to an alcohol. Terminal reductase domains generally use the energy from NAD(P)H to install a reactive aldehyde that serves as a warhead26 or as an intermediate that rearranges to yield a mature natural product.25 For the aspercryptins, the removal of a charge-bearing site from dominantly hydrophobic portion of the compound is a potential reason for the installation of this alcohol at the C-termini of aspercryptins. In fact, the putatively membrane-associated nature of the aspercryptins may be relevant to their function and is consistent with our ability to isolate these compounds mainly from the cell mass and little from the extracellular medium. Many microbes use lipopeptides as a means to adhere to or move across surfaces, and to establish biofilms.27 Because we could not assign an antibacterial activity, the aspercryptins may serve a similar structural/motility function for A. nidulans. We also note that atnH and atnJ are induced by ethanol (the other atn genes were not examined), which possibly reflects a role for this metabolite during hypoxic stress.28 Conclusion Here, we uncovered over a dozen family members of the recently discovered lipopeptides, the aspercryptins, and mapped the complete gene cluster responsible for their biosynthesis. Instrumental in the discovery of the aspercryptins was the marriage of HDAC inhibition with MS-based metabolite screening. By inhibiting HDAC function, we were able to tease the production of a new family of natural products. By taking advantage of MS/MS networking, we could characterize the structural variation of the aspercryptins. Our successful use of HDACi in the discovery of the aspercryptins demonstrates its potential to survey fungal extracts for new chemical matter, even in species that have been so intensively studied, such as A. nidulans. Discovery of new natural product scaffolds from the microbial world at rates far higher than in past decades now represents a promising path for reinvigorating the pipeline of compounds flowing into pharmaceutical screening platforms. METHODS Fungal Transformations The generation of the rpdAKD strain was previously described.11 All other strains in this study are described in Table S5. For construction of overexpression strains, 1 kb of AN7884 or AN7872 5′ flanking regions and 3′ coding regions were amplified and fused to A. parasiticus pyrG (amplified from pJW24) and a 401 bp fragment of the alcA promoter using double joint PCR.29,30 The resulting overexpression construct was transformed into RJMP1.1 to create strains TAAS393.2 and TAAS394.2. For deletion constructs, 1 kb of flanking regions were amplified and fused to A. fumigatus riboB using double joint PCR. The resulting knockout construct(s) were transformed into RJMP1.1 and/or TAAS393.2 to create strains TAAS176.3, TAAS395.1, and TAAS217.1. Transformants were examined for targeted replacement of the native loci by PCR and Southern blotting, and expression levels were confirmed by northern analysis. Growth and Extraction of Fungal Strains for LC-MS/MS Analysis All strains were grown with initial inoculations of 106 spores/mL in 250 mL of GMM in 1 L unbaffled flasks, grown in the dark at 37 °C for 4 days at 200 rpm. For the initial screening of the rpdAKD, overexpression and deletant mutants, extractions, and analysis were performed as previously described.11 For induction of over-expression strains, lactose minimal medium +30 mM cyclopentanone was used. For rescue experiments, media was supplemented with 10 μM octanoic and doceanoic acids in acetone. However, for stable isotope incorporation experiments, 5 mL of GMM in 13 mL culture tubes were inoculated with 106 spores/mL and grown as described above, with the additional step of spiking with 1 mM sterile-filtered stable-isotope amino acids after 48 h of growth. After a total of 4 days of growth, whole cultures were extracted with an equal volume of ethyl acetate with sonication and vortexing for several minutes, followed by overnight incubation at 4 °C. Organic layers were then dried down and analyzed as previously described.11 Isolation and Structural Determination of Aspercryptins A1 and A2 For isolation of aspercryptin from rpdAKD mutant, large-scale growths (500 mL growths in 2 L flask, total of 4 L) were performed as described above. Mycelium was separated from the spent media with coffee filters. The spent media was extracted with dichloromethane to generate an emulsion layer. The emulsion was dried down. The cellular material was washed with excess methanol. Methanol extracts and emulsions from DCM extraction were pooled in methanol and dried onto excess silica. Silica was washed with excess ethyl acetate, and aspercryptins were eluted from silica with methanol. The methanol fraction’s pH was set to ~8.5 and run over SAX resin (Dowex 1 × 2 chloride form); flowthrough contained aspercryptins. SAX Resin washed with MeOH (pH 5) and pooled with flowthrough. This was then fractionated over preparative RPLC to afford ~4 mg of aspercryptin A1 and A2 in a ~2:1 mixture that could not be chromatographically resolved. All NMR experiments were performed in DMSO-d6 on an Agilent 600 MHz DD2 with a HCN cryoprobe, except for 13C NMR, which was acquired with an AVANCE III 500 MHz with direct cryoprobe. Stereochemistry of amino acids was determined following a standard Marfey’s Test protocol.31 Briefly, 1 mg of a ~2:1 mixture of aspercryptins A1 and A2 was hydrolyzed in 6 N HCl overnight at 110 °C. The hydrolyzed mixture was dried in vacuo, to which 100 μL of 1% Marfey’s reagent (FDAA) in acetone and 20 μL 1 M NaHCO3 were added. Amino acid standards were derivatized as described above; 2.5 μmol were used of each DL-alanine, D-serine, L-serine, DL-threonine, DL-allo-threonine, DL-isoleucine, DL-aspartic acid (for asparagine), synthetic DL-2-amino-octanoic acid, and synthetic DL-2-amino-dodecanol. MS/MS Networking to Identify Aspercryptin Molecular Family All MS/MS spectra were preprocessed by removing the 25% lowest intensity peaks, applying a nonlinear transformation to the peak intensities by taking their square root, and normalizing peak intensities to the sum of the intensities of all remaining peaks in the spectrum. Spectra with less than 10 remaining peaks, along with those where the base peak constituted more than 75% of the total scan intensity were removed from the analysis. MS/MS spectra were compared using the cosine similarity method. When comparing two spectra, if more than six matching peaks were found, the remaining unmatched peaks were aligned by shifting their m/z by the difference in precursor mass. The resulting output is a cosine score between 0 and 1 that describes the similarity between two spectra, where 1 represents a perfect match. MS/MS spectra from the same precursor were determined by a cosine similarity of >0.7 and a precursor match within 0.01 m/z. In cases where multiple spectra from the same precursor were observed, the spectrum with the higher intensity was used and the lower intensity spectrum was discarded. Using the cosine comparisons, a network containing 1590 nodes and 3396 edges was constructed and visualized in Cytoscape, where each node is a representative spectrum from a unique precursor, and each edge is a cosine comparison with a value of 0.4 or greater. Synthesis of Lipid Amino Acid Monomers DL-2-Amino-Octanoic Acid ±-2-bromo-octanoic acid (1 g, 4.5 mmol) was dissolved in 10 mL of 1:1 H2O/acetone. NaN3 (0.5 g, 7.7 mmol) was added. The solution was vigorously stirred overnight at RT (reaction mostly complete after 1 h) to yield ± –2-azido-octanoic acid. The 2-azido-octanoic acid was then hydrogenated in dry THF with 20% Pd/C under 1 atm of H2 with constant stirring at RT overnight. The mixture was filtered through Celite to remove Pd/C.32 A small aliquot was purified by RP-LC to yield 36 μg of pure DL-2-amino-octanoic acid (m/z 160.1333 [M + H]+, C8H17NOH+, 0.6 ppm). White waxy solid.1H- and 13C NMR spectra can be found in the Supporting Information. DL-2-Amino-Dodecanol ±-2-bromo-dodecanoic acid (1.25 g, 4.5 mmol) was dissolved in 10 mL of 1:1 H2O/acetone. NaN3 (0.5 g, 7.7 mmol) was added. The solution was vigorously stirred overnight at RT (reaction mostly complete after 1 h) to yield ±-2-azido-dodecanoic acid. The 2-azido-dodecanoic acid was then reduced in dry THF with LiAlH4 (0.35 g, 9.34 mmol) with constant stirring on an ice bath for 2 h.33 The reaction was quenched with 5% KHSO4 and extracted with ethyl acetate. A small aliquot was purified by RP-LC to yield 70 μg of DL-2-amino-dodecanol (m/z 202.2166 [M + H]+, C12H27NOH+, 0.2 ppm). Off-white waxy solid. 1H- and 13C NMR spectra can be found in the Supporting Information. Bioactivity Assays A total of 15 μg of purified aspercryptins A1 and A2 (~2:1) were tested in a standard disk diffusion assay against M. luteus, E. coli, P. aeruginosa, S. epidermidis, K. pneumoniae, B. subtilis, A. nidulans, and P. citrinum. Supplementary Material Supplemental Figures We would like to thank the National Institutes of Health R01 AT009143 (N.L.K.), T32 GM105538 (R.A.M.), T32 GM07133 and NRSA AI55397 (A.A.S.), and NIH 5PO1GM084077 (N.L.K.). We also thank Y. Zhang of IMSERC at Northwestern University for NMR assistance (supported by the NIH 1S10OD012016-01 and 1S10RR019071-01A1). Figure 1 Workflow to compare the metabolomes of wildtype Aspergillus nidulans and a strain where the HDAC RpdA is constitutively repressed (rpdAKD) leading to hyperacetylation in bulk chromatin. Knockdown of RpdA leads to many newly observed metabolites (second panel), from which new metabolites can be viewed selectively (third panel) and targeted for structure determination when they display unique mass signatures. Aspercryptins A1 and A2 are lipopeptides made by A. nidulans (fourth panel at far right). Full NMR data can be found in the Supporting Information. Stable isotope incorporation experiments verify much of these structures (Supporting Information Figure S2). Annotated MS/MS spectra for both aspercryptin A1 and A2 (Supporting Information Figures S3 and S4) are included for reference. Figure 2 MS/MS networking to identify and characterize a large cluster of aspercryptins. (a) MS/MS networking clusters natural products into molecular families. Here the metabolites made by wildtype A. nidulans (blue) are compared to those made by the rpdAKD mutant (red). Some metabolites are expressed equally (gray). Some molecular families are seen in only one biological state. The aspercryptins are boxed, and the emericellamides are circled. (b) Aspercryptin A1 and A2 (larger yellow circles) fall into an MS/MS cluster of metabolites mostly expressed by rpdAKD mutant. Comparing the MS/MS spectra of 5 NMR-elucidated aspercryptins allows for characterization of an additional 13 aspercryptins in this molecular family for a total of 18 structurally characterized aspercryptins, which includes aspercryptins B1 and B3 (yellow). Figure 3 Putative structures for 8 of the 13 aspercryptins based on analysis of MS spectral differences. A complete version of this chart with the 18 structurally characterized aspercryptins is shown in the Supporting Information. Heavily annotated MS/MS spectra of aspercryptins solved by NMR (blue asterisk) were used as the basis for comparison to other metabolites (Supporting Information Figures S3 and S4). Conservatively, stereochemical assignment has not been made for the MS-based structures. Figure 4 Evidence for the association of the aspercryptins to their BGC. (a) The BGC responsible for the biosynthesis of the aspercryptins contains a nonribosomal peptide synthetase encoded by atnA (AN7884), which contains six adenylation domains and a terminal reductase domain; the two fatty acid synthase subunits atnF (AN7880) and atnM (AN7873); a transcription factor atnN (AN7872); two aminotransferases atnH (AN7878) and atnJ (AN7876); a cytochrome P450 atnE (AN7881); an oxidoreductase atnD (AN11028); and three for transport atnC (AN11031) and atnG (AN7879) and resistance atnI (AN7877). (b) Deletion of the NRPS gene atnA in the background of rpdAKD showed it is necessary for biosynthesis of the aspercryptins (only aspercryptin A1 is shown here for clarity). (c) Overexpression of the transcription factor atnN (OE atnN) led to a doubling of aspercryptin A1 levels. Subsequent deletion of the FAS gene atnF (OE atnN, ΔatnF) abolished levels of aspercryptins, which could be rescued by supplementing media with the fatty acids octanoic and dodecanoic acids (OE atnN, ΔatnF + FAs). Table 1 Formulae and Expression Ratios for the 18 Aspercryptins from Figure 2, Including the Five Aspercryptins Elucidated by NMR (bold) and the 13 Additional Aspercryptins Whose Putative Structures Are Supported by Comparing MS/MS Spectra name m/z [M + H]+ molecular formula [rpdAKD]/[wildtype] name m/z [M + H]+ molecular formula [rpdAKD]/[wildtype] aspercryptin A1 a 758.539 C37H71N7O9 20 aspercryptin B1 b 934.586 C47H79N7O12 60 aspercryptin A2 a 742.544 C37H71N7O8 90c aspercryptin B2 918.591 C47H79N7O11 130 epi-aspercryptin A2 a 742.544 C37H71N7O8 90c aspercryptin B3 b 920.571 C46H77N7O12 50 aspercryptin A3 744.523 C36H69N7O9 2 aspercryptin B4 906.555 C45H75N7O12 120 aspercryptin A4 730.508 C35H67N7O9 2 aspercryptin C1 800.549 C39H73N7O10 20 aspercryptin A5 728.528 C36H69N7O8 20 aspercryptin C2 784.554 C39H73N7O9 60 aspercryptin A6 714.513 C35H67N7O8 20 aspercryptin C3 786.534 C38H71N7O10 3 aspercryptin A7 671.507 C34H66N6O7 20 aspercryptin C4 772.518 C37H69N7O10 2 aspercryptin D1 814.566 C40H75N7O10 5 aspercryptin C6 756.523 C37H69N7O9 20 a NMR structures described in this work. b NMR structures described by Chiang et al.12 c Ratio is an unresolved combination of epimers. Author Contributions M.T.H. grew strains, analyzed extracts by LC-MS, determined the structures of aspercryptins A1 and A2 by NMR and the other 13 aspercryptins by MS/MS, and prepared the manuscript. A.A.S. generated all mutants used in this study. A.W.G. performed MS/MS networking. R.A.M. performed reduction reactions and Marfey’s derivatization. N.P.K. annotated the distribution of the atn gene cluster. A.A.S., N.L.K., and N.P.K. heavily revised several versions of the manuscript. Notes The authors declare no competing financial interest. ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acschem-bio.6b00398. Figures largely of structural validation of the aspercryptins, including MS/MS spectra and Tables (PDF) 1 Baltz RH 2006 Marcel Faber Roundtable: is our antibiotic pipeline unproductive because of starvation, constipation or lack of inspiration? J Ind Microbiol Biotechnol 33 507 513 16418869 2 Hoffmann T Krug D Huttel S Muller R 2014 Improving natural products identification through targeted LC-MS/MS in an untargeted secondary metabolomics workflow Anal Chem 86 10780 10788 25280058 3 Sidebottom AM Johnson AR Karty JA Trader DJ Carlson EE 2013 Integrated metabolomics approach facilitates discovery of an unpredicted natural product suite from Streptomyces coelicolor M145 ACS Chem Biol 8 2009 2016 23777274 4 Derewacz DK Covington BC McLean JA Bachmann BO 2015 Mapping Microbial Response Metabolomes for Induced Natural Product Discovery ACS Chem Biol 10 1998 2006 26039241 5 Khaldi N Seifuddin FT Turner G Haft D Nierman WC Wolfe KH Fedorova ND 2010 SMURF: Genomic mapping of fungal secondary metabolite clusters Fungal Genet Biol 47 736 741 20554054 6 Yaegashi J Oakley BR Wang CC 2014 Recent advances in genome mining of secondary metabolite biosynthetic gene clusters and the development of heterologous expression systems systems in Aspergillus nidulans J Ind Microbiol Biotechnol 41 433 442 24342965 7 Brakhage AA Schroeckh V 2011 Fungal secondary metabolites - strategies to activate silent gene clusters Fungal Genet Biol 48 15 22 20433937 8 Fisch KM Gillaspy AF Gipson M Henrikson JC Hoover AR Jackson L Najar FZ Wagele H Cichewicz RH 2009 Chemical induction of silent biosynthetic pathway transcription in Aspergillus niger J Ind Microbiol Biotechnol 36 1199 1213 19521728 9 Henrikson JC Hoover AR Joyner PM Cichewicz RH 2009 A chemical epigenetics approach for engineering the in situ biosynthesis of a cryptic natural product from Aspergillus niger Org Biomol Chem 7 435 438 19156306 10 Mao XM Xu W Li D Yin WB Chooi YH Li YQ Tang Y Hu Y 2015 Epigenetic genome mining of an endophytic fungus leads to the pleiotropic biosynthesis of natural products Angew Chem, Int Ed 54 7592 7596 11 Albright JC Henke MT Soukup AA McClure RA Thomson RJ Keller NP Kelleher NL 2015 Large-scale metabolomics reveals a complex response of Aspergillus nidulans to epigenetic perturbation ACS Chem Biol 10 1535 1541 25815712 12 Chiang YM Ahuja M Oakley CE Entwistle R Asokan A Zutz C Wang CC Oakley BR 2016 Development of Genetic Dereplication Strains in Aspergillus nidulans Results in the Discovery of Aspercryptin Angew Chem, Int Ed 55 1662 1665 13 Nguyen DD Wu CH Moree WJ Lamsa A Medema MH Zhao X Gavilan RG Aparicio M Atencio L Jackson C Ballesteros J Sanchez J Watrous JD Phelan VV van de Wiel C Kersten RD Mehnaz S De Mot R Shank EA Charusanti P Nagarajan H Duggan BM Moore BS Bandeira N Palsson BO Pogliano K Gutierrez M Dorrestein PC 2013 MS/MS networking guided analysis of molecule and gene cluster families Proc Natl Acad Sci U S A 110 E2611 2620 23798442 14 Brown DW Adams TH Keller NP 1996 Aspergillus has distinct fatty acid synthases for primary and secondary metabolism Proc Natl Acad Sci U S A 93 14873 14877 8962148 15 Ahuja M Chiang YM Chang SL Praseuth MB Entwistle R Sanchez JF Lo HC Yeh HH Oakley BR Wang CC 2012 Illuminating the diversity of aromatic polyketide synthases in Aspergillus nidulans J Am Chem Soc 134 8212 8221 22510154 16 Offenzeller M Santer G Totschnig K Su Z Moser H Traber R Schneider-Scherzer E 1996 Biosynthesis of the unusual amino acid (4R)-4-[(E)-2-butenyl]-4-methyl-L-threonine of cyclosporin A: enzymatic analysis of the reaction sequence including identification of the methylation precursor in a polyketide pathway Biochemistry 35 8401 8412 8679598 17 Bushley KE Raja R Jaiswal P Cumbie JS Nonogaki M Boyd AE Owensby CA Knaus BJ Elser J Miller D Di Y McPhail KL Spatafora JW 2013 The genome of Tolypocladium inflatum: evolution, organization, and expression of the cyclosporin biosynthetic gene cluster PLoS Genet 9 e1003496 23818858 18 Niehaus EM Janevska S von Bargen KW Sieber CM Harrer H Humpf HU Tudzynski B 2014 Apicidin F: characterization and genetic manipulation of a new secondary metabolite gene cluster in the rice pathogen Fusarium fujikuroi PLoS One 9 e103336 25058475 19 Cheng YQ Ahn JH Walton JD 1999 A putative branched-chain-amino-acid transaminase gene required for HC-toxin biosynthesis and pathogenicity in Cochliobolus carbonum Microbiology 145 3539 3546 10627051 20 Chooi YH Tang Y 2010 Adding the lipo to lipopeptides: do more with less Chem Biol 17 791 793 20797606 21 Dittmann J Wenger RM Kleinkauf H Lawen A 1994 Mechanism of cyclosporin A biosynthesis. Evidence for synthesis via a single linear undecapeptide precursor J Biol Chem 269 2841 2846 8300618 22 Lawen A Zocher R 1990 Cyclosporin synthetase. The most complex peptide synthesizing multienzyme polypeptide so far described J Biol Chem 265 11355 11360 2358465 23 Miller ED Kauffman CA Jensen PR Fenical W 2007 Piperazimycins: Cytotoxic Hexadepsipeptides from a Marine-Derived Bacterium of the Genus Streptomyces J Org Chem 72 323 330 17221946 24 Sanchez JF Entwistle R Corcoran D Oakley BR Wang CC 2012 Identification and molecular genetic analysis of the cichorine gene cluster in Aspergillus nidulans MedChemComm 3 997 1002 25 Evans BS Ntai I Chen Y Robinson SJ Kelleher NL 2011 Proteomics-based discovery of koranimine, a cyclic imine natural product J Am Chem Soc 133 7316 7319 21520944 26 Shigemori H Wakuri S Yazawa K Nakamura T Sasaki T Kobayashi Ji 1991 Fellutamides A and B, cytotoxic peptides from a marine fish-possessing fungus Penicillium fellutanum Tetrahedron 47 8529 8534 27 Raaijmakers JM De Bruijn I Nybroe O Ongena M 2010 Natural functions of lipopeptides from Bacillus and Pseudomonas: more than surfactants and antibiotics FEMS Microbiol Rev 34 1037 1062 20412310 28 Shimizu M Fujii T Masuo S Takaya N 2010 Mechanism of de novo branched-chain amino acid synthesis as an alternative electron sink in hypoxic Aspergillus nidulans cells Appl Environ Microbiol 76 1507 1515 20081005 29 Yu JH Hamari Z Han KH Seo JA Reyes-Domínguez Y Scazzocchio C 2004 Double-joint PCR: a PCR-based molecular tool for gene manipulations in filamentous fungi Fungal Genet Biol 41 973 981 15465386 30 Calvo AM Wilson RA Bok JW Keller NP 2002 Relationship between secondary metabolism and fungal development Microbiol Mol Biol Rev 66 447 459 12208999 31 Bhushan R Bruckner H 2004 Marfey’s reagent for chiral amino acid analysis: a review Amino Acids 27 231 247 15503232 32 Corey EJ Link JO 1992 A general, catalytic, and enantioselective synthesis of α-amino acids J Am Chem Soc 114 1906 1908 33 Boyer JH 1951 Reduction of Organic Azides to Primary Amines with Lithium Aluminum Hydride J Am Chem Soc 73 5865 5866

PMC005xxxxxx/PMC5119466.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101084138 22395 Infect Genet Evol Infect. Genet. Evol. Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases 1567-1348 1567-7257 27180895 5119466 10.1016/j.meegid.2016.05.014 NIHMS829531 Article Distinguishing the genotype 1 genes and proteins of human Wa-like rotaviruses vs. porcine rotaviruses Silva Fernanda D.F. a Gregori F. a McDonald Sarah M. bc* a Department of Preventive Veterinary Medicine and Animal Health, College of Veterinary Medicine, University of São Paulo, Brazil b Virginia Tech Carilion School of Medicine and Research Institute, Roanoke, VA, USA c Department of Biomedical Sciences and Pathobiology, Virginia-Maryland College of Veterinary Medicine, Blacksburg, VA, USA * Corresponding author at: 2 Riverside Circle, Roanoke, VA 24016, USA. mcdonaldsa@vtc.vt.edu (S.M. McDonald) 15 11 2016 12 5 2016 9 2016 22 11 2016 43 614 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Group A rotaviruses (RVAs) are 11-segmented, double-stranded RNA viruses and important causes of gastroenteritis in the young of many animal species. Previous studies have suggested that human Wa-like RVAs share a close evolutionary relationship with porcine RVAs. Specifically, the VP1-VP3 and NSP2-5/6 genes of these viruses are usually classified as genotype 1 with >81% nucleotide sequence identity. Yet, it remains unknown whether the genotype 1 genes and proteins of human Wa-like strains are distinguishable from those of porcine strains. To investigate this, we performed comprehensive bioinformatic analyses using all known genotype 1 gene sequences. The RVAs analyzed represent wildtype strains isolated from humans or pigs at various geographical locations during the years of 2004–2013, including 11 newly-sequenced porcine RVAs from Brazil. We also analyzed archival strains that were isolated during the years of 1977–1992 as well as atypical strains involved in inter-species transmission between humans and pigs. We found that, in general, the genotype 1 genes of typical modern human Wa-like RVAs clustered together in phylogenetic trees and were separate from those of typical modern porcine RVAs. The only exception was for the NSP5/6 gene, which showed no host-specific phylogenetic clustering. Using amino acid sequence alignments, we identified 34 positions that differentiated the VP1-VP3, NSP2, and NSP3 genotype 1 proteins of typical modern human Wa-like RVAs versus typical modern porcine RVAs and documented how these positions vary in the archival/unusual isolates. No host-specific amino acid positions were identified for NSP4, NSP5, or NSP6. Altogether, the results of this study support the notion that human Wa-like RVAs and porcine RVAs are evolutionarily related, but indicate that some of their genotype 1 genes and proteins have diverged over time possibly as a reflection of sequestered replication and protein co-adaptation in their respective hosts. Group A rotavirus Genomics Porcine Human Genotype 1 Sub-genotypic diversity Amino acid changes Host tropism 1. Introduction Group A rotaviruses (RVAs) are gastrointestinal pathogens of many animal species. In humans, RVAs are a leading cause of childhood diarrheal death, particularly in developing regions of the world (Tate et al., 2012). RVAs are also a major cause of acute viral diarrhea in suckling and weaned piglets, imparting significant financial losses to the pork industry (Chang et al., 2012). The RVA genome consists of 11 segments of double-stranded RNA, which are encapsidated within a non-enveloped, triple-layered virion particle (Estes and Kapikian, 2007). The outermost layer of the RVA virion is made up of VP4 and VP7, while the middle layer is comprised of VP6. The innermost core shell is formed of VP2, and it surrounds the viral genome and RNA processing enzymes (VP1 and VP3). Five or six viral non-structural proteins (NSP1-NSP5/6) are made within infected cells and play various roles during viral replication. RVAs are classified according to a system that designates a specific genotype for each of the 11 viral genome segments (i.e., genes) based on their nucleotide sequences and established percent identity cut-off values (Matthijnssens et al., 2008). In this system, the genotype constellation of the VP7-VP4-VP6-VP1-VP2-VP3-NSP1-NSP2-NSP3-NSP4-NSP5/6 genes is described as Gx-Px-Ix-Rx-Cx-Mx-Ax-Nx-Tx-Ex-Hx, where x is a number. The vast majority of human RVAs sequenced to date (n > 300) exhibit the genotype constellation of G1/2/3/4/9/12-P[8]-I1-R1-C1-M1-A1-N1-T1-E1-H1 (Matthijnssens and Van Ranst, 2012). Human RVAs with this genotype constellation are referred to as “Wa-like” because they are genetically similar to the archival reference strain Wa (Wyatt et al., 1980). While fewer porcine RVAs have been sequenced to date (n = 33), the available data suggests that these strains typically exhibit the genotype constellation G3/5/9/11-P[6]/[7]/[13]/[19]/[23]-I5-R1-C1-M1-A8-N1-T1/7-E1-H1 (Kim et al., 2012; Martel-Paradis et al., 2013; Matthijnssens et al., 2008; Monini et al., 2014; Nagai et al., 2015; Okitsu et al., 2013; Silva et al., 2015; Theuns et al., 2015). Thus, it seems that while human Wa-like RVAs and porcine RVAs differ in their VP7, VP4, VP6, and NSP1 gene genotypes, they tend to have similar genotype 1 VP1-VP3 and NSP2-NSP5/6 genes. This observation suggested that human Wa-like RVAs and porcine RVAs share an evolutionary relationship and perhaps a common ancestor (Matthijnssens et al., 2008). In the current study, we sought to more fully characterize the genotype 1 genes and proteins of human Wa-like RVAs and porcine RVAs and to determine whether specific genetic signatures differentiate those from each host. We constructed neighbor-joining phylogenetic trees to reveal the relationships between the genotype 1 genes of typical and atypical human Wa-like and porcine RVAs. We also employed amino acid sequence alignments to identify positions that varied in the genotype 1 viral proteins in a host-specific manner. Furthermore, we documented how several archival/unusual isolates differ at these host-specific variable positions. Altogether, these results enhance our understanding of RVA genetic diversity and elucidate putative evolutionary signatures of genotype 1 viral proteins from human versus porcine strains. 2. Materials and methods 2.1. Nucleotide sequencing of 11 Brazilian porcine RVAs Near-complete genome nucleotide sequences were determined for 11 porcine RVAs (ROTA02, ROTA03, ROTA04, ROTA05, ROTA13, ROTA16, ROTA18, ROTA25, ROTA27, ROTA30, and ROTA31) using the same approach described in Silva et al., 2015 (Silva et al., 2015). These porcine strains are considered to represent typical modern isolates as they were found in fecal specimens collected from diarrheic nursing and suckling piglets (<60 days of age) on various Brazilian farms during the years of 2012–2013. The 83 new full or partial gene sequences were genotyped using RotaC2.0 (Maes et al., 2009) and were deposited into GenBank (Table S1). Only the sequences for the genotype 1 VP1-VP3 and NSP2-NSP5/6 genes were analyzed in the current study. 2.2. Sub-genotypic neighbor-joining phylogenetic analyses During the initial stages of the analyses, we downloaded the nucleotide sequences of genotype 1 VP1-VP3 and NSP2-NSP5/6 genes of all known human Wa-like RVAs and porcine RVAs. Alignments were created for each gene using Geneious Pro v5.6.5 (Biomatters) and the ClustalW algorithm and were trimmed so that the sequences were the same length in each alignment. The following nucleotide regions were analyzed for each gene: VP1 (nts 162-1572), VP2 (nts 28-1431), VP3 (nts 52-1034), NSP2 (nts 76-907), NSP3 (nts 59-977), NSP4 (nts 48-577), and NSP5/6 (nts 43-535). Neighbor-joining phylogenetic trees were created for each gene using Geneious Pro v5.6.5 (Biomatters). The trees were out-group rooted to the genotype 2 genes of strain DS-1 and built using three different distance models (Jukes-Cantor, Tamura-Nei, and HKY) and 100 bootstrap replicates. The overall tree topologies and major groupings were identical irrespective of the distance model chosen. Based upon the clustering of sequences in the initial trees, we then selected 102 representative RVAs to include in the final trees, which were built using the Jukes-Cantor distance model (Table S2 and Figs. S1–S7). For selection of the final strains, we also considered the year, G/P-genotype, and geographical location of the strain. To prepare Fig. 1, major groupings were collapsed using FigTree v1.4 and colorized using Adobe Illustrator CS5 (Adobe Systems). 2.3. Identification of host-specific amino acid changes in genotype 1 proteins The deduced amino acid sequences of genotype 1 proteins from the 102 representative strains (Table S2) were aligned using Geneious Pro v5.6.5 and the BLOSUM-62 matrix of ClustalW. Amino acids that varied according to host species or viral isolate were identified by visual inspection of the alignments and confirmed by NCBI BLAST analysis. For VP2, which varies in length, the documented position numbers are based on those of strain RVA/Human-wt/PRY/1638SR/2008/G1P[8]. The three dimensional locations of the host-specific variable positions were determined using UCSF Chimera v1.8 and the predicted or known atomic structures of viral proteins: strain UK and SA11 VP1 and VP2 (PDB# 4F5X), strain RRV VP3 (PDB# 2IHP), strain SA11 NSP2 (PDB# 1L9V), and SA11 NSP3 (PDB# 1KNZ and PDB# 1LJ2) (Deo et al., 2002; Estrozi et al., 2013; Pettersen et al., 2004; Groft and Burley, 2002; Jayaram et al., 2002; Ogden et al., 2014). 3. Results 3.1. Genetic relationships between the genotype 1 genes of human Wa-like RVAs and porcine RVAs To investigate the genetic relationships between human Wa-like RVA and porcine RV genotype 1 genes, we constructed individual phylogenetic trees for VP1-VP3 and NSP2-NSP5/6. Initial trees were constructed using all human Wa-like RVA and porcine RVA genotype 1 nucleotide sequences available in GenBank (data not shown). In this study, we also deduced the full or partial gene sequences for 11 porcine RVAs isolated from diarrheic piglets in Brazil during the years 2012–2013. The final trees included the genotype 1 genes of these 11 new Brazilian porcine RVA sequences as well as those of 91 additional human or porcine RVAs that altogether reflected the genetic diversity seen in the initial trees (Fig. 1). The genotype constellations of the 102 representative strains are shown in Fig. 2. The human Wa-like RVAs chosen as representatives of typical modern strains (Fig. 2A) included those that were isolated during the years of 2004–2011 from diarrheic children at various geographical locations: Thailand (strains CU938-BK, CU747-KK, and CU460-KK), Africa (strains MRC-DPRU1424, MRC-DPRU1723, and MRC-DPRU1262), Belgium (strains BE00030, BE00043, BE00055, B4633, and B3458), the United States (strains 2008747100, 2008747288, 2007719720, VU08-09-20, VU06-07-21, VU06-07-32, VU05-06-2, and VU05-06-47), Australia (strains CK00100, CK00088, CK00034, and CK00005), China (strains R588 and Y128), Italy (strains JES11 and AV21), Germany (strains GER126-08 and GER172-08), India (strains 61060 and 6361), and Paraguay (strains 954SR and 1638SR) (Arora and Chitambar, 2011; Ianiro et al., 2013; Matthijnssens et al., 2008; McDonald et al., 2012; Nyaga et al., 2013; Pietsch and Liebert, 2009; Rahman et al., 2007; Shintani et al., 2012; Theamboonlers et al., 2014; Zeller et al., 2015). We also included in our phylogenetic analyses the genotype 1 gene sequences of archival/unusual human strains (Fig. 2B) that were either isolated prior to 1992 (strains RV3, 116E, YO, IAL28, DC1476, DC1600, and DC4608) or that were likely the result of pig-to-human interspecies transmission events (strains BE2001, Arg4605, Mc323, BP271, BP1227, BP1547, 1809SR, Dhaka6, Matlab36, Ryukyu-1120, and EC2184) (Banyai et al., 2009; Degiuseppe et al., 2013; Ghosh et al., 2012; Heiman et al., 2008; Komoto et al., 2013; Martinez et al., 2014; Matthijnssens et al., 2008; Matthijnssens et al., 2010; McDonald et al., 2011; McDonald et al., 2009; Papp et al., 2013; Rippinger et al., 2010; Zeller et al., 2012; Zhang et al., 2014). For the porcine RVAs, we included in the final trees typical modern strains for which complete or near-complete genome sequence information is available (Fig. 2C). These strains were considered to be wildtype porcine RVAs because they had G/P-genotypes normally associated with porcine RVAs and they were found in the feces of diarrheic or non-diarrheic piglets during the years of 2006–2014. The porcine RVAs were also from various geographical locations: Brazil (strains ROTA01-10, ROTA24-25, ROTA27, and ROTA30-31), Belgium (strains 12R022, 12R002, 12R006, 12R005, 12R041, and 12R046), Italy (strains 3BS, 2CR, and 7RE), Thailand (strains CMP45, CMP29, CMP40, and CMP48), Korea (strains PRG921, PRG9121, PRG9235, and PRG942), Canada (strains F8-4 and F7-4), and Japan (strains BU8 and BU2) (Kim et al., 2012; Martel-Paradis et al., 2013; Monini et al., 2014; Nagai et al., 2015; Okitsu et al., 2013; Silva et al., 2015; Theuns et al., 2015). We also included the gene sequences of several atypical strains (Fig. 2D) that were either archival and isolated prior to 1983 (strains RV277, OSU, YM, and Gottfried) or that were predicted to represent human-porcine RVA reassortants (strains RU172, A131, and NMTL) (Ghosh et al., 2010; Matthijnssens et al., 2008; Shi et al., 2012; Theuns et al., 2015). For VP1, we found that the genotype 1 genes of typical modern human Wa-like RVAs clustered together in the phylogenetic tree (Fig. 1A and Fig. 2A; green) and were generally separate from those of both porcine RVAs and unusual/archival strains. The genotype 1 genes of typical modern porcine RVAs, on the other hand, were found in several different phylogenetic clusters interspersed with those of archival/unusual strains (Fig. 1A and Fig. 2B–D; dark blue). Consistent with previous reports, the genotype VP1 genes of archival human strains YO, DC1476, and DC4608 clustered with those of porcine RVAs rather than with typical human strains, suggesting that they might have been acquired via gene reassortment (Fig. 1A and Fig. 2B; dark blue) (McDonald et al., 2011; Zhang et al., 2014). Moreover, as reported previously, the genotype 1 VP1 genes of several unusual human isolates (i.e., BE2001, Arg4605, Mc323, BP271, BP1227, BP1547, 1809SR, Ryukyu-1120, and EC2184) clustered with those of typical porcine RVAs, indicative of these viruses taking part in inter-species transmission events (Fig. 1A and Fig. 2B; dark blue) (Degiuseppe et al., 2013; Ghosh et al., 2012; Komoto et al., 2013; Martinez et al., 2014; Papp et al., 2013; Zeller et al., 2012). For VP2 and VP3, two major lineages were seen in each respective phylogenetic tree, which were comprised mostly of the genotype 1 genes of: (i) typical modern human Wa-like RVAs (Fig. 1B–C and Fig. 2A; green) or (ii) typical modern porcine RVAs (Fig. 1B–C and Fig. 2C; dark blue). The VP2 tree also showed several additional branches, which represented the genotype 1 genes of one modern Belgium porcine RVA (strain 12R046) and several archival/unusual RVAs found in pigs or humans (strains RV277, IAL28, BP271, BP1227, and 116E) (Fig. 1B andFig. 2B–D; cyan). Likewise, for VP3, additional branches and clusters were identified and found to be comprised of genotype 1 genes of modern porcine RVAs from Thailand (strains CMP29, CMP40, and CMP48) or Korea (strains PRG921, PRG9121, PRG9235, and PRG942), as well as genes of archival/unusual porcine or human isolates (strains RU172, NMTL, Mc323, BP271, BP1227, BP1547, Dhaka6, Matlab38, and 116E) (Fig. 1C and Fig. 2B–D; cyan). Consistent with previous reports, the VP2 and/or VP3 genotype 1 genes of several archival/unusual human RVAs (e.g., DC1476, BE2001, 1809SR, and Ryukyu-1120) clustered with those of typical porcine RVAs (Fig. 1B–C and Fig. 2B–D; dark blue), reflecting either gene reassortment events or pig-to-human transmission of the strains (Komoto et al., 2013; Martinez et al., 2014; McDonald et al., 2011; Zeller et al., 2012; Zhang et al., 2014). For NSP2, the phylogenetic tree showed the same two major lineages that were identified in the VP2 and VP3 trees, which were comprised of genotype 1 genes of: (i) typical modern human Wa-like RVAs (Fig. 1D and Fig. 2A; green) and (ii) typical modern porcine RVAs (Fig. 1D and Fig. 2C; dark blue). However, for NSP2, a third larger lineage was also identified, and it was made up of genotype 1 genes of both typical and atypical human and porcine strains (Fig. 1D and Fig. 2; cyan). Specifically, the NSP2 genes of 19 of 44 modern porcine RVAs, 5 archival/unusual porcine RVAs (strains RV277, OSU, YM, Gottfried, and A131), and 6 pig-to-human transmitted RVAs (strains BE2001 Arg4605, Mc323, BP1547, 1809SR, and EC2184) clustered in this unique grouping (Fig. 2B and 2D; cyan). Fewer modern human Wa-like RVA NSP2 genes were found in this grouping; yet, those that were are considered to be typical wildtype isolates from South Africa (strain MRC-DPRU1262), Belgium (strains BE00043, BE00055, and B3458), United States (strains 2007719720 and VU05-06-47), and Paraguay (strains 954SR and 1638SR) (Fig. 2A; cyan). The NSP3 genes of human Wa-like RVAs are almost exclusively designated as genotype 1; however, the NSP3 genes of porcine RVAs can either be genotype 1 or genotype 7 (i.e., T1 or T7, respectively). We found that 23 of the typical modern porcine RVAs and 19 of the archival/unusual porcine or human RVAs have genotype 1 NSP3 genes (Fig. 2B–D). The phylogenetic tree showed that the genotype 1 NSP3 genes of modern human Wa-like strains clustered together (Fig. 1E and Fig. 2A; green) and distinctly from those of porcine strains, which formed two separate sub-clusters (Fig. 1E and Fig. 2C; dark blue). The genotype 1 NSP3 genes of several archival/unusual human strains (i.e., IAL28, DC1600, DC4608, Mc323, BP1227, Ryukyu-1120, 116E, and EC2184) were found to be porcine RVA-like (Fig. 2B; dark blue), consistent with the reports of these strains being reassortants or the result of inter-species transmission events (Banyai et al., 2009; Ghosh et al., 2012; Komoto et al., 2013; McDonald et al., 2011; Papp et al., 2013; Rippinger et al., 2010). For NSP4, the genotype 1 genes of typical modern human Wa-like RVAs clustered together in the phylogenetic tree (Fig. 1F and Fig. 2A; green) and were largely separate from those of porcine RVAs and those of archival/unusual strains. The genotype 1 NSP4 genes of typical modern porcine RVAs, on the other hand, were found in several different phylogenetic clusters interspersed with those of archival/unusual strains (Fig. 1F and Fig. 2C; dark blue). A small grouping of genes from 8 Brazilian porcine RVAs formed a sub-cluster distinct from the genes of typical human RVAs and other porcine RVAs (Fig. 1F and Fig. 2C–D; cyan). Related to this sub-cluster were the NSP4 genes of the unusual porcine RVA from China (strain NMTL) and a single modern human Wa-like RVA from Thailand (strain CU460-KK) (Shi et al., 2012; Theamboonlers et al., 2014). For NSP5/6, we did not find any specific clustering pattern that corresponded with host or virus type (Fig. 1G and Fig. 2; cyan). Despite a few clusters in the tree, the genotype 1 genes of modern human Wa-like RVAs and porcine RVAs could not be delineated. Likewise, the genotype 1 NSP5/6/genes of modern strains were not separate from those of archival/unusual strains. 3.2. Amino acid residues distinguishing typical modern human Wa-like vs. typical modern porcine RVAs Having found that the genotype 1 VP1-VP3 and NSP2-NSP4 genes of typical modern human Wa-like RVAs generally clustered together in phylogenetic trees and were distinct from those of typical modern porcine RVAs, we next wondered whether their encoded proteins exhibited specific amino acid changes. To investigate this, we created amino acid sequence alignments and identified positions that were conserved among all typical modern human Wa-like RVAs but that varied compared to typical modern porcine RVAs and vice versa. Altogether, we found 34 host-specific variant positions for VP1-VP3, NSP2, and NSP3; no such positions were identified for NSP4, a nonstructural protein involved in virion morphogenesis and virulence. For VP1, the viral RNA-dependent RNA polymerase, we identified 2 positions (284 and 1049) that varied in a host-specific manner (Fig. 3A). Residue 284 is a buried residue located in the N-terminal domain of the polymerase, whereas residue 1049 is a surface-exposed residue in the C-terminal bracelet domain (Lu et al., 2008) (Fig. 3A and data not shown). Neither of these residues are within the catalytic motifs of VP1. For VP2, the inner core shell protein, we identified 6 positions (27, 75, 140, 258, 417, and 891) that varied in a host-specific manner (Fig. 3B). We also found that the VP2 proteins of typical modern human Wa-like strains, but not porcine strains, have an asparagine-lysine (N-K)-rich insertion at positions ~40–50 in the alignment. This insertion differs in length among human Wa-like strains and is not seen in human DS-1-like or AU-like strains with genotype 2 or 3 VP2 proteins, respectively (McDonald and Patton, 2008). Residues 140, 258, and 891 of VP2 reside in the central carapace subdomain of the core shell, whereas reside 417 is in the apical subdomain near the fivefold axis and near predicted VP1 contact sites (Fig. 3B and data not shown) (Estrozi et al., 2013). The extreme N-terminal domain of VP2 (residues ~1–100) is not resolved in any published rotavirus structure. However, this region of the core shell protein is predicted to reside within the particle interior, suggesting that host-specific variable positions 27, 40–50, and 75 may be proximal to the location of RNA, VP1, and VP3 (McClain et al., 2010). VP2 residues 258 and 891 are on the outer surface of the core shell at the VP6 interface (data not shown). For VP3, an RNA capping enzyme that has innate immune antagonist activities via a C-terminal 3′5′-phosphodiesterase domain (PDE), we identified 10 positions that varied (109, 167, 234, 275, 276, 419, 470, 517, 639, and 646) between typical modern human Wa-like RVAs and typical modern porcine RVAs (Fig. 3C) (Morelli et al., 2015). According a VP3 homology model, residues 109 and 167 are located in the N-terminal adapter domain, residues 234, 470, and 517 are in the guanine-N7-methyltransferase (N7-MTase) domain, residues 275, 276 and 419 are in the ribose 2′ O-methyltransferase (2′ O-MTase) domain, and residues 639 and 646 are located in the guanylyltransferase/RNA 5′ triphosphatase (GTase/RTPase) domain (Ogden et al., 2014) (Fig. 3C and data not shown). All of these residues are distant from the predicted catalytic motifs of VP3, with the exception of residue 470, which is located immediately adjacent to the major S-adenosyl-l-homocysteine (SAH)-binding surface in the N7-MTase domain (Ogden et al., 2014). These residues are predicted to be fully or partially surface-exposed in the context of the homology model (data not shown). No host-specific variable positions were identified in the C-terminal PDE domain of VP3. For NSP2, which is an octameric nonstructural protein that plays roles during viral genome replication and early particle assembly, we identified 9 positions (48, 97, 98, 191, 245, 256, 282, 286, and 314) that varied between typical modern human Wa-like RVAs and typical modern porcine RVAs (Fig. 3D). Based on the structure of simian rotavirus strain SA11 NSP2, residues 48, 97, and 98 are located in the N-terminal domain, residues 191, 245, 256, 282, and 286 are located in the C-terminal domain, and residue 314 is located in the extreme C-terminal alpha-helix (Jayaram et al., 2002) (Fig. 3D and data not shown). These residues are predicted to be surface-exposed in the NSP2 octamer (data not shown). For NSP3, a nonstructural protein involved in protein synthesis, we identified 6 positions (155, 235, 252, 268, 278, and 300) that varied in a host-specific manner (Fig. 3E). Residue 155 is surface-exposed and located in the N-terminal RNA-binding domain of NSP3. Residues 235, 252, 268, 278, and 300 are located within the C-terminal eIF4G-binding domain (Groft and Burley, 2002). (Fig. 3E and data not shown). Residues 235, 252, 278, and 300 are buried and contribute to NSP3 dimerization, while residue 258 is surface-exposed (Fig. 3E and data not shown) (Deo et al., 2002). 3.3. Amino acid changes associated with unusual/archival RVAs When creating the amino acid sequence alignments, we noticed that many of the 34 host-specific variable positions differed in the sequence of archival/unusual isolates (Fig. 4). Because the changes in these atypical RVAs might be influenced by the genetic context of the entire virus, we sought to document and interpret them in the context of the viral genotype constellations. Based on this analysis, we divided the archival/ unusual isolates into 3 groups: Group 1 consisted of 5 RVAs that are quite similar to typical modern porcine RVAs (Fig. 4A–B). Only one member of this group was isolated from a pig (strain YM), whereas the other 4 members (strains BE2001, Mc323, Ryukyu-1120, and BP1547) were isolated from humans following inter-species transmission events (Ghosh et al., 2012; Komoto et al., 2013; Matthijnssens et al., 2008; Papp et al., 2013; Zeller et al., 2012). In general, few changes were detected at the host-specific amino acid positions for this group. Strain YM showed a single change at position 275 in VP3, while strain BE2001 showed changes at position 470 in VP3 and position 191 in NSP2. Strains Mc323, Ryukyu-1120, and BP1547 did not show any changes at the host-specific positions, suggesting that little if any adaptation occurred during their replication in the human host. Group 2 consisted of 6 putative reassortant RVAs (strains NMTL, OSU, A131, RU172, and Gottfried, RV277) that were isolated from pigs but that have one or more genes more similar to those of human Wa-like RVAs (Fig. 4A and 4C) (Ghosh et al., 2010; Matthijnssens et al., 2008; Shi et al., 2012; Theuns et al., 2015). Four of the viruses in this group showed changes at host-specific variable positions in VP3. Additionally, the VP2 gene of the archival Belgium strain RV277, which clustered distinctly in the phylogenetic tree from the those of typical modern porcine RVAs, was found to match the cognate human Wa-like RVA residues at positions 258 and 417. Similar changes were found for the VP2 proteins of human strains IAL28, 116E, BP1227, and BP271 (see below). Group 3 consisted of 14 putative reassortant RVAs (strains RV3, YO, IAL28, DC1476, DC1600, DC4608, Dhaka6, Matlab36, 116E, BP1227, BP271, Arg4605, 1809SR, and EC2184) that were isolated from humans but that have one or more porcine RVA-like genes (Fig. 4A and 4D) (Banyai et al., 2009; Degiuseppe et al., 2013; Heiman et al., 2008; Martinez et al., 2014; Matthijnssens et al., 2008; Papp et al., 2013; Rippinger et al., 2010). We found several changes at host-specific amino acid positions in VP2, VP3, and NSP2 for many members of this group. For example, archival human strains DC1476 and DC4608 have porcine RVA-like VP1 and VP2 genes/proteins. While the NSP2 proteins of these two isolates are most similar to those of typical modern human Wa-like RVAs (with cognate amino acids at 6 of the 9 host-specific sites), they differed at positions 48, 97, and 282, showing amino acid residues typical of porcine RVAs. It is interesting to speculate that perhaps the amino acid changes seen in these human-porcine reassortant RVAs reflects ongoing adaptive evolution of the virus as a result of the new (i.e., reassortant) gene constellation. 4. Discussion Previous comparative genomic studies have reported a likely evolutionary relationship between human Wa-like RVAs and porcine RVAs (Matthijnssens et al., 2008; Matthijnssens and Van Ranst, 2012; Theuns et al., 2015). However, until now, there had not been a systematic analysis of the genes and proteins from these seemingly related RVAs strains mainly because we lacked sufficient porcine RVA genome sequences. In fact, prior to 2015, <20 complete/near-complete genome sequences had been reported for wildtype porcine RVAs (i.e., those found in the feces of symptomatic piglets), whereas >300 wildtype human Wa-like RVA genome sequences existed. Recently, we described the near-complete genome sequencing and analysis of 12 Brazilian porcine RVAs isolated in 2012–2013 (Silva et al., 2015). Here, we report an additional 11 near-complete genome sequences of porcine RVAs from this same sample collection, thereby adding significantly to the number of wildtype porcine RVA sequences available in GenBank. Importantly, the genotype and sub-genotype constellations of our Brazilian strains mirror those of RVAs isolated from pigs in Belgium, Italy, Thailand, Korea, Canada, and Japan suggesting that they are representative of porcine RVA genetic diversity (Martel-Paradis et al., 2013; Kim et al., 2012; Monini et al., 2014; Nagai et al., 2015; Okitsu et al., 2013; Theuns et al., 2015). As such, in the current study, we were well-positioned to ask the following questions: Do the genes and proteins of human Wa-like RVAs differ from those of porcine RVAs? If so, are there specific genetic signatures that distinguish human versus porcine strains? By comparing the genotype constellations of typical modern human Wa-like RVAs to those of typical modern porcine RVAs (Fig. 2), we confirmed previous reports that the most divergent genes are those encoding outer capsid proteins VP7 and VP4, the intermediate capsid protein VP6, and the nonstructural innate immune antagonist protein NSP1 (Kim et al., 2012; Martel-Paradis et al., 2013; Monini et al., 2014; Nagai et al., 2015; Okitsu et al., 2013; Silva et al., 2015; Theuns et al., 2015). Indeed, human Wa-like RVAs and porcine RVAs usually exhibit different genotypes for these four genes, which by definition share <85% nucleotide sequence identity (Matthijnssens et al., 2008). Consistent with previous reports, we also observed that the NSP3 gene can sometimes, but not always, differ in genotype between human Wa-like RVAs and porcine RVAs (Kim et al., 2012; Martel-Paradis et al., 2013; Monini et al., 2014; Nagai et al., 2015; Okitsu et al., 2013; Silva et al., 2015; Theuns et al., 2015). Thus, it is likely that the VP7, VP4, VP6, NSP1, and NSP3 genes/proteins dictate whether an RVA productively infects a human versus a pig. However, we hypothesized that host-specific changes might be found in the six other viral genes, which share the genotype 1 designation (i.e., VP1-VP3 and NSP2-5/6) and >81% nucleotide sequence identity. We posited that such host-specific changes could contribute to host tropism or could reflect evolutionary changes resulting from sequestered replication of the virus. In support of this notion, our phylogenetic analyses showed that the genotype 1 genes from typical modern human Wa-like strains generally clustered together and were separate from those of typical modern porcine RVAs (Figs. 1 and 2). We did not observe any host-specific clustering for the NSP5/6 gene, suggesting that either this gene is genetically identical in human and porcine strains, or that our current data is insufficient to resolve host-specific differences. To investigate whether the proteins encoded by the genotype 1 genes differed depending upon the host species (i.e., human vs. pig), we performed amino acid sequence alignments (Fig. 3). For this analysis, it was critical that we stratify typical modern strains away from archival/ unusual isolates because we noticed that the latter showed amino acid variations at positions in the alignment that were otherwise invariable within a host (Fig. 4). Altogether, we identified a total of 34 amino acid positions that are conserved among the typical modern human Wa-like RVAs, but that differed for typical modern porcine RVAs or vice versa. Of these 34 host-specific variable positions, 29 of them are located at surface-exposed regions of viral replication proteins VP1, VP2, VP3, NSP2, or NSP3. Given the three-dimensional locations of these host-specific residues, we hypothesize that protein co-adaptation may have been a selection pressure that influenced their emergence. Protein co-adaptation occurs when a fitness-diminishing mutation in one viral protein is compensated for by a fitness-restoring change in a second viral protein. As a consequence of protein co-adaptation, gene segments become genetically-linked and stable constellations are observed (Heiman et al. 2008; Iturriza-Gòmara et al., 2003; Miño et al., 2016; Zhang et al., 2015). This hypothesis may explain why several of the putative reassortant viruses (i.e., archival/unusual isolates) show variations at the identified host-specific positions; the viral proteins may be trying to re-adapt to a new genetic context. Indeed, strains with gene segments derived from animal rotaviruses do not appear to be major causes of infections in humans, suggesting that their overall gene/protein constellations confer decreased fitness (compared to wildtype human Wa-like strains) in the absence of compensatory changes. However, several of the archival/unusual isolates, including reassortants were grown in cell culture for many passages before sequencing, which could account for the observed changes. Moreover, the role of co-adaptive RNA-RNA or RNA-protein interactions cannot be excluded as additional possible selection pressure leading to the emergence of host-specific changes or variations following reassortment. For instance, although we observed some host-specific segregation of genotype 1 NSP4 genes, we did not find any amino acid positions that differentiated these proteins in a host-specific manner, suggesting a role for the viral RNA molecule itself. Of course, it is also possible that the host-specific amino acid (or nucleotide) residues emerged as a result of evolutionary pressure(s) from cellular factors, which would differ between humans and pigs. A fuller understanding of RVA evolution will require additional genome sequences of isolates found in humans, pigs, and other animal hosts. Nevertheless, the results of the current study provide a baseline to compare future sequences to as we continue to resolve RVA genetic diversity and identify determinants of host tropism. Supplementary Material Table S1 Table S2 We are grateful to members of the McDonald laboratory for scientific and editorial suggestions. FDFS was supported by funding from the Brazilian government/Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Project 1079750). SMM was supported by grants from the NIH/NIAID (R01-AI116815-01, R21-AI113402-01, R21-AI119588-01). FG was supported by Foundation of Support to Research of the São Paulo State (Project 2011/00870-0). Fig. 1 Genetic relationships among the genotype 1 genes of human Wa-like RVAs and porcine RVAs. The neighbor-joining phylogenetic trees are outgroup rooted to strain DS-1, and horizontal branch lengths are drawn to scale (nucleotide substitutions per base). Bootstrap values are shown as percentages for key nodes. Monophyletic groupings were collapsed and are shown as cartooned triangles. Groupings comprised mostly of typical modern human RVA genotype 1 genes are shown in green and those of typical modern porcine RVA genes are shown in dark blue. Other groupings that contained genotype 1 genes of archival/unusual isolates or that contained genotype 1 genes from both human and porcine RVAs are shown in cyan. Fig. 2 Genotype constellations of representative human Wa-like RVAs and porcine RVAs. Strain names are listed to the left of the corresponding genotype constellation for the strains analyzed in this study. Each gene is shown as a box and is color-coded according to genotype or according to the results of Fig. 1. Specifically, genes characteristic of human Wa-like RVAs are shown in green, and those characteristic of porcine RVAs are shown in dark blue. Genes that formed distinct clusters or those that could not be resolved in Fig. 1 are shown in cyan. Black boxes represent genes for which no sequence information is available, and grey boxes represent genes with genotypes not generally found in human Wa-like or porcine strains. (A) Typical modern human Wa-like RVAs. These strains are considered to be wildtype, as they were isolated from human clinical fecal specimens during the years of 2004–2011 and have G/P-genotypes normally associated with human RVAs. (B) Archival/unusual human Wa-like RVAs. These human strains are considered atypical as they were isolated >20 years ago, are putative reassortants, and/or were involved in inter-species transmission events. (C) Typical modern porcine RVAs. These strains are considered to be wildtype, as they were isolated from porcine clinical fecal specimens during the years of 2006–2014 and have G/P-genotypes normally associated with porcine RVAs. (D) Archival/unusual porcine RVAs. These porcine strains are considered atypical as they were isolated from porcine fecal specimens >30 years ago or are putative reassortants. Fig. 3 Host-specific variable positions in genotype 1 viral proteins. The rectangles represent linear schematics of genotype 1 viral proteins (not drawn to scale) and are color-coded based on the results of Fig. 1. Typical modern human RVA genotype 1 proteins shown in green, and typical modern porcine RVA genotype 1 proteins shown in dark blue. Genotype 1 proteins of archival/unusual isolates or those from additional phylogenetic groupings containing both human and porcine RVAs are shown in cyan. Known or predicted domains/subdomains of each protein are bracketed and labeled. Positions in the amino acid sequence alignments that varied in a host-specific manner are labeled and the corresponding residues are listed below each protein schematic. Fig. 4 Amino acid differences in archival/unusual RVA strains at host-specific variable positions. Genotype constellations of RVAs are shown as in Fig. 2. For proteins VP1-VP3, NSP2, and NSP3, the host-specific variable positions identified in Fig. 3 are listed to the right of the corresponding protein. Green circles represent amino acid residues identical to those of typical modern human Wa-like RVAs. Dark blue circles represent amino acid residues identical to those of typical modern porcine RVAs. Orange circle represent amino acid residues that are unique, differing from those of human/porcine RVAs. Grey circles are shown for non-genotype 1 proteins (i.e. T7). Filled circles indicate that the residue found at that position is not what would be expected based on the evolutionary origin of the protein, indicating amino acid changes in archival/unusual isolates at host-specific amino acid positions. (A) Typical modern human Wa-like RVA and typical modern porcine RVA constellations/positions are shown as references. (B) Group 1 is comprised of porcine RVAs isolated from pigs or humans. (C) Group 2 is comprised of putative reassortant RVAs isolated from pigs but containing one or more human RVA-like genes. (D) Group 3 is comprised of putative reassortants isolated from humans but containing one or more porcine RVA-like genes. Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.meegid.2016.05.014. References Arora R Chitambar SD Full genomic analysis of Indian G1P[8] rotavirus strains Infect. Genet. Evol 2011 11 504 511 21256248 Banyai K Esona MD Kerin TK Hull JJ Mijatovic S Vasconez N Torres C de Filippis AM Foytich KR Gentsch JR Molecular characterization of a rare, human-porcine reassortant rotavirus strain, G11P[6], from Ecuador Arch. Virol 2009 154 1823 1829 19763776 Chang KS Kim Y Saif LJ Zimmerman JJ Karriker LA Ramirez A Schwartz KJ Stevenson GW Rotavirus and reovirus Diseases of Swine 2012 United Kingdom Wiley-Blackwell 621 634 Degiuseppe JI Beltramino J Millan A Stupka JA Parra GI Complete genome analyses of G4P[6] rotavirus detected in Argentinean children with diarrhoea provides evidence of interspecies transmission from swine Clin. Microbiol. Infect 2013 19 367 371 Deo RC Groft CM Rajashankar KR Burley SK Recognition of the rotavirus mRNA 3′ consensus by an asymmetric NSP3 homodimer Cell 2002 108 71 81 11792322 Estes MK Kapikian AZ Knipe DM Howley PM Rotaviruses and their replication Fields Virology 2007 fifth Philadelphia Lippincott Williams and Wilkens 1917 1974 Estrozi LF Settembre EC Goret G McClain B Zhang X Chen JZ Grigorieff N Harrison SC Location of the dsRNA-dependent polymerase, VP1, in rotavirus particles J. Mol. Biol 2013 425 124 132 23089332 Ghosh S Kobayashi N Nagashima S Chawla-Sarkar M Krishnan T Ganesh B Naik TN Full genomic analysis and possible origin of a porcine G12 rotavirus strain RU172 Virus Genes 2010 40 382 388 20157771 Ghosh S Urushibara N Taniguchi K Kobayashi N Whole genomic analysis reveals the porcine origin of human G9P[19] rotavirus strains Mc323 and Mc345 Infect. Genet. Evol 2012 12 471 477 22226701 Groft CM Burley SK Recognition of eIF4G by rotavirus NSP3 reveals a basis for mRNA circularization Mol. Cell 2002 9 1273 1283 12086624 Heiman EM McDonald SM Barro M Taraporewala ZF Bar-Magen T Patton JT Group A human rotavirus genomics: evidence that gene constellations are influenced by viral protein interactions J. Virol 2008 82 11106 11116 18786998 Ianiro G Heylen E Delogu R Zeller M Matthijnssens J Ruggeri FM Van Ranst M Fiore L Genetic diversity of G9P[8] rotavirus strains circulating in Italy in 2007 and 2010 as determined by whole genome sequencing Infect. Genet. Evol 2013 16 426 432 23542095 Iturriza-Gòmara M Anderton E Kang G Gallimore C Phillips W Desselberger U Gray J Evidence for genetic linkage between the gene segments encoding NSP4 and VP6 proteins in common and reassortant human rotavirus strains J. Clin. Microbiol 2003 41 3566 3573 12904356 Jayaram H Taraporewala Z Patton JT Prasad BV Rotavirus protein involved in genome replication and packaging exhibits a HIT-like fold Nature 2002 417 311 315 12015608 Kim HH Matthijnssens J Kim HJ Kwon HJ Park JG Son KY Ryu EH Kim DS Lee WS Kang MI Yang DK Hyun BH Park SI Park SJ Cho KO Full-length genomic analysis of porcine G9P[23] and G9P[7] rotavirus strains isolated from pigs with diarrhea in South Korea Infect. Genet. Evol 2012 12 1427 1435 22613801 Komoto S Maeno Y Tomita M Matsuoka T Ohfu M Yodoshi T Akeda H Taniguchi K Whole genomic analysis of a porcine-like human G5P[6] rotavirus strain isolated from a child with diarrhoea and encephalopathy in Japan J. Gen. Virol 2013 94 1568 1575 23515025 Lu X McDonald SM Tortorici MA Tao YJ Vasquez-Del Carpio R Nibert ML Patton JT Harrison SC Mechanism for coordinated RNA packaging and genome replication by rotavirus polymerase VP1 Structure 2008 16 1678 1688 19000820 Maes P Matthijnssens J Rahman M Van Ranst M RotaC: a web-based tool for the complete genome classification of group A rotaviruses BMC Microbiol 2009 9 238 19930627 Martel-Paradis O Laurin MA Martella V Sohal JS L'Homme Y Full-length genome analysis of G2, G9 and G11 porcine group A rotaviruses Vet. Microbiol 2013 162 94 102 23017831 Martinez M Galeano ME Akopov A Palacios R Russomando G Kirkness EF Parra GI Whole-genome analyses reveals the animal origin of a rotavirus G4P[6] detected in a child with severe diarrhea Infect. Genet. Evol 2014 27 156 162 25075468 Matthijnssens J Van Ranst M Genotype constellation and evolution of group A rotaviruses infecting humans Curr. Opin. Virol 2012 2 426 433 22683209 Matthijnssens J Ciarlet M Heiman E Arijs I Delbeke T McDonald SM Palombo EA Iturriza-Gomara M Maes P Patton JT Rahman M Van Ranst M Full genome-based classification of rotaviruses reveals a common origin between human Wa-like and porcine rotavirus strains and human DS-1-like and bovine rotavirus strains J. Virol 2008 82 3204 3219 18216098 Matthijnssens J Rahman M Ciarlet M Zeller M Heylen E Nakagomi T Uchida R Hassan Z Azim T Nakagomi O Van Ranst M Reassortment of human rotavirus gene segments into G11 rotavirus strains Emerg. Infect. Dis 2010 16 625 630 20350376 McClain B Settembre E Temple BR Bellamy AR Harrison SC X-ray crystal structure of the rotavirus inner capsid particle at 3.8 A resolution J. Mol. Biol 2010 397 587 599 20122940 McDonald SM Patton JT Molecular characterization of a subgroup specificity associated with the rotavirus inner capsid protein VP2 J. Virol 2008 82 2752 2764 18216104 McDonald SM Matthijnssens J McAllen JK Hine E Overton L Wang S Lemey P Zeller M Van Ranst M Spiro DJ Patton JT Evolutionary dynamics of human rotaviruses: balancing reassortment with preferred genome constellations PLoS Pathog 2009 5 e1000634 19851457 McDonald SM Davis K McAllen JK Spiro DJ Patton JT Intra-genotypic diversity of archival G4P[8] human rotaviruses from Washington, DC Infect. Genet. Evol 2011 11 1586 1594 21712102 McDonald SM McKell AO Rippinger CM McAllen JK Akopov A Kirkness EF Payne DC Edwards KM Chappell JD Patton JT Diversity and relationships of cocirculating modern human rotaviruses revealed using large-scale comparative genomics J. Virol 2012 86 9148 9162 22696651 Miño S Barrandeguy M Parreño V Parra GI Genetic linkage of capsid protein-encoding RNA segments in group A equine rotaviruses J. Gen. Virol 2016 97 912 921 26758293 Monini M Zaccaria G Ianiro G Lavazza A Vaccari G Ruggeri FM Full-length genomic analysis of porcine rotavirus strains isolated from pigs with diarrhea in northern Italy Infect. Genet. Evol 2014 25 4 13 24704759 Morelli M Ogden KM Patton JT Silencing the alarms: innate immune antagonism by rotavirus NSP1 and VP3 Virology 2015 479–480 75 84 Nagai M Shimada S Fujii Y Moriyama H Oba M Katayama Y Tsuchiaka S Okazaki S Omatsu T Furuya T Koyama S Shirai J Katayama K Mizutani T H2 genotypes of G4P[6], G5P[7], and G9[23] porcine rotaviruses show super-short RNA electropherotypes Vet. Microbiol 2015 176 250 256 25724331 Nyaga MM Jere KC Peenze I Mlera L van Dijk AA Seheri ML Mphahlele MJ Sequence analysis of the whole genomes of five African human G9 rotavirus strains Infect. Genet. Evol 2013 16 62 77 23369762 Ogden KM Snyder MJ Dennis AF Patton JT Predicted structure and domain organization of rotavirus capping enzyme and innate immune antagonist VP3 J. Virol 2014 88 9072 9085 24899176 Okitsu S Khamrin P Thongprachum A Kongkaew A Maneekarn N Mizuguchi M Hayakawa S Ushijima H Whole-genomic analysis of G3P[23], G9P[23] and G3P[13] rotavirus strains isolated from piglets with diarrhea in Thailand, 2006–2008 Infect. Genet. Evol 2013 18 74 86 23681022 Papp H Borzak R Farkas S Kisfali P Lengyel G Molnar P Melegh B Matthijnssens J Jakab F Martella V Banyai K Zoonotic transmission of reassortant porcine G4P[6] rotaviruses in Hungarian pediatric patients identified sporadically over a 15 year period Infect. Genet. Evol 2013 19 71 80 23792183 Pettersen EF Goddard TD Huang CC Couch GS Greenblatt DM Meng EC Ferrin TE UCSF Chimera—a visualization system for exploratory research and analysis J. Comput. Chem 2004 25 1605 1612 15264254 Pietsch C Liebert UG Human infection with G12 rotaviruses, Germany Emerg. Infect. Dis 2009 15 1512 1515 19788829 Rahman M Matthijnssens J Yang X Delbeke T Arijs I Taniguchi K Iturriza-Gomara M Iftekharuddin N Azim T Van Ranst M Evolutionary history and global spread of the emerging G12 human rotaviruses J. Virol 2007 81 2382 2390 17166908 Rippinger CM Patton JT McDonald SM Complete genome sequence analysis of candidate human rotavirus vaccine strains RV3 and 116E Virology 2010 405 201 213 20580391 Shi H Chen J Li H Sun D Wang C Feng L Molecular characterization of a rare G9P[23] porcine rotavirus isolate from China Arch. Virol 2012 157 1897 1903 22729562 Shintani T Ghosh S Wang YH Zhou X Zhou DJ Kobayashi N Whole genomic analysis of human G1P[8] rotavirus strains from different age groups in China Viruses 2012 4 1289 1304 23012626 Silva FD Espinoza LR Tonietti PO Barbosa BR Gregori F Whole-genomic analysis of 12 porcine group a rotaviruses isolated from symptomatic piglets in Brazil during the years of 2012–2013 Infect. Genet. Evol 2015 32 239 254 25796358 Tate JE Burton AH Boschi-Pinto C Steele AD Duque J Parashar UD 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis Lancet Infect. Dis 2012 12 136 141 22030330 Theamboonlers A Maiklang O Thongmee T Chieochansin T Vuthitanachot V Poovorawan Y Complete genotype constellation of human rotavirus group a circulating in Thailand, 2008–2011 Infect. Genet. Evol 2014 21 295 302 24296010 Theuns S Heylen E Zeller M Roukaerts ID Desmarets LM Van Ranst M Nauwynck HJ Matthijnssens J Complete genome characterization of recent and ancient Belgian pig group A rotaviruses and assessment of their evolutionary relationship with human rotaviruses J. Virol 2015 89 1043 1057 25378486 Wyatt RG James WD Bohl EH Theil KW Saif LJ Kalica AR Greenberg HB Kapikian AZ Chanock RM Human rotavirus type 2: cultivation in vitro Science 1980 207 189 191 6243190 Zeller M Heylen E De Coster S Van Ranst M Matthijnssens J Full genome characterization of a porcine-like human G9P[6] rotavirus strain isolated from an infant in Belgium Infect. Genet. Evol 2012 12 1492 1500 22430049 Zeller M Donato C Trovao NS Cowley D Heylen E Donker NC McAllen JK Akopov A Kirkness EF Lemey P Van Ranst M Matthijnssens J Kirkwood CD Genome-wide evolutionary analyses of G1P[8] strains isolated before and after rotavirus vaccine introduction Genome Biol. Evol 2015 7 2473 2483 26254487 Zhang S McDonald PW Thompson TA Dennis AF Akopov A Kirkness EF Patton JT McDonald SM Analysis of human rotaviruses from a single location over an 18-year time span suggests that protein coadaption influences gene constellations J. Virol 2014 88 9842 9863 24942570

PMC005xxxxxx/PMC5119550.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 100962779 22269 Nat Rev Genet Nat. Rev. Genet. Nature reviews. Genetics 1471-0056 1471-0064 22179717 5119550 10.1038/nrg3129 NIHMS819401 Article Experimental and analytical tools for studying the human microbiome Kuczynski Justin 1 Lauber Christian L. 2 Walters William A. 1 Parfrey Laura Wegener 3 Clemente José C. 3 Gevers Dirk 4 Knight Rob 35 1 Department of Molecular, Cellular and Developmental Biology, University of Colorado at Boulder, 347 UCB, Boulder, Colorado 80309, USA 2 Cooperative Institute for Research in Environmental Sciences, University of Colorado at Boulder, 216 UCB, Boulder, Colorado, 80309, USA 3 Department of Chemistry and Biochemistry, University of Colorado at Boulder, 215 UCB, Boulder, Colorado 80309, USA 4 Microbial Systems & Communities, Genome Sequencing and Analysis Program, The Broad Institute, 7 Cambridge Center, Cambridge, Massachusetts 02142, USA 5 Howard Hughes Medical Institute, 215 UCB, Boulder, Colorado 80309, USA Correspondence to R.K. Rob.Knight@colorado.edu 28 9 2016 16 12 2011 16 12 2011 22 11 2016 13 1 4758 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. The human microbiome substantially affects many aspects of human physiology, including metabolism, drug interactions and numerous diseases. This realization, coupled with ever-improving nucleotide sequencing technology, has precipitated the collection of diverse data sets that profile the microbiome. In the past 2 years, studies have begun to include sufficient numbers of subjects to provide the power to associate these microbiome features with clinical states using advanced algorithms, increasing the use of microbiome studies both individually and collectively. Here we discuss tools and strategies for microbiome studies, from primer selection to bioinformatics analysis. Human-associated microbial communities are implicated in a variety of diseases. Altered microbiota (the microorganisms that are present in a community) have been linked to vaginosis1, obesity2 and inflammatory bowel disease (IBD)3,4, among other maladies5. Although the importance of the human microbiota and the human microbiome has been hypothesized for some time, recent advances in the technology used to identify and to analyse components of the microbiome have substantially improved our knowledge of the microbial communities that are associated with various habitats, including humans. The amount of variability in the microbiota and in the microbiomes both within a human subject and between different subjects is immense, and projects such as the Human Microbiome Project6,7 and MetaHIT8 have already done much to define this variability. Few microbial species are shared by most people at appreciable abundance levels9, at least among the well-sampled human skin, gut, oral and genital communities. Immense diversity is even observed between serial samples that have been taken from the same site in the same person9,10. Although human body sites and individuals do have distinct signatures, the differences in community composition across body sites are large, and within a body site, differences across individuals are, in general, much larger than differences between a single individual’s community across time11. The importance of the human microbiome is thus immense, and this emerging field is rife with opportunities for discovery. Approaches to microbiome research are increasingly diverse, so here we present an outline of the many investigations that identify microorganisms (community surveys) and genes (shotgun metagenomics, referred to hereafter simply as ‘metagenomics’) that are present in the human-associated microbial communities and relate these communities to the host phenotype. Investigations into the RNAs, proteins and metabolites that are present — processes that are referred to as metatranscriptomics, metaproteomics and metabolomics, respectively — can also provide valuable insights, especially when they are combined with community surveys and metagenomic data12,13,14. In this Review, we primarily focus on DNA-based approaches, as they have been the primary means of interrogating the microbiome in recent investigations. DNA-based microbiome studies DNA-based microbiome studies frequently fall into one of two categories. Targeted amplicon studies focus on one or a few marker genes and use these markers to reveal the composition and diversity of the microbiota. Other studies use an entire metagenomic approach. This is sometimes referred to as shotgun metagenomics owing to the randomness with which genomic sequences are obtained. FIGURE 1 provides an overview of both study types and how they may be combined. Metagenomics approaches have the advantage of providing much richer data on the functional potential present in microbial communities. However, compared to targeted amplicon studies, they sacrifice resolution into the composition (that is, the identity of the microorganisms present) of those communities. Both approaches are useful, and we consider them both in this Review. We begin by addressing how researchers can investigate various constituents of a microbial community, such as eukaryotes, viruses and various groups of bacteria. We then cover the processing of biological samples and DNA extraction, followed by DNA sequencing and the methodological options that are available. We continue by providing some discussion about the bioinformatics software used to analyse the sequencing data that emanate from both targeted amplicon and metagenomic sequencing studies, and we conclude with our outlook on the near future of this rapidly evolving field. The aim of this article is to provide a guide to the experimental designs and analytical tools used in microbiome studies and to discuss the important decisions that experimenters face when conducting investigations into the microbiome. Selecting microbial genetic targets Most microbial community studies include targeted amplicon sequencing of phylogenetically informative markers, such as the ribosomal small subunit (16S ribosomal DNA (rDNA)) gene. This allows researchers to compare the identities of the microorganisms that are present in the communities of interest. One advantage of rRNA genes is that ribosomes, and thus 16S rDNA, are present in all living organisms, whereas other commonly used markers have a limited taxonomic distribution. Furthermore, ribosomal genes contain both slowly evolving regions that can be used to design broad-spectrum PCR primers and fast-evolving regions that can be used to classify organisms at finer taxonomic levels (for example, at the family or genus levels), although species-level resolution might be unfeasible using this information alone. Another advantage that the 16S rDNA gene provides over other potential marker genes is the availability of several large databases of reference sequences and taxonomies, such as greengenes15, SILVA16 and the Ribosomal Database Project17. Although 16S rDNA is the most predominately used marker, the internal transcribed spacer region of the rRNA gene can be useful for some taxonomic groups, such as fungi, especially when resolution below the genus level is needed18. Choice of universal PCR primer There is a large choice of PCR primers (even for the widely used 16S rDNA), each of which has advantages and disadvantages. PCR primers should therefore be carefully chosen to take into account the taxonomic coverage desired, the extent of phylogenetic information generated by the fragment, the compatibility of the fragment length with the sequencing platform and the degree of specificity for amplifying microbial sequences compared to host sequences. For instance, the widely used 16S rDNA primer pair F27–R338 is highly specific for bacteria (as opposed to archaea and eukaryotes) but lacks sensitivity for taxa such as Bifidobacterium19, which is an important member of the gut microbiota. Furthermore, a complete bacterial phylum, Verrucomicrobia, is so poorly amplified by the F27–R338 primer set that this group was perceived to be a rare soil microorganism when it is, in fact, a dominant taxon that constitutes more than 20% of soil community members20. In addition to minimizing amplification bias, an optimal primer set should amplify a region that is informative both taxonomically and phylogenetically, depending on the desired analyses. One way of assessing the taxonomic usefulness of the various hypervariable regions of the 16S rDNA gene is to compare the taxonomic assignments of short fragments (for example, 100–250 bp) in these regions against those of the full 16S rDNA gene. This type of analysis demonstrated that the 16S rDNA hypervariable region V6 poorly replicates full-length taxonomic assignments compared with the V4 or V2 region21. An illustrative cartoon on the effects of primer choice is shown in FIG. 2. In addition to the taxonomic coverage of a primer set and the usefulness of its amplicon, primer selection should also take into consideration what is known about the constituents of a target community. For instance, the universal primer set F515–R806 amplifies a broad range of bacterial and archaeal phyla and would be a prudent choice for a diverse community, such as a soil sample for the purpose of mapping the entire bacterial and archaeal community. However, this primer set poorly amplifies Propionibacterium, a common skin microorganism, and another primer set such as F27–R338 could be a better choice for this particular sample type22. Amplifying eukaryotic and viral communities High-throughput analyses of eukaryotic communities and parasites in host-associated environments are currently limited23,24 and have generally been targeted to specific taxonomic groups, such as fungi25,26. Eukaryotes generally make up a small proportion of the biomass and so fully incorporating this class of organism into microbiome studies will require addressing the challenges that are presented by the low abundance of eukaryotes relative to bacteria, as well as addressing those challenges that are presented by the difficulty of avoiding amplification of host DNA. These are probably the reasons why many current studies have focused on specific taxa. One possible solution to avoid amplifying host DNA is to use universal primers together with blocking probes that are specific to the host sequence27. As more community-wide data are generated, it will be interesting to see whether the presence or absence of specific eukaryotic taxa is more or less informative than studying the overall community composition. Viruses also have a major role in shaping the human microbiome. A typical healthy human carries an abundance of viral particles (~1012 by some estimates)28, consisting mainly of bacteriophages but also of a substantial number of eukaryotic viruses29. Although an individual’s gut virome varies substantially over the first few weeks of life30, the virome of adults seems stable over time; this is remarkable in light of the strong variation between the viromes of different individuals31. This pattern may be similar to the rapid changes that are observed in the bacterial microbiota early in life32,33 followed by relative stability in adulthood9,11. The effects of these patterns in the human virome are mostly not understood, although certain bacteriophages in other animals are beneficial to the host34. The lack of a gene that is universally present in all viruses means that amplicon-based studies cannot be used to characterize the virome in its totality. Instead, clade-specific viral genes can be used to identify and resolve viral subtypes from DNA sequences28,35. In order to characterize the virome comprehensively, researchers can isolate virus-like particles before metagenomic sequencing31. Elucidating the important effects of the human virome is confounded by the fact that most human virome sequences remain ‘unknown’ when compared to public databases, such as the nr database from the US National Center for Biotechnology Information (NCBI)36. The challenges of annotating the functional roles of unknown organisms are therefore particularly acute in virome studies. Processing of biological samples and DNA Sample collection The rate-limiting step in obtaining samples of human-associated microorganisms is the approval of a human study protocol and the recruitment of subjects before the actual collection of the samples; the method of collection depends on the goals of the study and the willingness of the subjects to be sampled. Methods for collection are generally less controversial than other aspects of microbiome studies, but a few suggestions are presented in BOX 1 to make collecting samples (and, by extension, study design) more straightforward. Box 1 Recommendations for sample collection The level of invasiveness of the sampling procedure should always be minimized when designing a study of human-associated microbiota. For instance, skin-associated communities can be sampled by swabbing106,107, scraping and performing a punch biopsy108, each of which represents a different level of discomfort for the subject. Fortunately, each method produces a similar picture of the microbial community108; as sufficient amounts of bacterial DNA can be recovered for 16S rDNA analysis by the less invasive swab method, punch biopsies should only be considered in circumstances in which deeper layers of the dermis must be studied. This logic extends to sampling other body sites. Mouth- and gut-associated microorganisms can easily be collected on swabs, which pose little risk of physical discomfort to the subject. However, when the study is focused on the lining of the gastrointestinal tract, there is little choice but to collect samples by endoscopy, as faecal samples are not sufficiently similar to the intestinal mucosa109. When designing a study, we suggest erring on the side of collecting as many samples as possible, as the increased statistical power of larger samples sizes cannot be overlooked when the difficulties of gaining approval and recruiting subjects represent a substantial barrier. Longitudinal studies, in particular, can provide much needed insights into the temporal variability of the microbial communities that are associated with specific body sites by using more samples from few individuals9,11,110. DNA extraction There has been some debate about which method results in the best picture of the microbial community37. However, most extraction methods use the same basic steps (cellular lysis, removal of non-DNA macromolecules and collection of the DNA) to separate the DNA from the cellular components and the environmental matrix. In general, the lysis step receives the most scrutiny, as the intensity of the lysis can result in bias towards a particular taxonomic group38. To avoid this, many studies use a combination of chemical, physical and mechanical means to lyse cells efficiently, and we suggest that all three means should be used to lyse cells in complex microbial assemblages. Incubating samples with detergents and/or enzymes followed by bead beating is a commonly used lysis method1,39,40, but there are also modified protocols for commercially available extraction kits that use a strong base, a high temperature and bead beating9,39,41. Non-DNA macromolecules are then physically separated from the DNA, and the DNA is then concentrated into a volume that is suitable for downstream applications. However, we should add that no extraction protocol works equally well on all sample types nor produces completely unbiased results: for example, researchers studying soil microorganisms have debated the ‘best’ method of DNA extraction for decades. Minimizing contamination Regardless of the extraction protocol used, steps must be taken to minimize contamination from a ‘foreign’ source. This will involve using sensible precautions (such as operating in as sterile an environment as possible and using clean equipment) and using commercially available kits or laboratory reagents that have been thoroughly tested to be free of contaminating DNA. Likewise, care should be taken to minimize contamination during sample collection. People collecting samples should wear gloves to prevent their own microbiota from contaminating samples, and they should adhere to an overall strategy of collecting samples in order of increasing biomass (for example, skin, then mouth and then faecal) to prevent the contamination of the lower biomass samples with those that are known to contain many microorganisms. After they have been collected, samples can be stored at various temperatures for short periods without there being a noticeable impact on the bacterial community42,43. However, the long-term effects of various storage conditions on human-associated microorganisms are not well described, so it is best to extract samples as soon as possible after collection or to use consistent storage conditions within and across studies. As the field of human microbiome research is still young, we have the chance to be early adopters of emerging standards in sample preparation and to use largely homogenous methods to allow more convenient analysis of results across studies. DNA sequencing methodologies Following successful extraction of DNA from the communities of interest, there remains the important step of obtaining sequences from that DNA. Microbiome studies that use 16S rDNA-based taxonomic profiling, as well as shotgun metagenomic studies, have used several sequencing technologies, including capillary (Sanger) sequencing (such as the Applied Biosystems 3730xl DNA analyser), pyrosequencing (such as the Roche 454 Genome Sequencer GS, FLX and FLX Titanium) and Illumina’s clonal arrays (such as the Illumina GAIIx and HiSeq2000). To our knowledge, no microbial community study has been published using other sequencing technologies. As shown in TABLE 1, each technology has distinct characteristics, and getting the most out of each technology requires striking the best balance between insert size, read length, depth, sequence accuracy, usability and cost. Taxonomic profiling studies The first taxonomic profiling studies were based on the Sanger sequencing method, starting with the short 5S rRNA gene and then using increasingly long fragments of the 16S rDNA, ultimately leading up to near-full-length 16S rDNA sequences44. The study of increasing fragment lengths has been made easier by the recent use of high-throughput sequencing techniques (FIG. 3). These techniques allow more samples to be sequenced at a higher depth and lower cost, albeit at shorter read length (TABLE 1). In principle, longer sequences should yield higher resolution for applications such as classification45, defining novel taxa44 or identifying specific biomarker organisms based on a culture-independent approach46. However, changes in community composition can typically be assessed on the basis of a gene fragment that is as small as 100 bp47. The drop in the length of sequenced fragments resulted in the need to select a shorter region of the 16S rDNA as a proxy, although no consensus has been reached regarding the single best region of 16S rDNA to assay (as noted above, several regions of the 16S rDNA are almost equally effective). Within 5 years, the read lengths obtained by 454 pyrosequencing kits evolved from 100 bases (using the GS platform) to >250 bases (using the FLX platform) and then to 500 bases (using the FLX Titanium platform). Now these kits allow sequencing of almost the entire length of amplicons, which, because of the current chemistry of this technology, were thought to be limited to <600 bases (TABLE 1). This approach can span multiple hypervariable and conserved regions, increasing the performance at different phylogenetic depths. More recently, the promise of high depth at low cost introduced by the Illumina platform has resulted in the development of different strategies to compensate for the limitation posed by the shorter read lengths. Depending on the preferred region of 16S rDNA and amplification primers, fragments between ~100 and ~300 bases can be subjected to paired-end sequencing with read lengths of 76, 101 or 125 bases48–53. Some of these approaches result in overlapping reads, increasing the total fragment length and sequence quality, whereas others result in a gap in the middle of the sequence. However, all approaches potentially allow much larger-scale studies than are possible with the 454 pyrosequencing technology. The switch from Sanger sequencing to parallel pyrosequencing has resulted in a widespread increase to thousands of reads per sample and to hundreds of samples per run. An Illumina-based approach gains two to three orders of magnitude, achieving millions of reads per sample and thousands of samples per run (TABLE 1). This continued evolution in the number of reads per run leads to an increasing need for higher levels of automation in sample preparation and for improved software tools to handle the resulting massive data sets. The decreased cost per sample and per nucleotide sequence has allowed immensely deep coverage of hundreds of samples simultaneously52. It has also introduced a plethora of new hurdles to overcome, from decreased taxonomic classification sensitivity owing to short read lengths and sequencing errors54 to increased time and financial costs of sample preparation, which can be mitigated through new approaches51. Functional profiling studies Sequencing needs are somewhat different for metagenomic functional profiling approaches compared to the typical 16S-rDNA-based taxonomic-profiling approaches described in the previous section. Initial studies involved sequencing clones of large-insert libraries (for example, bacterial artificial chromosomes) that were derived from genomic DNA extracted from a microbial community55. Despite the great use of contextual data generated by these long fragments, such directed sequencing has largely been replaced by cheaper, high-throughput shotgun sequencing, both for metagenomics and for single-organism genomics. Ideally, large contiguous pieces of DNA would be generated that contain operons as well as phylogenetic markers or even that contain complete microbial genomes of previously uncultivated species. The first example of using a large-scale, random shotgun sequencing approach is the reconstruction of a simple environmental community making up an acid mine biofilm. This study involved 78 million bases of long Sanger reads and led to several near-complete genomes of the key organisms that are present56. However, metagenomics studies have also undergone substantial changes owing to the introduction of new sequencing technologies. Increasingly more complex communities have now been tackled, making it far more difficult to sample the full breadth of a population at a sufficient depth to result in large contiguous sequences, something that is possible in low-diversity environments, such as in acid mine drainage. Using next-generation sequencing to increase the amount of data up to billions of bases per sample (as was recently done in a large-scale microbiome study of the human gut8) still shows limitations in generating high-quality de novo assemblies: fewer than half of the short (75-base) reads could be assembled into contigs that were larger than 500 bases, and most of them were smaller than 2 kb. This depth issue is even more pronounced in those sample types that are not dominated by the microbial community and that contain large fractions of host DNA (for example, human mucosal sites can contain over 80% human DNA). Although terabase metagenomic data sets will be available in the near future, the challenge of metagenomic assembly is unlikely to be solely addressed by increasing the depth of coverage, and improvements will depend on both the nature and the quality of the sequence data used as input and/or on the introduction of novel sequencing technologies. Approaches that are used and/or considered for metagenomic studies currently include: increasing read lengths by creating composite reads from short, overlapping paired-end sequences57; applying a hybrid approach that combines abundant short reads patched with longer reads (for a first-generation application, see REF. 58; future applications could leverage the longer reads from the PacBio RS platform); or obtaining long-range connectivity using jumping libraries59. All of these approaches are heavily dependent on advancements in computational algorithms. Error rates An important distinction between genome sequencing — which drives the development of many sequencing methods — and microbiome sequencing is that in genome sequencing, the error rates are less important. In genome sequencing, each region of the genome is sequenced many times; by contrast, in microbiome sequencing, each fragment (derived from either amplicon or metagenomic sequencing) may be sequenced only once, and this intrinsic variation, which is conflated with the sequencing error on each read, is itself of interest. The effect of raw sequencing error rates on the observed microbial diversity is potentially great, as every sequencing error could be portrayed as arriving from a novel organism. This effect is now typically evaluated on the basis of a synthetic, or ‘mock’, community that is created by pooling genomic DNA or cloned 16S rDNA fragments of multiple isolates60,61. Ideally, near-full-length 16S rDNA gene sequences are available at reference quality for the organisms in the mock community; this allows a description of the effect of raw sequencing error rates on the number of operational taxonomic units (OTUs) and a calculation of the actual error rates and error types62. Research on reducing all types of errors — including both sequencing errors and PCR-based chimeric sequence formation63 — is an active area of study, especially with the regular introduction of novel technologies that require faster algorithms for downstream data analysis. In general, sequences that are suspected of including a high level of sequencing errors are discarded in an initial filtering step. Parameters such as the average quality score of the sequence, number and length of homopolymers, number of mismatches in the primers or total length of the sequence are tested to determine whether the sequence represents a true DNA fragment or a sequencing artefact. Chimeric sequences represent a different type of error: during PCR amplification, synthesis from a given template might be interrupted and then restarted using a different template that shares a certain degree of homology with the first template, resulting in a final DNA fragment that is formed from (at least) two different templates. Different tools exist to detect and remove chimeric sequences64–66. However, as these sequences are usually equally distributed across all samples that are subjected to the same protocol, chimaeras will primarily affect estimates of species richness (total number of OTUs in a given sample), not estimates of relative species diversity between samples. Bioinformatics data analysis tools After DNA sequences have been acquired, they must be analysed and interpreted. The large amount of data produced in modern investigations requires sophisticated analysis tools; direct manipulation of the data, such as manually aligning DNA sequences, is no longer feasible. There are many approaches for analysing such data. Here we distinguish between the analysis of targeted amplicon and metagenomic microbiome data, where the latter includes a sampling of the entire complement of the DNA that is present in the microbiome rather than relying on inference of gene content from the organisms present (FIG. 1). Analysis tools for targeted amplicon data For targeted amplicon sequencing (for example, sequencing of 16S rDNA profiles), tools such as QIIME67, mothur66 and VAMPS were developed to allow researchers to compare and analyse microbial communities using large amounts of DNA sequence data. As noted above, high-throughput sequencing technologies introduce errors, or ‘noise’, into sequence data. For 454 pyrosequencing technology, there are approaches to reduce these errors or to ‘denoise’ the sequencing results by clustering the flowgrams produced by the sequencer into a smaller number of DNA sequences that were probably present in the original biological samples. This can be done with tools such as AmpliconNoise64 and Denoiser69, which can be applied individually or used within the context of QIIME, or they can be applied using mothur-based reimplementations. One important decision to make when performing analysis is whether to use the original, published version of a tool or whether to use a tool integrator’s implementation. In QIIME, our laboratories use the original implementations; this makes installation more difficult but guarantees the use of ‘brand-name’ tools. In mothur, the algorithms are rewritten from scratch, which makes the package easier to install and, in many cases, makes it run faster, but it does not guarantee that the same results will be produced. Other tools may use either of these two strategies or a mixture of them. Assigning amplicon sequences to operational taxonomic units DNA sequences, whether or not they have been denoised, are then typically clustered into OTUs. This process is a proxy for assigning DNA sequences to microbial species and is an important step for the analysis of diversity using a variety of species-based or OTU-based ecological metrics. Because most microbial species cannot yet be cultured in the laboratory69, diversity estimates are generally based on DNA sequence data that are grouped into these OTUs. OTUs can be defined as containing as much diversity as a given taxonomic level (for example, species or genus) by changing the sequence similarity threshold. However, the sequence similarity thresholds used are imprecise measures of an imprecise concept of a microbial ‘species’, and the sequence identity of, for example, the V2 region of the 16S rDNA does not exactly reflect the sequence identity of the entire gene70. An alternative to grouping sequence data using OTU thresholds to cluster the sequences without any external information (also known as de novo OTU picking) is to analyse only those sequences that can be formally assigned to known taxa or to analyse only those sequences that are highly similar to known taxa. This alternative approach is known as reference-based OTU clustering (or picking), and it frequently yields similar patterns to de novo approaches. In reference-based OTU picking, new DNA sequences are compared to one of a variety of reference databases of typically long, high-quality DNA sequences, such as those that are maintained by greengenes15 or by SILVA16. Reference-based approaches have several key advantages over the de novo approaches that have just been described, although they lack the latter’s suitability for exploring uncharted territory in microbiome space. They can be especially useful for combining sequence data from different regions of the 16S rDNA gene by mapping disparate regions onto a database of full-length reference sequences or for combining sequence data that are generated from different sequencing technologies. In such cases, de novo OTU picking would not be appropriate, as identical microorganisms might wrongly be assigned to different OTUs based solely on differences in sequencing technology or in the DNA region chosen for amplification. The reference-based approach to OTU generation is increasingly valuable as the extent of publicly available data expands because it allows new investigations to be interpreted in the context of existing studies. Picking OTUs against a reference database can also reduce the impact of chimeric sequences and noisy data. For example, one study resequenced the microbiota of two individuals using Illumina sequencing and compared community composition over time to previous results obtained on the same individuals by pyrosequencing11. These approaches are becoming increasingly important, as databases such as VAMPS, MG-RAST71 and the QIIME database aggregate sequences that are produced by different platforms and technologies, although such comparisons can be highly effective, even using existing techniques11. As the amount of sequence data for microbial communities grows exponentially, it is becoming increasingly clear that detailed information about samples (metadata) is crucial for enabling comparison of studies in databases and for allowing meta-analyses of the sequence data. The Genomic Standards Consortium aims to make standard the collection and reporting of detailed and standardized ‘metadata’ about the sequences. Such data would include clinical information about the subjects and technical information about how the DNA extraction, sequencing and other steps were performed. Towards this goal, the Genomic Standards Consortium has recently introduced standards, such as Minimum Information about Any (x) Sequence (MIxS)72, that allow these important factors to be defined. These standards have already been adopted by large projects, such as the Human Microbiome Project73 and the Earth Microbiome Project74. Inferring phylogenetic relationships The phylogenetic relationships among DNA sequences can also be inferred, either by using an existing reference database with an associated phylogeny (such as greengenes or SILVA) or by inferring the phylogeny de novo using tools such as NAST75,76 (for sequence alignment) and FastTree77 (for phylogeny inference from aligned sequences). Phylogenies allow the use of phylogenetically aware analysis, such as UniFrac78 and Phylogenetic Diversity79, which have been extensively used. Again, such tools for phylogeny inference can be used in isolation or within the context of QIIME, mothur or other pipelines. Even after sequences have been assigned to OTUs and possibly related to one another using a phylogenetic tree, the scale of data is still extensive. Various approaches exist to interpret such data to reveal meaningful patterns in microbial diversity. Typical approaches include applying metrics of community-wide similarity and using analyses such as principal coordinates analysis (PCoA) to visualize the relationships between microbial communities. The generation of rarefaction curves to display within-community diversity (α-diversity) is also common, as are various approaches to assess informative taxa across communities and across categories of communities. VAMPS, QIIME and mothur all perform such analyses and frequently produce publication-ready figures11,80. Analytical tools for metagenomic data In addition to taxonomic analyses using marker genes, microbial community studies are increasingly using metagenomic sequencing to assess community membership in diverse environments8,10,81–83. One additional benefit of metagenomic sequencing is that it yields information on the encoded functional potential in the community DNA, and this functional profile can help to generate hypotheses on community dynamics and metabolic properties. To determine the phylogenetic membership of microbial communities based on metagenomic sequences, several freely available and popular software packages compare the metagenomic reads to a variety of full genome sequences using BLAST or interpolated Markov models. The identity of the best match then determines the likely phylogenetic origin of the sequence84,85. Another alternative is to find and extract informative phylogenetic markers from the metagenomic reads, which can be processed with similar methods to targeted gene surveys86,87. However, the taxonomic assignments from arbitrary metagenomic fragments remain a big challenge as much of the novelty in metagenomes still corresponds to organisms that lack a representative sequenced genome, and complementing metagenomic analyses with 16S rDNA analyses for which much larger reference databases exist are often useful. One advantage of metagenomic approaches is their ability to discriminate strains of common species by gene content beyond the resolution that is possible with 16S rDNA sequences8, although this approach requires high coverage and thus cannot be applied to rarer members of the community. Functional annotation of metagenomic sequences Community functional potential is almost always analysed by comparing the metagenomic sequences to large databases of metabolic annotations88–90. Depending on the sequencing strategy, coverage and community complexity, the sequences can either be used directly as short fragments, or they can be assembled into larger contigs for gene calling before annotation (for a review of metagenomic assembly, see REF. 91). Some software packages exist to tie together various components, and although no single standard exists yet, the Genomic Standards Consortium92 is currently working towards a consistent way for describing and comparing approaches. Depending on the need for customization and computational resources, annotation pipelines can be downloaded and run on a local computer cluster (for example, SmashCommunity93), or a user can upload their sequence data to a Web server and use a pre-structured pipeline, such as MG-RAST71, CAMERA94 or IMG/M95. One major benefit of online resources such as MG-RAST is the ability to compare publicly available metagenomic data sets, allowing users to perform comparative metagenomic analysis of their samples against a huge variety of environmental and host-associated metagenomic projects. As with targeted gene surveys, one of the greatest challenges for understanding metagenomic data sets is the identification of significantly different features between communities. Just like targeted gene surveys, the large number of DNA sequences and functional groups requires paying careful attention to false discovery rates and good knowledge of suitable statistical tests — simple parametric statistics are rarely appropriate. Additionally, it can be difficult to understand the biological importance of a list of gene functions that are differentially represented between two communities or between several groups of communities. An increasing number of tools and strategies are being developed that allow the identification of substantial differences in functional groups in metagenomes and, importantly, they can aggregate these gene-level differences into metabolic pathways that are differentially represented46,96,97. It should be noted that many DNA sequences from both metagenomic and targeted gene studies originate from poorly characterized genes and microorganisms, and our understanding of many of the microbial guests that are harboured in the human body is still shallow. There are thus ongoing efforts to sequence the complete genomes of a wide variety of microorganisms, in part to provide a framework from which to interpret DNA sequences from microbiome studies. These additional reference genomes that are provided by large sequencing efforts, such as the Human Microbiome Project, are allowing more insight into the nature of microorganisms that are identified in targeted gene surveys and are aiding in the functional annotation and assembly of metagenomic sequences. The availability of a larger number of reference genomes will also have a dramatic influence in our understanding of disease-associated strains and pathogenicity, as well as in providing a clearer picture of microbial evolution. Finally, the accumulation of more reference genomes will result in a faster pace at which we can reconstruct new genomes from short reads by using co-assembly methods98. Conclusions and prospectus Characterizing the taxonomic and functional characteristics of the human microbiome, despite considerable variation in methodology, is providing a much richer picture of our ‘normal’ microbial symbionts and, for several body sites, the association between microbial communities and human disease. One theme that is common to both taxonomic and functional analyses of the human microbiome is that, in order to find associations between genes, organisms and physiological or disease states, it is more informative to use larger numbers of samples than it is to sequence each sample at greater coverage. Depending on the subtlety of the patterns to be discovered, surprisingly few sequences per sample may be needed to reveal them99. However, associations that involve rare species and rare genes will require far deeper sampling than is currently typical100. For example, we know that many pathogens can be introduced with inocula consisting of only a handful of cells: the discovery of rare, persistent pathogens (either at the level of species or at the level of strains) or of rare genes, such as virulence factors, might require several additional orders of magnitude of sequencing effort and/or finer resolution of sampling sites. These deeper investigations might involve directly sampling gut mucosa rather than relying on stools as a proxy or sampling specific and anatomically defined regions of skin. Relevant to this issue is the possibility that rare ‘keystone’ genes or species that are themselves undetectable might produce widespread ripples in common members of the microbiota that are more readily detectable. For example, sampling a cubic kilometre of Yellowstone National Park would be unlikely to reveal much wolf DNA, but we know that wolves are crucial in structuring this ecosystem because their role is reflected in the distribution of common plants, animals and even fish101. In this context, it is especially important to distinguish correlations between the presence or abundance of genes and species with physiological states and causation of those states by these species. Causality has been established in a few cases: stool transplantations from one human to another can cure persistent Clostridium difficile infections102, and transfer of microbial communities together with associated physiological states has been seen between different mice103 or even between humans and mice104. However, manipulating the microbiome of humans involves overcoming substantial bureaucratic obstacles, which, in our view, are excessive, given the prevalence of unverified and largely unregulated products that aim to manipulate the microbiome and are marketed directly to non-expert consumers. Thus, we can expect much of this work to take place in animal models for the foreseeable future. Whether prospective time series studies will reveal which associations between genes or microorganisms and physiological states are causal and which are side effects remains an important unresolved question. In this Review, we have focused on 16S rDNA and metagenomic studies, which tell us about community membership and functional capacity. However, studies of gene expression at the RNA and protein level, of metabolites and of specific groups of lipids, carbohydrates and other markers will be increasingly important, as will studies of complete genomes of individual species in their natural environments. An especially exciting prospect is the ability to assemble personalized culture collections that describe the variability within an individual person105. Such investigations will allow more detailed manipulation of these communities, such as leave-one-out analyses and other perturbations, allowing us to untangle the effects of specific genes and organisms within complex communities. Manipulations of culture collections should help to determine which level of ‘multi-omics’ analysis is most useful for identifying biomarkers of human disease and also for generally increasing our understanding of the associations between microorganisms and human health. This Review covers work in the Knight laboratory that is supported by the US National Institutes of Health (NIH), the Bill and Melinda Gates Foundation, the Crohn’s and Colitis Foundation and the Howard Hughes Medical Institutes. D.G. was supported by a grant from the NIH (NIHU54HG004969), the Crohn’s and Colitis Foundation of America and the Juvenile Diabetes Research Foundation. Abbreviations Microbiota The collection of microbial organisms from a defined environment, such as a human gut. Microbiome The collection of genes that are harboured by microbiota. Metagenomics The study of the collective genome of microorganisms from an environment. Shotgun metagenomics refers to the approach of shearing DNA that have been extracted from the environment and sequencing the small fragments. Amplicon An amplified fragment of DNA from a region of a marker gene (such as 16S rDNA) that is generated by PCR. Bead beating A process used to lyse cells and to disrupt larger structures before DNA extraction. Paired-end sequencing An approach used in some sequencing platforms in which a single DNA clone is subjected to sequencing reads that originate from each of a set of primers, such that the direction of each sequencing reaction is directed to the origin of the other. Functional profiling approaches Studies in which the genomic DNA of the microbiome is assessed for functional potential. Jumping libraries Libraries that use molecular biology techniques to join together the ends of a larger DNA fragment, allowing sequencing on platforms that can only sequence a shorter fragment length. For example, 10 kb fragments might be reduced to 200 bases from each end, giving a final fragment size of 400 bp that can undergo paired-end sequencing. Operational taxonomic units (OTUs) Sequences are generally collapsed into OTUs based on sequence similarity thresholds for downstream analyses. The typical threshold is 97%, and this is taken as a proxy for species level divergence, although what constitutes a microbial species remains an open debate. Chimeric sequence An artificial sequence that juxtaposes gene regions from two or more unrelated organisms. It is produced by recombination between two or more DNA molecules during PCR amplification. Homopolymers Sequences that contain repetitions of identical bases. Metadata Information associated with sequences, including environmental conditions and the time and location of the sample collection site. Principal coordinates analysis (PCoA) A multivariate technique used in microbiome studies to visualize the relationships among communities. Each community is represented by a point in typically two- or three- dimensional space, and similar communities are located close to one another in the resulting PCoA plot. Rarefaction curves Plots of community diversity versus depth of sequencing (or, generally, observation). They are used to assess the amount of diversity and the extent to which it has been sampled at a given depth of sequencing. Interpolated Markov models A bioinformatics technique used here to classify DNA sequences using patterns of k-mer nucleotide strings that are present in a within a genome database. Leave-one-out analyses Studies of a microbial community that lacks one of its constituent microbial taxa. Figure 1 Bioinformatics analysis of microbiome sequence data Although variations exist, we show typical analysis paths for both targeted amplicon analysis (for example, analysis of 16S rDNA) and metagenomic analysis (for example, shotgun amplification). In targeted amplicon studies (left branch), raw sequences are usually passed through quality filtering and denoising algorithms to minimize the effects of sequencing artefacts. The resulting filtered sequence reads are clustered into operational taxonomic units (OTUs), representing similar organisms, and the phylogeny and taxonomic identity (when the organisms closely resemble named taxonomic groups) are inferred. At this stage, it is possible to incorporate sequence data from other relevant studies, or the data can be treated individually. The abundance of the various OTUs is then subjected to a variety of multivariate analyses and visualization procedures to elucidate the structure and patterns of the microbial communities. In metagenomic studies (right branch), the raw sequence fragments are sometimes assembled into contiguous sequences (contigs). The functional potential of those sequences is then typically assessed using a functional annotation database. The results are used to identify important metabolic pathways and are compared to the results of other metagenomic studies. The processed data are then subjected to multivariate analyses and visualizations, and they are often combined with the results of microbial profiling. Note that there are several opportunities for obtaining targeted gene data (similar to those produced in 16S rDNA gene surveys) in metagenomic studies, as indicated by the step labelled ‘identify target gene sequences’. KEGG, Kyoto Encyclopedia of Genes and Genomes; MG-RAST, Metagenomics Rapid Annotation using Subsystems Technology. Figure 2 Effects of primer choice in targeted amplicon sequencing In choosing primers, there is often a trade-off between being broadly inclusive and avoiding biases for or against specific groups, which might occur owing to variation in sequence conservation between lineages. Therefore, changes that increase the representation of one group may cause another group to drop out. This figure shows predicted taxonomic coverage for two 16S rDNA primer pairs: carefully selected universal primers (a) and commonly used primers with high bias (b). This results in variations in the observed taxonomic composition of communities owing to primer sensitivity, as shown in the pie charts for the communities in the mouth, the head and the skin and in the histograms on the right, representing the primer pairs’ sensitivity to various microbial phyla (the effect continues to finer-level taxa, such as genera (not shown)). As no primer set is without some degree of bias, it is important that a priori knowledge of the target microbiota is used along with knowledge of a primer’s performance when choosing a primer set. However, primer set 1 (which amplifies the F515–R806 fragment) is an especially good choice in terms of avoiding bias and allowing for good representation of known bacterial and archaeal groups. It has been adopted by the Earth Microbiome Project, among other projects. Data taken from REF. 9. Figure 3 How to get the most taxonomic information out of each sequencing technology Taxonomic profiling consists of generating an amplicon (in red) of the (partial) 16S ribosomal RNA (rRNA) gene (top) with selected PCR primers, followed by sequencing that amplicon with a preferred technology (grey arrows): Sanger ABI 3730xl, 454 (FLX and FLX Titanium) and Illumina 101 paired-end (PE) sequencing technologies are compared in the figure. Arrows emanating from the schematic 16S gene represent common forward (F) and reverse (R) primers, and the orange boxes denote the hypervariable regions (V1–9), which are known to be far less conserved than the surrounding sequence. Technologies differ in the maximum allowable amplicon size and read length (TABLE 1) and therefore result in a different view of the community. To increase overall length and/or quality, Sanger- and Illumina-based strategies involve sequencing amplicons in both directions; in Sanger sequencing, there is also the option of using a third read. By contrast, a preferred 454-based strategy sequences the amplicons in a single direction owing to the lack of standard paired-end sequencing and loss of pairing information. Getting the most taxonomic information requires careful selection of primers, 16S rDNA windows and technologies in order to obtain the most data27,47. The long length of Sanger reads diminishes the need for careful selection of amplicon primer pairs, which have been shown to have a large role in taxonomic assignment and community comparison results with 454 data. A variety of options exist, and studies such as the one described in REF. 47 provide suggestions on which sets to choose. Short read effects are exacerbated further in Illumina sequencing, and choosing amplicon size such that overlapping paired-end reads occur is an important consideration. Table 1 Comparison of sequencing technologies Read length Maximum insert size Run time (hours (h) or days (d)) Reads per run Relative cost factor (per Mb) Scale of reads per sample Scale of samples per run Raw error rate (%) Total Insertions Deletions Mismatches ABI 3730 800 b >1 Kb 2 h 96 100 102 101 0.001 ≪0.1 ≪0.1 ≪0.1 454 FLX Titanium 300–400 b 800 b 9 h 106 1 103 102 1 < 1 ≪ 0.1 ≪ 1 454 FLX+ 500–600 b 1200 b 23 h 106 0.7 103 102 Illumina GAIIx 76–101 b 500 b 6–9 d 4 × 108 0.1 105–106 103–104 <1 ≪1 ≪1 <1 Illumina HiSeq 2000 101–151 b 500 b 9–15 d 3 × 109 0.002 105–106 103–104 Illumina MiSeq 36–151 b 500 b 4h–27 h 107 0.06 104 102 PacBio 1100 b >1 Kb 1.5 h 3.5 × 107 1.5 103 101 15 13 1 1 IonTorrent 200 b 400 b 2–3 h 1.5 × 106–3 × 106 0.4 103 102 2 1 1 <1 The table shows a comparison of characteristics and average error rates and types of various sequencing technologies, both those that are commonly used in microbiome studies and promising but less-established technologies (MiSeq, PacBio, IonTorrent). The estimated relative cost factor presented here (relative to 454 Titanium sequencing) includes only sequencing costs and not the many other time and money costs that are involved in microbiome studies. Note that values are based on capabilities at the time of submission and that the next-generation sequencing technologies are likely to change quickly. Competing interests statement The authors declare no competing financial interests. 1 Ravel J Vaginal microbiome of reproductive-age women Proc Natl Acad Sci USA 108 Suppl 1 4680 4687 2011 20534435 2 Ley RE Obesity alters gut microbial ecology Proc Natl Acad Sci USA 102 11070 11075 2005 16033867 3 Aas J Gessert CE Bakken JS Recurrent Clostridium difficile colitis: case series involving 18 patients treated with donor stool administered via a nasogastric tube Clin Infect Dis 36 580 585 2003 12594638 4 Sartor RB Microbial influences in inflammatory bowel diseases Gastroenterology 134 577 594 2008 18242222 5 Kinross JM Darzi AW Nicholson JK Gut microbiome–host interactions in health and disease Genome Med 3 14 2011 21392406 6 Peterson J The NIH Human Microbiome Project Genome Res 19 2317 2323 2009 19819907 7 Blaser MJ Harnessing the power of the human microbiome Proc Natl Acad Sci USA 107 6125 6126 2010 20360554 8 Qin J A human gut microbial gene catalogue established by metagenomic sequencing Nature 464 59 65 2010 This is a large-scale study aimed at characterizing the functionality encoded in the gut microbiome. This work defined a minimal set of functions that are present in all of the sampled individuals 20203603 9 Costello EK Bacterial community variation in human body habitats across space and time Science 326 1694 1697 2009 This paper was the first to establish that the microbial communities harboured across the human body are personalized but vary substantially across body sites and over time 19892944 10 Turnbaugh PJ A core gut microbiome in obese and lean twins Nature 457 480 484 2009 19043404 11 Caporaso JG Moving pictures of the human microbiome Genome Biol 12 R50 2011 This is the densest time-series analysis of variation in the human microbiota that has been carried out so far. This study also proved the usefulness of newer DNA sequencers to provide deeper insights into the microbiota by recapturing previous results in variability across body sites and time using a different sequencing technology 21624126 12 Shi Y Tyson GW DeLong EF Metatranscriptomics reveals unique microbial small RNAs in the ocean’s water column Nature 459 266 269 2009 19444216 13 Maron PA Ranjard L Mougel C Lemanceau P Metaproteomics: a new approach for studying functional microbial ecology Microb Ecol 53 486 493 2007 17431707 14 Clayton TA Baker D Lindon JC Everett JR Nicholson JK Pharmacometabonomic identification of a significant host-microbiome metabolic interaction affecting human drug metabolism Proc Natl Acad Sci USA 106 14728 14733 2009 The authors of this paper suggest a link between a person’s microbiome and their ability to metabolize a common drug, paracetamol (acetaminophen) 19667173 15 DeSantis TZ Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB Appl Environ Microbiol 72 5069 5072 2006 16820507 16 Pruesse E SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB Nucleic Acids Res 35 7188 7196 2007 17947321 17 Cole JR The Ribosomal Database Project: improved alignments and new tools for rRNA analysis Nucleic Acids Res 37 D141 D145 2009 19004872 18 Bellemain E ITS as an environmental DNA barcode for fungi: an in silico approach reveals potential PCR biases BMC Microbiol 10 189 2010 20618939 19 Hayashi H Sakamoto M Benno Y Evaluation of three different forward primers by terminal restriction fragment length polymorphism analysis for determination of fecal Bifidobacterium spp. in healthy subjects Microbiol Immunol 48 1 6 2004 14734852 20 Bergmann GT The under-recognized dominance of Verrucomicrobia in soil bacterial communities Soil Biol Biochem 43 1450 1455 2011 22267877 21 Liu Z DeSantis TZ Andersen GL Knight R Accurate taxonomy assignments from 16S rRNA sequences produced by highly parallel pyrosequencers Nucleic Acids Res 36 e120 2008 18723574 22 Walters WA PrimerProspector: de novo design and taxonomic analysis of barcoded polymerase chain reaction primers Bioinformatics 27 1159 1161 2011 21349862 23 Marchesi JR Prokaryotic and eukaryotic diversity of the human gut Adv Appl Microbiol 72 43 62 2010 20602987 24 Parfrey LW Walters WA Knight R Microbial eukaryotes in the human microbiome: ecology, evolution, and future directions Front Microbiol 2 153 2011 21808637 25 Ott SJ Fungi and inflammatory bowel diseases: alterations of composition and diversity Scand J Gastroenterol 43 831 841 2008 18584522 26 Ghannoum MA Characterization of the oral fungal microbiome (mycobiome) in healthy individuals PLoS Pathog 6 e1000713 2010 20072605 27 Vestheim H Jarman SN Blocking primers to enhance PCR amplification of rare sequences in mixed samples — a case study on prey DNA in Antarctic krill stomachs Front Zool 5 12 2008 18638418 28 Haynes M Rohwer F Metagenomics of the Human Body Nelson KE 63 77 Springer New York 2011 29 Virgin HW Wherry EJ Ahmed R Redefining chronic viral infection Cell 138 30 50 2009 19596234 30 Breitbart M Viral diversity and dynamics in an infant gut Res Microbiol 159 367 373 2008 18541415 31 Reyes A Viruses in the faecal microbiota of monozygotic twins and their mothers Nature 466 334 338 2010 20631792 32 Palmer C Bik EM DiGiulio DB Relman DA Brown PO Development of the human infant intestinal microbiota PLoS Biol 5 e177 2007 17594176 33 Koenig JE Succession of microbial consortia in the developing infant gut microbiome Proc Natl Acad Sci USA 108 Suppl 1 4578 4585 2011 This paper describes a two-year longitudinal study of the development of the gut microbiota in an infant. This work provides a detailed analysis of the relationship between life events and changes in microbiome composition and function 20668239 34 Oliver KM Degnan PH Hunter MS Moran NA Bacteriophages encode factors required for protection in a symbiotic mutualism Science 325 992 994 2009 19696350 35 Caporaso JG Knight R Kelley ST Host-associated and free-living phage communities differ profoundly in phylogenetic composition PLoS ONE 6 e16900 2011 21383980 36 Willner D Metagenomic analysis of respiratory tract DNA viral communities in cystic fibrosis and non-cystic fibrosis individuals PLoS ONE 4 e7370 2009 19816605 37 McOrist AL Jackson M Bird AR A comparison of five methods for extraction of bacterial DNA from human faecal samples J Microbiol Methods 50 131 139 2002 11997164 38 Wang RF Beggs ML Erickson BD Cerniglia CE DNA microarray analysis of predominant human intestinal bacteria in fecal samples Mol Cell probes 18 223 234 2004 15271382 39 Wu GD Linking long-term dietary patterns with gut microbial enterotypes Science 334 105 108 2011 21885731 40 Turnbaugh PJ Bäckhed F Fulton L Gordon JI Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome Cell Host Microbe 3 213 223 2008 18407065 41 Fierer N Hamady M Lauber CL Knight R The influence of sex, handedness, and washing on the diversity of hand surface bacteria Proc Natl Acad Sci USA 105 17994 17999 2008 19004758 42 Wu GD Sampling and pyrosequencing methods for characterizing bacterial communities in the human gut using 16S sequence tags BMC Microbiol 10 206 206 2010 This study shows that long-term dietary patterns are associated with particular enterotypes. Bacteroides spp. were associated with a Western-like diet that is rich in proteins and animal fats, whereas Prevotella spp. were linked with high-carbohydrate diets 20673359 43 Lauber CL Zhou N Gordon JI Knight R Fierer N Effect of storage conditions on the assessment of bacterial community structure in soil and human-associated samples FEMS Microbiol Lett 307 80 86 2010 20412303 44 Amann RI Ludwig W Schleifer KH Phylogenetic identification and in situ detection of individual microbial cells without cultivation Microbiol Rev 59 143 169 1995 7535888 45 Wang Q Garrity GM Tiedje JM Cole JR Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy Appl Environ Microbiol 73 5261 5267 2007 17586664 46 Segata N Metagenomic biomarker discovery and explanation Genome Biol 12 R60 2011 21702898 47 Liu Z Lozupone C Hamady M Bushman FD Knight R Short pyrosequencing reads suffice for accurate microbial community analysis Nucleic Acids Res 35 e120 2007 17881377 48 Zhou HW BIPES, a cost-effective high-throughput method for assessing microbial diversity ISME J 5 741 749 2011 20962877 49 Hummelen R Deep sequencing of the vaginal microbiota of women with HIV PLoS ONE 5 e12078 2010 20711427 50 Lazarevic V Metagenomic study of the oral microbiota by Illumina high-throughput sequencing J Microbiol Methods 79 266 271 2009 19796657 51 Gloor GB Microbiome profiling by illumina sequencing of combinatorial sequence-tagged PCR products PLoS ONE 5 e15406 2010 21048977 52 Caporaso JG Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample Proc Natl Acad Sci USA 108 Suppl 1 4516 4522 2011 20534432 53 Bartram AK Lynch MD Stearns JC Moreno-Hagelsieb G Neufeld JD Generation of multimillion-sequence 16S rRNA gene libraries from complex microbial communities by assembling paired-end illumina reads Appl Environ Microbiol 77 3846 3852 2011 21460107 54 Claesson MJ Comparison of two next-generation sequencing technologies for resolving highly complex microbiota composition using tandem variable 16S rRNA gene regions Nucleic Acids Res 38 e200 2010 20880993 55 Gilbert JA Dupont CL Microbial metagenomics: beyond the genome Ann Rev Mar Sci 3 347 371 2011 56 Tyson GW Community structure and metabolism through reconstruction of microbial genomes from the environment Nature 428 37 43 2004 14961025 57 Rodrigue S Unlocking short read sequencing for metagenomics PLoS ONE 5 e11840 2010 20676378 58 Goldberg SMD A Sanger/pyrosequencing hybrid approach for the generation of high-quality draft assemblies of marine microbial genomes Proc Natl Acad Sci USA 103 11240 11245 2006 16840556 59 Gnerre S High-quality draft assemblies of mammalian genomes from massively parallel sequence data Proc Natl Acad Sci USA 108 1513 1518 2011 21187386 60 Huse SM Huber JA Morrison HG Sogin ML Welch DM Accuracy and quality of massively parallel DNA pyrosequencing Genome Biol 8 R143 2007 17659080 61 Kunin V Engelbrektson A Ochman H Hugenholtz P Wrinkles in the rare biosphere: pyrosequencing errors can lead to artificial inflation of diversity estimates Environ Microbiol 12 118 123 2010 19725865 62 Schloss PD Gevers D Westcott SL Reducing the effects of PCR and sequencing artifacts on 16S rRNA-based studies PLoS ONE in the press 63 Haas BJ Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons Genome Res 21 494 504 2011 21212162 64 Quince C Lanzen A Davenport RJ Turnbaugh PJ Removing noise from pyrosequenced amplicons BMC bioinformatics 12 38 2011 21276213 65 Edgar RC Haas BJ Clemente JC Quince C Knight R UCHIME improves sensitivity and speed of chimera detection Bioinformatics 27 2194 2200 2011 21700674 66 Schloss PD Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities Appl Environ Microbiol 75 7537 7541 2009 19801464 67 Caporaso JG QIIME allows analysis of high-throughput community sequencing data Nature Methods 7 335 336 2010 This paper introduces QIIME, an open-source software tool that performs the complete analysis of microbial communities. Among other functions, QIIME implements quality filtering of the input raw reads, OTU picking, α- and β-diversity estimates and prediction of OTUs that are significantly associated with categories in the data 20383131 68 Reeder J Knight R Rapidly denoising pyrosequencing amplicon reads by exploiting rank-abundance distributions Nature Methods 7 668 669 2010 20805793 69 Rappe MS Giovannoni SJ The uncultured microbial majority Annu Rev Microbiol 57 369 394 2003 14527284 70 Schloss PD The effects of alignment quality, distance calculation method, sequence filtering, and region on the analysis of 16S rRNA gene-based studies PLoS Comput Biol 6 e1000844 2010 20628621 71 Meyer F The metagenomics RAST server — a public resource for the automatic phylogenetic and functional analysis of metagenomes BMC Bioinformatics 9 386 2008 18803844 72 Yilmaz P Minimum information about a marker gene sequence (MIMARKS) and minimum information about any (x) sequence (MIxS) specifications Nature Biotech 29 415 420 2011 73 Turnbaugh PJ The human microbiome project Nature 449 804 810 2007 17943116 74 Gilbert JA The Earth Microbiome Project: meeting report of the “1 EMP meeting on sample selection and acquisition” at Argonne National Laboratory October 6 2010 Stand Genomic Sci 3 249 253 2010 21304728 75 Caporaso JG PyNAST: a flexible tool for aligning sequences to a template alignment Bioinformatics 26 266 267 2010 19914921 76 DeSantis TZ Jr NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes Nucleic Acids Res 34 W394 W399 2006 16845035 77 Price MN Dehal PS Arkin AP FastTree 2—approximately maximum-likelihood trees for large alignments PLoS ONE 5 e9490 2010 20224823 78 Lozupone C Knight R UniFrac: a new phylogenetic method for comparing microbial communities Appl Environ Microbiol 71 8228 8235 2005 This study introduces UniFrac, a phylogenetically aware measure of similarity, and one of the most widely used methods to establish the extent to which different microbial communities resemble each other 16332807 79 Faith DP Baker AM Phylogenetic diversity (PD) and biodiversity conservation: some bioinformatics challenges Evolutionary Bioinform Online 2 121 128 2006 80 Morowitz MJ Strain-resolved community genomic analysis of gut microbial colonization in a premature infant Proc Natl Acad Sci USA 108 1128 1133 2011 21191099 81 Tringe SG Comparative metagenomics of microbial communities Science 308 554 557 2005 15845853 82 Arumugam M Enterotypes of the human gut microbiome Nature 473 174 180 2011 In this study, faecal microbiomes were found to cluster into three distinct groups (‘enterotypes’) with minimal overlap 21508958 83 Muegge BD Diet drives convergence in gut microbiome functions across mammalian phylogeny and within humans Science 332 970 974 2011 21596990 84 Brady A Salzberg S PhymmBL expanded: confidence scores, custom databases, parallelization and more Nature Methods 8 367 2011 21527926 85 Mitra S Functional analysis of metagenomes and metatranscriptomes using SEED and KEGG BMC Bioinformatics 12 S21 2011 86 Sharpton TJ PhylOTU: a high-throughput procedure quantifies microbial community diversity and resolves novel taxa from metagenomic data PLoS Comput Biol 7 e1001061 2011 21283775 87 von Mering C Quantitative phylogenetic assessment of microbial communities in diverse environments Science 315 1126 1130 2007 17272687 88 Muller J eggNOG v2.0: extending the evolutionary genealogy of genes with enhanced non-supervised orthologous groups, species and functional annotations Nucleic Acids Res 38 D190 D195 2010 19900971 89 Kanehisa M Goto S Furumichi M Tanabe M Hirakawa M KEGG for representation and analysis of molecular networks involving diseases and drugs Nucleic Acids Res 38 D355 D360 2010 19880382 90 Finn RD The Pfam protein families database Nucleic Acids Res 36 D281 D288 2008 18039703 91 Wooley JC Godzik A Friedberg I A primer on metagenomics PLoS Comput Biol 6 e1000667 2010 20195499 92 Glass E Meeting report from the Genomic Standards Consortium (GSC) Workshop 10 Stand Genom Sci 3 225 231 2010 93 Arumugam M Harrington ED Foerstner KU Raes J Bork P SmashCommunity: a metagenomic annotation and analysis tool Bioinformatics 26 2977 2978 2010 20959381 94 Sun S Community Cyberinfrastructure for Advanced Microbial Ecology Research and Analysis: the CAMERA resource Nucleic Acids Res 39 D546 D551 2011 21045053 95 Markowitz VM IMG/M: a data management and analysis system for metagenomes Nucleic Acids Res 36 D534 D538 2008 17932063 96 Kristiansson E Hugenholtz P Dalevi D ShotgunFunctionalizeR: an R-package for functional comparison of metagenomes Bioinformatics 25 2737 2738 2009 19696045 97 Liu B Pop M MetaPath: identifying differentially abundant metabolic pathways in metagenomic datasets BMC Proc 5 S9 2011 98 Chen K Pachter L Bioinformatics for whole-genome shotgun sequencing of microbial communities PLoS Comput Biol 1 106 112 2005 16110337 99 Kuczynski J Microbial community resemblance methods differ in their ability to detect biologically relevant patterns Nature Methods 7 813 819 2010 20818378 100 Quince C Curtis TP Sloan WT The rational exploration of microbial diversity ISME J 2 997 1006 2008 18650928 101 Mcpeek MA Mcpeek MA The consequences of changing the top predator in a food web: a comparative experimental approach Ecol Monogr 68 1 23 1998 102 Khoruts A Dicksved J Jansson JK Sadowsky MJ Changes in the composition of the human fecal microbiome after bacteriotherapy for recurrent Clostridium difficile-associated diarrhea J Clin Gastroenterol 44 354 360 2010 20048681 103 West TE Toll-like receptor 4 region genetic variants are associated with susceptibility to melioidosis Genes Immun 2011 1 9 2011 104 Turnbaugh PJ The effect of diet on the human gut microbiome: a metagenomic analysis in humanized gnotobiotic mice Sci Transl Med 1 6ra14 2009 105 Goodman AL Extensive personal human gut microbiota culture collections characterized and manipulated in gnotobiotic mice Proc Natl Acad Sci USA 108 6252 6257 2011 This paper showed that a substantial proportion of an individual’s gut microbiota can be recaptured using anaerobic culturing conditions, both in vitro and in vivo 21436049 106 Paulino LC Tseng CH Strober BE Blaser MJ Molecular analysis of fungal microbiota in samples from healthy human skin and psoriatic lesions J Clin Microbiol 44 2933 2941 2006 16891514 107 Gao Z Tseng CH Pei Z Blaser MJ Molecular analysis of human forearm superficial skin bacterial biota Proc Natl Acad Sci USA 104 2927 2932 2007 17293459 108 Grice EA A diversity profile of the human skin microbiota Genome Res 18 1043 1050 2008 18502944 109 Zoetendal EG Mucosa-associated bacteria in the human gastrointestinal tract are uniformly distributed along the colon and differ from the community recovered from feces Appl Environ Microbiol 68 3401 3407 2002 12089021 110 Brotman RM Ravel J Cone RA Zenilman JM Rapid fluctuation of the vaginal microbiota measured by Gram stain analysis Sex Transm Infect 86 297 302 2010 20660593

PMC005xxxxxx/PMC5119551.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 9710002 22013 Clin Liver Dis Clin Liver Dis Clinics in liver disease 1089-3261 1557-8224 27742011 5119551 10.1016/j.cld.2016.07.001 NIHMS814005 Article Towards Elimination of Hepatitis B Virus Using Novel Drugs, Approaches, and Combined Modalities Sebastien Boucle PhD Leda Bassit PhD Maryam Ehteshami PhD Schinazi Raymond F. PhD, DSc * Center for AIDS Research, Laboratory of Biochemical Pharmacology, Department of Pediatrics, Emory University School of Medicine, Atlanta, GA 30322, USA * Corresponding author. Tel.: +1-404-727-1414; fax: +1-404-727-1330, rschina@emory.edu (RF Schinazi) 10 9 2016 30 8 2016 11 2016 01 11 2017 20 4 737749 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary With nearly 30% of the world population infected, hepatitis B virus (HBV) causes significant morbidity and mortality worldwide. Although several potent antiviral agents are currently in use against HBV infection, the majority of chronically infected individuals do not achieve a functional and complete cure, as measured by the clearance of HB surface antigen (HBsAg) from blood and eradication of the covalently closed-circular DNA (cccDNA) from the nuclei of hepatocytes. In addition, even treated persons who achieve a long-term (> 10-15 years) sustained virological response (undetectable HBV DNA), are still at high risk of developing morbidity and mortality from liver complications. This review focuses upon novel, mechanistically diverse anti-HBV therapeutic strategies that are currently in development or in clinical evaluation, and highlights new combination strategies which may contribute to full elimination of HBV DNA and cccDNA from the infected liver, leading to a complete cure of chronic hepatitis B. HBV Cure siRNA CRISPR/Cas9 Anti-HBV agents cccDNA HBsAg Antiviral therapy Introduction Over the past decades research efforts have led to the development of several potent nucleosides analog inhibitors (NA) such as lamivudine (Epivir), adefovir dipivoxil (Hepsera), entecavir (Baraclude), telbivudine (Tyzeka), and tenofovir disoproxil fumarate (Viread) allowing a large decrease of HBV viremia in chronically infected persons.1 NAs in their 5’-triphosphate form are potent inhibitors of DNA polymerase/reverse transcriptase activities of the viral polymerase enzyme. They compete with natural nucleotides and act on several steps of viral DNA synthesis including initial polymerization, protein priming or the subsequent DNA strand elongation. It has been suggested that combination therapy using one of these nucleoside analogs and interferon could have better virus elimination efficacy than NA monotherapy, but such studies are difficult to perform since the current monotherapy is already very effective at controlling HBV viral load. However, current NA treatment does not lead to HBV cure as indicated by low levels of HBsAg seroconversion (Box 1).2,3 Despite the success of current available therapy, subjects who cleared the virus (HBeAg negative, HBsAg negative) can experience reactivation of HBV on treatment interruption or following the use of anti-inflammatory or immunosuppressant medications.4 This strongly suggests that current anti-HBV therapeutics are unable to eradicate the virus from infected liver cells. These limitations have led researchers to continue their drug development efforts toward finding new viral targets that could potentially lead to the discovery of a functional or absolute cure (Box 1).5 Considering that HBV covalently closed circular DNA (cccDNA) serves as the template for pregenomic RNA (pgRNA) transcription, it is thought to be responsible for virus persistence. At the same time, integration of HBV DNA is thought to be associated with increase risk of hepatocellular carcinoma (HCC) development.6 Accordingly, new therapeutic approaches that target cccDNA directly or indirectly are in development (Fig. 1). Following the recent successes with HCV drug development, the field of viral hepatitis is turning its focus to another major threat to liver health, namely HBV.7 These new therapeutic strategies will have to address the problems of cccDNA elimination, intrahepatic innate immune response stimulation, HBV specific immune response restoration and will probably have to include combination of drugs to target multiple steps of the HBV replication cycle. New antivirals currently in the pipelines include entry inhibitors, relaxed circular (rc)DNA-cccDNA conversion inhibitors and capsid assembly effectors (Fig. 1). Besides these direct acting agents, there has also been a significant development in the area of host-targeting agents (HTA). Examples include small interfering RNA-based strategies (SiRNA), RNAi silencers, CRISPR/Cas9 approaches, HBsAg inhibitors, immunomodulators, therapeutic vaccines and toll like receptor (TLR) agonists.8,9 Using these new investigational approaches, it is hoped that a functional cure for chronic HBV is achieved within the next decade. This review highlights recent progress in developing novel anti-HBV drugs and their mechanisms of action. Current treatments and limitations for a cure Nucleoside analogs are relatively potent inhibitors used for the treatment of chronic hepatitis B. These HBV reverse transcriptase inhibitors are usually well tolerated and have excellent bioavailability. They are also cost-effective in comparison to interferon treatments such as peginterferon alfa-2a (Pegasys) & interferon alfa-2b (Intron A). Nevertheless, these drugs have some limitations in terms of HBsAg clearance and cccDNA suppression. Although drug resistance can occur clinically with some of the earlier oral treatment such as Epivir, drug resistant viruses are rarely selected with the more recent drugs such as Baraclude and Viread. It has been thought that combination of NAs could have an additive and synergistic antiviral effect and could reduce the rate of drug resistance. However, combination studies involving two nucleoside analogs did not increase virologic response, since the drugs are already very potent on their own.10 As a result, because pegylated- interferon (PEG-IFN) has a different mechanism of action than NA, their combination (TDF + PEG-IFN for 48 wk) showed a greater viral suppression and higher rates of HBsAg loss.11,12 As new therapeutic strategies are being developed for the treatment of chronic hepatitis B (CHB), uncovering new inhibitory mechanisms and potential targets, it is very likely that NA will have their place in future combinations with future drug candidates. Viral entry inhibitors The HBV viral replication cycle consists of a complex multistep mechanism (Fig. 1), starting with the virus entering the hepatocyte, followed by DNA replication, nucleocapsid formation and release of virions. HBV entry represents an essential step for spreading and maintaining virus replication. The process involves two major interactions between the viral envelope protein Pre-S1 and hepatocyte cellular receptors including first, HBV binding to the glycoproteins heparin sulfate proteoglycans followed by its interaction with the sodium taurocholate co-transporting polypeptide (NTCP). Recently, Hepatera developed a synthetic lipopeptide, called Myrcludex-B, which is derived from the HBV L-protein.13 Studies have shown that the peptide competes with the viral Pre-S1 motif for NTCP binding, blocking de novo HBV infection. Because of its early effect on the HBV replication cycle, according to the authors, the drug may also efficiently block the amplification of the HBV cccDNA. With this new concept, this inhibitor, which is currently in phase II clinical trials, could have a role in the development of an HBV cure regimen (Table 1).14 Although, there are several other HBV entry inhibitors that can block the in vitro interaction of HBV with NTCP such as Cyclosporin A, Ritonavir, Ezetimibe, Vanitaracin A, Irbesartan, among others; these inhibitors alone can not lead to a complete inhibition of cccDNA synthesis as observed with Myrcludex-B. However, they might still play an important role by preventing viral entry into cccDNA free hepatocytes when combined with other antiviral therapies.15 Therapies targeting cccDNA cccDNA formation inhibitor HBV has evolved a unique replication cycle that results in the production of large viral loads during active replication without actually killing the infected cell directly. Two of the key events in the viral replication cycle of HBV involve first, the generation of cccDNA transcriptional template, either from input genomic DNA or newly replicated capsid-associated DNA and second, reverse transcription of the viral pgRNA to form progeny HBV DNA genomes.16,17 The HBV cccDNA is associated with viral persistence in HBV-infected hepatocytes.18,19 Hepatocytes have a long half-life (> 6 months or even years); therefore, elimination of cccDNA by hepatocyte turnover is not a major means of clearance. The major limitation of current treatment is the failure to eliminate the preexisting cccDNA pool and/or prevent cccDNA formation from trace-levels of wild-type or drug-resistant virus.20 As a consequence, HBV commonly rebounds after cessation of treatment with NA, leading different groups to develop assays to screen libraries of compounds in order to discover new antiviral candidates that can inhibit cccDNA formation.20 Doing so, disubstituted sulfonamides (DSS), such as CCC-0975 and CCC-0346 have been identified as inhibitors of cccDNA production.21 These molecules are believed to interfere with rcDNA conversion to cccDNA in HepDES19 cells, also inhibiting de novo cccDNA formation. Further development of these DSS in combination with other antivirals such as NA might lead to the elimination of HBV cccDNA. cccDNA targeted endonuclease New promising systems that specifically use sequence-specific endonucleases to cleave cccDNA and eradicate it from infected hepatocytes have been developed, including the programmable RNA-guided DNA endonucleases (CRISPR/Cas9), transcription activator-like effector nuclease (TALEN) or zinc-finger nuclease. Promising studies in cell and mouse models with CRIPSR/Cas9 have shown that these systems have the potential to serve as effective tools for the depletion of the cccDNA pool in chronically HBV infected subjects.22,23,24 CRIPSR/Cas9 specifically reduced total viral DNA levels by up to ~1,000-fold and HBV cccDNA levels by up to ~10-fold, in addition, it also mutationally inactivated the majority of the residual viral DNA in the stably transfected HepAD38 system. Moreover, these Spy Cas9/sgRNA systems showed additive inhibition of HBV DNA accumulation when used in combination with known pharmacological inhibitors of the HBV RT enzyme in the Hep2.2.15 cells, and in the infected HepaRG cells, reduced both viral production and up to 67% cccDNA formation.23 In a HBV hydrodynamics-mouse model, CRISPR/Cas9 system was capable of disrupting the intrahepatic HBV genome (~28%), with significant reduction but not complete elimination of HBsAg.24 siRNA approach Persistence of chronic HBV infection is markedly demonstrated by an absence of antiviral immune response against the virus. As a result, a continuous production of surface antigen (HBsAg) in the plasma of chronically infected individuals is observed.25 Three forms of HBsAg are secreted from infected hepatocytes, comprising of filaments and spherical particles, with or without virion. The empty, non-infectious particles are the most abundant in the plasma, and may play a role in preventing the immune system from building a specific immune response against HBV. One way to stop secretion of HBsAg from infected hepatocytes is to cease transcription of messenger RNA (mRNA) by using small interfering RNA (siRNA). These short sequences of nucleotides (siRNA) knock- down expression of genes of interest by promoting gene silencing at the posttranscriptional level. Several siRNA-based regimens are currently being developed and evaluated. The promising ARC-520 from Arrowhead Pharmaceuticals is in a phase II/III clinical studies (Table 1). This new molecule is comprised of two distinct siRNA sequences, which was designed to reduce all transcripts of HBV cccDNA and with wide genotype coverage of the HBV genome. To enhance delivery to hepatocytes, ARC-520 was conjugated with cholesterol and then coinjected with a hepatocyte-targeted membrane-active peptide. In chimpanzees, ARC-520 treatment resulted in a remarkable 95% decline in HBV DNA levels and, as high as 90% inhibition of secreted HBeAg and HBsAg.26,27 Similar results were demonstrated in HBeAg-positive patients, however, insignificant suppression of HBsAg was observed in HBeAg-negative chimpanzees or patients, supporting the hypothesis that HBsAg in this case was produced from integrated DNA which is not targeted by ARC-520. TKM-HBV/ARB-001467 developed by Arbutus Biopharma is another siRNA regimen in a phase II clinical trial (Table 1), which is currently being evaluated for its safety and tolerability in HBeAg-negative or –positive subjects receiving nucleoside analog therapy. This molecule targets three conserved regions within the HBV genome and appears to clear HBsAg expression from both cccDNA and integrated HBV. Lipid nanoparticles are utilized to transport it to the hepatocytes giving it more stability against nucleases. siRNA-based approaches for HBV are especially beneficial for the fact that the HBV viral RNA transcripts have their sequences overlapped. This facilitates the synthesis of one single siRNA trigger that could degrade all viral transcripts simultaneously and prevent viral proteins secretion. However, there are three main drawbacks with regard to siRNA approach for HBV therapeutics: (i) Specific delivery to hepatocytes in vivo: because of their small size and highly negatively charged hydrophilic phosphate backbone, siRNA are rapidly filtrated by the kidney and are cleared from the blood stream before achieving their target. (ii) siRNA that reach the cell membrane of hepatocytes can be easily trapped in the endosome and undergo degradation by nucleolytic enzymes. (iii) Undesirable off-target effects of siRNA and innate system stimulation are also a concern. Despite the aforementioned obstacles, novel chemical modifications seem to minimize the chance of cross-reactivity with human mRNAs to occur. These approaches can also enhance efficient delivery of siRNA to the cytoplasm where they can react with RNA-induced silencing complex (RISC) and prompt specific degradation of the HBV mRNAs.27 More recently, Benitec Biopharma developed BB-HB-331 based on a similar approach pertaining to DNA-directed RNA interfering strategy (ddRNAi).28 BB-HB-331 is a recombinant DNA construct, capable of continuously expressing short hairpin RNA (shRNA), which in turn can permanently silence the targeted viral messenger RNA expression with a single treatment. They revealed the results of an in vivo study conducted in humanized mouse PhoenixBio (PXB), showing a 98.5% elimination of circulating HBV (reduced serum HBV DNA by 1.83 logs), a 94.5% reduction of intracellular liver HBV DNA, and almost complete suppression of serum antigens HBeAg and HBsAg (92.6% and 97.6%), and reduction of HBV viral RNA and cccDNA levels. Capsid assembly and core protein effectors The HBV nucleocapsid is well recognized to have an important role in the viral replication cycle. It is believed to play an essential role in HBV genome packaging, reverse transcription, intracellular trafficking and maintenance of chronic infection.29 Several small molecules including heteroarylpyrimidines (HAP) have been shown to target the capsid protein (Cp) homo-dimers that rearrange to form the nucleocapsid. They have been identified to disrupt the capsid assembly, thus leading to inhibition of HBV replication both in vitro and in vivo.30,31 BAY 41-4109 (AiCuris) was the first HAP to be developed and reached phase I, but because of toxicity, solubility and other issues, it seems to have been abandoned.8 Despite hepatotoxicity in rats at high dosage,32 it was shown to inhibit the virus replication in HBV transgenic mouse,33 and more importantly effectiveness against lamivudine (Epivir) and adefovir dipivoxil (Hepsera) resistant viruses.34,33 Based on these results, HEC Pharm developed more recently another heteroarylpyrimidine named morphothiadine mesilate GLS4 which entered a phase II clinical trial in China.35 Early studies have demonstrated that this new HAP was more potent and significantly less toxic than analog BAY41-4109.36 GLS4 was found to misdirect capsid assembly leading to the formation of aberrant capsids without primarily affecting core protein levels.37 As these molecules may also have an impact on cccDNA stability, it is suggested that they may contribute to discovery of an HBV cure.38 Another class of small molecules known as sulfamoylbenzamides has been identified to interfere with the capsid, and potently inhibit the formation of pgRNA-containing capsids.39 NVR 3-778 is a sulfamoylbenzamide compound having a pangenotypic antiviral activity, developed by Novira (later acquired by Johnson & Johnson) that recently reached human phase IIa producing significant virus loads reduction (1.7 log reduction of serum HBV DNA and 0.86 log for HBV RNA, at 600 mg BID for 41 days). NVR 3-778 has shown encouraging pharmacokinetic properties, and was well tolerated in human volunteers.40 It has also been shown to inhibit the production of HBV DNA and RNA particles, especially in combination with PEG-IFN. As their mechanism of action is still not completely clear, this new class of small molecules represent a promising cohort of molecules with curative potential when combined with other small molecule inhibitors. Toll-like receptor (TLR-7) Toll-like receptors (TLR) agonists have antiviral effects. TLR-7 agonist activates the innate immunity by stimulating plasmacytoid dendritic cells to produce IFN-alpha and other cytokines/chemokines and induce the activation of killer cells as well as cytotoxic lymphocytes. Therefore, this new approach with agonist-induced activation of TLR7 can trigger both innate and adaptive immune responses and may represent a new strategy to treat chronic viral infections. GS-9620 (Gilead) is a small molecule, which has agonist activity. It binds to TLR7 leading to subsequent activation of several transcription factors, including nuclear factor κB (NF-κB) and interferon regulatory factors. GS-9620 has recently entered phase II clinical trials in combination with Tenofovir versus Tenofovir monotherapy.41,42 Other therapeutics with potential Caspase activators, RIG 1 activators, cyclophilin inhibitors, RNase H inhibitors and therapeutic vaccines are also being evaluated (Table 1.). Some of these strategies such as therapeutic vaccines seem very promising, but are still in development and will have to overcome any possible toxicity and problems related to immune-enhancing approaches variable in treated subjects.9 An impressive reduction of HBsAg has been demonstrated with the novel nucleic acid polymer Rep 2139-Ca (Replicor) alone or in combination with pegylated interferon alpha 2a in subjects chronically infected with HBV or co-infected with hepatitis delta virus (HDV). 43 This compound is in a phase II clinical trial and has the ability to block the formation of surface antigen protein by inhibiting the interaction of apolipoproteins with these subviral particles.44 Recently, a new in vitro approach was developed to facilitate the direct interaction of small molecules with the human HBV polymerase. With a large-scale production of this enzyme coupled with its structural and biophysical characterizations,45 Voros et al., validated their new system using a small molecule – metal-dependent and -binding modulator of HBV polymerase, calcomine orange 2R – which inhibits not only the duck HBV polymerase, but also human HBV polymerase. It remains to be determined whether this drug would interact synergistically with NAs that also target the viral polymerase. Another approach targeting microRNA could also have a role toward an HBV cure. MicroRNA-122 (miR-122) is a non-coding RNA involved in liver development and hepatic function, which has also been found to play a role in the regulation of HBV replication. It has been shown that miR-122 plays a role in viral persistence, as a decrease of miR-122 is correlated with enhancement of HBV replication through a cyclin G1-P53 dependent pathway. Based on these observations, Li et al. found that all four HBV mRNAs were harboring an miR-122 complementary site, revealing a novel mechanism by which viral mRNAs mediate host miRNA activity, contributing to the regulation of liver cancer cell proliferation, invasion and tumor growth.46 Moreover, recent studies have shown that transfection of miR-122 expression vector into HepG2.2.15 cells repressed the transcription and expression of the protein N-myc downstream-regulated gene 3 (NDRG3), contributing to HBV-related hepatocarcinogenesis.47 Thus, given the broad interactions of miR-122 in HBV chronic infection and HBV-related hepatocarcinomas, this microRNA represents a target for the development of new anti-HBV therapies. Conclusion Compared to the current available therapies that decrease and suppress the HBV viral DNA levels to undetectable levels, the new investigational drugs and approaches described herein have the potential to decrease or eliminate cccDNA and/or HBsAg. It is believed that combinations of antiviral agents targeting HBV replication and drugs restoring or increasing the host immune response could lead to a functional and perhaps an absolute cure within a decade.9 Following the recent success of HCV therapy, the viral hepatitis community has turned its focus on the discovery of novel HBV-associated biomarkers and therapeutic targets. It is hoped that the recent surge in anti-HBV drug discovery efforts will lead to the development of novel therapeutic strategies that could represent a path to cure for the more than 300 million individuals who are suffering from chronic hepatitis B infection worldwide. Fig. 1 Schematic representation of the inhibitors of Hepatitis B virus replication cycle. cccDNA, covalently closed circular DNA; ER, endoplasmic reticulum; HBV, hepatitis B virus; INF, interferon; mRNA, messenger RNA; pgRNA, pregenomic RNA; rcDNA, relaxed circular DNA; RT, reverse transcriptase. Table 1 Therapeutics in development for the treatment of hepatitis B chronic Infection. Updated information can be found on the Hepatitis B Foundation website (http://www.hepb.org/professionals/hbf_drug_watch.htm). Box 1. Definitions of cures Apparent virological cure Sustained off-drug suppression of serum HBsAg, HBeAg and viral DNA. cccDNA = undetectable or repressed Normalization of liver function (Normal levels of serum ALT and AST). *Risk of death from liver disease: to be determined once long-term survival data have been obtained. Functional cure Sustained off-drug suppression of serum HBsAg, HBeAg, viral DNA and cccDNA. Normalization of liver function (Normal levels of serum ALT and AST). *Comparable to individuals with naturally resolved infection. Absolute cure – virological cure Sustained off-drug suppression of serum HBsAg, HBeAg and viral DNA. Normalization of liver function (Normal levels of serum ALT and AST). Elimination of cccDNA. Presence of HBsAb. * Comparable to uninfected individuals. Key Points Despite current treatments available for HBV infection, the inability of the host immune system to completely clear the virus can lead to the establishment of incurable chronic infection. Identification of new targets and development of novel, curative drugs are necessary. Development of new therapeutic strategies that address the problems of cccDNA elimination, intrahepatic innate immune response stimulation, and HBV-specific immune response restoration are critical components of a path toward a possible cure. Combination of antiviral agents targeting HBV replication, and drugs restoring or increasing the host immune response could lead to a functional cure. Novel modalities such as CRISPR/Cas9 could disrupt HBV cccDNA and also target integrated viral DNA. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. The authors have nothing to disclose. References 1 Kang L Pan J Wu J Hu J Sun Q Tang J Anti-HBV drugs: progress, unmet needs, and new hope. Viruses 2015 7 9 4960 77 26389937 2 Wei W Wu Q Zhou J Kong Y You H A better antiviral efficacy found in nucleos(t)ide analog (NA) combinations with interferon therapy than NA monotherapy for HBeAg positive chronic hepatitis B: a meta-analysis. International journal of environmental research and public health 2015 12 8 10039 55 26308024 3 Chang J Guo F Zhao X Guo JT Therapeutic strategies for a functional cure of chronic hepatitis B virus infection. Acta pharmaceutica Sinica B 2014 4 4 248 57 26579392 4 Wang YJ Yang L Zuo JP Recent developments in antivirals against hepatitis B virus. Virus Res 2016 213 205 13 26732483 5 Block TM Gish R Guo H Mehta A Cuconati A Thomas London W Chronic hepatitis B: What should be the goal for new therapies? Antiviral Research 2013 98 1 27 34 23391846 6 Hai H Tamori A Kawada N Role of hepatitis B virus DNA integration in human hepatocarcinogenesis. World journal of gastroenterology 2014 20 20 6236 43 24876744 7 Lucifora J Trepo C Hepatitis: After HCV cure, HBV cure? Nature Reviews Gastroenterology & hepatology 2015 12 7 376 8 8 Block TM Rawat S Brosgart CL Chronic hepatitis B: A wave of new therapies on the horizon. Antiviral Research 2015 121 69 81 26112647 9 Zeisel MB Lucifora J Mason WS Sureau C Beck J Levrero M Towards an HBV cure: state-of-the-art and unresolved questions--report of the ANRS workshop on HBV cure. Gut 2015 64 8 1314 26 25670809 10 Sung JJY Lai JY Zeuzem S Chow WC Heathcote E Perrillo R A randomised double-blind phase II study of lamivudine (LAM) compared to lamivudine plus adefovir dipivoxil (ADV) for treatment naïve patients with chronic hepatitis B (CHB): Week 52 analysis. J Hepatol 2003 38 Suppl 2 25 6 11 Perrillo RP Current treatment of chronic hepatitis B: benefits and limitations. Seminars in liver disease 2005 25 S1 20 8 16103978 12 Marcellin P Ahn SH Ma X Caruntu FA Tak WY Elkashab M Combination of tenofovir disoproxil fumarate and peginterferon alpha-2a increases loss of hepatitis B surface antigen in patients with chronic hepatitis B. Gastroenterology 2016 150 1 134 44 26453773 13 Urban S Schulze A Schieck A Gähler C Ni Y Meier A 10 preclinical studies on Myrcludex B, a novel entry inhibitor for hepatitis B and hepatitis delta virus (HDV) infections. J Hepatol 2010 52 Supplement 1 S5 14 Volz T Allweiss L Ben MM Warlich M Lohse AW Pollok JM The entry inhibitor Myrcludex-B efficiently blocks intrahepatic virus spreading in humanized mice previously infected with hepatitis B virus. J Hepatol 2013 58 5 861 7 23246506 15 Verrier ER Colpitts CC Sureau C Baumert TF Hepatitis B virus receptors and molecular drug targets. Hepatology International 2016 1 7 16 Locarnini S Hatzakis A Chen D-S Lok A Strategies to control hepatitis B: Public policy, epidemiology, vaccine and drugs. J Hepatol 2015 62 S76 S86 25920093 17 Schadler S Hildt E HBV life cycle: entry and morphogenesis. Viruses 2009 1 2 185 209 21994545 18 Block TM Guo H Guo JT Molecular virology of hepatitis B virus for clinicians. Clin Liver Dis 2007 11 4 685 706 17981225 19 Guo H Jiang D Zhou T Characterization of the intracellular deproteinized relaxed circular DNA of hepatitis B virus: an intermediate of covalently closed circular DNA formation. J Virol 2007 81 22 12472 84 17804499 20 Zhou T Guo H Guo JT Cuconati A Mehta A Block TM Hepatitis B virus e antigen production is dependent upon covalently closed circular (ccc) DNA in HepAD38 cell cultures and may serve as a cccDNA surrogate in antiviral screening assays. Antiviral Res 2006 72 2 116 24 16780964 21 Cai D Mills C Yu W Yan R Aldrich CE Saputelli JR Identification of disubstituted sulfonamide compounds as specific inhibitors of hepatitis B virus covalently closed circular DNA formation. Antimicrob Agents chemother 2012 56 8 4277 88 22644022 22 Kennedy EM Bassit LC Mueller H Suppression of hepatitis B virus DNA accumulation in chronically infected cells using a bacterial CRISPR/Cas RNA-guided DNA endonuclease. Virology 2015 476 196 205 25553515 23 Weber ND Stone D Sedlak RH AAV-mediated delivery of zinc finger nucleases targeting hepatitis B virus inhibits active replication. PLoS One 2014 9 5 e97579 24827459 24 Lin SR Yang HC Kuo YT The CRISPR/Cas9 system facilitates clearance of the intrahepatic HBV templates in vivo. Mol Ther Nucleic Acids 2014 3 e186 25137139 25 Marcellin P Castelnau C Martinot-Peignoux M Boyer N Natural history of hepatitis B. Minerva Gastroenterol. Dietol 2005 51 1 63 75 26 Gish RG Yuen MF Chan HL Synthetic RNAi triggers and their use in chronic hepatitis B therapies with curative intent. Antiviral Res 2015 121 97 108 26129970 27 Sebestyén MG Wong SC Trubetskoy V Targeted in vivo delivery of siRNA and an endosome-releasing agent to hepatocytes. Methods Mol Biol 2015 1218 163 86 25319651 28 Mao T Graham M Kao SC Strings V Cock TA Lindell P Roelvink P Suhy D BB-HB-331, a DNA-directed RNA interference agent (ddRNAi) for the treatment of subjects infected with hepatitis B virus (HBV), can effectively suppress HBV in a primary hepatocyte model. Global Antiviral Journal 2015 11 Suppl 3 Abstract 111 29 Alaluf MB Shlomai A New therapies for chronic hepatitis B. Liver international: official journal of the International Association for the Study of the Liver 2016 30 Bourne C Lee S Venkataiah B Lee A Korba B Finn MG Small-molecule effectors of hepatitis B virus capsid assembly give insight into virus life cycle. J Virol 2008 82 20 10262 70 18684823 31 Stray SJ Zlotnick A BAY 41-4109 has multiple effects on hepatitis B virus capsid assembly. J Mol Rec 2006 19 6 542 8 32 Shi C Wu CQ Cao AM Sheng HZ Yan XZ Liao MY NMR spectroscopy-based metabonomic approach to the analysis of Bay41-4109, a novel anti-HBV compound, induced hepatotoxicity in rats. Toxicol Lett 2007 173 161 167 17826925 33 Weber O Schlemmer KH Hartmann E Hagelschuer I Paessens A Graef E Deres K Goldmann S Niewoehner U Stoltefuss J Haebich D Ruebsamen-Waigmann H Wohlfeil S Inhibition of human hepatitis B virus (HBV) by a novel non-nucleosidic compound in a transgenic mouse model. Antiviral Res 2002 54 69 78 12062392 34 Billioud G Pichoud C Puerstinger G Neyts J Zoulim F The main hepatitis B virus (HBV) mutants resistant to nucleoside analogs are susceptible in vitro to nonnucleoside inhibitors of HBV replication. Antiviral Res 2011 92 271 276 21871497 35 Manzoor S Saalim M Imran M Resham S Ashraf J Hepatitis B virus therapy: What's the future holding for us? World j gastroenterol 2015 21 44 12558 75 26640332 36 Wu G Liu B Zhang Y Li J Arzumanyan A Clayton MM Preclinical characterization of GLS4, an inhibitor of hepatitis B virus core particle assembly. Antimicrob Agents chemother 2013 57 11 5344 54 23959305 37 Wang XY Wei ZM Wu GY Wang JH Zhang YJ Li J In vitro inhibition of HBV replication by a novel compound, GLS4, and its efficacy against adefovir-dipivoxil-resistant HBV mutations. Antiviral ther 2012 17 5 793 803 38 Belloni L Li L Palumbo GA Chirapu SR Calvo L Finn M HAPs hepatitis B virus (HBV) capsid inhibitors block core protein interaction with the viral minichromosome and host cell genes and affect cccDNA transcription and stability. AASLD Liver Meeting 2013 Abstract 138 39 Lam A Ren S Vogel R Inhibition of hepatitis B virus replication by the HBV core inhibitor NVR 3-778. AASLD Liver Meeting 2015. San Francisco November 13-17, 2015 Abstract 33 40 Gane EJ Schwabe C Walker K Flores L Hartman G Klumpp K Phase 1a safety and pharmacokinetics of NVR 3-778, a potential first-in-class HBV core inhibitor. AASLD 2014 LB 19 41 Gane EJ Lim YS Gordon SC Visvanathan K The oral toll-like receptor-7 agonist GS-9620 in patients with chronic hepatitis B virus infection. J Hepatol 2015 63 2 320 8 25733157 42 Rebbapragada I Birkus G Perry J Molecular determinants of GS-9620-dependent TLR7 activation. PLoS One 2016 11 1 e0146835 26784926 43 Al-Mahtab M Bazinet M Vaillant A Effects of nucleic acid polymer therapy alone or in combination with immunotherapy on the establishment of SVR in patients with chronic HBV infection. J Clin Virol 2015 69 228 44 REP 2139-Ca / Pegasys™ combination therapy in hepatitis B / hepatitis D co-infection 2015 U.S. National Institutes of Health ClinicalTrials.gov Available from: clinicaltrials.gov/ct2/show/NCT02233075 45 Vörös J Urbanek A Rautureau GJP Large-scale production and structural and biophysical characterizations of the human hepatitis B virus polymerase. J Virol 2014 88 5 2584 99 24352439 46 Li C Wang Y Wang S Wu B Hao J Fan H Hepatitis B virus mRNA-mediated miR-122 inhibition upregulates PTTG1-binding protein, which promotes hepatocellular carcinoma tumor growth and cell invasion. J Virol 2013 87 4 2193 205 23221562 47 Fan CG Wang CM Tian C Wang Y Li L Sun WS miR-122 inhibits viral replication and cell proliferation in hepatitis B virus-related hepatocellular carcinoma and targets NDRG3. Oncology Reports 2011 26 5 1281 6 21725618

PMC005xxxxxx/PMC5119578.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101253275 36512 Asian Med (Leiden) Asian Med (Leiden) Asian medicine (Leiden, Netherlands) 1573-420X 1573-4218 27885323 5119578 10.1163/15734218-12341360 EMS70401 Article Contested Issues of Efficacy and Safety between Transnational Formulation Regimes of Tibetan Medicines in China and Europe Schrempf Mona University of Westminster m.schrempf@westminster.ac.uk 17 11 2016 2015 22 11 2016 10 1-2 273315 This file is available to download for the purposes of text mining, consistent with the principles of UK copyright law. Tibetan medicines are key material objects for medical treatment and have become part of a global trend of ‘pharmaceuticalisation’, playing increasingly important political and socio-economic roles in an ‘alternative modernity’. As I argue in this paper, they also have become key ‘sites of contestation’ between different epistemic values and styles of practice related to efficacy and safety that are reproduced in and through specific formulation regimes. Based on my multisited ethnography of production, prescription, and use practices of Tibetan medicines in China and Europe, this paper conceptualises three distinct formulation regimes, offering a heuristic model for transnational comparison—a classical, an industrialised or reformulated, and a polyherbal regime. The first two are the major orientations while the polyherbal is a conjoint hybrid with either the classical or the industrialised formulation regime. Globalised national drug safety regulations legalise and confer legitimacy to industrialised Tibetan formulas that follow biomedically defined efficacy, safety, and disease categories, while classical formulas produced by private physicians or small-scale cottage pharmacies are increasingly marginalised as producing ‘unsafe’ and at times illegal medicines, and need to find new ways for adapting and circulating their formulas. Tibetan medicines transnational formulation regimes efficacy safety legality China Europe Introduction Tibetan medicines are key material objects for medical treatment and have become a key ‘site of contestation’ among different stakeholders in a global trend of ‘pharmaceuticalisation’.1 While Tibetan medicines have taken on ‘social lives’ of their own in various formulations,2 as patent industrialised pharmaceuticals they also play increasingly important socio-economic and political roles in a global ‘alternative modernity’.3 This development is particularly salient in China and also in Europe where growing pharmaceutical industries produce standardised, quality-controlled, industrialised Tibetan drugs based on particular ‘reformulations’ of classical multicompound formulas successfully tapping into national, increasingly globalised niche markets of ‘natural’, ‘alternative’, or ‘integrative’ medicine.4 At the same time, classical Tibetan formulas are exclusively prescribed and often still produced by private physicians trained in ‘Tibetan medicine’ (Tib. bod sman) alias Sowa Rigpa (Tib. gso ba rig pa), commonly translated as the ‘science of healing’,5 diagnosing and treating their patients according to fundamental Tibetan medical principles in relation to the disease-causing imbalance in the individual patient’s body and mind. Most physicians of Tibetan medicine (called henceforth tm physicians) practice within rural Tibetan or Himalayan communities in Tibetan populated areas of China (Tibet Autonomous Region tar, Qinghai Province, Gansu Province, Sichuan Province, Yunnan Province), in Indian exile and in the historically influenced spheres of Tibetan culture of the Himalayas (Ladakh, Nepal, Bhutan), as well as in Mongolia, and Buryatia. The globalisation of Tibetan culture and religion, more recently propelled by exile-Tibetans in or from India, has also fostered the transnational movements of physicians and medicine(s). Individual tm physicians, most of them from India (fewer directly from China), were able to establish themselves in private clinics or as teachers in Europe or America. Some are also itinerant visitors prescribing classical Tibetan formulas that are circulating within specific transnational therapeutic networks. I argue in this paper that in contrast to classical Tibetan formulations, ‘modern traditional’, industrialised medicines targeting conventional diseases and a clientele outside of the Tibetan cultural sphere, tend to become ‘independent’ of the Tibetan medical knowledge system. They are mostly prescribed by biomedically trained physicians, those of ‘complementary and alternative medicine’ (cam) or ‘traditional Chinese medicine’ (tcm), and are marketed for niche markets of ‘natural’ or ‘herbal’ medicines in China and Europe.6 In other words, symptoms and diseases are diagnosed and configured according to biomedical parameters, such as diabetes, cardio-vascular, digestive, or stress-related disorders. Testing and validating clinical efficacy and safety of branded Tibetan pharmaceuticals for such symptoms and diseases is primarily done via ‘randomised controlled trials’ (rct), while some clinical studies in both Europe and China additionally included Tibetan medical diagnostics.7 In China, however, conspicuously more heterogenous regimes are followed in such rcts, including a ‘syndrome’ or ‘pattern differentiation’ (Ch. bianzheng 辩证) modelled on tcm. Similarly, Good Manufacturing Practices (gmp) employed to produce industrialised Tibetan pharmaceuticals follow the tcm drug regime and are often perceived as incompatible with Tibetan medical values of efficacy.8 These new producers, prescribers, and markets follow the global ‘scientific evidence’ and pharmaceutical research validation regimes; they have thus reconfigured what is considered efficacious and safe, legal, and legitimate in Tibetan medicine(s) even though most of their parameters are based on biomedicine. Such industrialised Tibetan medicines are, in turn, legalised by national (globalised) drug regulations and laws. Together with national regulatory regimes and the ever-growing, global wellness market these ‘pharmacoscapes’,9 as I argue in this paper, at the same time reinforce the marginalisation and, one could add, the making dispensable of the formula-producing and/or individually prescribing physician of Tibetan medicine, known as amchi (Tib. e mchi) or menpa (Tib. sman pa) and of his personal localised knowledge. Those who still have learnt how to produce Tibetan formulas on their own from their teachers in so-called teacher-student lineages, menpa gyüpa (Tib. sman pa rgyud pa), are classically trained tm physicians and mostly senior menpa or amchi. They use their skilled sensory abilities based on the five elements—water, earth, fire, air, space—that manifest not only in the environment but specifically in plants and the human body. While I will explain the principles of the classical formulation regime later on, it is important to note that physician-cum-pharmacists are personally known and highly esteemed by colleagues and patients for producing effective medicines; in fact, their renown is often based on this.10 Institutionalised education, in contrast, tends to separate and specialise training in Tibetan medicine in either physician or pharmacist training, thereby decoupling skills that are important for a fine-grained prescription practice. In contrast to the knowledge and experience for making medicines that are located in the personal skills of the tm physician, industrialised Tibetan pharmaceuticals locate and encapsulate an objectified ‘scientific’ knowledge within the materiality of the industrialised formula per se. It seems to embody ‘the best’ of both Tibetan and bio-medicine, i.e. bioscientifically valorised efficacy and safety based on the ‘traditional’ efficacy of Tibetan formulas tested to heal internationally and biomedically defined diseases.11 The scientisation of the formulas enables and legalises the producers of Tibetan pharmaceuticals to commercialise them in larger medical fields of biomedicine, cam, or even tcm settings in China or Europe. In Europe, and increasingly so in China as well, private physicians-cumpharmacists of Tibetan medicine have to find innovative ways in order to ensure that they can continue to produce their own formulas or have access to those which they trust to be produced in such a way as to ensure their maximum efficacy and safety. In contrast to the pharmaceuticalisation of Tibetan and Asian medicines more generally, little attention has so far been paid to how these recent socio-political and economic developments have contributed to asymmetrical shifts in power, legitimacy, and legality in the diverse transnational landscapes in which Tibetan medicine is practised today.12 Little is also known on how these changes have impacted the production and prescription of classical Tibetan formulations directly or indirectly, as well as on therapeutic practices of Tibetan medicine more generally in both China and Europe. In order to better understand these shifts from the perspectives of the different stakeholders involved, I examine and compare the production, prescription, and use of Tibetan medicines in situated contexts and in relation to their respective national regulatory regimes in China (specifically Tibetan-populated areas of Qinghai Province, known among Tibetans as Amdo) and Europe (Switzerland, Germany).13 Based on my multisited ethnography, this transnational macro-perspective consists in a heuristic model of three distinct formulation regimes that exist in parallel and partly in hybridised form—a ‘classical’, a ‘reformulated’, and a ‘polyherbal’—that I developed out of my fieldwork data.14 This study is following up on the ‘materiality of drugs’ in the anthropology of pharmaceuticals focusing on Tibetan medicines.15 It has in particular profited from comprehensive academic work undertaken on transnational studies of Asian medical practices,16 the emergence of the Tibetan pharmaceutical industry in China and the ‘social ecologies’ in Tibetan medical practices, examining the multivalent practices of efficacy and safety, including issues surrounding gmp, in China, Nepal, Tibet (tar), and Qinghai Province.17 The globalisation of Tibetan medicines has been directly addressed as a topic,18 as well as the multiple perspectives and practices of Tibetan medicine related to efficacy in the broadest sense, and to science and religion.19 Before I explain in more detail the different formulation regimes and who their major stakeholders are, I briefly outline the salient regulatory frameworks in both China and Europe that have shaped the way in which these formulations are produced in various ways. The Changing Landscapes of Tibetan Medicine in China and Europe In China, from about the mid- to late 1970s onwards, Tibetan medicine was slowly ‘integrated’ or rather institutionalised and transformed into a state-supported ‘nationality medicine and pharmacy’ (Ch. minzu yiyao 民族医药)20 with the intitial aim to provide primary health care to rural Tibetan areas.21 Deng Xiaoping’s reforms opened up the possibility for physicians of Tibetan medicine who had survived the Cultural Revolution (1966–76) to be retrained as ‘barefoot doctors’ (Tib. rkang rjen sman pa) and learn some of the basics of Chinese biomedicine, such as giving injections for administering antibiotic drugs. The latter is still common practice in most Tibetan medicine clinics and hospitals.22 tm physicians initially received small township posts and were thus integrated into local government health systems in Tibetan populated areas. Producing again their own classical formulas, often with the help of a growing number of students, was usually the only medicine available. Many of these physicians stayed in government health posts, while some opened private clinics. From the early 1980s onwards, Tibetan medical hospitals with attached pharmaceutical departments were built up in Tibetan populated areas of China that started to use mechanised production techniques in order to cater to the growing demands of their patients. As Saxer argues, these were the first steps leading to industrialised Tibetan medicines in China.23 At the same time, classically trained senior physicians became the teachers for a new generation of students of Tibetan medicine. Initial training courses based mainly on the Gyüshi were later incorporated into newly established Tibetan medical departments or schools as part of university medical school curricula. From the 1990s onwards, these were increasingly based on new textbooks. Also, national pharmacopoeias were produced to comply with state efforts to integrate Tibetan medicine and drugs, at times were rather hastily incorporated into Tibetan drug standards, with monographs on common ingredients and over 200 formulas. Also a provincial Qinghai pharmacopoeia was created.24 About a decade later they were integrated into the Drugs Standards of the Ministry of Public Health of the People’s Republic of China and the Pharmacopoeia of the People’s Republic of China.25 From the late 1990s onwards, and following the privatisation of public health services and the new food and drug policies on the production of medicines in China in 2001, the pharmaceutical industry of Tibetan medicines started to boom.26 An increasing ‘scientification’ and commodification of Tibetan medicines also began to impact the curricula and medical practices as taught in governmental tm schools and at hospitals, in particular the increasing professionalisation and thus separation of a training as physician and as pharmacist, happening in parallel to, and obviously catering to, the growth of private pharmaceutical enterprises and their markets needing specialised personnel.27 Several former pharmaceutical departments of hospitals of Tibetan medicine became private enterprises in Tibetan populated areas of China. In Europe the 1960s witnessed the very beginnings of Tibetan medicine, following the first exodus of Tibetan refugees from India arriving in Switzerland, among them also some Tibetan physicians.28 While already a century earlier the physician family lineage Badmayev, originally from Buryatia, had opened up Europe’s first Tibetan medical clinic in St. Petersburg, it was the decisive encounter between Tsultrim Badma’s grand-nephew Dr Peter Badmayev and the Swiss pharmacist Karl Lutz who had a strong interest in Tibetan medicine that triggered the making of new Tibetan pharmaceuticals in Europe based on recipes from this Buryat physician lineage. After establishing an international study and research group in Zurich on Tibetan formulas using different classical Tibetan texts and sounding out their potential applications in the West including some trials with herbal pills among Swiss physicians, Karl Lutz founded the Padma ag in 1969. The first Tibetan pharmaceutical, Padma Lax, was registered with the Swiss drug regulation agency (today Swissmedic) in 1970, followed by the flagship formula Padma 28 seven years later.29 Still, Padma had to develop an elaborate network of distribution pathways between cantonal, national and international borders in order to be able to distribute their products in Switzerland and beyond. Only from about the mid-1990s onwards were exile-Tibetan physicians able to establish their first clinics or schools in Europe.30 Nevertheless, and unlike tcm, until today Tibetan medicine lacks an officially acknowledged status as an Asian medical system in Europe. This has also consequences for the practice of Tibetan medicine. An incisive regulatory law that affected the import of Asian medicines in 2001, the same year that China’s regulatory agencies employed the demand for gmp, was the European Union-directive 2001/83/ec. It was implemented by the European Parliament and the Council, requiring ‘traditional herbal medicinal products’ (thmp) to comply with globalised gmp quality standards and a scientific evaluation regarding efficacy and safety via clinical studies.31 Even though the later amendment of the eu directive 2004/24/ec has skipped these regulatory challenges, it insists on compound herbal formulas needing to be licenced by a Market Authorization (ma) or a Traditional Herbal Registration (thr). A thmp—strictly excluding metals and animal ingredients—is now required to demonstrate: bibliographical or expert evidence to the effect that the medicinal product in question, or a corresponding product has been in medicinal use throughout a period of at least 30 years preceding the date of the application, including at least 15 years within the Community.32 Schwabl and Vennos, working for and at the pharmaceutical company Padma ag, the sole producer of about 11 formulas of Tibetan medicines and dietary supplements in Europe, recommend to amend the strict rule of 15 years of documented use, since otherwise ‘European citizens will be excluded from access to high quality medical traditions with their accumulated empirical knowledge’.33 As Gerke points out, European legislation already impacts the production of Tibetan ‘dietary supplements’ and ‘health tonics’ in India, allowing for their access to the European market.34 Private physicians—if they fully want to practice Tibetan medicine in Europe—usually find Tibetan medicines as produced by Padma ag too limited and also too expensive. On the other hand, the import of Tibetan classical compound formulas into the European Union from Asia remains a difficult issue. Physicians need to be innovative in their prescription practices, and how they access medicines due to shortage of classical formulas, and having to avoid border controls. Nevertheless, in the eyes of most practising tm physicians, no matter whether in Europe or in China, these medicines are a necessity for efficacious and safe healing. Their use also keeps the costs of medicines low for their patients and allows for their own small profit margin. In Europe, tm physicians now charge for their diagnostic skills which in many rural settings in Tibetan cultural areas were not remunerated. In the past and at present, however, classical formulations of Tibetan medicines build the very basis for the economic and professional survival of tm physicians. In Europe classical Tibetan formulas circulate under the radar of eu regulations mainly within the personal therapeutic networks of some tm physicians with relation to both India (or Nepal) and Europe. The one notable exception is the certification of physicians of Tibetan Medicine under the regulations of ehtpa (European Herbal Traditional Practitioners Association) in the uk. It allows physicians trained in Asia to practise prescribing herbal formulas that, following tcm-modalities, need to be imported as quality-controlled single ingredients that are compounded at the time of consultation for each patient.35 Transnational Formulation Regimes in China and Europe Moving on from nationally specific regulatory regimes, we broaden our view to transnational and analytic domains. I distinguish between three distinct formulation regimes—a ‘classical’, a ‘reformulated’, and a ‘polyherbal’. These three formulation regimes are often hybridised and reconfigured in, as well as circulated through, distinct social networks of producers, distributors, physicians, and patients in both China and Europe and at different scales. To illustrate this, in the diagram above (Fig. 1) I have used vector-type arrows indicating both the relative coherence as well as dynamic flexibility of Tibetan medical production, prescription, and use practices that are ongoing in and between these three formulation regimes. These formulation regimes are conceptualised both as analytical tools and as major orientations for practices in which specific epistemes of efficacy and safety prevail and thus will structure this paper. They serve as primary orientations in relation to the production, prescription and use of Tibetan medicines involving different groups of actors and a range of dynamic practices that co-exist in various forms. While classical formulations are the foundation of the other regimes, the primary tension, as I argue in this paper, lies between the two major orientations—the classical and the industrialised reformulation regime. The various actors involved in these formulation regimes orient themselves primarily towards culturally distinct sources and values of authenticity, legtimation, and trust that hinge on predominant values of efficacy and safety produced and reiterated through these regimes. The third regime, the polyherbal, is a conjoint hybrid with either the classical or the industrialised reformulation as primary orientation. The classical formulation regime is neither a uniform practice nor a thing of the past. It is (and always was) dynamic and needs to be adapted to local environments, availability of formula ingredients and, in modern times, also to changing regulations. It remains the primary orientation among physicians trained in Tibetan medicine following the fundamental principles of their living tradition. The classical formulation regime also shows the strongest continuity with those premodern practice environments that we know have existed within living memory, as well as how they used to be before the Cultural Revolution in China. I agree with Blaikie that we must critically assess the ‘classical’ as a heterogenous and fluid, rather than homogenous, category and stress its dynamic character.36 Yet, as I argue, the classical formulation regime coalesces as a rather coherent practice, one based upon Tibetan medical principles with particular actors, practices, and identities attached to it and characterised by predominant sets of epistemic values of efficacy and safety that are very different from the industrialised regime. The classical formulation also builds the very basis for the professional and economic survival of physicians fully trained in Tibetan medicine and the Tibetan medical system at large, as I will demonstrate. The reformulation regime, in contrast, is employed by a very different set of actors—pharmaceutical factories, mainly physicians trained in biomedicine, cam, or tcm, and targeted consumer markets of an alternative medicine. This regime is built upon what Pordié and Gaudillière have coined a ‘reformulation’ for ayurvedic pharmaceuticals in India, defining it as being based on ‘what its actors call “reverse engineering”’.37 They describe innovative ways by which the ayurvedic herbal industry transforms ‘shastric’ (or classic) polyherbal formulas for massproduction according to gmp, and for targeting biomedical disease categories conforming to the pharmaceutical market.38 I adopt their term for the reformulation regime of industrialised Tibetan pharmaceuticals that are produced in China and in Europe because they conform pretty much to the same set of globalised regulations.39 In Europe, however, the regime is much more strictly implemented, and not only includes gmp but generally ‘good practice’ standards.40 Also, the outcomes of clinical studies, including rcts, conducted in China for validating efficacy and safety of Tibetan pharmaceuticals are often not verifiable or useful.41 More than half of the rcts carried out to test Tibetan industrialised patent medicines were conducted in China’s biomedical hospitals, while only 22.9 % were undertaken in Tibetan medical hospitals. Biomedical disease categories, such as diabetes, or simple symptoms, were mainly used as indication factors, supplemented by some ‘traditional Tibetan medicine’ (ttm), as well as tcm ‘syndromes’ or ‘pattern differentiation’.42 Also slightly opaque is the fact that, even though patent or branded Tibetan medicines fall under the category of ‘Chinese new herbal medicines’,43 this merely seems to be an official statement following the nationally adapted category of ‘traditional medicines’ influenced by the who definition.44 Moreover, in China and among physicians of Tibetan medicine, the famous ‘jewel pills’ or rinchen rilbu (Tib. rin chen ril bu) are highly esteemed as being the most effective Tibetan formulas. Despite the fact that they contain ‘purified mercury’ tsotel that is excluded as heavy metal or toxic material in the Pharmacopoeia of the People’s Republic of China, they are also produced as industrialised patent medicines. This demonstrates one of several blind spots that in this case are covered under the label ‘national heritage drugs’.45 Also, animal ingredients characteristic of the Tibetan Plateau ecology are excluded from this official Pharmacopoeia, yet in several classical Tibetan formulas they are important and used because of their special efficacy, as I will show below in the case of musk. Animal ingredients are generally maintained within both classical and reformulated formulations in China. Therefore, and in contrast to Pordié and Gaudillière’s definition, I separate the industrialised reformulation from the polyherbal regime. Thus, the third polyherbal regime primarily concerns the exclusive use of usually at least five different herbal ingredients. It is slightly different in character from the two other major regimes because it is primarily oriented towards the ingredients being only herbs and some minerals. The polyherbal regime is a hybrid per se, one that merges either with the classical multicompound formulation into formulas that are produced and prescribed by physicians of Tibetan medicine in India and in Europe,46 or that merges with the reformulation regime being produced as a multicompound pharmaceutical, such as the Tibetan medicines and supplements produced by the Padma AG in Switzerland for Europe.47 Concerning the polyherbal regime, I adapted Naraindas’ analysis of a creole formulary logic of herbal ayurvedic pharmaceuticals in India to my transnational material in a different way. He distinguishes between three forms of logic that are hybridised in specific ways—a ‘biomedical formulary’, that is based on an active ingredient (or the synthesis of several) and a specific (biomedical) cause of disease, the multicompound ‘polyherbal formulary’ of the West with an underlying biomedical nosology, and the ‘ayurvedic formulary’ based on the classical shastra literature.48 However, rather than separating the biomedical from the classical (or shastric) as a distinct form of an industrialised ayurvedic formulary logic, I understand the three formulation regimes of Tibetan medicines in the contexts of China and Europe as transnational orientations driven by particular actors, and in which different epistemic values of efficacy and safety as well as medical principles are in use. At the same time biomedical and Tibetan medical notions and practices are negotiated in each of these regimes in specific ways. In Europe, herbal medicines have a long history of traditional therapeutic use. According to the European directives following who-regulatory regimes of ‘traditional medicines’ they should only contain herbs. Furthermore, cites-regulations to protect endangered animal and plant species49 are strictly followed in Europe, reinforced by rigorous documentation of sourcing and enforced legal penalties. In India, these regulations are known and applied to classical formulations by mostly exile-Tibetan physicians and/or pharmacists within the context of the Tibetan Medical and Astro Institute (tmai) in Dharamsala, better known as the Men-Tsee-Khang. However, because Tibetan medicines are not yet produced as industrialised pharmaceuticals following gmp in India, the extent to which these regulations are actually followed and by whom remains to be researched. The two major formulation regimes, the classical and the reformulated, also have the tendency to evoke as well as reiterate what Ian Hacking has defined as a ‘style of practice’.50 I differentiate between two major styles of practice in close relation to the classical and reformulated regimes, respectively. The first is a sensory based, personalised, complex diagnostic and prescription style following the fundamental principles of Tibetan medicine as defined in the classical formulation regime, and according to the constitution and imbalance of the ‘bodily dynamics’ nyépa (Tib. nyes pa) of each individual patient. The second is a one-formula disease-centred style, centring on one particular formula produced by pharmaceutical companies with standardised biotechnologies, and tested to target and cure a specific, usually biomedically defined disease.51 Interestingly, both styles are also found with regard to the application of certain formulas in historical Tibetan medical texts. For example, specific formulas for treating wounds or particular organs, such as eye medicines, are administered beyond a specific patient’s imbalance.52 The Classical Formulation Regime The classical formulation regime is clearly characterised by the fundamental Tibetan medical principle of five elements (Tib. byung ba lnga) that make up all phenomena in the universe. As expounded in the Tibetan classical medical text, the Gyüshi, the five elements manifest in plants and other materia medica as the ‘six tastes’ (Tib. ro drug), the ‘eight potencies’ (Tib. nus pa brgyad), and the ‘three post-digestive tastes’ (Tib. zhu rjes gsum) as well as the 17 ‘attributes’ (Tib. yon tan bcu bdun).53 These criteria are used for identifying single materia medica and compounding medicines. The five elements also make up the ‘three bodily dynamics’ or ‘three faults’ nyépa sum (Tib. nyes pa gsum) in the human body.54 Together, taste, potency, and post-digestive taste constitute the ‘benefit of effect’ (Tib. phan nus)55 for the overall efficacy of a specific formula. The Fourth or Subsequent Tantra (Phyi ma’i rgyud) of the Gyüshi clearly describes taste as the central sense for evaluating the material efficacy of raw medicinals, in order to validate their maximum ‘natural potency’ (Tib. ro nus). This means learning how to utilise the senses—taste, touch, smell, hearing, and vision—for differentiating and recognising the material quality and efficacy of materia medica (Tib. rdzas gyi nus pa). This is combined with the ability to diagnose the patient’s disease and underlaying individual imbalances via pulse, urine, and visual check-up, including investigation by questioning the patient on the onset and development of the disease. The classical formulation regime therefore implies a sensory-based, personalised prescription style. The natural potency of medicinal plants that are either imported from Himalayan, Indian, and Chinese environments, or growing in the wild on the Tibetan Plateau, is dependent upon the so-called ‘seven limbs’ (Tib. yan lag bdun) used as the traditional gold standard for ensuring their efficacy and safety. Knowing when and where to collect the most potent plants, which part of the plant should be used, how to collect, clean, detoxify, and store them properly, as well as knowing how to compound them for individual patients, constitutes the art by which a medicine-producing physician personally ensures efficacy and safety. The physician-pharmacist stands in the centre of the classical formulation regime of Tibetan medicine. Efficacy and safety are primarily connected within the physician’s knowledge and experience in diagnosing and prescribing, and also compounding medicines in a personalised style. While I have so far focused on the materiality of drugs, their efficacy extends of course into the immaterial sphere as well. Additional sources of efficacy in the classical formulation can include tantric ritual empowerment of medicines mendrub (Tib. sman grub). For example, small quantities of formerly blessed medicines are used for empowering new medicine production, and these embody a physician-cum-pharmacist’s lineage.56 Even within modern medical institutions, such as Tibetan medicine hospital pharmacies, where such practices are out of place on a formal level, senior physicians ensure, if necessary through private practice, the ritual empowerment of medicines used to treat their patients more efficiently. Among physicians of Tibetan medicine and their patients, these empowered medicines are perceived of as the only ‘true’ (Tib. ngo ma) ones in the sense of them being authentic and more efficacious. They are often juxtaposed with industrialised pharmaceuticals that are—depending upon the producer57—perceived of as being ‘fake’ (Tib. rdzun ma) by comparison.58 To illustrate the classical formulation regime and its relation to these fundamental principles, I now offer three ethnographic examples from Amdo, Qinghai Province. Already back in 2005, Ani Khandroma, a petite and delicate-looking, yet very energetic and self-assured 68-year-old female physician of Tibetan medicine from Amdo, felt the deep rift widening between what she called ‘old’ and ‘new’ practices. She was a famous tantric chö (Tib. gcod) practitioner, and had learnt Tibetan medicine back in the 1950s directly within a personal master-disciple apprenticeship in her home place of Rebgong (Tib. Reb gong, Ch. Tongren 同仁),59 a local centre of Tibetan medicine in Amdo.60 After the Cultural Revolution, she was able to continue her practice again, and succeeded in building up a specialised private clinic for women’s diseases—her main field of expertise—with medicinal baths near Chabcha (Tib. Chab cha, Ch. Gonghe 共和).61 With an awe-inspiring, slightly intimidating, and matter-of-fact manner of speaking, she was determined to tell me first what she thought of contemporary Tibetan medical practice in Qinghai: ‘It’s all modern these days. Even in private clinics of Tibetan medicine, they practise the modern way.’ She elaborated: In the (old) days, medical knowledge was different—lots of seeds and leaves were dried separately in the sun. There were no students, and no hospital. The ‘modern way’ is too fast, too big,62 and not careful enough. Yet, everything is medicine—it just depends on the quality, and the amount and the sequence of mixing—like cooking, really. This is the old way; it is basically not practised anymore.63 She pointed out that these old ways of knowledge were the most effective, something many private physicians of Tibetan medicine also stressed during my enquiries. For example, in the past physicians would only administer enough medicine for three days. If the patient was not better by then they were convinced they could not treat the patient, implying that their medicines were so strong and effective that of course they should have healed the patient within that time. Ani Khandroma used to produce her own medicines until 1958. This was the most decisive year of twentieth-century history in the lives of Amdo Tibetans, the threshold that sharply divided the ‘old’ Tibetan society from the ‘new’ one that was strongly influenced by Chinese modernity. Ani Khandroma continued: It is also important to know how to clean the medicinal ingredients properly, how to dry and store them, such as knowing whether to put them into a wooden container or into an animal skin pouch—which has a different effect. All this knowledge is gone these days. She reaffirmed that she kept her ‘women’s medicines’ in special animal skin pouches, which are the best for this purpose (see also Fig. 2).64 Interrupting her explanations all of a sudden, and with a challenging look on her face, she asked me what seemed to be a leading question, ‘What do you think is the most important thing in healing, the right diagnosis or the right medicine?’ I answered diplomatically that I think both would be important, but she interjected, ‘No, no, to diagnose correctly is the most important! You can heal in so many ways, even without medicines. Yet without the right diagnosis, medicine is useless and can even be harmful’. Ani Khandroma’s statement made me think that this kind of endangered knowledge might capture the actual core of any kind of efficacy and safety in healing. Change of scene. Moving out of Xining and travelling through rural Amdo, the area of Qinghai Province where Tibetan nomadic and farming communities live on the Tibetan Plateau, I also met a locally well-known, private physician who still handcrafted his own medicines. Based on his own experience and that of his patients, he stressed the importance of ‘true’ or ‘real’ efficacy: I will give you an example. Patients come to me and claim that Tibetan medicine does not work. Of course that is not true in general. They show me the medicines they purchased and were taking, and I check them by tasting them. Factory-produced Agar 35 is never as effective as when I produce it myself—the factory owners cut corners where they can, so will not use the most potent, rare but also expensive best quality of ar nag.65 They will substitute it with another cheap ingredient or won’t even realise that they bought a ‘fake’ ingredient.66 This local physician from a nomadic area was renowned for producing very efficacious medicines for his patients at a small rural clinic, where he kept his compounded powders safe in glass bottles on shelves. Some of his pills were packaged but only to ensure that they would not dry out. The local county government had asked him to move to the county town and into a local hospital of Tibetan medicine, in order to attract patients and make the place more popular. However, he did not want to move away from his own home and clinic, yet in the end had to comply. Nevertheless, he was able to bargain for a deal that allowed him to keep his private clinic (that otherwise would have been closed down if he had not consented) and occasionally work at the county hospital. Such recruiting is a practice often used by hospitals, at times even by Peoples’ Hospitals, which shows the social importance of locally known and trusted senior physicians of Tibetan medicine for attracting patients. Another example of using the classical formulation regime was found in a large and innovative government institution, the Tibetan Medicine Hospital of Qinghai Province (Tib. Mtsho sngon zhing chen bod sman khang; Ch. Qinghai sheng zangyiyuan 青海省藏医院) in Xining City (西宁市), which below I will refer to simply as the Qinghai Tibetan Hospital. It is considered number one among ‘seven most productive study centres’ of Tibetan medicine in China.68 Founded in 1983, the Qinghai Tibetan Hospital includes biotechnological diagnostic features and a general organisation into specialised departments similar to a biomedical hospital. It maintains Good Clinical Practices (gcp) standards, being the only hospital in Qinghai with such a high status. Somewhat ironically, the Qinghai Tibetan Hospital mainly conducts clinical research and studies on biomedical and tcm drugs, rather than Tibetan pharmaceuticals.69 Only very few studies at this institution were able to examine the efficacy of classical Tibetan formula because they lacked both the interest and money of large pharmaceutical factories usually sponsoring such studies. Despite all these modern biotechnological influences, Qinghai Tibetan Hospital has a pharmaceutical department that is primarily oriented towards the classical formulation regime.70 This department prides itself on being based upon lineage knowledge. This is due to its co-founder, the 84 year-old Dr Nyima, the most famous senior physician at the Qinghai Tibetan Hospital.71 Over the years, he had built up this pharmaceutical department on the basis of his personal knowledge in materia medica and formula production. Dr Nyima still personally tastes the quality of all the medical ingredients that arrive at the hospital’s pharmaceutical department, in order to check their quality and potency. He was also recently awarded the title of a knowledge holder of an intangible cultural heritage item of Tibetan medicine known as sertel (Tib. gser thal), a specific efficacy-enhancing practice employed for making certain jewel pills. In fact, the pharmaceutical company Arura Tibetan Medicine Co. Ltd. (Ch. Qinghai Jinhe zangyao gufen youxian gongsi 青海金诃藏药股份有限公司), that had evolved out of the pharmaceutical department of the Qinghai Tibetan Hospital, and that produces reformulated Tibetan medicines, had applied for his acknowledgment.72 Pharmacists at the respective Qinghai Tibetan Hospital department proudly refer to Dr Nyima as the ultimate authority and perceive themselves as followers of his lineage. Dr Nyima still practises as a physician three days a week, diagnosing and treating on average more than 60 patients per day. He is also often addressed by younger or less experienced physicians at the hospital, as a last resort for helping to treat difficult cases. Dr Nyima’s unique knowledge is based on his immense experience of many decades treating thousands of patients. He also possesses the unique skill of adjusting formulas via kha tshar, a specially potent and efficacious practice that uses the ‘add-on’ of a single ingredient to an established base formula. Such add-ons can target a specific organ, for example, the cooling effect of gur gum (L. Crocus sativus) targets both heat and liver in ‘hot’ liver disorders. It is also used as an add-on together with the digestive base formula Sendu 4, which thus becomes Sendu Drangné, which is suitable to warm the stomach and kidneys, but at the same time does not overheat an (already ‘hot’) liver.73 If several prescribed formulas share the same ingredient(s), this can increase their specific potency, but also increase their effects too much. For these cumulative effects to be balanced properly, the physician must compound his own formula in powder form, and know the exact dosage of each single ingredient within a set formula, in order to increase or decrease the dosage of a single ingredient precisely. Only a few senior physicians of Tibetan medicine, such as Dr Nyima, still know how to do this properly. To ensure the highest efficacy of treatment in more difficult patient cases, a complex style of prescription practice is used at the in-patient ward of the Qinghai Tibetan Hospital in Xining. In those cases, at least five different formulas are usually prescribed five times a day. Adjustment of dosage and medicines, if necessary, is undertaken based upon regular daily checks of pulse and urine, the dosage or type of formula being adjusted accordingly. Patient record keeping (Tib. nad tho) is meticulously maintained, as required of a government hospital. Some of it was written in Tibetan where pulse and urine diagnosis and Tibetan formulas were concerned, while technoscientific diagnostics, including blood tests, were written in Chinese. In the out-patient ward, patients would bring small patient booklets with them into which Tibetan formula names were written in Tibetan cursive script by Dr Nyima, while an assistant would type these Tibetan formula names into a computer that automatically translated them into Chinese drug names. Clearly, there are a wide range of practices adapted to a ‘Tibetan way of doing science’,74 each incorporated into what I call the classical formulation regime which constitutes the main framework in this example. Hybrid Classical-Polyherbal Formulations The ‘classical-polyherbal’ formulation is an important hybrid formulation based on very similar classical principles of Tibetan medicine for evaluating and ensuring efficacy and safety—yet with one crucial difference to those described above in that it completely excludes all use of animal products. This hybrid type of formulation among Tibetan medicines is mainly found in Europe and India. Polyherbal formulations have come to play an important role in meeting European and Indian patients’ dietary requirements, as well as their ethical and conservationist concerns. The main actors producing it are exiled Tibetan physicians-cum-pharmacists from India, some of whom also practise in Europe or whose products circulate in transnational therapeutic networks between the two regions. Although the Gyüshi does mention polyherbal formulations, historically Tibetan formulas based exclusively upon medicinal herbs (Tib. sngo sbyor) appear to have been rather limited to specific localities and their ecologies, and limited or no access to crucial animal ingredients. Blakie reports that, mainly for pragmatic and economic reasons, until the late 1970s some Ladakhi physicians who were still trained within family lineages used simple, locally available herbal ingredients and recipes inherited from their forefathers.75 More recently among exiled Tibetans in India, these classical-polyherbal formulations have become related to ethical concerns. In 2004, the pharmaceutical department of the Tibetan Medical and Astro Institute (tmai) in Dharamsala, India, better known as Men-Tsee-Khang, officially stopped using animal products at the behest of the Dalai Lama.76 Additionally, there are quite a considerable number of vegetarians among Indian and Western patients of Tibetan medicine. Thus, for ethical reasons formulas were adapted by substituting animal with plant ingredients of similar potency based upon the classical ‘taste’ ro criteria.77 Conservationist concerns regarding cites-listed endangered species were also taken up as necessary changes to be followed. The tradition of ahimsa or ‘nonviolence’, that some Hindus and Buddhists have aligned themselves to, conveniently converges here with dietary and conservationist concerns, as well as with the who’s definition of ‘traditional medicine’. The polyherbal regime also feeds into the ‘alternative’ modern sensitivities and consumerism of a global health-conscious urban elite concerned with lifestyle and disease-prevention by consuming health promoting, and at best vegan, ‘slow foods’, and dietary supplements including Asian herbal medicines. Nevertheless, as many physicians whom I had interviewed repeatedly stressed, due to the highly effective potency of certain animal ingredients, some are impossible to substitute. Not all tm physicians agree with compromising efficacy for reasons of ethics. The formula Agar 35, for example, contains musk or latsi (Tib. gla rtsi) in its classical formulation, and it is one of the most prescribed formulas in both Asia and Europe. The most potent and efficacious musk, however, comes from the wild musk deer (Tib. gla ba). Physicians of Tibetan medicine whom I have interviewed on this topic stressed conjointly that the potency and efficacy of wild musk is unsurpassed and that there is nothing more effective for treating inflammation. Thus, the farmed musk extracted from sedated animals produced by certain pharmaceutical companies in China, and which both government-run Tibetan hospital pharmacies and pharmaceutical factories in China are officially permitted to use, is not considered an equivalent substitution.78 This is an open secret among tm physicians in both Asia and Europe, but one not freely talked about due to conservation policies and ethical concerns. When I asked a senior tm physician from Amdo about polyherbal formulas, he laughed spontaneously, and said that such formulations would only be those used among poor farmer physicians who did not have the means to buy the very potent and also expensive animal ingredients (such as musk and bear bile) for making medicines. This echoes Blakie’s findings in Ladakh cited above and those by Craig in Nepal who also found a strong consonance expressed on amchi between ‘wildness’ and high potency of plants (and animals) more generally.79 In contrast to this statement, a senior tm physician from Indian exile was very upset when I even dared to question that the polyherbal formulation regime (presently practised) had ever been different in the past, and she denied outright the fact that classical Tibetan formulas can and do include animal ingredients. Clearly, experience, regulations, and ideology all play a role in formulation regimes, and constitute aspects of a physician’s medical and cultural-cum-ethical identity. In Europe, the late Akong Rinpoche, a famous Tibetan lama-physician from Kham, who had promoted Tibetan medicine within the uk as well as in his homeland for several decades,80 had specifically ordered the production of quality-controlled, polyherbal Tibetan medicines manufactured in Xining for consumption by Western patients in his Tara Clinics in the uk.81 The formulas used for this followed a little known formulary booklet written by the famous scholar Ju Mipham Jamyang Namgyal Gyamtso (’Ju mi pham rnam rgyal rgya mtsho, 1846–1912) that lists purely herbal formulas. Before production, Ju Mipham’s formulas were also screened in order to exclude or replace endangered plant species, targeting single ingredients that were cites-listed, and the formulas had to be adapted accordingly.82 To summarise, such polyherbal formulations are not necessarily new, but they have different implications today given the scaled up and more market-based nature of the industrialisation regime and its spillover effects on all of the others which will be examined in the following. The Reformulation Regime The Jinhe or Arura Tibetan Medicine Pharmaceutical Co., Ltd. based in Xining, Qinghai Province, mainly produces Tibetan pharmaceuticals for the Chinese market.83 A short glimpse into the recent past of this company shows that it developed out of the pharmaceutical department of the Qinghai Tibetan Hospital, following the socialist market reforms that triggered the decentralisation and privatisation of the public health sector in China. Since the early 1990s, with the help of its enterprising director Dr Ao, a Tibetan biomedical physician, Tibetan medicines have been produced for the larger Chinese market under the brand name of ‘Jinhe’ (金诃; ‘golden myrobalan’) or ‘Arura’. Arura (Tib. a ru ra) is the ‘king’ of medicinal plants in Tibetan medicine, being endowed with all six tastes, and derived from the Sanskrit arura classifying a specific myrobalan fruit (L. Terminalia chebula Retz). In 1999, the so-called Arura Tibetan Medical Group was founded, comprising the hospital, pharmaceutical factory, research department, college of Tibetan medicine, and a museum of Tibetan culture and medicine in Xining.84 As stated earlier, the year 2001 was decisive for the production of tm medicines in China, because the country entered into the World Trade Organization (wto), with gmp and a new drug administration law being introduced for validating the efficacy and safety of drugs primarily based upon biomedical parameters and modelled upon those used for tcm. While this exempted Chinese single ‘crude drugs’ from strict quality controls, Tibetan multicompound formulas that are already formulated as pills came under tighter control. From 2003 to 2004, the Arura pharmaceutical factory opened new premises complying with the latest gmp standards. Another decisive event occurred in 2010, when the gmp law was revised under China’s general gmp-standard regulations. In 2013, Arura was the first company in Qinghai to have obtained a gmp certificate under the revised Specifications for Quality Management of Drug Production (2010 Revision).85 Arura medical production combines technological know-how with ‘ancient’ Tibetan medical knowledge. Its products are encapsulated in shiny blister packages produced for the Chinese market, with indications written in Chinese. These indications at times resemble a hodge-podge of biomedical and Chinese medical terms that can more easily be understood by Chinese-speaking customers. These Arura products are taken directly as ‘over the counter medicines’ (otc), or can be prescribed by physicians not fully or at all trained in Tibetan medicine, and are thus much more versatile and commodifiable than ever before. The Arura Tibetan Medicine Co. Ltd has its registered production centre in Xining, while its marketing centre is located in Shanghai. The company has nearly 800 employees, over 200 shops throughout China, and maintains a complete R&D (Research and Development), production, marketing, and industrial chain. Using effective marketing, the Arura alias Jinhe Pharmaceutical Company has developed its own pharmacoscape in China, called ‘Jinhe Culture’ (Ch. jinhe wenhua 金诃文化). The first line one encounters on their Jinhe Culture website is a mission statement that could have come from the Men-Tsee-Khang in Indian exile, and where the stated aim is ‘to enhance Tibetan medicine, for the benefit of mankind’.86 Strategic objectives are stated more plainly, ‘to engage in the business of Tibetan medicine products’; ‘to expand the field of Tibetan medicine’,87 and ‘to cultivate the Tibetan medicine culture industry’. The Buddhist altruistic goal of doing things for the benefit of all beings is a well-known and oft-cited slogan which has become part of a Tibetan medical identity that Saxer has called the ‘moral economy of Tibetanness’, and one also shared by the Men-Tsee-Khang in Dharamsala, for example.88 Furthermore, Arura publically promote their own hybrid style of practice as innovative, technologically advanced, and combining global values of efficacy and safety—according to biomedical parameters—with Tibetan cultural patterns. The company has won several honorary titles, including ‘Chinese Famous Trademark’, ‘National Geographic Indications Protection Product’, ‘The Enterprise of National List of Intangible Cultural Heritage’, ‘National Contract Compliance Enterprise’, and ‘National Innovative Technology Enterprise’.89 Their website states: The company has 1000 types of products including drugs, supplements, caterpillar fungus,90 health products, daily-life regime, and external therapy for 6 types of categories. Of the 1000, 68 are nationally approved medicines with 21 ‘exclusive drugs’, 5 exclusive dosage forms, 10 medicines approved for reimbursement under national health insurance, 1 national essential drug, and 10 otc drugs.91 Agar 35 alias Sanshiwu wei chenxiang wan (三十五味沉香丸) or ‘Eaglewood Pill of 35 Tastes’ is promoted as an otc drug by the Arura Pharmaceutical Company, for example. It is a very popular Tibetan medicine prescribed by tm physicians in both Asia and Europe as well as taken in times of stress by patients directly. It is considered to be generally effective for stress-related symptoms, such as sleeplessness and concentration problems regardless of which underlying imbalances the patient has. It contains 35 mostly herbal ingredients with the primary one being eaglewood (Ch. chenxiang 沉香; L. Aquilaria agallocha).92 It also contains two animal products, ‘yak heart’ (Ch. yeniuxin, 野牛心) and ‘artificial musk’ (Ch. rengong shexiang, 人工麝香).93 tm physicians in Europe avoided the question of how these important animal ingredients are substituted while tm physicians and pharmacists in China pointed towards animal substitutions. The Hybrid Polyherbal Reformulation in Europe In Europe, the production of Tibetan pharmaceuticals and dietary supplements has been successfully established by the company Padma ag based in Switzerland, the sole pharmaceutical company producing Tibetan medicines in Europe.94 Despite the fact that Padma have only developed some 11 different polyherbal Tibetan formulas since the start of their production in 1970, Padma’s products are now the official legal face of Tibetan medicine(s) in Europe. Except for their specific formula composition, Padma’s Tibetan medical products are akin to other industrialised herbal remedies on the European market that fall under the rubric of traditional herbal medicinal products, with a standardised dosage and quality-controlled ingredients, and production methods following the pharmaceutical gold standard of good practice rules. The efficacy and safety of their products is clinically tested, and due to rcts, some of their products even attained the (biomedical) status of a ‘Tibetan medicine’ (German ‘Tibetisches Arzneimittel’).95 However, and similar to the reformulated pharmaceuticals produced by Arura for the Chinese market, Padma’s products are mainly prescribed by physicians of biomedicine and cam, or taken directly by patients as self-medication for common symptoms or often for chronic diseases that are difficult to heal and where use of biomedical pharmaceuticals can have side effects when taken longer term. Other Padma products circulate within a growing European cam and wellness consumer market as dietary supplements with low dosages and indicated by general symptoms. A side remark by an employee from Padma, plainly stating that such low dosage standard indications, as required by law for dietary supplements, tend to fail to be effective at all, aroused my curiosity. Consequently, I was supposed to ‘forget about proving the “efficacy” of Tibetan medicines’. The remark was meant to signify that at the time, when clinical trials were still a requirement for proving (biomedically validated) efficacy and safety of Padma’s products, it was a very lengthy, difficult and costly process to conduct rcts in the proper manner. Fortunately, this hurdle was overcome by the amended European directive of 2004. Depending upon a formula’s respective regime, and its legal status—and at times on any single ingredient within a given formula which might be classified as ‘toxic’ in a particular country of the eu, and also in Switzerland—its distribution may be officially restricted to cantonal or national borders. New pathways of circulation of products are international internet orders, yet again they become increasingly tricky to use due to border control regimes between countries. Even legalisation has its limits, it seems. Padma’s Tibetan polyherbal pharmaceuticals and dietary supplements do strictly comply with European legislations and international regulations of efficacy and safety. Thus, for example, one ingredient in the so-called Padma Nerves Tonic formula (‘Nerventonikum’), alias Sogdzin 10, a slightly adapted form of the classic formula Sogdzin 11 (Srog ’dzin 11), is omitted and not substituted for obvious reasons, since it is an animal product, i.e. rabbit’s heart. Schwabl and Vennos explain that because the European diet is already high in animal protein, the missing effect of this animal ingredient in their Tibetan formula is easily compensated or additionally supporting the overall warming effect of Sogdzin 10, counterbalancing the ‘cool’ nature of the disturbed ‘wind energy’ (Tib. rlung) by eating a meat bone soup instead. They elucidate the efficacy of Sogdzin 10 in both modern scientific terms of how it addresses stress systems by improving ‘better sleep’ and a ‘focused mind’, enumerating the single ingredients, in part their biochemical make-up but also their ‘Tibetan medical’ effects, such as nutmeg (Tib. dza ti) that ‘harmonizes the Lung [Tib. rlung] energy’.96 In particular, herbal ingredients that are endangered plant species listed in cites, even those of only doubtful botanical identity, are substituted in Padma’s formulations.97 Therefore, the generally endangered cites-listed species of eaglewood, in particular Aquilaria agallocha Roxb. (Tib. a gar; Ch. chenxiang) is substituted with guaiacum (Lat. Guaiacum sanctum L.), like eaglewood, an aromatic wood known as ingredient for incense but obtained from South America.98 According to Herbert Schwabl, director of Padma AG, guaiacum was already used as a substitute for eaglewood in the earlier Buryat recipes of the Badma family physicians that serve as basis for Padma formulas.99 Nevertheless, different species of eaglewood continue to be used in other classical eaglewood formulas in Asia, such as Agar 8, Agar 15, Agar 20, Agar 35, and Sogdzin 11. These formulas are all based on the root formula of Agar 8, and are usually prescribed for what according to Tibetan medical concepts is meant by an imbalance related to the ‘wind’ (Tib. rlung). From a Tibetan medical point of view, this concerns in particular the pathological ‘heart wind’ (Tib. snying rlung) and the ‘life-sustaining wind’ (Tib. srog rlung) that flows through the ‘life-sustaining channel’ (Tib. srog rtsa).100 In contrast to this classification, Padma Nerves Tonic (Sogdzin 10) is described by Padma AG in alternative medical terms simply as an anti-stress formula, and as having astringent and warming effects that strengthen the nerves and calm the mind: ‘It is used in the treatment of nervousness, irritability, restlessness, and nervous tension, as well as in difficulty falling asleep and remaining asleep’.101 Recently, Padma Nerves Tonic was evaluated with a positive outcome for similar symptoms related to menopause by the Institute for Naturopathy, University Hospital Zurich, Switzerland.102 Using scientific explanations for the overall synergistic ‘multitarget’ effects of the polyherbal classic Tibetan formulas, researchers associated with Padma stress a Tibetan formula’s ‘signature’ effects. ‘Signature’ is a characteristic cam term that is used to explain simultaneous ‘holistic’ effects that whole—rather than extracts from—medicinal plants can have upon the body and mind. Furthermore, systems biology is consulted as an explanatory model for the ‘energetic understanding’ of diseases in Tibetan medicine, and how diseases are both conceptualised in the body and how phytotherapeutics can work on several levels simultaneously.104 Polyherbal Tibetan formulas are explained as ‘multi-target medicines with a pleiotropic effect profile’ endowed with a particularly complex structure that can address more than one chronic disease.105 Padma is clearly trying hard to make up for their small number of products by stressing their polyvalent application potential. At the same time, they deliver new explanations on complexities of chronic and multimorbid diseases that are still little understood biomedically, or difficult to treat with conventional biomedicines. However, a study undertaken by Antonio and several physicians at the Dharamsala Men-Tsee-Khang focusing on ‘neuro-psychiatric disorders’, demonstrated just how challenging it can be to explain the complexity and efficacies of different Tibetan formulas in relation to specific biomedical disease categories by deducing these via their prescription.106 Ultimately, this study found that it was not possible to make any correlation between one formula being prescribed and a particular biomedical disease category. While such explanatory models and their cultural translations between different epistemologies and systems of knowledge as used by Padma remain to be studied more closely, it is clear that Padma’s transformations and innovations of formulas do not conform to any reductionist styles of biomedical and biochemical practice that only focus upon the efficacy of active agents, and this case is similar to that of industrialised ayurvedic pharmaceuticals.107 Instead, and increasingly so, at least in the Western cam milieu, concepts of efficacy pertaining to signature medicines, synergy, quantum physics, and systems biology are used to explain the ‘holistic’ and multivalent aspects of Asian medicines. Conclusions In this paper, I have argued that the official standards of bioscientifically validated efficacy and safety—as these are embodied in globalised, nationally specific regulatory regimes, and reiterated in the production of licensed reformulated Tibetan pharmaceuticals—transform Tibetan medical practices at large in both China and Europe in both direct and indirect ways. The heuristic model of formulation regimes I have proposed allows us to compare these developments in terms of transnational production and prescription practices of a similar style and scale. The comparison encompasses two principle groups of stakeholders and their formulation regimes. One is the classical formulation regime of small-scale producers and physicians of Tibetan medicine reconfiguring classical formulas in a personalised style, while focusing on addressing the individual imbalance of the patient, which from their perspective is the underlying cause of the disease. Their personal medical identity is intimately articulated with Tibetan medicine as their living tradition. The other is the reformulation regime reproduced by large-scale pharmaceutical companies, who create their own pharmacoscapes that are reconfiguring in relation to, and are at the same time validated by, nationally legitimised, bioscientific regulatory regimes of efficacy and safety. The reformulated Tibetan pharmaceuticals they produce are mainly used by a majority of physicians oriented towards biomedical or cam-parameters for targeting specific diseases. What has also been shown is how these formulation regimes configure a direct relationship between modes of production and ways of prescribing by specific groups of physicians with distinct medical identities. Sharing the same or very similar epistemic values of efficacy and safety link actors, production, and prescription practices together in these two different professional environments. Both of their distinctive ways of sourcing, production, and registration can entail complex translation processes between classical empirical knowledge on which these formulas are based and biomedical standards of regulations. This is true not only within the reformulation regime, but also where the classical formulation is practised within governmental institutional frames, such as in the case of the Qinghai Tibetan Hospital. In China, the reformulation regime of Tibetan formulas is clearly more influenced by tcm, while in Europe it follows the herbal cam-market dynamics. In both cases, the formula-disease-centred style of practice allows Tibetan formulas to be taken directly by patients themselves, catering to ever-growing wellness and consumer markets of alternative medicine. Formulas seem to embody an objectified efficacy per se that allows them to become independent from the medical system out of which they were originally developed, opening up new markets and also possibilities for application. In any case, physicians of Tibetan medicine—no matter in which country they practice—prefer to use a wide spectrum of formulas produced by trusted pharmacists. They also prefer to find their own ways of continuing the circulation of medicines via personal therapeutic networks or transitional grey market zones through which the substances and formulas can pass. Another way of continuing circulation within the same national context, for example, is to rename a formula already produced by a pharmaceutical company so as to avoid impinging upon intellectual property rights (ipr), or to adapt a formulation (e.g. by substituting or omitting certain ingredients, and/or lowering the dosage) in order to legalise its existence. By way of such processes, Tibetan formulas then also become teas, tonics, or dietary supplements in Europe, or ‘over-the-counter’ medicines in China. Within the general field of practice I have outlined and from which ethnographic examples were given, conundrums of safety and efficacy are manifest in various ways. First of all, strict food and drug policies in Europe, as well as those implemented with increasing rigour in some Tibetan areas of China, redefine classically formulated, hand-crafted, or small-scale manufactured medicines as being neither ‘safe’ nor legal. In the wake of this, in Tibetan areas at least, some private pharmacies-cum-clinics were already temporarily closed or have been threatened with closure.108 Gerke has called the ‘moving from efficacy to safety’ a generally increasing legal concern (also in medical anthropological approaches to the subject).109 Secondly, the official legal situation in Europe clearly condemns the classical formulation regime as generally ‘unsafe’ due to its non-standardised quality and dosage controls. Consequently, the few Tibetan pharmaceuticals produced by Padma are currently the only safe and efficacious licensed Tibetan medicines in Europe. They serve, in turn, to treat a large scope of chronic ailments, much larger than their classical indications would have suggested, and are supplemented by dietary and external therapies—such as massage—of Tibetan medicine that are integrated into other nationally legalised, mainstream therapeutic practices. Tibetan medicine, extended by Tibetan forms of yoga, and a host of other mind-body-techniques, has now become part of the European cam-landscape. Yet another conundrum is the consensus among classically trained tm physicians that pharmaceutical companies are not able to ensure the gold standards of efficacy and safety according to the logic of the classical formulation regime. This critique entails several concerns, first and foremost of which is that the marketing of medicines as over-the-counter products can jeopardise patient safety and undermine the possibility of efficacious treatment. Furthermore, they say that the maximisation of profits in commercial production gives rise to many compromises being made in terms of collection and production of medicines, all of which can adversely impact the overall quality, safety, and thus efficacy of medicines. Finally, there is general concern about over-harvesting, shortages, and price rises of materia medica. One general trend this research draws attention to is that increasing pharmaceuticalisation marginalises the classic formulation regime and its non-institutionalised agents who use a personalised style of producing and prescribing Tibetan formulas. They are being increasingly relegated to a grey zone in both Europe and China by regulatory regimes that ultimately support the production style of ‘big pharma’. The unfolding dynamics of this development should be the subject of future research. Figure 1 Diagram of Transnational Formulation Regimes of Tibetan Medicines. drawn by mona schrempf, 2015. Figure 2 The compounded mixture of the common Tibetan formula Sendu Drangné (Se ’bru dvangs gnas) being blessed in the altar room of a small monastic clinic in Amdo. photo by mona schrempf, 2013. Figure 3 Deer skin leather bags of a nomadic physician of Tibetan medicine from Amdo used for storing his handcrafted medicinal powders.67 photo by mona schrempf, 2013. Figure 4 The Tibetan reformulated formula Agar 35 produced by the Arura alias Jinhe Pharmaceutical Company. photo by mona schrempf, 2014. Figure 5 The Tibetan reformulated formula Sogdzin 10 produced by Padma AG.103 1 I apply the term ‘sites of contestation’ to Tibetan medicines as core therapeutic objects of a ‘living tradition’ that Tibetan medicine in all its diversity and styles of practices represents, cf. Scheid and Lei 2014. I understand ‘pharmaceuticalisation’ as a driving global force based on ‘the technological, material, and social specificity of pharmacy as a world of practices’, cf. Pordié and Gaudillière 2014b, p. 7, based on Biehl 2007. I thank my reviewers and, in particular, Sienna Craig for fruitful feed-back on earlier versions of this article. 2 See Whyte, van der Geest, and Hardon 2002. In relation to Tibetan medicines, cf. Craig 2012 (chapter 7, pp. 215–51) focusing on the ‘birth-helping pill’ Zhi byed 11; Gerke 2013a, b on the practice of purifying mercury tsotel (Tib. btso thal); Blaikie 2015 on the ‘jewel pill’ (Tib. rin chen ril bu) Samphel Norbu; also Nianggajia on the ‘White Pill’, in this volume. 3 Petryna, Lakoff, and Kleinman (eds) 2006. On the concept of ‘alternative modernity’, see Knauft 2002; and in relation to industrialised ‘reformulated’ ayurvedic Asian pharmaceuticals, see Pordié and Gaudillière 2014a. I use the term ‘Tibetan medicines’ as well as ‘Tibetan formulas’ as umbrella terms for multicompounds produced on the basis of classical Tibetan formulas according to the three different formulation regimes outlined in this paper, and applied as therapeutic objects for medical treatment in different social, cultural, clinical, and geographical contexts. 4 On the term ‘reformulation’ in relation to the making of industrialised herbal ayurvedic medicines in India, see Pordié and Gaudillière 2014a, b; Pordié and Hardon 2015. 5 I use the English term ‘science’ as a direct translation of Tibetan rig gnas in the sense of Tibetan medicine being part of the classical ‘five major sciences’ taught in monastic curricula (Tib. rig gnas che ba lnga). Since the physicians of Tibetan medicine whom I have interviewed in Europe and China and also some pharmacists in India, were all ethnically Tibetan, I intentionally use the term ‘Tibetan medicine’ (Tib. bod sman) rather than the term Sowa Rigpa. The latter designates primarily the living tradition as it is practised in the Himalayas by non-Tibetan ethnic groups, although Sowa Rigpa has also recently become a globalised umbrella term for Tibetan medicine more generally. See Craig and Gerke (forthcoming). 6 On the use of the term ‘traditional’ medicines in the European cam regulatory context, see Schwabl 2009. On Tibetan pharmaceuticals and gmp compliance in China, see Saxer 2013. 7 For a systematic review of rcts with Tibetan medicine, here called ‘traditional Tibetan medicine (ttm) modelling the name on tcm conducted in Europe, see Reuter et al. 2013; and for China, see Luo et al. 2015. 8 See, in particular, Saxer 2012. For critical social science studies on gmp in Tibetan medicine, see Craig 2011b; Craig 2012, pp. 146–82; Saxer 2013, pp. 59–94. On challenges in undertaking cross-cultural clinical trials research with Tibetan medicines, see Adams et al. 2005. 9 Based on Appadurai’s (1990) definition of ‘(land)scapes’, I use the term ‘pharmacoscape’ describing the specific professional and social networks of sourcing, distribution, and circulation built up by pharmaceutical companies that produce Tibetan pharmaceuticals, sponsor clinical research of their products, as well as target niche alternative markets of potential clients, i.e. mainly biomedically trained physicians, cam, or tcm hospitals, patients looking for alternative ‘side-effect-free’ medicines, consumers of ‘wellness’, and so forth. 10 Cf. Schrempf 2007 on lineage physicians-cum-pharmacists in Nagchu (tar). 11 For the International Classification of Diseases-10 (icd-10) set up by the who in 1990, see url: <http://www.who.int/classifications/icd/en/>, last accessed 12 December 2015. 12 At present, Stephan Kloos’ collaborative research project funded by the European Research Council focuses on industrialised Tibetan pharmaceuticals in Asia, see url: <http://www.ratimed.net>, accessed on 15 December 2015. 13 I am grateful to The Wellcome Trust for funding my research (2012–15) as part of the collaborative project ‘Beyond Tradition—Styles of Practice and Ways of Knowing in East Asian Medicine 1000 to Present’, located at the EASTmedicine Research Group, University of Westminster, London. My monograph on the topic is forthcoming (Schrempf forthcoming). 14 I used mainly participant observation and semi-structured interviews with producers and physicians of Tibetan medicine, as well as observations during several conferences where issues of efficacy and safety were discussed. Earlier fieldwork was funded within the framework of the collaborative research centre sfb 640 by the German Research Foundation (dfg). A separate fieldtrip was undertaken to India in 2014 conducting fieldwork with some small-scale producers of Tibetan medicines whose formulas are used in Europe. I thank all my informants for sharing their knowledge and time. 15 Cf. Hsu 2010, p. 23. On Tibetan medicines, see in particular Blaikie et al. 2015 on co-producing efficacious medicines among physicians of Tibetan medicine alias Sowa Rigpa from Nepal, Ladakh, and Tibetan areas of China. See also Craig and Glover 2009; Blaikie 2011, 2013, 2014, 2015. 16 Zhan 2009. 17 Craig 2012; Saxer 2013. 18 See, in particular, Janes 2002; Craig and Adams 2008; Craig 2012, 2014; also Hofer 2014; Schwabl 2013. 19 Cf. Adams, Schrempf, and Craig 2011a, (eds) 2011b; Adams 2002a, b; Adams, Le, and Dhondup 2010, 2011; Craig 2011a, 2012; Czaja, this volume. 20 I thank Lena Springer for pointing out this literal translation of minzu yiyao. While the term connotes that a particular minority nationality has its own ethnic system of medical knowledge and practice it also clearly implies to be part of China’s culture and public health system. The present politics of China’s cultural heritage including Tibetan medicine clearly point into this direction. 21 Cf. Janes 1995, 1999. For comparison, see Hsu 2009 on previous processes of standardisation and institutionalisation leading to the making of tcm in China that already began during the late 1950s. 22 On the plurality and diversity of healing practices in Amdo, the Tibetan populated area of Qinghai Province, see Schrempf 2011. 23 Saxer 2013, p. 35. For a detailed account on the development of a Tibetan pharmaceutical industry, see Saxer 2013, chapter 2, pp. 22–58. 24 See Krung go’i sman mdzod 2000; also sqtf 1992; see also Czaja and Schrempf (forthcoming). 25 See Ministry of Health 1995; State Pharmacopoeia Commission (ed.) 2015. 26 See sfda 2001. 27 On the complex encounters of Tibetan medicine and Chinese biomedicine, see Adams, Schrempf, and Craig (eds) 2011b. 28 At about the same time, in 1961, the Men-Tsee-Khang (today called the Tibetan Medical and Astro Institute) was founded in Dharamsala. On the beginnings and the development of Tibetan medicine in Indian exile, see Kloos 2010, 2013. 29 See url: <http://www.padma.at/padma/padma-ag-schweiz/geschichte/>, last accessed on 12 December 2015. 30 See, for example, the biography of the exile-Tibetan physician Tamdin Sither Bradley, url: <http://www.aruratibetanmedicine.com/biography.asp>; and of Pasang Yontan Arya, <http://www.tibetanmedicine-edu.org/index.php/dr-pasang-y-arya/11-dr-pasang-aryas-autobiography>; both last accessed on 12 December 2015. 31 url: <http://ec.europa.eu/health/files/eudralex/vol-1/dir_2001_83_consol_2012/dir_2001_83_cons_2012_en.pdf>, last accessed on 15 December 2015. See also Schwabl 2009 who points out that the definition of traditional herbal products is most likely following tcm phytotherapy via who definitions; see also Kadetz, this volume. 32 See Chapter 2a, Specific provisions applicable to traditional herbal medicinal products, Article 16c, at url: <http://ec.europa.eu/health/files/eudralex/vol-1/dir_2001_83_consol_2012/dir_2001_83_cons_2012_en.pdf>, last accessed on 15 December 2015; cf. Schwabl and Vennos 2015. 33 Schwabl and Vennos 2015, p. 108. 34 The uk allows herbal products from the Men-Tsee-Khang in Dharamsala to be imported; see Gerke 2012. Tibetan Therapeutics is a London-based outlet of their Sorig Tibetan Health Products, see url: <http://www.tibetan-therapeutics.com>, last accessed on 15 December 2015. 35 Personal communication with Brion Sweeney, London, University of Westminster, 8 May 2015. Cf. Millard 2008; Soktsang and Millard 2013. 36 Blaikie 2015. 37 Pordié and Gaudillière 2014a, p. 59. 38 ‘[… T]his regime consists in reformulating and simplifying ayurvedic medicinal compositions in order to create new “traditional” drugs for the biomedical disorders of an international as well as Indian clientele with a holistic claim’, Ibid. 39 While Tibetan medicines in India are not part of my research focus, it is noteworthy that in India a Tibetan pharmaceutical industry is only very slowly emerging, cf. Blaikie 2015. 40 For details, see Schwabl 2009. 41 Cf. Luo et al. 2015. Cf. also Craig 2011b. 42 Ibid., p. 453. 43 Ibid., referring to Zheng 2002. 44 Cf. World Health Organization 2005; Schwabl 2009; see also Kadetz, this volume, on the confluence of who and tcm definitions. 45 Cf. State Pharmacopoeia Commission (ed.) 2015; Saxer 2013, p. 71ff. Ironically, in Europe jewel pills are perceived as being poisonous according to bioscientific standards of safety since they contain some form of mercury, cf. Gerke 2013a. 46 I have included pharmacists producing Tibetan medicines in India in this study, because their medicines play a crucial role in the treatment of Western patients in Europe. 47 I am excluding here teas and ‘vitalising dietary supplements’ produced by the Men-Tsee-Khang in India for export to Europe. 48 Naraindas defines shastra (Skt. śāstra) in the context of Ayurveda as the classical ‘science of formulating drugs’ based on the principles of the ‘five elements’ (Skt. panca mahābhūta), the three ‘humours’ (Skt. doṣa) and the seven ‘tissues’ (Skt. dhātu), as well as on their applications. See Narinadas 2014, p. 13. 49 cites or Convention on International Trade in Endangered Species of Wild Fauna and Flora is an international treaty that aims at preventing overexploitation and regulating international trade in endangered species listed in the convention. 50 The notion of ‘styles of practice’ is based on ‘styles of scientific thinking and doing’ as defined by Hacking 2010, 2012. It is applied here as a general analytical approach to understanding two major ‘styles of practice’ in Tibetan medicine, i.e. in the sense of ways of reasoning that we call ‘scientific’ and encompassing both biomedical and Tibetan medical principles and practice and understandings of the body; cf. Hacking 2010, p. 235. They are self-authenticating, emerging with new objects and truth claims. 51 The reason why I do not define the second style as always being biomedically oriented is that some classically formulated formulas, as produced, for example, by the Gartse monastery clinic in Rebgong (see Nianggajia, this volume), are labelled with indications in specifically Tibetan medical and Chinese medical disease categories, rather than biomedical ones. Such formulas are not produced according to gmp either, yet they target specific vernacular symptoms and diseases surrounding stomach and digestive disorders. It appears that in the case just cited, the disease orientation for producing an effective digestive medicine for a locally common disease was specifically oriented to the local market in order to increase the sales among a specific multiethnic clientele. 52 I thank Olaf Czaja for pointing this out, personal communication July 25, 2015. 53 ‘The six basic tastes are sweet, sour, salty, bitter, hot, and astringent, and the three postdigestive tastes are sweet, sour, and bitter. […] Each of the six basic tastes is a composite of two of the five elements (earth, water, fire, air, and space) that define the characteristics of each taste, resulting in a particular potency of each taste and its subsequent attributes’, Sonam Dolma 2013, p. 109. 54 Water and earth manifest in ‘phlegm’ (Tib. bad mkan), fire in ‘bile’ (Tib. mkhris pa), air in ‘wind’ (Tib. rlung), while space gives rise to and pervades all four elements. 55 Phan nus is a compound derived from phan tog ‘benefit’ and nus pa ‘potency’, i.e. literally the ‘benefit of potency’. In English, I think this understanding comes closest to what we can call ‘efficacy’ in the broadest sense and including various epistemic values. 56 Blaikie 2014, pp. 293ff. On empowerment rituals, see Craig 2011a, and Czaja, this volume. 57 Saxer 2013 speaks here of a moral economy of Tibetanness where ethnically Tibetan owners of pharmaceutical factories are still perceived as more legitimate than Chinese. Since Saxer’s fieldwork with producers of Tibetan pharmaceuticals in China, shareholder ownership of several of these pharmaceutical companies has changed, in particular the Arura Medicine factory in Xining, Qinghai Province, has been owned for the past three years by a Chinese majority, which was perceived as very disappointing by many Tibetans I spoke with. In contrast, Jigme Phuntsok’s pharmaceuticals were less criticised and believed to be more authentic. On the Qinghai Jiumei Tibetan Medical Co. Ltd. (Ch. Qinghai Jiumei zangyao yaoye youxian gongsi 青海久美藏藥業有限公司), see Saxer 2013. 58 On a bifurcated view of reality into ‘real’ and ‘fake’ among Tibetans living in China, see Schrempf 2008, p. 131. 59 Rebgong is both a county and a county town, as well as the prefectural capital of the Malho Tibetan Autonomous Prefecture (Tib. Rma lho bod rigs rang skyong khul; Ch. Huangnan zangzu zizhizhou 黄南藏族自治州). 60 Ani Khandroma was said to be a reincarnation of a female enlightened being, a so-called ‘sky goer’ or khandroma (Tib. mkha ’gro ma) from Labrang Monastery (Tib. Bla brang bkra shis ’khyil, Ch. Labuleng si 拉卜楞寺) in Xiahe County (Ch. 夏河县) Gansu Province, and was very respected and revered within the Tibetan community. On Rebgong as a local hub of Tibetan medicine in Amdo, see Nianggajia, this volume. 61 Located in the Tsolho Tibetan Autonomous Prefecture (Tib. Mtsho lho bod rigs rang skyong khul, Ch. Hainan 海南藏族自治州) of Qinghai Province. When I met Ani Khandroma during 2005 in a small apartment in Xining where she occasionally stayed, she was planning to build another clinic at the shore of Tsongonpo alias Lake Kokonor or Qinghai Hu. She sadly passed away in April 2010. Shortly after, her beloved project of building up a clinic at the shore of Lake Kokonor was dismissed and the initial buildings were torn down, allegedly because of a newly declared law on ‘nature preservation’ in this area; cf. the ngo report by Schweiger (url: http://www.asia-ngo.de/pdf/ReiseAmdo2010.pdf, last accessed 24 November 2015). 62 As she explained, she meant the quantity and number of college students, size of schools, and hospitals, and the sheer scale of Tibetan medicine factories in big cities, such as Xining. 63 Personal communication in Xining, 25 August 2005. I thank my interlocutor at the time, Lhamo Drolma. 64 Next to deer skin, deer blood, and musk as well as bear’s gall bladder, and in general, wild animal blood and flesh are said to be of particular importance for healing women’s diseases (Tib. mo nad) of which the Gyüshi enumerates about 60 specific kinds. 65 Ar nag is the most potent form of eaglewood, and usually identified as Aquilaria agallocha Roxb, a dark, heavy, and densely resinous and fragrant hardwood that is more costly than other forms of Aquilaria used by the pharmaceutical department in Xining, for example. 66 Agar 35 should contain a blackish and oily type of eaglewood, the most potent version of ar nag or Aquilaria agallocha Roxb. This is a common formula frequently used in both China and in Europe for general relaxation, better sleep, and less anxiety. However, as local tm physicians stated, and reconfirmed by the medical botanist Christine Leon from Kew Gardens in London, there are many adulterations of eaglewood because it is rare and expensive, as well as being listed in cites as an endangered species (personal communication with Christine Leon during a workshop on ‘Developing an Interdisciplinary and Multilingual Digital Knowledge Base on Tibetan Medical Formulas with a Focus on Stress-related (rlung) Disorders’, EASTmedicine Research Group, University of Westminster, London, May 8, 2015); url: <https://www.westminster.ac.uk/news-and-events/news/2015/eastmedicine-international-workshop-0>, last accessed 15 December 2015. 67 He kept them, however, locked away, fearing unexpected food and drug safety controls that already once before had forced him to close his clinic down in 2003. 68 Luo et al. 2015, p. 453. 69 See Craig 2012, p. 57. 70 One should point out that ultimately, it depends on the preference and personal style of a physician whether and how biotechnogical and Tibetan medical principles are hybridised in practice and on a case-to-case study. 71 On the biography of Akhe Nyima (Tib. A khu Nyi ma; ‘Uncle Nyima’) who holds the highest Tibetan medical degree of menpa bumrampa (Tib. sman pa ’bum rams pa), see url: <http://arurahp.com/tibetan/?p=341>, last accessed 15 December 2015; see also Craig 2012, pp. 66–70. Among famous teaching physicians of Tibetan medicine, their personal knowledge, lineage authority, and legitimacy is embodied within the knowledge of compounding and recipes of certain Tibetan formulas. For example, the famous scholar-physician-teacher Khyenrab Norbu (Mkhyen rab nor bu, 1883–1962) taught the practice of tsotel to many tm physicians, and is used by many physicians of Tibetan medicine refer to his knowledge having been transmitted to them personally as part of their legitimation to produce jewel pills. 72 Personal communication with Akhe Nyima via Tsering Namgyal, 20 April 2016. 73 As I was told many times by private physicians, hot liver-related problems due to an excess of ‘bile’ are very common among Tibetan patients in Amdo and China, who tend to consume, next to too much meat and diary products, also too much oily food and hot chillies from the Chinese kitchen. See also Bassini 2013; Nianggajia, this volume. 74 Adams, Le, and Dhondup 2011. 75 Blaikie 2015, p. 14. 76 Ibid. 77 Personal communication with a physician-pharmacist of Tibetan medicine from India, Dharamsala, 4 November 2014. 78 Additionally, there is much fake musk in circulation, and the ‘real’ wild substance is difficult to access and expensive, thus physicians are very secretive about where they source their musk from. 79 Cf. Craig 2012, pp. 183–214. 80 On Akong Rinpoche’s life, his training as a Tibetan medical practitioner, role as a co-founder of the first Tibetan Buddhist monastery in Europe Samye Ling, and also position as strong promoter of Tibetan medicine in Europe, see url: <http://www.khenpo.eu/tara/andtara.html>, last accessed 2 December 2015. 81 On the Tara Clinics of Traditional Tibetan Medicine, see url: <http://www.tararokpa.org/medicine/index.html>, last accessed 2 December 2015. 82 Cf. Millard 2008, p. 192f. 83 url: <http://www.arura.cn/about.aspx?classid=1>, last accessed 10 February 2016. In the following, I call this company by the shortened name Arura. 84 On the pharmaceutical factory and its development, see Adams, Le, and Dhondup 2010, 2011; Saxer 2013, pp. 45ff. 85 url: <http://eng.sfda.gov.cn/WS03/CL0768/65113.html>, last accessed 10 February 2016. 86 弘扬藏医药, 造福全人类! url: <http://www.arura.cn/about.aspx?classid=3>, last accessed 15 February 2016. According to Lena Springer, ‘enhance’ connotes here also ‘praise’ and ‘spread’. I also thank Matthias Bauer and Nianggajia for the translations of this website from Chinese into English. 87 Verbatim: ‘to expand the field of medical treatment with Tibetan medicine’. 88 Cf. Saxer 2013; Kloos 2010. 89 See url: <http://www.arura.cn/about.aspx?classid=1>, last accessed 15 February 2016. 90 Until very recently, caterpillar fungus (Tib. dbyar rtsa dgun ’bu; L. Cordyceps sinensis) used to be the main cash income-generating wild crop collected by Tibetans due to a large demand on the Chinese market. It was mainly used in guanxi transactions among business people, yet prices fell in the past two years due to policy changes, and collectors have consequently suffered large losses in income. 91 公司拥有药品、保健品、冬虫夏草、健康产品、起居养生、特色外治6大类1000 余种产品, 其中药品共有68个国药准字号产品, 独家品种 21 个, 独家剂型 5 个, 国家医保品种 10 个, 国家医保品种1个, otc 品种 10。 url: <http://www.arura.cn/about.aspx?classid=1>, last accessed 2 December 2015. 92 According to Pasang Yontan Arya, ‘There are three types of a ga ru: Aquilaria sinensis (Lour) Cuilg (ar skya) Aquilaria agallocha Roxb. (ar nag) and Cinnanomum parthenoxylon (Tack.) Nees (ar dmar).’ Ar dmar is also known under the Tibetan name of a gar go snod. ‘In general, they are beneficial for heart and life channel.’ Pasang Yontan Arya 1998, p. 299. 93 The way in which the formulation Agar 35 is culturally translated into Chinese via hybrid Tibetan, biomedical, and tcm terminologies will be expounded in a joint article by Czaja and Schrempf (forthcoming). 94 On this company, see their elaborate website url: <http://www.padma.ch/en/products/medicinal-products/>, last accessed on 17 January 2015. 95 The first Padma pharmaceutical to attain this status was Padma Lax in 1970. On in vivo, in vitro and ex vitro studies on the effects of Padma Lax, see Gschossmann et al. 2010. On the in vitro study of the effects of Padma Digestin that more recently has attained the status of a ‘Tibetan medicine’ on certain muscle tissues in rats, see Balsiger et al. 2013. On clinical studies of Padma 28, see Vennos, Melzer, and Saller 2013; for a systematic review of clinical studies on the efficacy and safety of Padma 28 written as a doctoral thesis, see Röösli 2010. 96 Schwabl and Vennos 2009, p. 11. It should be noted that the addressees of this journal publication mostly are Western therapists trained in Tibetan medicine by Nida Chenaktsang, the director of the largest European school of ttm (traditional Tibetan medicine). He is actively retranslating Tibetan medical treatments into Western contexts with an emphasis on the ‘energetic levels’ of body and mind, including Tibetan Buddhist teachings of the Yuthok Nyingthik (g.Yu thog snying thig); cf. Garrett 2009. 97 The problem with the correct botanical identification of eaglewood—despite its large use in Tibetan formulas across the Himalayas, Tibet, and China—is that there are similar yet different species that are difficult to differentiate from each other and therefore their correct identification and use is questionable. 98 Schwabl and Vennos 2009, p. 11. 99 Personal communication with Herbert Schwabl, Director of the Padma ag, Zurich, January 2014. 100 On the complex of wind disorders in Tibetan medicine, see Yoeli-Tlalim 2010. 101 Leaflet of Padma Nerve Tonic (alias Padma Nerven-Tonikum). 102 Kneip 2015. 103 Image source: the website Padma ag, url: <http://www.padma.ch/en/products/other-swiss-products/other-padma-formulas/padma-nerves-formula/>, accessed 05 December 2015. 104 In systems biology, the body is conceptualised as a whole complex system consisting of interactive parts that are organised on different hierarchical levels and systems of cells, organs, and functions, and therefore in need of being addressed simultaneously. The complexity of the body (and disease) is related to the complexity of the synergistic efficacies of a single plant, as well as to a complex formula pattern; see Verpoorte, Choi, and Kim 2005. 105 Cf. Schwabl, Vennos, Saller 2013, p. 35. On the ‘holistic’ concept of ‘signature’ medicines in European cam-medicine as entailing a combination of material and ethereal or ‘essential’ (‘wesenhafte’) efficacy that addresses both the patient’s body and mind, cf. Kalbermatten 2015. 106 Cf. Antonio et al. 2013. 107 This coincides with the studies by Pordié and Gaudillière 2014a, b on industrialised polyherbal ayurvedic reformulations. 108 As always in China, there is no uniform implementation of policies, even at the level of a single county. I have met three individual physicians in Amdo with well-established private clinics independent of one another, partly with a lineage-based background, but also with official certificates from medical schools. They were locally or even regionally well-known for the quality of their medicines. Their clinics have either been closed down temporarily or they were threatened with closure because they had no licence for their medicines. The one whose clinic was closed down was only allowed to reopen if ‘safely’ produced, licensed Tibetan and Chinese medicines were sold on the premises. I also heard of similar reports from Lhasa in central Tibet (Xizang). However, other physicians elsewhere had never even heard of such policies. 109 Gerke 2012. Bibliography Adams V The Sacred in the Scientific: Ambiguous Practices of Science in Tibetan Medicine Cultural Anthropology 2002a 16 4 542 75 Adams V Establishing Proof: Translating “Science” and the State in Tibetan Medicine Lock M Nichter M New Horizons in Medical Anthropology: Essays in Honour of Charles Leslie 2002b London Bergin and Garvey 200 20 Adams V Miller S Craig S Samen A Nyima Sonam Droyoung Lhakpen Varner M The Challenge of Cross-Cultural Clinical Trials Research: Case Report from the Tibetan Autonomous Region, People’s Republic of China Medical Anthropology Quarterly 2005 19 3 267 89 16222962 Adams V Le P Dhondup R Translating Science: The Arura Medical Group at the Frontiers of Medical Research Craig S Cuomu M Garrett F Schrempf M Studies of Medical Pluralism in Tibetan History and Society Andiast International Institute for Tibetan and Buddhist Studies GmbH 2010 111 36 Proceedings of the 11th Seminar of the International Association for Tibetan Studies, Bonn 2006 Adams V Le P Dhondup R A Tibetan Way of Science: Revisioning Biomedicine as Tibetan Practice Adams V Schrempf M Craig S Medicine between Science and Religion 2011 107 26 Adams V Schrempf M Craig S Introduction: Medicine in Translation between Science and Religion Adams V Schrempf M Craig S Medicine between Science and Religion 2011a 1 28 Adams V Schrempf M Craig S Medicine between Science and Religion―Explorations on Tibetan Grounds 2011b London, New York Berghahn Publishers Antonio RL Kozasa EH Carlos J Galduróz F Dawa Dorjee Yeshi Kalsang Tsultrim Norbu Tsering Tenzin Tashi Rodrigues Eliana Formulas Used by Tibetan Doctors at Men-Tsee-Khang in India for the Treatment of Neuropsychiatric Disorders and Their Correlation with Pharmacological Data Phytotherapy Research 2013 27 552 63 10.1002/ptr.4749 22674653 Appadurai A Disjuncture and Difference in the Global Cultural Economy, Public Culture 1990 2 2 1 24 Balsiger BM Krayer M Rickenbacher A Flogerzi B Vennos C Gschossmann JM Tibetan Herbal Formula Padma Digestin Modulates Gastrointestinal Motility in Vitro World Journal of Gastrointestinal Pharmacology and Therapeutics 2013 4 1 9 15 23515138 Bassini P The Hierarchy of Food Consumption and Tibetan Experiences of Gastric and Gallbladder Disorders in Amdo East Asian Science, Technology and Society 2013 7 3 453 66 Biehl J Pharmaceuticalization: AIDS Treatment and the Global Health Politics Anthropological Quarterly 2007 80 4 1083 126 Blaikie C Critically Endangered? Medicinal Plant Cultivation and the Reconfiguration of Sowa Rigpa in Ladakh Craig SR Glover DM Conservation, Cultivation, and Commodification of Medicinal Plants in the Greater Himalayan-Tibetan Plateau (special issue), Asian Medicine—Tradition and Modernity 2011 5 243 72 Blaikie C Currents of Tradition in Tibetan Medicine Pharmacy East Asian Science, Technology and Society 2013 7 3 425 51 Blaikie C Making Medicine: Pharmacy, Exchange and the Production of Sowa Rigpa in Ladakh unpublished PhD dissertation School of Anthropology and Conservation, University of Kent 2014 Blaikie C Wish-Fulfilling Jewel Pills: Tibetan Medicines from Exclusivity to Ubiquity Anthropology and Medicine 2015 22 1 7 22 10.1080/13648470.2015.1004504 25633307 Blaikie C Craig S Hofer T Gerke B Co-Producing Efficacious Medicines: Collaborative Ethnography with Tibetan Medicine Practitioners in Kathmandu, Nepal Current Anthropology 4 2015 56 2 178 204 Craig SR From Empowerments to Power Calculations: Notes on Efficacy, Value and Method Adams V Schrempf M Craig SR Medicine between Science and Religion 2011a 215 44 Craig SR “Good” Manufacturing by whose Standards? Remaking Concepts of Quality, Safety, and Value in the Production of Tibetan Medicines Anthropological Quarterly 2011b 84 2 331 78 Craig SR Healing Elements: Efficacy and the Social Ecologies of Tibetan Medicine 2012 Berkeley University of California Press Craig SR Tibetan Medicine in the World: Local Scenes, Global Transformations Hofer Theresia Bodies in Balance. The Art of Tibetan Medicine New York: Rubin Museum of Art with London and New York University of Washington Press 110 25 Craig S Adams V Global Pharma in the Land of Snows: Tibetan Medicines, SARS, and Identity Politics across Nations Asian Medicine—Tradition and Modernity 2008 4 1 1 28 Craig SR Glover DM Conservation, Cultivation, and Commodification of Medicinal Plants in the Greater Himalayan-Tibetan Plateau Craig SR Glover DM Conservation, Cultivation, and Commodification of Medicinal Plants in the Greater Himalayan-Tibetan Plateau (special issue), Asian Medicine—Tradition and Modernity 2009 5 2 219 42 Craig SR Gerke B Naming and Forgetting: Sowa Rigpa and the Territory of Asian Medical Systems Medicine Anthropology Theory (in press) Czaja O Schrempf M Tibetan Industrialised Formulas in China—Cultural Translations between Tibetan, Chinese and Biomedicine (forthcoming) Dolma S Understanding Ideas of Toxicity in Tibetan Medical Processing of Mercury Gerke B Mercury in Ayurveda and Tibetan Medicine (special issue), Asian Medicine—Tradition and Modernity 2013 8 1 106 19 Garrett F The Alchemy of Accomplishing Medicine (sman sgrub): Situating the Yuthok Heart Essenc (G.yu thog snying thig) in Literature and History Journal of Indian Philosophy 2009 37 207 30 Gerke B Treating Essence with Essence: Re-inventing bcud len as Vitalising Dietary Supplements in Contemporary Tibetan Medicine Asian Medicine 2012 7 1 196 224 Gerke B The Social Life of tsotel: Processing Mercury in Contemporary Tibetan Medicine Gerke B Mercury in Ayurveda and Tibetan Medicine (special issue), Asian Medicine—Tradition and Modernity 2013a 8 1 120 52 Gerke B Mercury in Ayurveda and Tibetan Medicine (special issue), Asian Medicine—Tradition and Modernity 2013b 8 1 Brill Publishers Gerke B Moving from Efficacy to Safety: A Changing Focus in the Study of Asian Medical Systems Hsu E Ulijaszek S Potter C Medical Anthropology at Oxford: The First Decadeand Beyond(special issue), Journal of the Anthropological Society of Oxford (JASO) 2015 370 84 Gschossmann JM Krayer M Flogerzi B Balsiger BM Effects of the Tibetan Herbal Formula Padma Lax on Visceral Nociception and Contractility of Longitudinal Smooth Muscle in a Rat Model Neurogastroenterology and Motility 2010 22 9 1036 41 e269 70 10.1111/j.1365-2982.2010.01512.x 20518857 Hacking I Lloyd, Daston, Nurture, and “Style” Interdisciplinary Science Reviews 2010 35 3–4 231 40 Hacking I Language, Truth and Reason 30 Years Later Studies in History and Philosophy of Science 2012 43 599 609 Hofer T Socio-Economic Dimensions of Tibetan Medicine in the Tibet Autonomous Region, China—Part 1 + 2 Asian Medicine—Tradition and Modernity 2008 a, b 4 1+2 174 200 492 514 Hofer T Foundations of Pharmacology and the Compounding of Tibetan Medicines Hofer Theresia Bodies in Balance The Art of Tibetan Medicine New York: Rubin Museum of Art with London and New York University of Washington Press 2014 46 63 Hsu E The History of Chinese Medicine in the People’s Republic of China and Its Globalization East Asian Science, Technology and Society 2009 2 4 465 84 Hsu E Introduction. Plants in Medical Practice and Common Sense: on the Interface of Ethnobotany and Medical Anthropology Hsu E Harris S Plants, Health and Healing: On the Interface of Ethnobotany and Medical Anthropology 2010 New York, Oxford Berghahn Books 1 48 Janes CR The Transformations of Tibetan Medicine Medical Anthropology Quarterly 1995 9 6 39 7697551 Janes CR The Health Transition, Global Modernity and the Crisis of Traditional Medicine: The Tibetan Case Social Science and Medicine 1999 48 1–2 1803 20 10405018 Janes CR Buddhism, Science, and Market: The Globalisation of Tibetan Medicine Anthropology and Medicine 2002 9 3 267 89 26869120 Kalbermatten R Wesen und Signatur von Heilpflanzen. Die Gestalt als Schlüssel zu Heilkraft der Pflanzen 2015 Aarau and Munich AT Verlag Kloos S The History and Development of Tibetan Medicine in Exile Tibet Journal 2008 33 3 15 49 Kloos S Tibetan Medicine in Exile. The Ethics, Politics, and Science of Cultural Survival unpublished PhD thesis University of California, San Francisco & Berkeley 2010 Kloos S How Tibetan Medicine in Exile Became a “Medical System” East Asian Science, Technology and Society 2013 7 3 381 95 Knauft B Critically Modern: An Introduction Knauft B Critically Modern: Alternatives, Alterities, Anthropologies Bloomington Indiana University Press 2002 1 56 Kneip B Eine Therapieevaluation zur Behandlung von Menopausalen Symptomen mit Padma-Nerventonikum Schweizerische Zeitschrift für Ganzheitsmedizin 2015 27 165 7 10.1159/000430943 Krung go’i sman mdzod 2000 lo’i par gzhi’I bod sman ched bsgrigs. ’Year 2000 Edition of China’s Pharmacopoeia Specific for Tibetan Medicine’, mTsho sngon mi rigs dpe skrun khang: rGyal khab sman mdzod u yon lhan khang. Luo H Zhong G Yue L Wang Q Ma L Luobu Z Traditional Tibetan Medicine in China: A Systematic Overview of Randomized Clinical Trials European Journal of Integrative Medicine 2015 7 5 450 9 Millard C The Integration of Tibetan Medicine in the United Kingdom: The Clinics of the Tara Institute of Tibetan Medicine Pordié L Tibetan Medicine in the Contemporary World. Global Politics of Medical Knowledge and Practice 2008 London, New York Routledge 189 214 Ministry of Health Drugs Standards of the Ministry of Public Health of the People’s Republic of China: Tibetan Medicine 1995 Xining Ministry of Health of the PRC Naraindas H Of Shastric Yogams and Polyherbals—Exogenous Logics and the Creolisation of the Contemporary Ayurvedic Formulary Pordié L Gaudillière J-P The Herbal Pharmaceutical Industry in India (special issue), Asian Medicine—Tradition and Modernity 2014 9 1–2 12 48 Arya Pasang Yontan Gyatso Yonten Dictionary of Tibetan Materia Medica 1998 Delhi Motilal Banarsidas Publishers Ltd Petryna A Lakoff A Kleinman A Global Pharmaceuticals. Ethics, Markets, Practices 2006 Durham, NC Duke University Press Pordié L Gaudillière J-P The Reformulation Regime in Drug Discovery: Revisiting Polyherbals and Property Rights in the Ayurvedic Industry East Asian Science, Technology and Society 2014a 8 1 57 79 Pordié L Gaudillière J-P Industrial Ayurveda: Drug Discovery, Reformulation and the Market Pordié L Gaudillière J-P The Herbal Pharmaceutical Industry in India (special issue), Asian Medicine 2014b 9 1–2 1 11 Pordié L Hardon A Drugs’ Stories and Itineraries. On the Making of Asian Industrial Medicines Anthropology and Medicine 2015 22 1 1 6 10.1080/13648470.2015.1020745 25783429 Reuter KP Weißhuhn TER Witt CM Tibetan Medicine: A Systematic Review of the Clinical Research Available in the West Evidence Based Complementary Alternative Medicine 2013 2013 213407 10: eCAM 10.1155/2013/213407 23662117 Röösli MB Systematische Übersichtsarbeit: Klinische Studien zur Wirksamkeit und Sicherheit des Phytotherapeutischen Kombinationspräparats PADMA 28 2010 unpublished PhD dissertation Medizinische Fakultät der Universität Zürich Saxer M A Goat’s Head on a Sheep’s Body? Manufacturing Good Practices for Tibetan Medicine Medical Anthropology 2012 31 497 513 10.1080/01459740.2011.631958 22985109 Saxer M Manufacturing Tibetan Medicine. The Creation of an Industry and the Moral Economy of Tibetanness 2013 Oxford, New York Berghahn Publishers Scheid V Lei SH-L Introduction East Asian Science, Technology and Society: An International Journal 2014 8 1 7 10.1215/18752160-2413221 Schrempf M Lineage Doctors and the Transmission of Local Medical Knowledge and Practice in Nagchu Schrempf M Soundings in Tibetan Medicine. Historical and Anthropological Perspectives 2007 Leiden Brill Academic Publishers 91 126 Proceedings of the 10th Seminar of the International Association for Tibetan Studies [PIATS], Oxford 2003 Schrempf M Planning the Modern Tibetan Family Houben V Schrempf M Figurations of Modernity–Global and Local Representations in Comparative Perspective 2008 Frankfurt am Main: Campus 121 51 Schrempf M Between Mantra and Syringe: Healing and Health Seeking Behavior in Contemporary Amdo Adams V Schrempf M Craig S Medicine Between Science and Religion 2011 157 83 Schrempf M Transnational Tibetan Medicine—Formula Regimes, Therapeutic Networks and Styles of Practice in China and Europe (monograph) (forthcoming) Schwabl H It is Modern to be Traditional: Tradition and Tibetan Medicine in the European Context Asian Medicine―Tradition and Modernity 2009 5 373 84 Schwabl H The Mercury Puzzle. Tibetan Medicines and the Pharmaceutical Regulations in the European Union—Assessments and Opportunities Gerke B Mercury in Ayurveda and Tibetan Medicine (special issue), Asian Medicine—Tradition and Modernity 2013 8 1 181 98 Schwabl H Vennos C Srog ’zin 10 a Tibetan Formula for Times of Crisis and Unrest—keeping one’s nerve in spite of high pressure TTM (Traditional Tibetan Medicine) Journal 2009 3 1 8 13 Schwabl H Vennos C From Medical Tradition to Traditional Medicine: A Tibetan Formula in the European Framework Journal of Ethnopharmacology 2015 167 5 108 14 25446636 Schwabl H Vennos C Saller R Tibetische Rezepturen als pleiotrope Signaturen—Einsatz von Netzwerk-Arzneien bei Multimorbidität Forschende Komplementärmedizin 2013 20 suppl 2 35 40 10.1159/000351718 23860113 SFDA (State Food and Drug Administration of the PRC) Drug Administration Law of the People’s Republic of China last accessed 15 December 2015 2001 URL: http://eng.sfda.gov.cn/WS03/CL0766/61638.html SFDA (State Food and Drug Administration of the PRC) Good Manufacturing Practice for Pharmaceutical Products (2010 Revised Edition) last accessed 15 December 2015 2011 URL: http://eng.sfda.gov.cn/WS03/CL0768/65113.html Soktsang LD Millard C Diversity in Unity: The Changing Forms of Tibetan Medicine East Asian Science, Technology and Society: An International Journal 2013 7 467 86 10.1215/18752160-2333510 SQTF ‘Standard of Qinghai Tibetan Formulas’, Mtsho sngon bod sman tshad gzhi (Chin. Qing hai sheng zang yao biao zhun) 1992 Qinghai Province Health Department State Pharmacopoeia Commission weiyuanhui Guojia yaodian 国家药典委员 Pharmacopoeia of the People’s Republic of China (Zhongguo Renmin Gongheguo Yaodian 中国人民共和国药典 2015 Beijing China Medical Science and Technology Press Van der Geest S Reynolds Whyte S Hardon A The Anthropology of Pharmaceuticals: A Biographical Approach Annual Review of Anthropology 1996 25 153 78 Vennos C Melzer J Saller R Clinical Studies on the Efficacy and Safety of Padma 28, a Complex Herbal Formulation from Tibetan Medicine: An Overview Forschende Komplementmedizin 2013 20 suppl 2 25 30 10.1159/000351722 Verpoorte R Choi YH Kim HK Ethnopharmacology and Systems Biology: A Perfect Holistic Match Journal of Ethnopharmacology 2005 100 53 6 16026949 Whyte SR van der Geest S Hardon A Social Lives of Medicines. Cambridge Studies in Medical Anthropology 2002 Cambridge, UK Cambridge University Press World Health Organization National Policy on Traditional Medicine and Regulation of Herbal Medicines: Report of a WHO Global Survey 2005 Geneva WHO Yoeli-Tlalim R Tibetan “Wind” and “Wind Illnesses”: Towards a Multicultural Approach to Health and Illness Studies in History and Philosophy of Biological and Biomedical Sciences 2010 41 318 24 21112005 Zhan M Other-Worldly Making Chinese Medicine through Transnational Frames 2009 Durham, N.C Duke University Press Zheng XY Guidelines for Clinical Research on Chinese New Herbal Medicines 2002 Beijing Medical Science and Technology Publishing House of China

PMC005xxxxxx/PMC5119645.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101634614 42750 Microbiol Spectr Microbiol Spectr Microbiology spectrum 2165-0497 27780018 5119645 10.1128/microbiolspec.MCHD-0027-2016 NIHMS811569 Article Inflammation – a critical appreciation of the role of myeloid cells Iqbal Asif J 1 Fisher Edward A 2 Greaves David R. 1 1 Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE, UK 2 Deaprtments of Medicine (Cardiology) and Cell Biology, NYU School of Medicine, Smilow 7, 522 First Avenue, New York, NY 10016 Contact: David R. Greaves - david.greaves@path.ox.ac.uk 20 8 2016 10 2016 01 1 2017 4 5 10.1128/microbiolspec.MCHD-0027-2016This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. “The receptor concept is to pharmacology as homeostasis is to physiology, or metabolism to biochemistry. They provide the basic framework, and are the ‘Big Ideas’ without which it is impossible to understand what the subjects are about.” H.P. Rang. The receptor concept: pharmacology’s big idea. British Journal of Pharmacology (2008) 17: S9–S1 ‘What is inflammation’s big idea?’ In this brief overview of the role of myeloid cells in inflammation we will critically discuss what drives the initiation, amplification and resolution of inflammation in different anatomical sites in response to different pathological stimuli. It can be argued that we have a good understanding of the basic principles that underlie myeloid cell activation and the mobilization of innate immune cells to sites of injury and infection in acute inflammation. The challenge now for inflammation biologists is to understand how resolution of this normal physiological response goes wrong in hyper-acute and chronic inflammation. A better understanding of how inflammation is regulated will allow us to develop new anti-inflammatory drugs that will reduce the burden of inflammatory disease without compromising the patient’s immune defenses against infectious disease. Ideally such drugs should encourage a return to homeostasis and enhance tissue repair processes. Introduction – a historical and evolutionary synthesis Multicellular organisms have had to develop a rapid response to infection and tissue injury. In all animals on our planet this involves mobilization of specialized cells to the focus of the infection or injury. This important insight was beautifully illustrated by Elie Metchnikoff whose detailed drawings of cells being recruited to the site of injury caused by a rose thorn in a starfish embryo gave us the first glimpse of cells he termed macrophages and neutrophils, which he termed ‘microphages’. If not the first person to observe phagocytosis and leukocyte diapedesis, Metchnikoff was probably the first person to fully appreciate the role of these two important cellular processes in ‘natural’ or innate immunity (1). An important question arises from Metchnikoff’s observations in model organisms and histological images of human tissues infected by pyogenic bacteria - what are the locally produced molecular signals that mediate innate immune cell mobilization? A priori, we would expect these signals and the receptors that recognize them to have arisen early in evolution of multicellular organisms and for their function to be maintained by selection pressure exerted by everyday exposure to infectious disease and physical injury. A glimpse of how local signals to both pathogens and physical injury might have arisen comes from the Atlantic horseshoe crab, Limulus polyphemus. The horseshoe crab is often described as a ‘living fossil’ due to its near identical form to species present in the Triassic period 230 million years ago. The blood (or haemolymph) of Limulus species coagulates in response to intact bacteria and bacterial endotoxin effectively walling off invading pathogens. The Limulus protein that recognizes lipopolysaccharide (LPS) and Lipid A present in the outer membrane of Gram negative bacteria is a 132kDa protein called Factor C. LPS binding to the LPS / Lipid A recognition domain of Factor C activates a serine protease domain within the same protein. Activated Factor C activates haemolymph protein, Factor B, which initiates the clotting cascade to cause local haemolymph coagulation. This observation formed the basis of the Limulus amoebocyte lysate (LAL) assay for detecting low-level endotoxin contamination of tissue culture reagents, biologicals and medical devices (2). Limulus haemolymph also contains two conserved oligomeric serum proteins, the ‘short’ pentraxins C-reactive protein (CRP) and Serum Amyloid P component (SAP) (3). These two pentraxins, together with the evolutionarily conserved ‘long’ pentraxin PTX3, play important roles in mammalian host defense (4). A recent analysis of CRP knockout mice showed a marked sensitivity to Streptococcus pneumoniae infection in animals lacking endogenous CRP production that could be rescued by infusion of purified human plasma CRP or generation of anti- Strep Pneumoniae antibodies (5). These experiments strongly suggest that CRP has evolved to protect neonatal mammals from specific virulent bacterial pathogens. Specific roles for SAP in mammalian host defence have been harder to identify (possibly due to functional redundancy) but are likely to centre around recognition of bacterial peptidoglycan, damaged host membranes and complement activation, reviewed in (6). PTX3 was originally described as a non-redundant mammalian pattern recognition receptor essential for defence against the fungal pathogen Aspergillus fumigatus and later recognised to bind the bacterial pathogens Pseudomonas aeruginosa and uropathic E. coli as well as influenza virus (7, 8). Mammalian hepatocytes synthesize a range of other host defense proteins as part of the acute phase response, most notably ~30 proteins of the complement cascade. Complement is another evolutionary ancient defense against pathogens that shares with the coagulation cascade local activation and local amplification via serine protease cleavage of inactive enzymes (zymogens). It is clear that the complement system has evolved to be much more than a simple plasma pathogen recognition system that can kill microbes through deposition of a membrane attack complex (C5b, C6–9) (9). Proteolytic cleavage of the plasma protein C3 leads to deposition of the C3b protein fragment on target cells greatly enhancing phagocytosis by professional phagocytes of the innate immune system (neutrophils and macrophages). Cleavage of the C5 complement protein by the C3 convertase complex generates a high local concentration of C5a, a potent chemoattractant for innate immune cells via the G protein coupled receptor (GPCR) C5aR1 (10). The unrelenting evolutionary pressure exerted by the twin drivers of infectious disease and tissue injury means that any germline encoded signaling molecule or cellular response system that enhances tissue defense can be rapidly fixed, duplicated and mutated within eukaryote genomes. Multiple examples are provided from comparative genomics. One striking example is the Drosophila dorsoventral regulatory gene network, spätzle/Toll/cactus, which has been ‘re-engineered’ and ‘re-purposed’ during vertebrate evolution to give the cytokine activated NF-κB signaling pathway (11). Gene duplication has generated a family of Toll-like receptors (TLRs), which act as cellular pattern recognition receptors (PRRs) for highly conserved molecules on microbial pathogens termed pathogen associated molecular patterns (PAMPs) by Charles Janeway and Ruslan Medzithov (12, 13). Another striking example of duplication and diversification of immune defense genes comes from consideration of chemokines and their receptors. Comparative genomics reveals that the chemokine-chemokine receptor system has proven a useful module for directing cell-type specific chemotaxis and activation. In addition to mediating T cell chemotaxis the CXCR4 - CXCL12 interaction is used to keep haematopoietic stem cells (HSCs) within a specific bone marrow niche (14). Indeed, a small molecule CXCR4 antagonist (AMD3100) has found clinical application in mobilizing donor HSCs from bone marrow to peripheral blood for more efficient and less painful harvesting. The chemokine-chemokine receptor system has been exploited by the adaptive immune system for dendritic cell migration to lymph nodes, homeostatic leukocyte trafficking, lymphocyte homing to different tissues (e.g. the gut) and recruitment of specific lymphocytes subsets to sites of inflammation (15). The original description of the chemokine system as a inflammatory cell recruitment system, and the impressive results obtained using chemokine receptor gene knockout animals in models of chronic inflammation e.g. Boring et al 1998 (16), suggested that small molecule drugs that inhibit chemokine receptor signaling would make potent, cell type-specific anti-inflammatory drugs. To date this initial optimism has not (yet) been converted into clinically useful drugs (17, 18). No critical discussion of the role of myeloid cells in inflammation would be complete without consideration of inflammasome activation. The seminal contribution of Jurg Tschopp to immunology and inflammation biology was the recognition that the secretion of active Interleukin-1 (and Interleukin-18) is critically dependent on the formation of a large (>=700kD), cytoplasmic, multi-subunit caspase-activating complex (19). Since Tschopp’s first description of this macromolecular structure it has been shown that inflammasome activation can be triggered by a wide range of bacterial, viral, fungal and even helminth PAMPs as well as by a range of host damage associated molecules. Inflammasome activation leads to high local concentrations of IL-1 and inflammatory cell death by pyroptosis and pyronecrosis. Recent experiments in murine models have revealed tantalizing glimpses of a link between inflammasome activation and the inflammatory component of metabolic diseases such as obesity and Type 2 diabetes. Finally, it is important to remember that the prototypic acute inflammatory response can be triggered by non-microbial stimuli. A good example of such ‘sterile inflammation’ is provided by administration of substances such as monosodium urate (MSU) crystals and calcium pyrophosphate (CPP) crystals, the pathogenic drivers of gout and pseudo-gout respectively. Sensing of these and other crystalline insults is absolutely dependent upon the presence of a functional NALP3 gene and subsequent caspase activation, showing the central role of inflammasome activation and IL-1β in this neutrophil-dominated response to tissue injury (20). Necrotic cell death releases specific intracellular molecules that can induce activation of innate immune cells in vitro and a prototypic acute inflammatory response in vivo (21). Following the intellectual lead provided by Charles Janeway, signaling molecules released by necrotic tissue have been termed Damage Associated Molecular Patterns (DAMPs) by some, ‘Alarmins’ by others (22). Such molecules include extracellular ATP, mitochondrial DNA, uric acid and chromatin associated proteins, including the chromatin high mobility group-1 protein HMG-1. In a recent study Kataoka et al (23) directly compared the role of multiple signaling pathways in two mouse models of leukocyte recruitment in response to intraperitoneal injection of necrotic cells or induction of hepatocyte necrosis with paracetamol. The authors showed that neutrophil mobilization at early time points was significantly decreased in the absence of Complement C3, natural antibodies and the protease-activated receptor PAR2. Local depletion of ATP or deletion of the P2X7 receptor gene had no effect on leukocyte mobilization in these two models of necrotic cell induced inflammation, in contrast to findings in cultured cells or other models of sterile injury. Another potential ‘danger signal’ in the context of tissue injury is damage and modification of the extracellular matrix (ECM) (24). Hyaluronan (HA) is an abundant nonsulphated glycosaminoglycan component of the ECM found in many tissues. Multiple biological functions have been ascribed to HA and pro-inflammatory functions have been assigned to low molecular weight forms of HA (LMW HA) generated by the action of a range of hyauronidase enzymes. An interesting pre-clinical study by Huang et al (25) demonstrated that commercially available LMW HA and hyaluronidase enzyme preparations are contaminated with endotoxin and other proteins. This contamination of key reagents had lead to the erroneous conclusion that LMW HA is a ligand of the TLR2 and TLR4 receptors leading to pro-inflammatory cytokine production. By using an endotoxin-free pure preparation of the human hyaluronidase enzyme PH20 (rHuPH20) in an LPS dorsal air pouch model of acute inflammation Huang et al. demonstrated a marked anti –inflammatory effect of hyaluronidase characterised by no change in pro-inflammatory cytokine production but marked reduction of neutrophils. The pro-inflammatory role of ECM damage merits further study in the context of both acute and chronic inflammation as well as in the context of tissue repair and fibrosis. Experimental Models of Inflammation Animal models of inflammation are always open to criticism of how well they mimic clinical events in human disease. For instance mouse models of atherosclerosis do not display the classical clinical sequelae of human atherosclerosis, i.e. myocardial infarction and ischaemic stroke. Another obvious concern is that drugs that work well in pre-clinical models of human disease do not always translate into successful clinical trials. An early example was the success of anti-TNF therapy in murine and primate models of endotoxic shock and the subsequent failure of anti-TNF monoclonal antibodies to impact on morbidity and mortality in human septic shock and sepsis in randomized clinical trials (RCTs). Despite these obvious limitations, our current knowledge of inflammatory mediators, inflammatory cell biology and the resolution of inflammation owes much to a wide range of well characterized pre-clinical models some of which are outlined below. Animal models of acute inflammation Peritonitis is the inflammation of the peritoneum, a thin membrane lining the abdominal cavity. Experimentally it can be triggered by infectious stimuli, which if not treated immediately can spread to the blood and lead to septic shock. Peritonitis can also be induced by injection of sterile inflammagens or implantation of necrotic cells or tissues. Rodent models of peritonitis have provided insight into the generation of local mediators that aid the resolution and return to tissue homeostasis, such as Annexin-A1, lipoxins, resolvins, protectins and maresins. A variety of inflammatory stimuli have been used to induce peritonitis including zymosan, IL-1β and Brewers thioglycollate. Zymosan induced peritonitis is a simple and reproducible model of self-resolving inflammation. It has become a ‘go to’ model to study not only the kinetics of leukocyte recruitment and pro-inflammatory mediator production, but also pro-resolving actions on processes such as macrophage efferocytosis. Zymosan is a yeast cell wall extract of Saccharomyces cerevisiae and is recognised by TLR2 and Dectin-1. Intraperioneal injection with low dose zymosan (0.1–1mg/ml) leads to an initial wave of PMN recruitment into the peritoneal cavity which peaks between 6–8 hours followed by a second wave of mononuclear cells (>16 hours) during the resolution phase(26). Carageenan (CG) induced paw oedema is a well-established model of acute inflammation used to test the effects of a variety of anti-inflammatory drugs/compounds and to understand the role of mediators during an acute inflammatory response. CG is a gelling agent consisting of sulphated galactans and three main forms have been identified; iodo-CG, kappa-CG and lambda CG. The lambda species is most widely used in mice and rats and is given as sub-plantar injection in one paw. A similar volume of saline is injected into the contralateral paw as a control and oedema is usually measured by plethysmometry (27). Upon sub-plantar injection with 1–3% CG, a biphasic inflammatory response ensues with pain, increased vascular permeability, oedema and PMN influx observed as a result of the release of range of mediators being generated locally including substance P, histamine, bradykinin, prostaglandins, complement and reactive oxygen species. Peak oedema levels in the first phase of the response is observed between 4–6 hours followed by a second more intense phase developing between 48–72 hours. A study carried out by D’Agostino et al, (2007) examined the role of peroxisome proliferator-activated receptor (PPAR)-α agonistsin modulating CG induced paw oedema in mice. The authors found that intracerebroventricular administration of an endogenous PPAR-α agonist, palmitoylethanolamide (PEA) 30 minutes before CG administration reduced oedema formation. This reduction was linked to a decrease in COX-2, iNOS synthase expression and IκB degradation. Mice lacking PPARα showed no reduction in oedema formation following pre-treatment with PEA. This study elegantly demonstrated for the first time that activation of PPAR-α in the CNS could control peripheral inflammation (28). The air-pouch model is another simple model widely used to study inflammation in vivo. Two subcutaneous injections of air (day 0 and 3) into the dorsal intrascapular region leads to the formation of a discrete pouch. Injecting inflammatory stimuli such as zymosan, MSU or cytokines into the air-pouch results in rapid influx of PMNs and the local generation of mediators, including IL-8 and C5a. Depending upon the dose and the type of inflammagen used a second wave of mononuclear cells can follow. The air-pouch offers many advantages over the peritonitis model including ease of multiple dosing and application of compounds with poor solubility profiles. It also offers users the ability to recover relatively clean exudate samples, which can be used to measure mediators with low levels of abundance(29). Animal models of chronic inflammation The collagen induced arthritis model (CIA) has been an invaluable tool in furthering our understanding of some of the key mechanisms involved in the pathogenesis of human rheumatoid arthritis (RA). CIA is an inflammatory polyarthritis that shares many of the clinical and histological manifestations associated with human RA, i.e. disease being centered on the joints and the destruction of cartilage and bone. CIA is induced following sensitization with heterologous type II collagen (CII) in complete Freund’s adjuvant (CFA) followed by a second immunization (day 21) with CII in incomplete Freund’s adjuvant (IFA). This results in the initiation of cell mediated and humoral adaptive immune responses, which are characterized by increased levels of anti-CII IgG, complement fixation in joints, leukocyte infiltration and activation into the joint space and synovium, which leads to the generation of pro-inflammatory mediators such as TNFα and IL-17/23 that continue to drive the disease. A study carried out by Notley et al in 2008 utilised the CIA model to assess the effects of TNF blockade on IL-17 production. They found that mice treated with TNFR-Fc fusion protein or anti-TNF monoclonal antibody had reduced arthritis severity and showed reduced accumulation of Th1 and Th17 cells in the joint, but expanded Th1 and Th17 cell numbers in lymph nodes. Collectively, these findings suggested two opposing roles for TNF blockade in CIA, first blocking the accumulation of Th1 and Th17 cells in joints and secondly sequestration of pathogenic T cell numbers in peripheral lymphoid organs(30). In recent years, inducing rheumatoid arthritis in mice using serum transfer has become a more widely used model of RA, in part because the CIA model only works well in selected inbred mouse strains (a common feature of several mouse models of inflammation). The K/BxN serum transfer arthritis model was first described in the mid-90s as an inadvertent byproduct of crossing KRN T-cell receptor transgenic and nonobese diabetic (NOD) mice that lead to the development of spontaneous arthritis (31). Transfer of pathogenic anti-glucose-6-phosphate isomerase (GPI) antibodies from these mice to normal recipients resulted in the development of a severe arthritis as a result of alternative complement pathway activation and continual recruitment of neutrophil and mast cells into joints. A study from the laboratory of Mathis and Benoist employed this model to delineate the role of chemokines in leukocyte recruitment during the initiation of auto-antibody mediated arthritis. The authors induced arthritis by transferring serum into several chemokine and chemokine receptor deficient mice (CCR1–7, CCR9, CXCR2, CXCR3, CXCR5, CX3CR1, CCL2, or CCL3) and found that only the absence of CXCR2, a classical neutrophil chemokine receptor, was critical for the development of autoantibody-mediated joint inflammation and arthritisin C57/Bl6 mice (32). The ability of ApoE−/− mice on a C57BL6/J background to develop atherosclerotic lesions was first reported in 1992 and a similar phenotype in Ldlr−/− mice fed a high fat-high cholesterol diet was reported a few years later. These two mouse strains have been used extensively to study the cell and molecular biology of atherosclerosis in vivo (33). Mouse models of atherosclerosis involve many of the key features of human atherosclerosis including trapping of apoB-containing lipoproteins in the sub-endothelial space of major arteries, monocyte recruitment and macrophage differentiation. Acute Inflammation is a Process Generally ascribed to the Roman physician Celsus, the cardinal signs of acute inflammation in response to infection or tissue injury have been recognized for 2,000 years i.e. redness (rubor), heat (calor), pain (dolor) and swelling (oedema). Acute inflammation, like grieving, is a process (Billy Crystal, Robert de Niro, Analyze This, movie 1999 http://www.imdb.com/title/tt0122933/). The coordinated recruitment of plasma proteins, lipid mediators and myeloid cells in the inflammatory exudate can be defined by careful examination of sterile inflammation in the experimental model systems outlined above and can be organized under three headings. Initiation The earliest event in acute inflammation is sensing of pathogens and tissue damage, typically by tissue resident macrophages and mast cells. The most important role of these sentinel cells, which are relatively few in number, is to generate signaling molecules that lead to local endothelial cell activation in post capillary venules and autocrine and paracrine activation of macrophage effector functions within tissues. Amplification Locally generated signaling molecules cause important changes in the properties of nearby endothelium, including the release of the contents of Wiebel-Palade bodies, endothelial cell contraction, up-regulation of cell adhesion molecules (e.g. ICAM-1) and synthesis and presentation of chemokines. Acting in concert, all these changes in the endothelium allow for the local elaboration of an inflammatory exudate of plasma proteins and myeloid cells, especially neutrophils. Changes in the properties of endothelial cells cause the observed clinical signs of inflammation – redness and heat via increased local blood flow, oedema through elevated oncotic pressure caused by elevated albumin in tissues and localized pain through the action of locally produced mediators including bradykinin and Prostaglandin E2. It is the recruitment of plasma proteins and their subsequent proteolytic cleavage at sites of tissue damage or suspected pathogen invasion that leads to the massive amplification phase of inflammation that is essential for the localized recruitment of myeloid cells from the blood and their subsequent activation. Local activation of serine protease cascades in response to DAMPs and PAMPs (using the same molecular mechanism adopted by the horseshoe crab 220 million years ago) generates large quantities of potent protein and peptide mediators such as complement C3a, C5a and bradykinin. Resolution This is probably the least well-understood aspect of the acute inflammatory process. Termination of further leukocyte recruitment requires catabolism of inflammatory mediators, neutrophil apoptosis, macrophage efferocytosis of apoptotic cells, lymphatic drainage and the initiation of tissue repair processes. Described by many as an ‘active’ rather than a ‘passive’ process, definitive experiments addressing the cell biology of inflammation resolution using in vivo model systems are at a premium. An under-researched area of inflammation biology is the role of the lymphatic system in clearance of the inflammatory exudate during the resolution phase. One important anatomical arena where future research seems merited is the infarcted myocardium following recent studies showing changes to the cardiac lymphatic system following injury(34). Animal models of sterile peritonitis provide some idea of the potential timescale of the key events in acute inflammation. In the widely used mouse zymosan peritonitis model resident F4/80hi macrophages leave the serosal cavity to local lymph nodes within 60 minutes of inflammagen injection and neutrophil recruitment peaks at around 4 hours. PMN recruitment is followed by a wave of inflammatory Ly6Chi monocytes, which peaks between 16 and 24 hours. In this model neutrophil and monocyte numbers in the cavity return to baseline within 96 hours but these timings and the absolute number of myeloid cells depends on the dose and nature of the inflammagen. A recent paper from the group of Derek Gilroy extended analysis of the classic zymosan peritonitis model out past this 96-hour window and showed that a true return to tissue homeostasis following clearance of zymosan particles took more than three weeks (35). After disappearance of PMNs from the peritoneum, Newson et al observed changes in monocyte derived macrophage subsets and a significant increase in the number of B and T lymphocytes in the peritoneum. The recruitment and retention of this collection of lymphoid and myeloid cells long after the disappearance of the inciting sterile stimulus is intriguing. It will be important to see exactly how the mixture of cell types present post-inflammation contribute to tissue repair and defence against infectious disease. An important myeloid cell type that we will not consider in this brief review is the dendritic cell. An important question for immunologists and pathologists is, ‘Do dendritic cells play a unique role in the initiation, amplification or resolution of inflammation that cannot be provided monocyte derived macrophages recruited to the site of inflammation?’ Perhaps the unique, non-redundant role of DCs in inflamed tissues is to engage the anti-inflammatory/ pro-repair arm of the adaptive immune response, most notably Treg cells and perhaps some lesser-studied B-lymphocyte subsets such as B1 and Breg cells. Published studies on the role of DCs and B cell subsets in animal models of atherosclerosis may lead the way in this regard. A typical time course of an acute inflammatory response is shown in Figure 1 (after Christopher Buckley, University of Birmingham, U.K.). Representing the intensity of the inflammatory response on the Y-axis (measured as leukocyte recruitment, leukocyte activation or local inflammatory cytokine concentrations) the initiation, amplification and resolution phases of a stereotypic, ‘healthy’ acute inflammatory response can be clearly demarcated. Failure to clear the initial inflammatory insult or failure of inflammation resolution leads to chronic inflammation, represented by the horizontal arrow. Hyper-acute inflammation is represented by a continuing escalation of leukocyte recruitment, leukocyte activation (locally and systemically) and unrestrained inflammatory cytokine production. A couple of important discussion points arise from consideration of this simplistic representation of the ‘classical’ acute inflammatory response. The first is whether there is ever a complete absence of inflammation in host tissues i.e. should the Y-axis be set to zero at t=0? A paper published in 2010 by Jeffrey Weiser’s laboratory showed that that low-levels of systemic peptidoglycan derived from the gut microbiota primes the host innate immune system via the Nod1 receptor leading to more efficient neutrophil killing of Streptococcus pneumonia and Stahylococcus aureus (36). A second important discussion point arising from consideration of Figure 1 is ‘What regulates the magnitude of the acute inflammation response?’ and more specifically ‘Does the magnitude of the acute inflammation to the same inciting stimulus differ between tissues in the same individual and between individuals in the same population?’ A brief consideration of the mechanisms that might drive hyper-acute inflammation as presented in Figure 1 is also warranted. One simple explanation for the continuing amplification of acute inflammatory amplification seen in say, septic shock, could simply be the continuing proliferation of the initial bacterial infection that acted as the stimulus for the initiation of inflammation at t=0. Circumstantial evidence supporting such an explanation comes from a retrospective analysis of mortality data for patients admitted to hospital with suspected sepsis. In a cohort of 17,990 patients given antibiotics upon admission to 165 intensive care units with severe sepsis or septic shock the probability of in-hospital mortality increased steadily with time to antibiotic administration (37). An alternate explanation for hyper-acute inflammation could be failure of endogenous pathways that act to limit myeloid cell responses to PAMPs, a phenomenon that has been termed endotoxin tolerance (38) or the more recently appreciation of altered myeloid cell metabolism induced by sepsis (39). In Figure 2 we represent this question in a simple diagram. We propose that the magnitude of the host response to an inflammatory insult in any given anatomical location is the net result of the action of local pro-inflammatory mediators (upward arrows) and the opposing action of endogenous anti-inflammatory mediators (downward arrows). We have termed our model the ‘Inflammatory Set point Hypothesis’ and it stands apart from a previous theoretical consideration of the acute inflammatory response, which placed more weight on the kinetics of inflammation resolution by deriving a ‘Resolution Interval’ for comparing the effects of different therapeutic interventions in experimental animal models, typically zymosan peritonitis (40). The inflammatory set point model as presented in Figure 2 immediately suggests that pharmacological interventions that lower the activity of specific pro-inflammatory mediators (e.g. anti TNF antibodies or chemokine receptor antagonists) will reduce the maximal intensity of the acute inflammatory response. Another approach to reduce the peak level of inflammation would be to augment the activity of relevant endogenous anti-inflammatory mediators. This therapeutic rationale could be particularly efficacious in treatment of human diseases where the magnitude of the initial inflammatory response overwhelms the inflammation resolution machinery i.e. chronic (‘non-resolving’) inflammation and hyper-acute inflammation (see Figure 1). Consideration of experimental evidence for the importance of the downward arrows in Figure 2 comes from the enhanced inflammatory response seen in mice carrying gene deletions for the anti-inflammatory mediator Annexin A1 and its receptor Fpr2 (41). Finding evidence in support of the model presented in Figure 2 using human rather than murine models will be challenging but use of the beetle blister model could be instructive. Derek Gilroy and colleagues performed an interesting experiment where they gave human volunteers aspirin (75mg, oral once a day for 10 days) or placebo and then induced used a fixed dose of beetle cantharidin toxin to induce an acute inflammatory response in the skin characterized by dermal oedema and localized leukocyte recruitment. Volunteers taking low dose aspirin showed no change in blister oedema volume at 24 hours compared to volunteers taking placebo but did show reduced neutrophil and monocyte numbers in the inflammatory exudate (42). In terms of our set point model of Figure 2 the effect of low dose aspirin is likely two-fold, reducing pro-inflammatory drive by reducing local PGE2 production and simultaneously enhancing endogenous anti-inflammatory 15-epi-lipoxin A4 production. Monocytes and Inflammation Macrophages have been described as the “mature forms of circulating monocytes that have left the blood and taken up residence in the tissues” (The Immune System, Peter Parham, 3rd edition, Garland Science, NY and London, 2009). We now know that the origin of tissue macrophages is not exclusively from the circulating pool of monocytes. Under homeostatic conditions, many tissue resident macrophage populations are probably of embryonic origin and capable of self-renewal (reviewed in Sieweke and Allen (43). Most relevant to this chapter, however, is that an inflammatory stimulus will significantly increase the contribution of circulating monocytes to the tissue macrophage pool. In mice, the pool of circulating monocytes has been broadly divided into Ly6Chiand Ly6C lo subsets, which are also referred to as classical and non-classical monocytes, respectively reviewed in (44). In humans, the corresponding subsets are CD14+ CD16− and CD14lo CD16+, but the proportions of classical:non-classical monocytes varies by species (1:1 in mice and 9:1 in humans). Non-classical monocytes are thought to be derived from classical monocytes, but there is some evidence also for independent origin (45). The functions of circulating monocyte subsets in inflammation have been extensively studied, particularly in mice. CCR2-mediated mobilization and chemotaxis are major drivers of classical monocyte recruitment from bone marrow to blood, and from blood to inflammatory sites. This has been borne out in multiple studies of CCR2−/− mice. CCR2 deficient mice have fewer monocyte/macrophages in models of both acute e.g., thioglycollate-induced peritonitis (46) and chronic inflammation e.g., atherosclerosis (16). Likely this reflects both the impaired release of classical monocytes from the bone marrow as well as the reduced chemokine mediated recruitment of circulating monocytes to the site of inflammation. Though non-classical monocytes are low expressors of CCR2, in contrast to the classical cells they express high levels of CX3CR1. This chemokine receptor is thought to contribute not only to the migratory behaviour of the cells, but also to enhance the survival of both the non-classical monocytes in the blood and the tissue macrophages derived from them (47). Like classical monocytes, non-classical cells are recruited to sites of inflammation, but less abundantly so in both acute and chronic models (48). Also in contrast to the classical subset, they perform “patrolling” functions in the vasculature and can be recruited to non-inflamed tissues (49). Another contrast between the two subsets is the polarization of the tissue macrophages derived from each type. It is thought to be towards the activated M1 state for those of classical, and towards the anti-inflammatory, tissue repair M2 state for those of non-classical origin (50). There are a number of exceptions, however, to this “rule” (51), suggesting that the phenotype of the macrophages derived from each subset is likely to be context dependent. Though the bone marrow is typically the major source of circulating monocytes, in certain circumstances, there can be acute and substantial contributions from extra-medullary sources. One example is reported by Swirski, Nahrendorf, and their colleagues in a series of elegant papers reviewed in (52). In a mouse myocardial infarction (MI) model, in which the myocardium experiences ischemia-reperfusion damage, they observed, not surprisingly, that the inflammatory process followed that in wound healing; namely that there was biphasic entry of monocytes into the injured tissue, the first wave being classical (Ly6Chi) cells responding to a burst of locally produced CCL2, the ligand of CCR2. These cells became activated, M1-like macrophages. The two surprises were that the second wave also consisted of Ly6Chi cells. These converted to Ly6Clo cells in the injured tissue, where they became tissue-repair, M2-like macrophages (53) in a variation to the “rule” that tissue M2 macrophages are derived from circulating non-classical monocytes. The other surprise was that the source of the circulating monocytes was not the bone marrow directly, but rather the spleen, where monocyte progenitors (HSPCs) that travelled from bone marrow to specialized niches were stimulated to proliferate and to give rise to monocytes that entered the circulation and were recruited to the injured tissue (54). These observations leave open the question of whether drugs that inhibit CCR2 activity will ever find therapeutic utility post-MI patients. The Swirski-Nahrendorf results raise the possibility that blocking CCR2+ cell recruitment post MI will interfere with tissue healing and remodeling ultimately resulting in reduced cardiac output(55). Inflammatory Mediators Our brief consideration of acute inflammation and its resolution has highlighted the importance of sequential mobilization and differentiation of innate immune cells. Could sequential recruitment and differentiation of leukocyte subsets fit the bill for Henry Dale’s ‘Big Idea’ for inflammation biology? If so, it is clear that we will need to understand how different classes of inflammatory mediators are generated, how they change leukocyte migration and activation and ultimately how these mediators work together to effect tissue repair programmes. Different classes of mediators (briefly) Vasoactive amines - Activation of mast cells leads to rapid degranulation and the release of their potent pro-inflammatory granule contents e.g. vasoactive amines histamine and serotonin. These mediators act via specific GPCRs that can lead to vasodilatation and very rapid changes in cellular behaviour, e.g. endothelial cell contraction. Systemic mast cell degranulation can be fatal, for instance binding of food allergens to specific IgEs bound to mast cells via Fcε receptors. Cytokines and chemokines Cytokine genes are transcribed within minutes of exposure of macrophages to DAMPs and PAMPs and these cytokines can act in an autocrine and paracrine manner to further amplify the transcription of pro-inflammatory cytokines. Typically cytokine production by macrophages is measured in response to a single PAMP using primary cells (often bone marrow derived macrophages) cultured ex vivo. The advent of single cell transcriptomics will allow us to follow changes in the transcription pattern of the whole genome in sentinel cells responding to pathogens in infected tissues in vivo. To take full advantage of this surge in information we will need to stop thinking about cytokines in isolation and start thinking in terms of cytokine networks. Lipid mediators- In macrophages TLR signaling increases prostaglandin synthesis by activating cytosolic phospholipase A2 (cPLA2). cPLA2 releases arachadonic acid from membrane phospholipids and up-regulates cyclooxygenase-2 and microsomal prostaglandin E synthase-1 (mPGES-1) expression. These changes in intracellular lipid pools and lipid metabolizing enzymes set the scene for generation of multiple pro-inflammatory prostaglandins and leukotrienes. Later in the acute inflammatory response there is a marked ‘eicosanoid class switching’ which is marked by a switch to production of anti-inflammatory, pro-resolution lipid mediators such as Lipoxin A4 (56, 57). Gases as mediators - Since the Nobel Prize winning discovery of nitric oxide NO as a signaling molecule we now better appreciate two other gaseous signaling molecules that can act as anti-inflammatory mediators, carbon monoxide CO and hydrogen sulfide H2S (58). Vascular endothelial cell production of NO from L-arginine is catalyzed by endothelial nitric oxide synthase (eNOS, NOS3) and this almost ephemeral, very short-lived signaling molecule strongly influences vascular tone via cGMP signaling. CO is generated by heme catabolism by the enzyme heme oxidase-1 (HO-1) and CO exerts its anti-inflammatory effects via the MAPK pathway (59). The use of multiple H2S donors and selective inhibitors of H2S synthesis has helped to define the cellular actions of this gaseous signaling molecule. The therapeutic effects of H2S releasing drugs seen in animal models of inflammation has lead to H 2S releasing drugs been taken forward into clinical trials (60). A Brief Note on Inflammasomes and Autoinflammation As alluded to above, the field of inflammation biology owes much to the seminal papers of Jürg Tschopp (1951–2011) who first identified the cytoplasmic molecular machinery for secretion of active Interleukin 1β, a structure that he termed the inflammasome (61). The majority of inflammasomes are formed with one or two Nod-like receptor proteins (NLRs). Other non-NLR proteins including absent in melanoma 2 (AIM2) and pyrin can also form inflammasomes, as reviewed in (62). The N-terminal pyrin domain (PYD) within NLRs associates with apoptosis-speck like protein containing CARD (ASC) and this permits the recruitment of pro-caspase 1 to the inflammasome. The NLRP3 inflammasome is the most extensively studied inflammasome to date. NLRP3 activation can occur is response to a wide range of stimuli including infectious agents including intracellular bacterial products e.g. Shigella shiga toxin, extracellular ATP, monosodium urate or cholesterol crystals as well as changes in osmolarity or pH (63, 64). MCC950 is a highly selective inhibitor of NLRP3 that blocks canonical (ATP, monosodium urate) and non-canonical (cytosolic LPS) NLRP3 inflammasome activation at nanomolar concentration. Administration of MCC950 to mice with EAE, a murine pre-clinical model of human multiple sclerosis, has been shown to improve clinical symptoms and attenuate IL-1β production (65). The ability to pharmacologically inhibit NLRP3 inflammasome activation in pre-clinical models will greatly aid investigation of the role of this signaling complex in the pathogenesis of inflammation and could be the starting point for development of novel small molecule anti-inflammatory drugs. Over the past 25 years rheumatologists have come to recognize that autoinflammation as a distinct disease pathology from autoimmunity. Tumour Necrosis Factor Receptor Associated Periodic Syndrome (TRAPS) is a rare, genetic disease that causes recurrent episodes of fever that are associated with chills and muscle pain. In 1999 Kastner et al showed that TRAPS did not involve T or B lymphocyte activation but rather inappropriate innate immune activation caused by germline mutations in the 55kDa TNF receptor 1 (66). Kastner et al coined the term ‘autoinflammation’ for the observed defects in the regulation of systemic inflammation. Molecular analysis of other rare genetic diseases characterized by systemic inflammation with no obvious infectious disease or autoimmune component identified a group of rare monogenic diseases with defects in IL-1β production and the NALP3 inflammasome including Cryopyrin-Associated Periodic Syndromes (CAPS), Muckle Wells syndrome and familial cold autoinflammatory syndromes (67). Detailed analysis of other rare monogenic disorders continues to provide further examples of how loss of endogenous regulatory pathways can lead to inappropriate inflammatory and innate immune responses (68). Hyper-acute Inflammation – Bacterial Septic Shock and Viral Cytokine Storm Local activation of macrophages, mast cells and endothelial cells is essential to mobilize an acute inflammatory response to sites of pathogen invasion and tissue injury. However, systemic activation of these cells by bacterial PAMPs can lead to the life-threatening septic shock. An excessive systemic host reaction to viral pathogens such as influenza, frequently termed a ‘cytokine storm’, can also have life-threatening consequences, often in young people with a robust immune system (shown diagrammatically in Figure 1). The clinical sequelae of septic shock and cytokine storm show the importance of regulating the magnitude of the initial inflammatory response (Figure 2). The recent appreciation of the need to distinguish sepsis from septic shock serves only to emphasise the importance of return to tissue homeostasis following the initial triggering of a hyper-acute inflammatory response (69). There is a substantial unmet clinical need to develop better treatments for septic shock and sepsis but a better understanding of the disease process is being held back by the lack of good experimental models and the continuing failure to translate basic science findings into effective new treatments (70). One fruitful area for future research might be to identify and augment endogenous pathways that can rapidly decrease the maximal host inflammatory response without ‘paralysing’ the innate immune system altogether. Chronic Inflammation– Pathogenesis and Current Treatments In an excellent 2010 review article Carl Nathan and Aihhao Ding wrote, “The problem with inflammation is not how often it starts, but how often it fails to subside. Non-resolving inflammation is not a primary cause of atherosclerosis, obesity, cancer, chronic obstructive pulmonary disease, asthma, inflammatory bowel disease, multiple sclerosis, or rheumatoid arthritis, but it contributes significantly to their pathogenesis.” (24). In their review the authors lay out multiple overlapping and competing models for how chronic inflammation can arise in vivo. Failure to clear a pathogen that was the original trigger for inciting acute inflammation. Classic examples would be failure to clear mycobacterial infection or chronic virus persistence in hepatitis. Response to continuing tissue injury, for instance necrotic cell damage and the release of DAMPs caused by ischaemia reperfusion injury. Continuing presence of antigen, e.g. antigens recognized by autoantibodies in rheumatoid arthritis. Non-resolving inflammation, for instance the failure to clear macrophages and macrophage derived foam cells from atherosclerotic lesions in major arteries leading to the build up of stable and unstable atherosclerotic plaques, a form of chronic inflammation that persists for decades. Consideration of the multiplicity of pathogenic mechanisms in human diseases caused by chronic inflammation reminds one of the opening lines of Leo Tolstoy’s novel Anna Karenina ‘All happy families resemble one another, each unhappy family is unhappy in its own way’. All successfully resolved bouts of inflammation resemble one another in showing a return to ‘happy’ tissue architecture and essentially normal tissue function. In contrast, each chronic inflammatory disease shows ‘unhappy’ tissue architecture caused by multiple defects in the return to homeostasis. Current treatments for chronic inflammation include steroids, non steroidal anti-inflammatory drugs (NSAIDs), disease modifying anti rheumatic drugs (DMARDs) and ‘biologicals’; i.e. recombinant monoclonal antibodies or decoy receptors that block inflammatory cytokine function. An important mode of action shared by these drugs is that they reduce inflammatory leukocyte recruitment and dampen down adaptive immune responses. Although current anti-inflammatory drugs have different targets and widely different modes of action, they all share a common side effect, namely a potentially life-threatening reduction in the host immune response to infectious disease. In a recent review Ira Tabas and Chris Glass give a good discussion of the challenges of developing new anti-inflammatory drugs and how we might achieve selective dampening of myeloid cell recruitment without compromise of ‘first responders’ to infectious disease and wound repair (71). One promising approach might be to target pro-resolving agents specifically to sites of chronic inflammation, a proof of concept of this approach using nanoparticles was recently published by Tabas and co-workers(72). Unanswered Questions and Future Directions in Inflammation Biology Following our brief overview of the role of myeloid cells in inflammation we identify seven areas where further research is needed and offer seven questions for those working in the field. (1) Macrophage differentiation and plasticity in vivo For an excellent historical overview of the macrophage M0, M1, M2 concept and a critical assessment of the current literature the authors thoroughly recommend a recent review by Fernando Martinez and Siamon Gordon (73). It is tempting to think about macrophage subsets in the same way we think about T cell subsets in disease. However, advances in flow cytometry, including single cell detection of cytokine production, have thrown up many more T cell classifications than the simplistic Th1, Th2, Th17 classifications that we previously used to explain the role of T cells in disease pathogenesis. Key questions for the field include ‘Do M1 macrophages, defined by surface expression of a handful of different markers really do something substantially different from ‘common or garden’ F4/80hi tissue resident macrophages in the context of an ongoing inflammatory response?’ Similarly, ‘Do M2 (alternatively activated) macrophages, defined in vitro by a cluster of markers, really enhance tissue repair in the context of inflammation resolution or wound repair? Put another way; ‘How plastic is macrophage differentiation within tissues? The issue of macrophage plasticity in vivo has been placed centre stage by two important papers published in 2014 (74, 75). Lavin et al. elegantly demonstrated the plasticity of tissue resident macrophage differentiation by taking F4/80hi peritoneal macrophages from CD45.1 donor mice and instilling them into the lungs of CD45.2 recipient mice (without irradiation). In parallel they transferred F4/80hi alveolar macrophages from CD45.1 donor mice and injected them into the peritoneum of CD45.2 recipient mice. CD45.1 donor cells were recovered from their new tissue microenvironments three weeks later and their transcriptomes and chromatin marks were analysed and compared to those of resident peritoneal and alveolar macrophages in the same tissue. Strikingly Lavin et al showed a complete re-programming of chromatin marks and gene expression patterns in donor macrophages so as to match the macrophages already resident in the recipient tissue. Combined with the intellectual framework provided by Chris Glass and co-workers from their studies of enhancer transcripts and transcription factor binding in myeloid cells we now have a much better understanding of how macrophage subset gene expression patterns are established and re-programmed(76). Building on these seminal studies, we need develop robust methodologies for switching specific macrophage M1 and M2 functions on or off in situ. Such an experimental approach will allow us to move from observing changes in gene expression to actually changing macrophage cellular behaviours within a site of chronic inflammation. The application of optogenetic technologies, until now largely confined to the CNS, might be one way to achieve this ambitious goal. In the future could changing macrophage behaviour in chronic inflammation be used as a novel therapeutic modality, e.g. enhancing macrophage emigration from atherosclerotic plaques, or changing the behaviour of tumour associated macrophages? (2) Functional assays for pro-resolution macrophages Over the last 30 years intense study of the molecular biology of inflammatory mediators, be they pro- or anti- inflammatory, has occurred at the expense of studying the cell biology of tissue repair and healing. Inflammation research would benefit greatly from developing useful ex vivo and in vivo models of tissue repair and fibrosis. Ideally such models should incorporate features of (patho)physiological ECM and tissue architecture, e.g. 3D cultures rather 2D tissue cultures on plastic. In recent papers Dalli and Serhan have shown that sulfido-conjugates of the pro-resolution lipid mediator maresin hasten the resolution of acute inflammatory responses. To test the ability of these mediators to promote tissue regeneration they have turned to tissue regeneration models using a Planaria flatworm model (77, 78). Tissue repair in mammals exhibits significant differences to the tissue regeneration seen in flatworms and urodeles following amputation. This important caveat emphasizes the need for developing more model systems to study the important process of tissue repair and healing. Will a better understanding of successful tissue repair processes in mammals allow us to develop treatments to prevent the irreversible effects of chronic inflammation - tissue destruction and fibrosis? (3) Time of day and time of life – circadian rhythms and Inflamm-aging Nearly all aspects of mammalian physiology have been shown to be strongly influenced by the time of day. A series of experiments from the laboratory of Ajay Chawla elegantly demonstrated that even something as critical to the host as response to Listeria monocytogenes infection is susceptible to circadian rhythm. Having demonstrated a 2–3 fold circadian oscillation in Ly6Chi monocyte numbers in blood and spleen Nguyen et al. infected two groups of C57BL/6J mice being kept on a 12 hour light / dark cycle with the same intraperitoneal dose of L. monocytogenes8 hours apart in terms of their Zeitgeber time (ZT). Mice infected at ZT8 had reduced bacterial counts in blood and tissues 2 days post infection compared to mice infected at ZT0 and this was correlated with improved recruitment of Ly6Chi monocytes to the site of infection. Mice with myeloid-specific deletion of the clock gene Bmal1 showed no rhythmic alterations in Ly6Chi monocyte numbers in blood, spleen or bone marrow and none of the circadian variation in chemokine gene expression seen in wild-type animals (79). The impact of circadian oscillation on innate immune cell numbers and myeloid cell gene expression patterns may explain the link between shift work, altered sleep patterns and increased risk of obesity, diabetes, certain cancers and cardiovascular disease (80). Aging populations, including our own, are characterized by chronic, low-grade inflammation often accompanied by elevated cortisol levels. This phenomenon merits further investigation because with few exceptions all age related diseases have a strong inflammatory component. The term ‘Inflamm-aging’ was coined by Franchesi and colleagues in 2000 and has become somewhat of a ‘poster child’ for systems biology ever since (81). In Figure 3 we have tried to represent key features of inflamm-aging in an idealized time course of inflammatory responses to a single inflammatory stimulus. The first point to note is that the inflammatory response in older subjects starts from an elevated baseline at t=0 but an important and key question is whether older experimental subjects show an elevated maximal response to the same inflammatory stimulus and whether the resolution of inflammation occurs more slowly in comparison with younger test subjects. In contrast to studying how the inflammatory response varies in inbred mouse strains over a 24-hour period, experimental investigation of how inflammatory responses are altered in older cohorts of experimental animals will be challenging but could bring novel mechanistic insights. Studies on exactly how inflammatory responses in humans change with age will be purely comparative (for instance GWAS of people living to 100 years of age), but studying the efficacy of immune responses to say influenza vaccines in different human cohorts might be a promising place to start teasing apart changes in the immune response with age. Future clinical and pre-clinical studies should take more account of the age of participants and the time of day when studies are performed. For instance does the efficacy of anti-inflammatory drugs vary with the time of delivery and/or age of the patient? Another important challenge will be to decide whether anti-inflammatory / pro-resolution agents should be given in extended versus short release formulation. The optimal drug formulation may well depend upon the ‘target’ and the specific disease. (4) Micro RNAs – Nature’s own cytokine network regulators? Micro RNAs (miRNAs) are estimated to modulate the expression of ~30% of all protein encoding genes in the mammalian genome so it is not surprising that multiple miRNAs have been shown to modulate myeloid cell inflammatory responses in a number of different settings. Silencing miRNA expression in vivo through the use of chemically modified oligonucleotides called ‘antagomirs’ has emerged as a new therapeutic modality for the development of novel anti-inflammatory drugs. The path to randomized clinical trials faces obvious problems, not least the mode of delivery and targeting antagomirs or miRNAs to the site of inflammation. However, the recent FDA approval of KYNAMRO, mipomersen sodium of a closely related therapeutic class of molecules- anti-sense oligonucleotides (ASOs)- gives rise to optimism for this approach. In a recent report Wang et al compared the efficacy of systemic delivery of an anti-miR21 compound versus the same reagent delivered via a drug eluting stent in an experimental animal model of in stent restenosis. The authors showed that an anti-miR21 oligonucleotide could prevent in stent restenosis in a balloon injured human internal mammary artery transplanted into nude rats equally well with either delivery method, but an anti-miR21 coated stent gave rise to fewer side effects(82). Will a better understanding of the roles miRNA splay in modulating inflammatory responses lead to the development of new drugs for regulating cytokine networks? (5) Metabolic inflammation and obesity -mediators and microbiota Increasing levels of obesity in the developed world have led to an alarming increase in the incidence of type 2 diabetes and cardiovascular disease. These conditions are associated with increased systemic inflammation and leukocyte infiltration into white adipose tissue (WAT). Healthy WAT in lean subjects is characterized by a population of M2 macrophages with the type 2 cytokines IL-4 and IL-13 for M2 polarisation being provided by eosinophils, which are present in reduced numbers in obese WAT. Furthermore, pro-inflammatory monocytes recruited into WAT differentiate into M1 macrophages that contribute to insulin resistance and dyslipidemia (83). It is not yet clear whether targeting leukocyte recruitment to and leukocyte activation within adipose tissue will be a viable strategy to prevent the conversion of obesity and metabolic syndrome into type 2 diabetes. Obesity is one of the more obvious manifestations of a “Western-Lifestyle” but some immunologists have postulated a link between the increased incidence of allergies, asthma and autoimmunity with changes in diet and host microbiota generated metabolites (84). Studies in pre-clinical models of inflammation have revealed the importance of a series of metabolite receptors including GPR41, GPR43, GPR109 and GPR120 expressed by innate immune cells that can have potent anti-inflammatory actions. One of the first papers to reveal a link between diet, the gut micobiota, short chain fatty acid (SCFAs) and inflammation was provided by the laboratory of Charles Mackay in 2009. Maslowski et al showed that Gpr43−/− mice had excessive inflammation in models of colitis, arthritis and asthma. Germ free mice showed a similar exacerbation of inflammatory responses consistent with bacterial fermentation of complex carbohydrate generating SCFAs for stimulation of GPR43’s anti-inflammatory actions (85). In the future will we be able to exploit our increasing knowledge of crosstalk between the microbiota, microbial metabolites and host metabolite receptors to modulate host innate and adaptive immune responses for therapeutic benefit ? (6) Towards a more molecular definition of inflammation In all pathology textbooks inflammation is defined by timing, being either short-term (acute) or long-standing (chronic). An alternative classification of inflammation could be envisaged that takes account of the inciting stimulus, e.g. metabolic inflammation, or the inflammatory cytokines that drive a specific chronic inflammatory disease process e.g. TNF versus IL-6 driven disease (86). Rapid advances in combined liquid chromatography mass spectroscopy (LC-MS) techniques for inflammatory exudates may soon allow us to draw up alternative classifications based on tissue responses to infectious diseases. A striking example is provided by two recent papers that followed the time course of lipid mediator appearance in the lungs of humans hospitalized with influenza infection and mice infected with the same titer of different strains of influenza that differed in their virulence. Morita at al and Tam et al. undertook a detailed analysis of host lipid mediators and their regulation during the course of influenza infection and correlated these changes with viral replication, host immune responses including transcriptomics and measurement of cytokine and chemokine profiles. One important result from these two extensive lipidomics studies in clinical cohorts and pre-clinical infection models was the identification of the endogenous lipid mediator protect in D1 as a potent inhibitor of influenza infection. In the longer term, advances in LC-MS technology and our ability to interpret large data sets may allow us to better assess the severity of respiratory distress and virus pathology through analysis of nasal swabs of patients hospitalized with influenza (87, 88). Ultimately, will better identification of the molecules and receptors that drive acute and chronic inflammation lead to better treatments and better patient outcomes? (7) Future anti-inflammatory drug targets and clinical trials In twenty years time when the current frontline biologics du jour such as anti TNFα, anti IL-1β, anti IL-6, anti IL-17 etc. have gone generic, we may well view these pioneering monoclonal antibodies and decoy receptors as excellent ‘test reagents’ used to identify the specific cytokine networks or specific cell types that cause chronic inflammatory responses. Currently a sub-group of RA and MS patients seem to respond better to B-lymphocyte depleting antibodies, such as the anti-CD20 monoclonal Rituximab, than they do to anti-TNFα biologics. In the future we would like to identify patients whose disease will respond better to anti B-lymphocyte therapy as soon as possible to avoid ‘hit or miss’ dosing with powerful and expensive drugs. Careful analysis of current clinical outcomes combined with biomarker analysis and pharmacogenomics studies will have the twin benefit of improved outcomes for patients and new insights in the pathogenesis of chronic inflammatory disease in human cohorts(86). Rheumatologists have long sought after biomarkers, haplotypes, environmental factors and genetic markers that can identify patients at increased risk of developing rheumatoid arthritis. Recently biomarker panels have been expanded to include serum titers of anti-citrullinated peptide antibodies (ACPA). Early reports have given rise to the idea that this class of auto-antibodies may be driving chronic inflammation in tissues other than the joints, most notably within the lungs of patients before they present with joint inflammation and are diagnosed with RA (86). If ACPA screening could identify people who will go on to develop debilitating RA up to a decade later, what will we do with this knowledge? Should we treat this pre-RA lung inflammation aggressively with systemic anti-cytokine biologics or should we use inhaled glucocorticoids? Alternatively, should we call back people with high ACPA titers for radiological assessment every year and initiate aggressive anti-inflammatory treatment as soon as we see any sign of joint disease? An analogy can perhaps be drawn with the link between elevated plasma LDL, accelerated atherosclerosis and the increased risk of cardiovascular disease where primary prevention emphasizes lifestyle changes and prescription of statins. In the future will we start prescribing anti -inflammatory drugs for people with pre-diabetes, pre-RA or even pre-dementia? DRG thanks Siamon Gordon for introducing him to macrophage biology and Thomas Schall for introducing him to chemokines and their receptors. We thank Lewis Taylor, and Sophia Valaris for constructive criticisms. Work in the laboratory of DRG is supported by the British Heart Foundation (RG/10/15/28578, RG/15/10/23915). Work in the laboratory of EAF is supported by the US National Institutes of Health (HL098055 and DK095684). We are grateful to the Royal Society for an International Exchange Grant (IE120747). Figure 1 Time course of a typical acute inflammatory response A schematic representation of the ideal outcome of an acute inflammatory response, i.e. resolution, is shown as a dashed line. Two potential outcomes leading to significant clinical sequelae are shown, hyper-acute inflammation e.g. septic shock and non-resolving, chronic inflammation e.g. rheumatoid arthritis. Figure 2 The inflammatory set point hypothesis This schematic representation highlights the balance between locally produced pro-inflammatory and endogenous anti-inflammatory / pro-resolution mediators in determining the magnitude of the inflammatory response in response to a given stimulus. Figure 3 Inflamm-aging, do inflammatory response change with age? This schematic representation compares the ‘normal’ inflammatory response (solid line) with the inflammatory response seen in aged populations (dashed line). Aged populations show increased basal levels of systemic inflammation and may show differences in the magnitude of the response to inflammatory stimuli and/or altered resolution. Recent papers 1 Gordon S Elie Metchnikoff: father of natural immunity Eur J Immunol 2008 38 12 3257 64 10.1002/eji.200838855 19039772 2 Shrive AK Metcalfe AM Cartwright JR Greenhough TJ C-reactive protein and SAP-like pentraxin are both present in Limulus polyphemus haemolymph: crystal structure of Limulus SAP J Mol Biol 1999 290 5 997 1008 10.1006/jmbi.1999.2956 10438598 3 Tharia HA Shrive AK Mills JD Arme C Williams GT Greenhough TJ Complete cDNA sequence of SAP-like pentraxin from Limulus polyphemus: implications for pentraxin evolution J Mol Biol 2002 316 3 583 97 10.1006/jmbi.2001.5356 11866519 4 Mantovani A Valentino S Gentile S Inforzato A Bottazzi B Garlanda C The long pentraxin PTX3: a paradigm for humoral pattern recognition molecules Ann N Y Acad Sci 2013 1285 1 14 10.1111/nyas.12043 23527487 5 Simons JP Loeffler JM Al-Shawi R Ellmerich S Hutchinson WL Tennent GA Petrie A Raynes JG de Souza JB Lawrence RA Read KD Pepys MB C-reactive protein is essential for innate resistance to pneumococcal infection Immunology 2014 142 3 414 20 10.1111/imm.12266 24673624 6 Du Clos TW Pentraxins: structure, function, and role in inflammation ISRN Inflamm 2013 2013 379040 10.1155/2013/379040 24167754 7 Garlanda C Hirsch E Bozza S Salustri A De Acetis M Nota R Maccagno A Riva F Bottazzi B Peri G Doni A Vago L Botto M De Santis R Carminati P Siracusa G Altruda F Vecchi A Romani L Mantovani A Non-redundant role of the long pentraxin PTX3 in anti-fungal innate immune response Nature 2002 420 6912 182 6 10.1038/nature01195 12432394 8 Reading PC Bozza S Gilbertson B Tate M Moretti S Job ER Crouch EC Brooks AG Brown LE Bottazzi B Romani L Mantovani A Antiviral activity of the long chain pentraxin PTX3 against influenza viruses J Immunol 2008 180 5 3391 8 18292565 9 Oikonomopoulou K Ricklin D Ward PA Lambris JD Interactions between coagulation and complement--their role in inflammation Semin Immunopathol 2012 34 1 151 65 10.1007/s00281-011-0280-x 21811895 10 Gerard NP Gerard C The chemotactic receptor for human C5a anaphylatoxin Nature 1991 349 6310 614 7 10.1038/349614a0 1847994 11 Lemaitre B Nicolas E Michaut L Reichhart JM Hoffmann JA The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults Cell 1996 86 6 973 83 8808632 12 Janeway CA Jr Approaching the asymptote? Evolution and revolution in immunology Cold Spring Harb Symp Quant Biol 1989 54 Pt 1 1 13 13 Janeway CA Jr Medzhitov R Innate immune recognition Annu Rev Immunol 2002 20 197 216 10.1146/annurev.immunol.20.083001.084359 11861602 14 Broxmeyer HE Orschell CM Clapp DW Hangoc G Cooper S Plett PA Liles WC Li X Graham-Evans B Campbell TB Calandra G Bridger G Dale DC Srour EF Rapid mobilization of murine and human hematopoietic stem and progenitor cells with AMD3100, a CXCR4 antagonist J Exp Med 2005 201 8 1307 18 10.1084/jem.20041385 15837815 15 Zabel BA Rott A Butcher EC Leukocyte chemoattractant receptors in human disease pathogenesis Annu Rev Pathol 2015 10 51 81 10.1146/annurev-pathol-012513-104640 25387059 16 Boring L Gosling J Cleary M Charo IF Decreased lesion formation in CCR2−/− mice reveals a role for chemokines in the initiation of atherosclerosis Nature 1998 394 6696 894 7 10.1038/29788 9732872 17 Mackay CR Moving targets: cell migration inhibitors as new anti-inflammatory therapies Nat Immunol 2008 9 9 988 98 10.1038/ni.f.210 18711436 18 White GE Iqbal AJ Greaves DR CC chemokine receptors and chronic inflammation--therapeutic opportunities and pharmacological challenges Pharmacol Rev 2013 65 1 47 89 10.1124/pr.111.005074 23300131 19 Martinon F Burns K Tschopp J The inflammasome: a molecular platform triggering activation of inflammatory caspases and processing of pro IL-beta Mol Cell 2002 10 2 417 26 12191486 20 Martinon F Petrilli V Mayor A Tardivel A Tschopp J Gout-associated uric acid crystals activate the NALP3 inflammasome Nature 2006 440 7081 237 41 10.1038/nature04516 16407889 21 Chen CJ Kono H Golenbock D Reed G Akira S Rock KL Identification of a key pathway required for the sterile inflammatory response triggered by dying cells Nat Med 2007 13 7 851 6 10.1038/nm1603 17572686 22 Chan JK Roth J Oppenheim JJ Tracey KJ Vogl T Feldmann M Horwood N Nanchahal J Alarmins: awaiting a clinical response J Clin Invest 2012 122 8 2711 9 10.1172/JCI62423 22850880 23 Kataoka H Kono H Patel Z Kimura Y Rock KL Evaluation of the contribution of multiple DAMPs and DAMP receptors in cell death-induced sterile inflammatory responses PLoS One 2014 9 8 e104741 10.1371/journal.pone.0104741 25127469 24 Nathan C Ding A Nonresolving inflammation Cell 2010 140 6 871 82 10.1016/j.cell.2010.02.029 20303877 25 Huang Z Zhao C Chen Y Cowell JA Wei G Kultti A Huang L Thompson CB Rosengren S Frost GI Shepard HM Recombinant human hyaluronidase PH20 does not stimulate an acute inflammatory response and inhibits lipopolysaccharide-induced neutrophil recruitment in the air pouch model of inflammation J Immunol 2014 192 11 5285 95 10.4049/jimmunol.1303060 24778442 26 Cash JL White GE Greaves DR Chapter 17. Zymosan-induced peritonitis as a simple experimental system for the study of inflammation Methods Enzymol 2009 461 379 96 10.1016/S0076-6879(09)05417-2 19480928 27 Di Rosa M Biological properties of carrageenan J Pharm Pharmacol 1972 24 2 89 102 4402944 28 D’Agostino G La Rana G Russo R Sasso O Iacono A Esposito E Raso GM Cuzzocrea S Lo Verme J Piomelli D Meli R Calignano A Acute intracerebroventricular administration of palmitoylethanolamide, an endogenous peroxisome proliferator-activated receptor-alpha agonist, modulates carrageenan-induced paw edema in mice J Pharmacol Exp Ther 2007 322 3 1137 43 10.1124/jpet.107.123265 17565008 29 Sedgwick AD Sin YM Edwards JC Willoughby DA Increased inflammatory reactivity in newly formed lining tissue J Pathol 1983 141 4 483 95 10.1002/path.1711410406 6663391 30 Notley CA Inglis JJ Alzabin S McCann FE McNamee KE Williams RO Blockade of tumor necrosis factor in collagen-induced arthritis reveals a novel immunoregulatory pathway for Th1 and Th17 cells J Exp Med 2008 205 11 2491 7 10.1084/jem.20072707 18936235 31 Kouskoff V Korganow AS Duchatelle V Degott C Benoist C Mathis D Organ-specific disease provoked by systemic autoimmunity Cell 1996 87 5 811 22 8945509 32 Jacobs JP Ortiz-Lopez A Campbell JJ Gerard CJ Mathis D Benoist C Deficiency of CXCR2, but not other chemokine receptors, attenuates autoantibody-mediated arthritis in a murine model Arthritis Rheum 2010 62 7 1921 32 10.1002/art.27470 20506316 33 Moore KJ Tabas I Macrophages in the pathogenesis of atherosclerosis Cell 2011 145 3 341 55 10.1016/j.cell.2011.04.005 21529710 34 Klotz L Norman S Vieira JM Masters M Rohling M Dube KN Bollini S Matsuzaki F Carr CA Riley PR Cardiac lymphatics are heterogeneous in origin and respond to injury Nature 2015 522 7554 62 7 10.1038/nature14483 25992544 35 Newson J Stables M Karra E Arce-Vargas F Quezada S Motwani M Mack M Yona S Audzevich T Gilroy DW Resolution of acute inflammation bridges the gap between innate and adaptive immunity Blood 2014 124 11 1748 64 10.1182/blood-2014-03-562710 25006125 36 Clarke TB Davis KM Lysenko ES Zhou AY Yu Y Weiser JN Recognition of peptidoglycan from the microbiota by Nod1 enhances systemic innate immunity Nat Med 2010 16 2 228 31 10.1038/nm.2087 20081863 37 Ferrer R Martin-Loeches I Phillips G Osborn TM Townsend S Dellinger RP Artigas A Schorr C Levy MM Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: results from a guideline-based performance improvement program Crit Care Med 2014 42 8 1749 55 10.1097/CCM.0000000000000330 24717459 38 Collins PE Carmody RJ The Regulation of Endotoxin Tolerance and its Impact on Macrophage Activation Crit Rev Immunol 2015 35 4 293 323 26757393 39 Arts RJ Gresnigt MS Joosten LA Netea MG Cellular metabolism of myeloid cells in sepsis J Leukoc Biol 2016 10.1189/jlb.4MR0216-066R 40 Bannenberg GL Chiang N Ariel A Arita M Tjonahen E Gotlinger KH Hong S Serhan CN Molecular circuits of resolution: formation and actions of resolvins and protectins J Immunol 2005 174 7 4345 55 15778399 41 Yazid S Norling LV Flower RJ Anti-inflammatory drugs, eicosanoids and the annexin A1/FPR2 anti-inflammatory system Prostaglandins Other Lipid Mediat 2012 98 3–4 94 100 10.1016/j.prostaglandins.2011.11.005 22123264 42 Morris T Stables M Hobbs A de Souza P Colville-Nash P Warner T Newson J Bellingan G Gilroy DW Effects of low-dose aspirin on acute inflammatory responses in humans J Immunol 2009 183 3 2089 96 10.4049/jimmunol.0900477 19597002 43 Sieweke MH Allen JE Beyond stem cells: self-renewal of differentiated macrophages Science 2013 342 6161 1242974 10.1126/science.1242974 24264994 44 Gautier EL Jakubzick C Randolph GJ Regulation of the migration and survival of monocyte subsets by chemokine receptors and its relevance to atherosclerosis Arterioscler Thromb Vasc Biol 2009 29 10 1412 8 10.1161/ATVBAHA.108.180505 19759373 45 Geissmann F Manz MG Jung S Sieweke MH Merad M Ley K Development of monocytes, macrophages, and dendritic cells Science 2010 327 5966 656 61 10.1126/science.1178331 20133564 46 Si Y Tsou CL Croft K Charo IF CCR2 mediates hematopoietic stem and progenitor cell trafficking to sites of inflammation in mice J Clin Invest 2010 120 4 1192 203 10.1172/JCI40310 20234092 47 White GE McNeill E Channon KM Greaves DR Fractalkine promotes human monocyte survival via a reduction in oxidative stress Arterioscler Thromb Vasc Biol 2014 34 12 2554 62 10.1161/ATVBAHA.114.304717 25359863 48 Mildner A Yona S Jung S A close encounter of the third kind: monocyte-derived cells Adv Immunol 2013 120 69 103 10.1016/B978-0-12-417028-5.00003-X 24070381 49 Geissmann F Jung S Littman DR Blood monocytes consist of two principal subsets with distinct migratory properties Immunity 2003 19 1 71 82 12871640 50 Woollard KJ Geissmann F Monocytes in atherosclerosis: subsets and functions Nat Rev Cardiol 2010 7 2 77 86 10.1038/nrcardio.2009.228 20065951 51 Lu H Huang D Saederup N Charo IF Ransohoff RM Zhou L Macrophages recruited via CCR2 produce insulin-like growth factor-1 to repair acute skeletal muscle injury FASEB J 2011 25 1 358 69 10.1096/fj.10-171579 20889618 52 Swirski FK Nahrendorf M Leukocyte behavior in atherosclerosis, myocardial infarction, and heart failure Science 2013 339 6116 161 6 10.1126/science.1230719 23307733 53 Hilgendorf I Gerhardt LM Tan TC Winter C Holderried TA Chousterman BG Iwamoto Y Liao R Zirlik A Scherer-Crosbie M Hedrick CC Libby P Nahrendorf M Weissleder R Swirski FK Ly -6Chigh monocytes depend on Nr4a1 to balance both inflammatory and reparative phases in the infarcted myocardium Circ Res 2014 114 10 1611 22 10.1161/CIRCRESAHA.114.303204 24625784 54 Leuschner F Rauch PJ Ueno T Gorbatov R Marinelli B Lee WW Dutta P Wei Y Robbins C Iwamoto Y Sena B Chudnovskiy A Panizzi P Keliher E Higgins JM Libby P Moskowitz MA Pittet MJ Swirski FK Weissleder R Nahrendorf M Rapid monocyte kinetics in acute myocardial infarction are sustained by extramedullary monocytopoiesis J Exp Med 2012 209 1 123 37 10.1084/jem.20111009 22213805 55 Dutta P Sager HB Stengel KR Naxerova K Courties G Saez B Silberstein L Heidt T Sebas M Sun Y Wojtkiewicz G Feruglio PF King K Baker JN van der Laan AM Borodovsky A Fitzgerald K Hulsmans M Hoyer F Iwamoto Y Vinegoni C Brown D Di Carli M Libby P Hiebert SW Scadden DT Swirski FK Weissleder R Nahrendorf M Myocardial Infarction Activates CCR2(+) Hematopoietic Stem and Progenitor Cells Cell Stem Cell 2015 16 5 477 87 10.1016/j.stem.2015.04.008 25957903 56 Dennis EA Norris PC Eicosanoid storm in infection and inflammation Nat Rev Immunol 2015 15 8 511 23 10.1038/nri3859 26139350 57 Norris PC Gosselin D Reichart D Glass CK Dennis EA Phospholipase A2 regulates eicosanoid class switching during inflammasome activation Proc Natl Acad Sci U S A 2014 111 35 12746 51 10.1073/pnas.1404372111 25139986 58 Wallace JL Ianaro A Flannigan KL Cirino G Gaseous mediators in resolution of inflammation Semin Immunol 2015 27 3 227 33 10.1016/j.smim.2015.05.004 26095908 59 Otterbein LE Bach FH Alam J Soares M Tao Lu H Wysk M Davis RJ Flavell RA Choi AM Carbon monoxide has anti-inflammatory effects involving the mitogen-activated protein kinase pathway Nat Med 2000 6 4 422 8 10.1038/74680 10742149 60 Wallace JL Wang R Hydrogen sulfide-based therapeutics: exploiting a unique but ubiquitous gasotransmitter Nat Rev Drug Discov 2015 14 5 329 45 10.1038/nrd4433 25849904 61 O’Neill LA Retrospective. Jurg Tschopp (1951–2011) Science 2011 332 6030 679 10.1126/science.1207046 21551056 62 Guo H Callaway JB Ting JP Inflammasomes: mechanism of action, role in disease, and therapeutics Nat Med 2015 21 7 677 87 10.1038/nm.3893 26121197 63 Ip WK Medzhitov R Macrophages monitor tissue osmolarity and induce inflammatory response through NLRP3 and NLRC4 inflammasome activation Nat Commun 2015 6 6931 10.1038/ncomms7931 25959047 64 Vanaja SK Rathinam VA Fitzgerald KA Mechanisms of inflammasome activation: recent advances and novel insights Trends Cell Biol 2015 25 5 308 15 10.1016/j.tcb.2014.12.009 25639489 65 Coll RC Robertson AA Chae JJ Higgins SC Munoz-Planillo R Inserra MC Vetter I Dungan LS Monks BG Stutz A Croker DE Butler MS Haneklaus M Sutton CE Nunez G Latz E Kastner DL Mills KH Masters SL Schroder K Cooper MA O’Neill LA A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases Nat Med 2015 21 3 248 55 10.1038/nm.3806 25686105 66 McDermott MF Aksentijevich I Galon J McDermott EM Ogunkolade BW Centola M Mansfield E Gadina M Karenko L Pettersson T McCarthy J Frucht DM Aringer M Torosyan Y Teppo AM Wilson M Karaarslan HM Wan Y Todd I Wood G Schlimgen R Kumarajeewa TR Cooper SM Vella JP Amos CI Mulley J Quane KA Molloy MG Ranki A Powell RJ Hitman GA O’Shea JJ Kastner DL Germline mutations in the extracellular domains of the 55 kDa TNF receptor, TNFR1, define a family of dominantly inherited autoinflammatory syndromes Cell 1999 97 1 133 44 10199409 67 Henderson C Goldbach-Mansky R Monogenic autoinflammatory diseases: new insights into clinical aspects and pathogenesis Curr Opin Rheumatol 2010 22 5 567 78 10.1097/BOR.0b013e32833ceff4 20671522 68 de Jesus AA Canna SW Liu Y Goldbach-Mansky R Molecular mechanisms in genetically defined autoinflammatory diseases: disorders of amplified danger signaling Annu Rev Immunol 2015 33 823 74 10.1146/annurev-immunol-032414-112227 25706096 69 Deutschman CS Tracey KJ Sepsis: current dogma and new perspectives Immunity 2014 40 4 463 75 10.1016/j.immuni.2014.04.001 24745331 70 Ward PA New approaches to the study of sepsis EMBO Mol Med 2012 4 12 1234 43 10.1002/emmm.201201375 23208733 71 Tabas I Glass CK Anti-inflammatory therapy in chronic disease: challenges and opportunities Science 2013 339 6116 166 72 10.1126/science.1230720 23307734 72 Fredman G Kamaly N Spolitu S Milton J Ghorpade D Chiasson R Kuriakose G Perretti M Farokhzad O Tabas I Targeted nanoparticles containing the proresolving peptide Ac2-26 protect against advanced atherosclerosis in hypercholesterolemic mice Sci Transl Med 2015 7 275 275ra20 10.1126/scitranslmed.aaa1065 73 Martinez FO Gordon S The M1 and M2 paradigm of macrophage activation: time for reassessment F1000Prime Rep 2014 6 13 10.12703/P6-13 24669294 74 Gosselin D Link VM Romanoski CE Fonseca GJ Eichenfield DZ Spann NJ Stender JD Chun HB Garner H Geissmann F Glass CK Environment drives selection and function of enhancers controlling tissue-specific macrophage identities Cell 2014 159 6 1327 40 10.1016/j.cell.2014.11.023 25480297 75 Lavin Y Winter D Blecher-Gonen R David E Keren-Shaul H Merad M Jung S Amit I Tissue-resident macrophage enhancer landscapes are shaped by the local microenvironment Cell 2014 159 6 1312 26 10.1016/j.cell.2014.11.018 25480296 76 Glass CK Genetic and genomic approaches to understanding macrophage identity and function Arterioscler Thromb Vasc Biol 2015 35 4 755 62 10.1161/ATVBAHA.114.304051 25745059 77 Dalli J Ramon S Norris PC Colas RA Serhan CN Novel proresolving and tissue-regenerative resolvin and protectin sulfido-conjugated pathways FASEB J 2015 29 5 2120 36 10.1096/fj.14-268441 25713027 78 Serhan CN Dalli J Karamnov S Choi A Park CK Xu ZZ Ji RR Zhu M Petasis NA Macrophage proresolving mediator maresin 1 stimulates tissue regeneration and controls pain FASEB J 2012 26 4 1755 65 10.1096/fj.11-201442 22253477 79 Nguyen KD Fentress SJ Qiu Y Yun K Cox JS Chawla A Circadian gene Bmal1 regulates diurnal oscillations of Ly6C(hi) inflammatory monocytes Science 2013 341 6153 1483 8 10.1126/science.1240636 23970558 80 Castanon-Cervantes O Wu M Ehlen JC Paul K Gamble KL Johnson RL Besing RC Menaker M Gewirtz AT Davidson AJ Dysregulation of inflammatory responses by chronic circadian disruption J Immunol 2010 185 10 5796 805 10.4049/jimmunol.1001026 20944004 81 Castellani GC Menichetti G Garagnani P Giulia Bacalini M Pirazzini C Franceschi C Collino S Sala C Remondini D Giampieri E Mosca E Bersanelli M Vitali S Valle IF Lio P Milanesi L Systems medicine of inflammaging Brief Bioinform 2016 17 3 527 40 10.1093/bib/bbv062 26307062 82 Wang D Deuse T Stubbendorff M Chernogubova E Erben RG Eken SM Jin H Li Y Busch A Heeger CH Behnisch B Reichenspurner H Robbins RC Spin JM Tsao PS Schrepfer S Maegdefessel L Local MicroRNA Modulation Using a Novel Anti-miR-21-Eluting Stent Effectively Prevents Experimental In-Stent Restenosis Arterioscler Thromb Vasc Biol 2015 35 9 1945 53 10.1161/ATVBAHA.115.305597 26183619 83 Carvalheira JB Qiu Y Chawla A Blood spotlight on leukocytes and obesity Blood 2013 122 19 3263 7 10.1182/blood-2013-04-459446 24065242 84 Thorburn AN Macia L Mackay CR Diet, metabolites, and “western-lifestyle” inflammatory diseases Immunity 2014 40 6 833 42 10.1016/j.immuni.2014.05.014 24950203 85 Maslowski KM Vieira AT Ng A Kranich J Sierro F Yu D Schilter HC Rolph MS Mackay F Artis D Xavier RJ Teixeira MM Mackay CR Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43 Nature 2009 461 7268 1282 6 10.1038/nature08530 19865172 86 Schett G Elewaut D McInnes IB Dayer JM Neurath MF How cytokine networks fuel inflammation: Toward a cytokine-based disease taxonomy Nat Med 2013 19 7 822 4 10.1038/nm.3260 23836224 87 Morita M Kuba K Ichikawa A Nakayama M Katahira J Iwamoto R Watanebe T Sakabe S Daidoji T Nakamura S Kadowaki A Ohto T Nakanishi H Taguchi R Nakaya T Murakami M Yoneda Y Arai H Kawaoka Y Penninger JM Arita M Imai Y The lipid mediator protectin D1 inhibits influenza virus replication and improves severe influenza Cell 2013 153 1 112 25 10.1016/j.cell.2013.02.027 23477864 88 Tam VC Quehenberger O Oshansky CM Suen R Armando AM Treuting PM Thomas PG Dennis EA Aderem A Lipidomic profiling of influenza infection identifies mediators that induce and resolve inflammation Cell 2013 154 1 213 27 10.1016/j.cell.2013.05.052 23827684

PMC005xxxxxx/PMC5119647.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8006349 6600 Placenta Placenta Placenta 0143-4004 1532-3102 27697215 5119647 10.1016/j.placenta.2016.07.005 NIHMS808656 Article Differential expression of Toll-like receptors in the human placenta across early gestation Pudney Jeffrey 2 He Xianbao 1 Masheeb Zahrah 2 Kindelberger David W. 3 Kuohung Wendy 2* Ingalls Robin R. 1* 1 Section of Infectious Disease, Department of Medicine, Boston Medical Center and Boston University School of Medicine, Boston, MA, USA 2 Division of Reproductive Immunology, Department of Obstetrics and Gynecology, Boston Medical Center and Boston University School of Medicine, Boston, MA, USA 3 Department of Pathology and Laboratory Medicine, Boston Medical Center and Boston University School of Medicine, Boston, MA, USA * To whom correspondence should be addressed: Dr. Robin R. Ingalls, EBRC Room 610, 650 Albany St., Boston, MA 02118 USA, ringalls@bu.edu. Dr. Wendy Kuohung. 85 East Concord St., 6th floor, Boston, MA 02118 USA, wkuohung@bu.edu * Contributed equally to this work. 6 8 2016 26 7 2016 10 2016 01 10 2017 46 110 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Toll-like receptors (TLRs) are an essential component of the innate immune system. While a number of studies have described TLR expression in the female reproductive tract, few have examined the temporal expression of TLRs within the human placenta. We hypothesized that the pattern of TLR expression in the placenta changes throughout the first and second trimester, coincident with physiological changes in placental function and the demands of innate immunity. We collected first and second trimester placental tissue and conducted quantitative PCR analysis for TLRs 1-10, followed by immunohistochemistry to define the cell specific expression pattern of a subset of these receptors. Except for the very earliest time points, RNA expression for TLRs 1-10 was stable out to 20 weeks gestation. However, the pattern of protein expression evolved over time. Early first trimester placenta demonstrated a strong, uniform pattern predominantly in the inner villous cytotrophoblast layer. As the placenta matured through the second trimester, both the villous cytotrophoblasts and the pattern of TLR expression within them became disorganized and patchy, with Hofbauer cells now identifiable in the tissue also staining positive. We conclude from this data that placental TLR expression changes over the course of gestation, with a tight barrier of TLRs forming a wall of defense along the cytotrophoblast layer in the early first trimester that breaks down as pregnancy progresses. These data are relevant to understanding placental immunity against pathogen exposure throughout pregnancy and may aid in our understanding of the vulnerable period for fetal exposure to pathogens. innate immunity reproductive immunology pregnancy Toll-like receptors placenta INTRODUCTION The mammalian placenta is a highly specialized organ composed of both maternal and fetal cells that performs vital functions for the developing fetus during gestation. In addition to transfer of nutrients, gas exchange, elimination of waste, and hormone production, the placenta is an important site of immune defense for the fetus against maternal rejection as well as exogenous microbial insults. During pregnancy, the maternal immune system adapts to permit tolerance of the fetal allograft while maintaining defenses against harmful pathogens [reviewed in (1, 2)]. Any disturbance in this tightly regulated balance may lead to a breach in immune defenses, intrauterine infection by bacteria or viruses, and obstetrical complications such as miscarriage, preterm labor, intrauterine growth restriction, and preeclampsia. Elucidating the etiology of intrapartum infection requires identifying not only the infectious agent but also the route of entry into the amniotic cavity. In contrast to the formerly held belief that the amniotic tissues are sterile, a number of studies over the past two years suggest the placenta harbors a unique, low-level microbiome that may be linked to birth outcome (3–7). Surprisingly, these studies demonstrate that the taxonomic profile of the gravid vaginal microbiome was quite dissimilar to that of the placental microbiome, favoring hematogenous dissemination over intrauterine ascension as the more common route of entry of infectious agents. It is therefore likely that immune defenses contained within the placenta, the barrier between maternal and fetal circulation, are critical in preventing the transmission of both commensals and pathogens. Toll-like receptors (TLRs) are a fundamental component of the innate immune system (8) and constitute an important host defense in the placenta against intrapartum infection. This family of transmembrane receptors plays a pivotal role in the recognition of microbial ligands by the innate immune system (9, 10). Toll, initially identified as a receptor involved in embryonic development in Drosophila (11), was found to regulate important antimicrobial responses against fungi in the adult fly (12), an observation that set off an explosion of research to identify similar receptors in humans and mice. At least ten mammalian orthologues of Toll have been identified, and most have been implicated in cellular responses to microbial pathogens [reviewed in (13–17). For example, TLR4 and the secreted protein myeloid differentiation factor 2 (MD-2; also known as lymphocyte antigen 96 or Ly96) are required for recognition of lipopolysaccharide (LPS); TLR2, paired with TLR1 or TLR6, recognizes bacterial lipoproteins and lipoteichoic acid; TLR5 recognizes bacterial flagellin; TLR9 recognizes CpG enriched double-stranded DNA; TLR7 and TLR8 recognize single-stranded RNA (ssRNA); and TLR3 recognizes double stranded RNA (dsRNA) [reviewed (14)]. The TLRs can be sub-divided into surface expressed TLRs (TLR1, -2, -4 and -5) and endosomal expressed TLRs (TLR3, -7, -8, and -9). Upon recognition of pathogen-associated ligands, the TLRs dimerize, initiating a signaling cascade that leads to the secretion of proinflammatory cytokines and antimicrobial peptides, induction of interferon stimulated genes (primarily from the endosomal TLRs and TLR4), as well as activation of the adaptive immune response. In contrast to the abundant data on TLR function, the temporal expression of toll-like receptors (TLRs) throughout pregnancy has not been as well studied. It is known that TLRs 1-10 are expressed in term human placenta, and that TLR2 and TLR5 transcripts appear to rise in association with labor (18). A recent survey of expression and function of TLRs in first trimester cytotrophoblasts suggests that TLRs are broadly expressed at 6–12 weeks gestation (19), although a comparison to second and third trimester placental tissue was not done. Knowledge of temporal shifts in TLR expression at the maternal-fetal interface may be important clinically to establish periods of increased maternal susceptibility to transmit infections such as CMV and Zika virus or optimal windows for treatment of infections that could impact the fetus. In this report, we describe the expression of TLRs in the human placenta at the level of gene expression and protein, reviewing the relevant details of placental histology over the first and second trimester. MATERIALS AND METHODS Study samples First and second trimester placentas were obtained from elective termination of pregnancies under a protocol approved by the Institutional Review Board (IRB) for the Boston University Medical Campus and in accordance with the principles expressed in the Declaration of Helsinki. Criteria for exclusion of placental samples included known current maternal viral or pelvic infection, preexisting fetal demise, and known abnormal fetal karyotype. Termination of pregnancy was performed by suction dilation and evacuation. Samples were divided according to trimester of collection as follows: first trimester, up to 12 weeks; and second trimester, 12–24 weeks gestation. Gestational age was dated according to the subject’s last reported menstrual period and/or ultrasound dating as well as foot measurement when possible. At least 2 placental specimens representing each week of gestation were obtained during the first trimester of pregnancy from 6–12 weeks and during the second trimester from 13–22 weeks. More specific numbers of samples per analysis are provided below. Placental tissue utilized for RNA analysis was stored in RNAlater (Qiagen) for later processing for RNA extraction. The remaining samples for histology were fixed in 10% unbuffered methanol-free formaldehyde and processed for embedding in wax; 5 μm thick sections were cut from each sample and placed on glass slides for histological analysis. RNA isolation Placental tissue was homogenized by using a Tissue Homogenizer LT (Qiagen) and TRIzol (Life Technologies). After pelleting tissue debris, tissue homogenates were transferred to a fresh tube and chloroform was added. After precipitation of protein, the RNA-rich upper aqueous phase was transferred, mixed with an equal volume of 70% ethanol, and then loaded onto RNeasy Mini Kit columns (Qiagen). All RNA was treated with RNase-free DNase (Qiagen) prior to cDNA synthesis. Quantitative reverse transcriptase PCR analysis A total of 11 samples were available for PCR analysis (first trimester, n=6 samples; second trimester, n=5 samples). Quantitative reverse transcription PCR analysis (qRT-PCR) for TLR expression was performed with the TaqMan Expression Assay system (Life Technologies), which utilizes FAM-MGB probes for each target. Primer IDs for the target genes [TLRs 1-10, MD-2, GAPDH, and N-myc downstream-regulated gene 1 (NDRG1)] are shown in Table 1. cDNA preparation was carried out using High Capacity RNA-to-cDNA kit (Life Technologies). The final 20 μl PCR reaction mixture consisted of 10 μl of 2x TaqMan Gene Expression Master mix, 1 μl of cDNA, 8 μl of RNase-free water, and 1 μl of TaqMan Gene Expression Assay mixture containing the target primer. Reactions were performed in a 96-well plate and the housekeeping gene GAPDH was used as internal control. Real-Time PCR was run on an Applied Biosystems StepOnePlus™ Real-Time PCR System using a standard program (50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15s and 60°C for 1 min). PCR data was analyzed using StepOne Software v2.3, and the comparative Ct method to calculate relative quantitation. The Ct from the target genes of each sample was normalized to the Ct of the reference gene GAPDH, to generate the sample delta Ct (ΔCt sample = Ct reference gene GAPDH − Ct target). To compare the gene expression levels between samples of different gestational age, the target genes were normalized to sample 1, which we defined as the earliest gestational age (sample ΔΔCt = sample ΔCt − sample 1ΔCt). To compare the different TLR gene expression levels, the TLR genes were normalized to the expression of TLR10 sample 1, which was the lowest expressed gene by both semi-quantitative PCR and real-time PCR. The Relative Quantitation (RQ) was calculated as: RQ = log2(−ΔΔCt). Graphpad Prism software 6.0 was used for statistical analysis of the data. For statistical comparison of a given target between the first and second trimester, a student’s t-test was used. A p value < 0.05 was considered statistically significant. The R-squared (R2) value for RQ vs. gestational age was calculated using a simple linear regression model. Histology and immunohistochemistry Two placental samples per gestational week studied were analyzed for TLR expression except as noted here: week 7 utilized 4 samples, and weeks 18 and 19 utilized 3 samples. For general histological analysis, sections from each gestational week studied were stained with routine hematoxylin and eosin (H&E). A trained pathologist in Obstetrics and Gynecology routinely screened these slides for the presence of inflammation, infection or abnormal placental development. No morphological evidence for any pathological conditions or changes were detected in any of the placental tissues examined. All placental samples that were studied for expression of the TLRs were therefor judged to be normal. The immunohistochemical analysis of TLR expression in placental tissue was carried out as previously described (20). Briefly, an antigen retrieval step was carried out by immersing the sections in a citrate buffer (pH 6.0), and heating them to 125°C for 30 seconds in a pressure cooker (Biocare Medical, Concord, CA, USA). The sections were rinsed thoroughly in distilled water and incubated with a blocking reagent (Background Sniper, Biocare Medical) prior to the application of primary antibodies directed against TLR2 (Novus Biologicals, Littleton, CO, USA) at 1:100 dilution; TLR3 (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) at 1:100 dilution; TLR4 (Novus Biologicals) at 1:120 dilution; TLR5 (Santa Cruz Biotechnology) at 1:80 dilution; and TLR9 (Santa Cruz Biotechnology) at 1:110 dilution for 60 minutes at room temperature. This was followed by washing in Tris buffer containing 0.1% Tween 20 (TBST). The antibodies were detected with a proprietary secondary reagent (MACH 4 Universal Alkaline Phosphatase, Biocare Medical) that detects both mouse and rabbit primary antibodies. The sections were then washed in TBST, and the antibodies visualized by incubating with a substrate for alkaline phosphatase (Vulcan Fast Red Chromogen, Biocare Medical) that stains positive cells a bright red. Development of the staining was monitored by light microscopy. Following a final wash, the sections were counterstained in aqueous hematoxylin mounted in a glycerin-based medium and mounted on a coverslip. Sections were analyzed under an Olympus microscope (Olympus America Inc., Center Valley, PA, USA) and images captured with a DP70 digital camera. Negative controls for the immunohistochemistry were processed in the absence of the primary antibodies. These were replaced with either the antibody diluent or a mouse or rabbit control antibody (Vector Laboratories, Burlingame, CA, USA) at the same concentration. RESULTS Expression of TLR genes across the first and second trimester is stable In order to quantify the changes in TLR gene expression in the placenta during the first and second trimester of pregnancy, we conducted a quantitative analysis for placental expression of the genes for TLRs 1-10 and the TLR4-associated protein, MD2. Comparisons were made to the housekeeping gene GAPDH, as well as the placental gene NDRG1, which has been shown to be expressed throughout the placenta in a relatively stable manner (21). We found mRNA for all 10 TLRs and MD2 could be detected, and for the majority of the TLRs, expression increased with gestational age. This is illustrated in Table 2, where raw Ct values are shown for each of the TLR targets tested, and Fig. 1, where relative gene expression (RQ) is graphed against trimester and gestational age. We found expression of TLR3 was the highest compared to the other TLRs in both the first and second trimester without a significant change in expression over time. We found a statistically significant increase in the RQ value in the second trimester compared to the first trimester samples for TLR1, TLR4, TLR7 and TLR8, with a non-significant trend towards increased RQ for TLR2 and TLR10 (Fig. 1A). For some of the TLRs (TLR1, TLR4, TLR7, TLR8), the increase in gene expression could be predicted by the gestational age, as shown by the calculated R squared value (Fig. 1B). In comparison, expression of the placental gene NDRG1 is stable across gestation. The limitation of this data is that it reflects gene expression not protein expression within a tissue of mixed cellular composition. To further examine cell specific expression in polarized cells within a tissue, we turned to immunohistochemistry. Histological changes within the placental villi during the first and second trimester We began by characterizing the histological changes that occur within the chorionic villi of the placenta over the first and second trimester using routine H&E staining of tissues, analyzing at least two tissues obtained at each time point. As shown in Fig. 2, there was considerable variation in the morphological composition of the placental villi with progression of gestation. As has been described, during the earliest weeks of gestation that we could evaluate (6–8 weeks), the inner villous cytotrophoblast layer was essentially continuous, consisting of small, regularly shaped cuboidal cells with round to oval nuclei (Fig. 2A). With further progression (9–12 weeks), the villous cytotrophoblast layer of some villi was noted to be more disorganized, whereas in some others it appeared as an attenuated layer of cells that were more flattened (Fig. 2B, 2C). Towards the end of the first trimester, a number of villi could be seen with scattered, large, cells within the cytotrophoblast layer (Fig. 2D). As gestation proceeded into the second trimester, the morphological changes observed to be taking place in the first trimester progressed such that few villi now possessed an intact layer of villous cytotrophoblasts. Although an attenuated cytotrophoblast layer could still be detected in some villi, in the majority of cases, cell-cell contact between cytotrophoblasts was lost leaving a loosely organized cellular layer beneath the outer syncytiotrophoblast (Fig. 2E). This resulted in a greater level of structural diversity between the individual villi for each placental sample compared to that observed in the first trimester. Moreover, it appeared that the few identifiable cytotrophoblasts were now increased in size and irregular in shape when compared to those present in the first trimester, perhaps suggesting increased differentiation towards a syncytiotrophoblast phenotype (Fig. 2F). With further maturation of the placenta during the last weeks of the second trimester, it was observed that many villi still possessed small but varying numbers of large villous cytotrophoblast cells scattered at the periphery. For many other villi, however, it was difficult to precisely identify the presence of any villous cytotrophoblast layer, and the villi instead appeared to be surrounded by a single layer of syncytiotrophoblast cells (Fig. 2G). Expression of TLRs within the placental villi during the first trimester of pregnancy Using immunohistochemistry protocols developed by our laboratory we examined the expression of TLR2, TLR3, TLR4, TLR5, and TLR9 in the available placental tissues. Negative controls of placental tissue samples observed at any stage of gestation showed no to little evidence of background or non-specific staining when the primary antibody was replaced with the antibody diluent (Fig. 3A). There was however a slight increase in background staining in placental tissues when incubated with either the rabbit IgG or mouse IgG negative controls (Fig. 3B). We found that in the early first trimester, at 6–7 weeks of gestation, the inner villous cytotrophoblast layer expresses all the TLRs that were studied (Fig. 4). Staining for TLR2 appeared especially intense on the cell membrane of the villous cytotrophoblast adjacent to the syncytiotrophoblast layer (Fig. 4A). This often gave the appearance of an apparent solid band of TLR2-positive staining dividing these two layers of cells (Fig. 4B). Staining for TLR4 and TLR5 was also membrane associated, but in contrast to TLR2, expression was not polarized (Fig. 4D and 4E). As expected for the endosomal TLRs, expression of TLR3 and TLR9 was observed as positive cytoplasmic staining within cytotrophoblasts (Figs. 4C and 4F), although unlike the surface expressed TLRs, the positive cells were discontinuous in some areas. The expression of TLR2, TLR3, TLR4, and TLR5, as evidenced by the intensity and pattern of positive staining in the cytotrophoblasts for all these TLRs, appeared to be fairly consistent, not only between the different placental samples studied at this stage, but also between the individual villi present within the same sample. The exception was TLR9, for which some villi, even from the same placenta, demonstrated abundant positive cytoplasmic granules throughout the sample, while others showed few or no cytoplasmic granules within the cells of the cytotrophoblast layer. Later in the first trimester (8–12 weeks), there was more variability in the intensity and distribution of the staining (Fig. 5). Villi that still possessed an intact cytotrophoblast layer consisting of conspicuous, well-developed, round cells were found to stain positive for the TLRs 2, 3, 4, 5 and 9 in a manner resembling that observed at 6–7 weeks gestation. The flattened, discontinuous cytotrophoblasts present in other villi also expressed TLR2 in a similar fashion to the earlier stages of gestation, with what appeared to be a continuous band of intense staining along the outer membrane bordering the syncytiotrophoblast (Fig. 5A). However, staining for TLR3 in many areas was more intense on the outer villous cytotrophoblast cell membrane adjacent to the syncytiotrophoblast layer, similar to what we had observed with TLR2, with some cytosolic staining as well (Fig. 5B). In contrast, the staining for both TLR4 and TLR5 appeared to be more uniform on the cell membrane of these cells, with some staining visible in the cytosol (Fig. 5C and 5D). The villous cytotrophoblasts, either as a discontinuous or flattened layer of cells, also stained positive for TLR9 with some variability in the expression similar to what we had observed earlier in gestation (Fig. 5E and 5F). Expression of TLRs within the placental villi during the second trimester of pregnancy During the early stages of the second trimester (13–17 weeks), the villous cytotrophoblasts in most villi were present as a loosely organized layer of large, often irregularly shaped cells (Fig. 6). It was observed that only a small number of these isolated cells were now expressing TLR2, although the pattern still resembled that observed in the first trimester, with intense expression on the membrane adjacent to the syncytiotrophoblasts (Fig. 6A). Other cytotrophoblast cells showed no TLR2 staining (Fig. 6B). TLR3 also appeared to be expressed by the villous cytotrophoblasts at this stage with both cytosolic and membrane staining detected (Fig. 6C). Intense staining was detected for TLR4 in the villous cytotrophoblasts, whether they were present as a disorganized layer or as individual cells scattered around the villi (Figs. 6D and 6E). Strong expression of TLR5 was also observed (Fig. 6F). As in the earlier time points, there appeared to be variability in the expression of TLR9 by the villous cytotrophoblasts. Whereas in some villi only a few cells stained positive for TLR9, in others it was apparent that most if not all of the cells contained intensely TLR9 positive granules (Fig. 6G). As the placenta developed during the last weeks of the second trimester (18–22 weeks), it was observed that some of the disorganized and often isolated cells of the villous cytotrophoblast layer still maintained the polarized staining for TLR2 with greater expression present on the cell membrane adjacent to the syncytiotrophoblast (Fig. 7A). Interestingly, the staining for TLR3 by the villous cytotrophoblasts was found to vary. For many villi, the remnant villous cytotrophoblast layer of large cells displayed uniform cytoplasmic expression of TLR3 (Fig. 7B), while in other villi, cells displayed TLR3 expression in a polarized fashion, similar to TLR2, with more intense staining of the cell membrane adjacent to the syncytiotrophoblast (Fig. 7C). Expression of TLR4 was consistently detected on all villous cytotrophoblast cells (Fig. 7D), while expression of TLR5 and TLR9 was not regularly detected (Fig. 7E and 7F). Hofbauer cells Cells that morphologically resembled Hofbauer cells were detected in villi during the early weeks (6–7) of gestation. As pregnancy proceeded however increasing numbers of putative Hofbauer cells appeared to be present in placental villi. These cells occurred in varying numbers in villi and many were found to stain positive for TLR2, 3, 4, 5 and 9. Examples of this are TLR2 (Fig. 6B arrow), TLR3 (Fig. 5B arrow) and TLR5 (Fig. 6F arrow). DISCUSSION Pregnancy is often perceived as an immune suppressed state that is necessary for the mother to tolerate the developing fetus. However, this over-simplified view underestimates the complex immunological exchange that occurs at the maternal-fetal interface within the placenta, particularly when microbial pathogens or commensals could be encountered. An inadequate immune response would allow the infection to spread to the fetus, while an excessive inflammatory response could lead to miscarriage, preterm labor, or other consequences of placental dysfunction. The innate immune system plays an essential role in balancing these often conflicting priorities with the goal of maintaining an appropriate environment for the developing fetus while protecting the health of the mother. The temporal regulation of TLRs in the placenta has not been widely studied. Ours is the first study to examine placental expression of the TLRs across the first and second trimester by both mRNA and protein. TLR gene expression was surprisingly consistent between individuals in the second trimester, although there was some variability within the first trimester for a subset of the TLRs. For example, placental TLR2 expression in the latter half of the first trimester was variable between individuals, as was TLR7 and TLR8. This coincides with the histological changes we observed in the first trimester, where the villous cytotrophoblast layer becomes more disorganized, suggesting a very dynamic tissue at this early stage. Of the TLRs studied, TLR3 was most highly expressed at the level of mRNA. In contrast, TLR9 and TLR10 mRNA expression was the lowest among the TLRs studied. Because our gene expression studies reflect mRNA levels throughout a tissue of mixed cellularity, it is impossible to determine the relative contribution of the individual cell types, such as villous cytotrophoblasts vs. Hofbauer cells. Clarification comes from the immunohistochemistry studies, where protein expression can be determined in situ. The most striking expression pattern was the band of TLR2 staining at the outer plasma membrane of the intact villous cytotrophoblast layer in the very early first trimester placenta. In fact, all the surface-expressed TLRs we examined (TLR2, TLR4, TLR5) demonstrated a similarly contiguous pattern of staining in this early stage placenta. TLR3 was generally detected in the cytosol; however, there were situations where TLR3 appeared to be expressed at the plasma membrane. While TLR3 is commonly considered an endosomal TLR, its expression has been described outside this compartment. For example, fibroblasts (but not dendritic cells) were shown to express TLR3 at the cell surface (22, 23), and it has been hypothesized that cell surface TLR3 binds dsRNA and becomes activated upon internalization (24). TLR9 staining was also localized to the cytosol, although by RT-PCR its expression was relatively lower than that seen for the other TLRs. The strong TLR3 expression by both RT-PCR and IHC is worth noting as TLR3 is a receptor for dsRNA and particularly important for danger signals. In addition to dsRNA viruses, dsRNA is formed as an intermediate during replication of some ssRNA viruses (such as respiratory syncytial virus, West Nile, influenza A, coxsackievirus, etc.) as well as DNA viruses (25). Cellular dsRNA can also be released during necrotic cell death [reviewed in (26)]. TLR3 activation is a strong inducer of type I IFNs and an antiviral immune response through the activation of IRF3 and NF-κB [reviewed in (27, 28)], and mouse models suggest that TLR3 activation can lead to preterm birth (29, 30). It was recently reported that placental IFN-β regulates inflammation by inhibiting responses to LPS, suggesting that inhibition of type I IFNs could remove the brake from inflammation induced by bacterial commensals and pathogens and lead to placental dysfunction and pregnancy loss (31). Thus, the presence of TLR3 could play a regulatory role in modulating inflammation through the induction of type I IFN. Our data is consistent with what others have reported. For example, Abrahams et al. demonstrated that TLR2 and TLR4 are expressed by villous cytotrophoblasts and extravillous trophoblasts, but not by syncytiotrophoblasts in first trimester placenta (32). Tangerås et al reported expression of functional TLRs 2–5 and TLR9 in isolated first trimester cytotrophoblasts (19). Ours is the first study to correlate expression of TLRs with histological changes in the placenta. It is interesting to speculate on the evolutionary adaptations that would shape the temporal and spatial patterns of TLR gene expression in the placenta. Early in the first trimester when the fetus is most vulnerable, the villous cytotrophoblasts appear to provide a firm line of defense, demonstrating the active role played by these cells as an arm of the innate immune system and countering the argument that pregnancy is a state of immune suppression. The presence of the full complement of TLRs in this tissue reflects the pressure to protect the fetus against invading pathogens as it weathers the first-trimester tests of chromosomal fitness, maternal antibody attack, endocrine imbalances, and defective implantation. As pregnancy progresses, however, breaches in this line of defense appear, perhaps reflecting improved fetal resilience as well as the need to temper an excessive inflammatory response that can be both harmful and beneficial to mother and fetus, leading to consequences such as miscarriage or preeclampsia. An improved understanding of the temporal and spatial regulation of TLRs in the placenta and the role they play in maintaining homeostasis could be exploited clinically. Much to the disappointment of clinicians, treatment of infections such as vaginitis (33, 34) or periodontal disease (35–38) during pregnancy has not reduced the incidence of preterm birth. Selective use of TLR antagonists in conjunction with more traditional antimicrobial treatments could prove useful as an approach to managing pregnancy-associated infections. The authors would like to thank the patients and clinical staff of the Obstetrics and Gynecology Department at Boston Medical Center for their support of this project. We thank Dr. Joe Politch for assistance with formatting figures, Dr. Deborah J. Anderson for insightful discussions, and the BUMC Analytical Instrumentation Core (AIC) for access to core instruments. This work was supported by funding from the NIH/NIAID through grant number AI101088. Fig. 1 Real time PCR analysis of TLR expression in placental tissue is shown. RNA isolated from first and second trimester placental tissue was subjected to RT-qPCR and normalized as described in the Methods section. Each data point represents an individual placenta, with n = 6 samples analyzed for first trimester, and n = 5 samples analyzed for second trimester except as follows: n = 5 for TLR5, TLR8 and TLR10 in the first trimester as no product was detected for TLR8 and TLR10 in a 6 week sample and TLR5 in a 7 week sample. (A) RQ is graphed on the y-axis according to trimester, and data shown is the mean with the SEM. PCR for each sample was repeated twice. Statistical significance (p-value) was calculated using a t-test: *, p < 0.05; **, p < 0.01; ***, p < 0.001. (B) Scatter plot analysis of TLR expression over 6–19 weeks gestation is depicted. RNA isolated from placentas of 6–19 weeks of gestation was subjected to Real-Time PCR for TLRs 1-10, MD2 and the control gene NDRG1, as described in the Methods. Each data point represents an individual placenta. Data shown above is the relative quantitation (RQ) of each gene vs. gestational age in the first and second trimester. The R2 values were calculated using a simple linear regression model. Fig. 2 Variations in the morphological appearance of the cytotrophoblast in villi during gestation. Cytotrophoblasts are seen as (A) a competent layer of small cells at 6 weeks; (B) a disorganized layer of cells at 12 weeks; (C) an attenuated layer of cells at 12 weeks; (D) large cells present in some villi at 12 weeks; (E) a loosely organized layer of cells at 13 weeks; (F) very large and irregular shaped cells now present in many villi at 13 weeks; and (G) mostly absent from villi that appear now to be surrounded by only the syncytiotrophoblast at 22 weeks. Original magnification: A, 10x; B–E, 20x; F, 40x; G 20x. Fig 3 Representative examples of negative controls. (A) Primary antibody was replaced with the antibody diluent (9 week sample). (B) Primary antibody was replaced with a negative control rabbit IgG (7 week sample). Original magnification: A–B, 10x. Fig 4 TLR expression by the villous cytotrophoblasts during early (6–7 weeks) first trimester of pregnancy is depicted. (A) Intense staining for TLR2 at 6 weeks with (B) polarized expression of TLR2 at the outer (apical) plasma membrane adjacent the syncytiotrophoblast layer (7 weeks). (C) Cytoplasmic staining of TLR3 (7 weeks). Membrane staining of (D) TLR4 (7 weeks) and (E) TLR5 (7 weeks). (F) Cytoplasmic staining of TLR9 (7 weeks). Original magnification: A, 10x; B, 40x; C–E, 20x; F, 10x. Fig 5 TLR expression by the villous cytotrophoblasts during late (8–12 weeks) first trimester of pregnancy is depicted. Attenuated layer of cells expressing (A) TLR2 (9 weeks) and (B) TLR3 (8 weeks) on the apical cell membrane adjacent the syncytiotrophoblast layer. Putative Hofbauer cell with positive staining for TLR3 (arrow). Expression of (C) TLR4 (12 weeks) and (D) TLR5 (12 weeks) by disorganized round cells. Expression of TLR9 by either (E) disorganized (12 weeks) or (F) attenuated cells (9 weeks). Original magnification: A, 10x; B–F, 20x. Fig 6 TLR expression by the villous cytotrophoblasts during early (13–17 weeks) second trimester of pregnancy is depicted. At this stage the cytotrophoblasts appear as a disorganized layer of large irregularly shaped cells. (A) TLR2 is expressed by a small number of cells (13 weeks), and some villi (B) stain negative for TLR2 (15 weeks; small arrow). Note adjacent putative Hofbauer cell that stained positive for TLR2 (large arrow). Expression of (C) TLR3 (13 weeks); TLR4 (D, 13 weeks; E, 15 weeks); (F) TLR5 (13 weeks); and (G) TLR9 (15 weeks). Original magnification: A–B, 40x; C–D, 20x; E, 10x; F–G, 20x. Fig. 7 TLR expression by the villous cytotrophoblasts during late (18–22 weeks) second trimester of pregnancy is depicted. (A) Remnant cells still expressed TLR2 with often intense staining of the cell membrane adjacent the syncytiotrophoblast (18 weeks). (B) TLR3 was expressed either as distinct positive cytoplasmic staining (22 weeks) or (C) as polarized staining with higher expression by the cell membrane next to the syncytiotrophoblast (21 weeks). Isolated cytotrophoblast cells expressing (D) TLR4 (20 weeks) and (E) TLR5 (22 weeks) and (F) TLR9 (18 weeks). Original magnification: A–F, 20x. Table 1 RT-qPCR primer ID, TaqMan® Gene Expression Assay Target Assay ID TLR1 Hs00413978_m1 TLR2 Hs01014511_m1 TLR3 Hs00152933_m1 TLR4 Hs00152939_m1 TLR5 Hs01019558_m1 TLR6 Hs01039989_s1 TLR7 Hs00152971_m1 TLR8 Hs00152972_m1 TLR9 Hs00152973_m1 TLR10 Hs01675179_m1 NDRG1 Hs00608387_m1 GAPDH Hs99999905_m1 Table 2 Raw Ct values First Trimester Second Trimester p-value Raw Ct (Mean ± SEM) n Raw Ct (Mean ± SEM) n TLR1 30.96 ± 0.6694 6 27.21 ± 0.5113** 5 0.0020 TLR2 32.56 ± 0.9181 6 28.13 ± 0.7483** 5 0.0055 TLR3 26.07 ± 0.3476 6 25.27 ± 0.2160 5 0.0947 TLR4 29.91 ± 0.4826 6 26.44 ± 0.2186*** 5 0.0002 TLR5 30.26 ± 0.5319 5 28.60 ± 0.1483* 5 0.0168 TLR6 30.62 ± 0.4225 6 28.24 ± 0.3195** 5 0.0019 TLR7 30.48 ± 0.7499 6 27.13 ± 0.2657** 5 0.0037 TLR8 35.47 ± 0.5801 5 31.89 ± 0.5597** 5 0.0022 TLR9 35.79 ± 0.1746 6 35.02 ± 1.045 5 0.445 TLR10 37.64 ± 0.2855 5 35.44 ± 0.6617* 5 0.0157 MD2 28.36 ± 0.3990 6 25.86 ± 0.1996*** 5 0.0005 NDRG1 24.70 ± 0.2163 6 24.63 ± 0.5569 5 0.9056 GAPDH 20.63 ± 0.5354 6 20.05 ± 0.3347 5 0.4127 * p<0.05; ** p<0.005; *** p<0.001 compared to first trimester. Highlights Cytotrophoblasts express multiple TLRs. TLR expression across early gestation is dynamic, especially in the first trimester. The villous cytotrophoblast layer becomes more disorganized as the placenta matures. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Land WG How evolution tells us to induce allotolerance Exp Clin Transplant 2015 13 Suppl 1 46 54 2 Svensson-Arvelund J Ernerudh J The Role of Macrophages in Promoting and Maintaining Homeostasis at the Fetal-Maternal Interface Am J Reprod Immunol 2015 74 2 100 9 10.1111/aji.12357 25582625 3 Aagaard K Ma J Antony KM Ganu R Petrosino J Versalovic J The placenta harbors a unique microbiome Sci Transl Med 2014 6 237 237ra65 6/237/237ra65 [pii] 10.1126/scitranslmed.3008599 4 Cao B Stout MJ Lee I Mysorekar IU Placental Microbiome and Its Role in Preterm Birth Neoreviews 2014 15 12 e537 e545 10.1542/neo.15-12-e537 25635174 5 Collado MC Rautava S Aakko J Isolauri E Salminen S Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid Sci Rep 2016 6 23129 10.1038/srep23129 27001291 6 Dong XD Li XR Luan JJ Liu XF Peng J Luo YY Liu CJ Bacterial communities in neonatal feces are similar to mothers’ placentae Can J Infect Dis Med Microbiol 2015 26 2 90 4 26015791 7 Zheng J Xiao X Zhang Q Mao L Yu M Xu J The Placental Microbiome Varies in Association with Low Birth Weight in Full-Term Neonates Nutrients 2015 7 8 6924 37 10.3390/nu7085315 26287241 8 Medzhitov R Janeway C Jr Innate immunity N Engl J Med 2000 343 5 338 44 10922424 9 Medzhitov R Janeway CA Jr Innate immunity: the virtues of a nonclonal system of recognition Cell 1997 91 3 295 8 9363937 10 Rock FL Hardiman G Timans JC Kastelein RA Bazan JF A family of human receptors structurally related to Drosophila Toll Proc Natl Acad Sci U S A 1998 95 2 588 93 9435236 11 Morisato D Anderson KV Signaling pathways that establish the dorsal-ventral pattern of the Drosophila embryo Annu Rev Genet 1995 29 371 99 8825480 12 Lemaitre B Nicolas E Michaut L Reichhart JM Hoffmann JA The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults Cell 1996 86 6 973 83 8808632 13 Akira S Mammalian Toll-like receptors Curr Opin Immunol 2003 15 2 238 14 Akira S Uematsu S Takeuchi O Pathogen recognition and innate immunity Cell 2006 124 4 783 801 16497588 15 Kawai T Akira S The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors Nat Immunol 2010 11 5 373 84 10.1038/ni.1863 20404851 16 Lien E Ingalls RR Toll-like receptors Crit Care Med 2002 30 1 Suppl S1 11 17 O’Neill LA The interleukin-1 receptor/Toll-like receptor superfamily: signal transduction during inflammation and host defense Sci STKE 2000 2000 44 RE1 18 Patni S Wynen LP Seager AL Morgan G White JO Thornton CA Expression and activity of Toll-like receptors 1-9 in the human term placenta and changes associated with labor at term Biol Reprod 2009 80 2 243 8 biolreprod.108.069252 [pii] 10.1095/biolreprod.108.069252 18815357 19 Tangerås LH Stodle GS Olsen GD Leknes AH Gundersen AS Skei B Vikdal AJ Ryan L Steinkjer B Myklebost MF Langaas M Austgulen R Iversen AC Functional Toll-like receptors in primary first-trimester trophoblasts J Reprod Immunol 2014 10.1016/j.jri.2014.04.004 20 Pudney J Anderson DJ Expression of toll-like receptors in genital tract tissues from normal and HIV-infected men Am J Reprod Immunol 2011 65 1 28 43 10.1111/j.1600-0897.2010.00877.x 20528831 21 Wyatt SM Kraus FT Roh CR Elchalal U Nelson DM Sadovsky Y The correlation between sampling site and gene expression in the term human placenta Placenta 2005 26 5 372 9 10.1016/j.placenta.2004.07.003 15850641 22 Matsumoto M Funami K Tanabe M Oshiumi H Shingai M Seto Y Yamamoto A Seya T Subcellular localization of Toll-like receptor 3 in human dendritic cells J Immunol 2003 171 6 3154 62 12960343 23 Matsumoto M Kikkawa S Kohase M Miyake K Seya T Establishment of a monoclonal antibody against human Toll-like receptor 3 that blocks double-stranded RNA-mediated signaling Biochem Biophys Res Commun 2002 293 5 1364 9 10.1016/S0006-291X(02)00380-7 12054664 24 Murakami Y Fukui R Motoi Y Kanno A Shibata T Tanimura N Saitoh S Miyake K Roles of the cleaved N-terminal TLR3 fragment and cell surface TLR3 in double-stranded RNA sensing J Immunol 2014 193 10 5208 17 10.4049/jimmunol.1400386 25305318 25 Jensen S Thomsen AR Sensing of RNA viruses: a review of innate immune receptors involved in recognizing RNA virus invasion J Virol 2012 86 6 2900 10 10.1128/JVI.05738-11 22258243 26 Chattopadhyay S Sen GC dsRNA-Activation of TLR3 and RLR Signaling Gene Induction-Dependent and Independent Effects J Interferon Cytokine Res 2014 34 6 427 36 10.1089/jir.2014.0034 24905199 27 Matsumoto M Funami K Oshiumi H Seya T Toll-like receptor 3: a link between toll-like receptor, interferon and viruses Microbiol Immunol 2004 48 3 147 54 15031527 28 Schroder M Bowie AG TLR3 in antiviral immunity: key player or bystander? Trends Immunol 2005 26 9 462 8 10.1016/j.it.2005.07.002 16027039 29 Cardenas I Means RE Aldo P Koga K Lang SM Booth CJ Manzur A Oyarzun E Romero R Mor G Viral infection of the placenta leads to fetal inflammation and sensitization to bacterial products predisposing to preterm labor J Immunol 2010 185 2 1248 57 10.4049/jimmunol.1000289 20554966 30 Koga K Cardenas I Aldo P Abrahams VM Peng B Fill S Romero R Mor G Activation of TLR3 in the trophoblast is associated with preterm delivery Am J Reprod Immunol 2009 61 3 196 212 10.1111/j.1600-0897.2008.00682.x 19239422 31 Racicot K Kwon JY Aldo P Abrahams V El-Guindy A Romero R Mor G Type I Interferon Regulates the Placental Inflammatory Response to Bacteria and is Targeted by Virus: Mechanism of Polymicrobial Infection-Induced Preterm Birth Am J Reprod Immunol 2016 75 4 451 60 10.1111/aji.12501 26892235 32 Abrahams VM Bole-Aldo P Kim YM Straszewski-Chavez SL Chaiworapongsa T Romero R Mor G Divergent trophoblast responses to bacterial products mediated by TLRs J Immunol 2004 173 7 4286 96 15383557 33 Carey JC Klebanoff MA Hauth JC Hillier SL Thom EA Ernest JM Heine RP Nugent RP Fischer ML Leveno KJ Wapner R Varner M Metronidazole to prevent preterm delivery in pregnant women with asymptomatic bacterial vaginosis. National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units N Engl J Med 2000 342 8 534 40 10.1056/NEJM200002243420802 10684911 34 Klebanoff MA Carey JC Hauth JC Hillier SL Nugent RP Thom EA Ernest JM Heine RP Wapner RJ Trout W Moawad A Leveno KJ Miodovnik M Sibai BM Van Dorsten JP Dombrowski MP O’Sullivan MJ Varner M Langer O McNellis D Roberts JM Failure of metronidazole to prevent preterm delivery among pregnant women with asymptomatic Trichomonas vaginalis infection N Engl J Med 2001 345 7 487 93 10.1056/NEJMoa003329 11519502 35 Macones GA Parry S Nelson DB Strauss JF Ludmir J Cohen AW Stamilio DM Appleby D Clothier B Sammel MD Jeffcoat M Treatment of localized periodontal disease in pregnancy does not reduce the occurrence of preterm birth: results from the Periodontal Infections and Prematurity Study (PIPS) Am J Obstet Gynecol 2010 202 2 147 e1 8 S0002-9378(09)02118-8 [pii] 10.1016/j.ajog.2009.10.892 20113691 36 Michalowicz BS Hodges JS DiAngelis AJ Lupo VR Novak MJ Ferguson JE Buchanan W Bofill J Papapanou PN Mitchell DA Matseoane S Tschida PA Treatment of periodontal disease and the risk of preterm birth N Engl J Med 2006 355 18 1885 94 355/18/1885 [pii] 10.1056/NEJMoa062249 17079762 37 Newnham JP Newnham IA Ball CM Wright M Pennell CE Swain J Doherty DA Treatment of periodontal disease during pregnancy: a randomized controlled trial Obstet Gynecol 2009 114 6 1239 48 10.1097/AOG.0b013e3181c15b40 00006250-200912000-00013 [pii] 19935025 38 Offenbacher S Beck JD Jared HL Mauriello SM Mendoza LC Couper DJ Stewart DD Murtha AP Cochran DL Dudley DJ Reddy MS Geurs NC Hauth JC Effects of periodontal therapy on rate of preterm delivery: a randomized controlled trial Obstet Gynecol 2009 114 3 551 9 10.1097/AOG.0b013e3181b1341f 00006250-200909000-00011 [pii] 19701034

PMC005xxxxxx/PMC5119654.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2985117R 4816 J Immunol J. Immunol. Journal of immunology (Baltimore, Md. : 1950) 0022-1767 1550-6606 27864551 5119654 10.4049/jimmunol.1600872 NIHMS810515 Article Immunological Outcomes of Antibody Binding to Glycans Shared Between Microorganisms and Mammals1 Patel Preeyam * Kearney John F. * * Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294 USA Corresponding Author: John F. Kearney, Office: (205) 934-6557, Fax: (205) 996-9908, jfk@uab.edu 16 8 2016 1 12 2016 01 12 2017 197 11 42014209 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Glycans constitute basic cellular components of living organisms across biological kingdoms (Table 1), and glycan-binding antibodies participate in many cellular interactions during immune defense against pathogenic organisms (Figure 1). Glycan epitopes are expressed as carbohydrate-only entities or as oligomers or polymers on proteins and lipids. Such epitopes on glycoproteins may be formed by post-translational modifications or neoepitopes resulting from metabolic/catabolic processes, and can be altered during inflammation. Pathogenic organisms can display host-like glycans to evade the host immune response. However, antibodies to glycans, shared between microorganisms and the host, exist naturally. These antibodies are able to not only protect against infectious disease, but are also involved in host housekeeping functions and can suppress allergic disease. Despite the reactivity of these antibodies to glycans shared between microorganisms and host, diverse tolerance-inducing mechanisms permit the B cell precursors of these antibody-secreting cells to exist within the normal B cell repertoire. Introduction Glycans, polymers of glycosidically linked sugars, are one of the most basic cellular components, and exist as carbohydrate-only entities as well as covalently attached modifications of proteins (glycoproteins) or lipids (glycolipids). Here, we use “glycan” to indicate both oligosaccharides and polysaccharides. In mammals, glycans have diverse functions, such as marking apoptotic cells for clearance, immune self/non-self discrimination, cell-cell communication, and intracellular signaling (1, 2). Glycosylation defects in humans are linked to disease (3), and the expressed glycome can be altered during inflammation, cellular stress, as well as cancer (4). Although the combinatorial composition of a saccharide array can generate an immense number of structures, the composition of mammalian glycans is well-conserved (5). Some microbial pathogens including bacteria, fungi, and protozoans display mammalian-associated glycans on their surfaces as an evolutionary adaptation to evade detection by the host’s immune system (6). This property can also directly contribute to pathogenicity of these organisms (7, 8). Additionally, expression of host-similar glycans by allergens may promote their engagement of innate receptors expressed by antigen-presenting cells (APCs) and epithelial cells in the lung (9, 10). In mammals, B cells and antibodies that react with self glycans exist naturally and function to promote homeostasis (11) by facilitating the clearance of dangerous and potentially inflammatory components, such as apoptotic cells (12), senescent red blood cells (13), and metabolic products such as oxidized lipids (14). Aside from these homeostatic functions, naturally occurring antibodies specific for mammalian glycoproteins or glycolipids recognize these structures when displayed by microorganisms as well as allergens, and can facilitate their clearance (10, 12, 15, 16). Many cellular processes such as engagement of SiglecG/CD22 (17), sequestration of autoreactive antigens (12), and induction of cellular anergy (18) exist to regulate and maintain autoreactive B cells within the B cell repertoire. In this review, we discuss the expression and interactions of B cells with selected glycan epitopes that are expressed on host cells, microbes, and allergens. These epitopes include N-acetylglucosamine (GlcNAc), sialyl-lacto-N-tetraose, and α-1,3-glucan. We then give some examples of how antibodies to these glycans mediate housekeeping functions and provide protection against pathogens and allergens (Figure 1, Table 1). Antibodies to Glycans, Implications for Polyreactivity, and Infection-Induced Autoimmunity Much has been written regarding antibodies with extensive “polyreactivity” (19–21). The term polyreactive has generally been used to describe antibody reactivity with seemingly structurally unrelated antigen targets and has been attributed to both germline and somatically mutated gene-encoded contributions to their antigen-binding sites (22, 23). The polyreactive nature of some antibodies has often been characterized by antibody binding to recombinant antigens or mimotopes in solid-phase ELISA-type assays or by western blotting of denatured complexes from microbes, mammalian tissues, or cell extracts. These assays may also detect low-avidity binding of antibodies against neoepitopes generated through processing of the antigen-containing material that may not normally be exposed on the native non-denatured molecules. By contrast, the representative antibodies discussed in this review exhibit exquisite specificity for defined oligosaccharide moieties expressed as glycan epitopes. These structures are themselves similar among bacterial, fungal, and parasitic components as well as allergens and multiple mammalian cell types. We, and others, have demonstrated that anti-GlcNAc antibody binding to bacterial cell wall peptidoglycan complexes is inhibited by soluble monomeric GlcNAc monosaccharide, but not its enantiomer GalNAc (24, 25). Thus, these antibodies are “polyreactive” only in the sense that they bind an identical or a very similar associated glycan epitope present in various molecular entities expressed by a wide range of living organisms. Host-resembling epitope expression by microorganisms has been referred to as molecular mimicry, and infections caused by some of these organisms, especially viruses and bacteria, have been implicated as the precipitating event for development of autoimmunity (26–29). Most of the studies in this field imply that the immune response to these infectious organisms drives the development of autoreactive T or B cells. However, evidence for most of these claims are sparse, and some of the suggested immune mechanisms by which this occurs are debatable. The association between Streptococcus pyogenes infection and rheumatic heart disease (RHD) is commonly discussed, and although 30–40% of humans that recover from rheumatic fever develop RHD (30), the cellular mechanism driving this disease remains controversial. Some prominent studies have implicated anti-GlcNAc antibody activity in mice immunized with myosin or GlcNAc-BSA conjugates as being a causative agent of RHD (31, 32). However, the specificity of antibody binding to GlcNAc-protein conjugates was not demonstrated by inhibition with free monosaccharide. Additionally, these T-dependent hybridoma antibodies reactive with myosin and GlcNAc-BSA did not express the canonical VHJ606 gene segment characterizing the highly conserved antibody repertoire of GlcNAc-specific monoclonal antibodies raised by immunizing mice with S. pyogenes (33). Further, other studies have clearly demonstrated that serum antibodies from rabbits immunized with S. pyogenes cell wall that bound to heart muscle were only inhibitable with the insoluble, but not soluble (GlcNAc-containing), portion of the bacterial cell wall (34). Instead, many studies have showed that streptococcal M protein and cardiac myosin express similar epitopes and peptides (35–37). Therefore, T and B cells with specificity for streptococcal M protein that also react with myosin are most likely the initiating factor in S. pyogenes-induced RHD. Although pathogenic antibodies against M protein are involved in RHD pathology, they are not likely to be directed towards defined GlcNAc epitopes as some studies suggest (31). Instead, these antibodies may be cross-reactive with neoepitopes possibly formed artificially by conjugating GlcNAc to BSA (31). In contrast, the GlcNAc residues on Group A streptococcal (GAS) carbohydrate are structurally ordered (38) and monoclonal antibodies against GlcNAc peptide conjugates do not bind native S. pyogenes bacteria or the purified GAS polysaccharide (unpublished observations). Therefore, studies demonstrating that antibodies generated against bacterial glycan epitopes cause autoimmunity should be interpreted cautiously. Bacterial infections may drive the development of other autoimmune conditions such as Guillain-Barré syndrome (GBS) (39). GBS is characterized by neuromuscular paralysis caused by T and B cells with specificity for peripheral nerve myelin gangliosides (40). Gangliosides are sialylated glycolipids that are widely distributed in mammalian neuronal membranes, where they are sequestered in islands in the lipid bilayer (41, 42). In some rare cases, GBS can be preceded by an infection with ganglioside-bearing organisms (39) such as Streptococcus agalactiae (Group B Streptococci) (43, 44), Mycoplasma pneumoniae (45), or Campylobacter jejuni (46–48), the most common organism implicated in this disease. Many C. jejuni strains have cell wall lipopolysaccharides containing sialylated gangliosides such as GM1, GT1a, GD1b, or GD3 that are identical to human gangliosides (47). It has been proposed that infection with C. jejuni can induce ganglioside-reactive antibodies that cause host nervous system damage (46–49). C. jejuni-induced autoreactive antibodies are specific for the oligosaccharide sequences attached to the polar groups in membranes, but sequence similarities and their membrane distribution may yield cross reactivities with different gangliosides such as GM1, GT1a, GD1b, or GD3 (46, 48). Due to the sequestration of gangliosides in membrane islands (41), the reactivity of these autoantibodies with gangliosides would most likely depend on their density, distribution, and clustering patterns in host membranes. Although the similarity of these ganglioside-like epitopes on the C. jejuni lipopolysaccharide may explain some GBS-inducing pathology, development of GBS among individuals infected with C. jejuni (50) and other ganglioside-expressing organisms, such as cytomegalovirus (CMV) (51), varicella zoster virus (VZV) and Epstein Barr virus (HBV) is low (52), making this cause and effect controversial (39). Therefore, in addition to antibodies to gangliosides, there may be other unknown immune factors contributing to incidence of disease. In addition to C. jejuni, it has long been known that the type-specific capsular polysaccharides of Group B streptococci (which will be discussed in detail later in this review) share structural similarities with epitopes on human glycoproteins (43) and the oligosaccharide head groups of mammalian membrane sphingolipids (43). The repeating units of the sialylated Group B streptococci type III polysaccharide are identical to gangliosides GD3 and GT3 (43). Similarly, the Group B streptococci type Ib repeating unit (53) is identical to the oligosaccharide sialyl-lactoneotetraosyl-ceramide (44), and has the potential to share epitopes with other gangliosides, including GD1a, expressed on mouse TH2 cells and GD1α expressed preferentially on mouse TH1 cells (54). Anti-Group B streptococcal antibodies raised by immunization of mice with the Group B streptococci type III or type Ib does not appear to cause neuromuscular paralysis or affect T cell functions (55). Another striking characteristic of these antibodies is that their binding to target acidic polysaccharides is calcium dependent (44) such that these antibodies may not bind to oligosaccharides expressed as gangliosides on host cells. The sialyllactose oligosaccharide epitope of the ganglioside GM3 is found on the plasma membrane of many mammalian cell types (56) and is highly expressed in melanoma, and has been extensively studied as a potential anti-tumor target (57, 58). Conventional wisdom suggests that antibodies against GM3 and other gangliosides may target and damage many GM3-bearing cells. However, binding of certain induced anti-GM3 monoclonal antibodies occurs in an all-or-none fashion depending on the threshold density of GM3 exposed at the cell surface (59). These observations suggest that bacteria could possibly induce antibodies that would bind to the target bacterial epitope without affecting functions of normal cells. Therefore, antibody recognition of cell-surface GM3 may depend on the density and spatial distribution of oligosaccharides on complex membrane bound structures and thus be a general characteristic regulating the relative binding of antibodies to similar epitopes on bacteria and host cells. Although bacteria display epitopes that can also be expressed on mammalian cells, the implications that antibodies to self glycan epitopes cause autoimmunity should be interpreted carefully. Naturally Occurring Antibodies In this review, we discuss the immunological activities of antibodies with specificity for GlcNAc, sialyl-lacto-N-tetraose, and α-1,3-glucan. Such antibodies are detectable in the serum of mice as well as humans and are referred to as natural antibodies (24). This terminology is often used to imply that these antibodies are detectable in the absence of deliberate immunization or vaccination. Glycan epitopes inducing natural antibodies may include those expressed by bacterial organisms comprising the microbiome, whereas others represent neoepitopes that are normally sequestered, but can be exposed by altered glycosylation related to cell apoptosis or death. The B cell repertoire encoding natural antibodies is generated early in life, and its individual clonal components can fluctuate throughout the lifetime of the host depending on exposure to self, microbiota, or environmental antigens expressing these epitopes (24). Mechanisms that both limit and maintain these autoreactive B cells within the B cell repertoire are discussed later in this review. Anti-GlcNAc-Based Therapeutics For Prevention of Bacterial and Protozoan Infections and Aspergillus-Induced Allergic Airway Disease Diverse groups of bacteria, protozoans, and fungi utilize N-acetylglucosamine (GlcNAc) as a building block in glycan-based structures (60). Helminths such as Heligmosomoides polygyrus and Strongyloides stercoralis express GlcNAc-containing cellular and structural components (24, 61), and helminth exposure can drive the production of high-titer serum antibodies against GlcNAc (24). Additionally, the biochemical composition of bacterial cell walls is highly conserved (62). Gram-positive bacteria incorporate alternating units of GlcNAc and N-acetylmuramic acid (MurNAc) connected by a β-1,4-glycosidic bond to assemble peptidoglycan for their cell walls (63). In addition to peptidoglycan, gram-negative bacteria possess an outer membrane containing a unique lipopolysaccharide that promotes immune evasion by shielding peptidoglycan GlcNAc moieties from serum antibodies and lectins (64). Certain gram-positive organisms such as Streptococcus pneumoniae evade the host immune system by generating a serotype-specific cell wall-linked polysaccharide capsule that prevents complement-mediated opsonophagocytosis (65). Regardless of serotype, S. pyogenes (Group A streptococcus, GAS) expresses a cell wall polysaccharide with a helical rhamnose backbone decorated with repetitive terminal GlcNAc residues (38) that is highly immunogenic in mice and humans. Infection with GAS induces antibodies against GlcNAc (25), and these antibodies are somewhat protective in mouse models of GAS infection (66–68). The development of a GlcNAc-based vaccine for GAS infection has been proposed for many years. However, the likelihood of pathogenic self reactivity of such antibodies and relatively poor protection in mouse models may have disfavored efforts to develop such a vaccine (38, 69). Because multiple bacterial species express variable levels of GlcNAc-containing molecules, efforts have also been made to design a GlcNAc-based vaccine that could protect against a broad array of organisms, including antibiotic-resistant Staphlococcus aureus (70). Recently, Genentech developed a novel anti-GlcNAc-based therapeutic agent in which human IgG1 anti-GlcNAc antibodies were linked to the antibiotic, rifalogue. In vivo, these modified antibodies bound to S. aureus, and upon phagocytosis, endosomal activation of the antibiotic agent mediated killing of intracellular S. aureus (71). The conserved microbial expression of molecules containing GlcNAc epitopes suggests that these approaches may have wider implications for drug delivery to intracellular compartments for killing other GlcNAc-expressing pathogenic organisms, such as Mycobacterium tuberculosis. GlcNAc is the basic subunit in the form of β-1,4 GlcNAc linked residues constituting the homopolymer chitin (72), which is the second most abundant biopolymer on earth. Chitin is present in the cell walls of various fungi, including yeast (73, 74), and is a major component of crustacean and insect exoskeletons (75) as well as protozoan and insect gut walls (76, 77). The physicochemical properties of different chitins may vary, and chitin particles purified from crab shells or different fungal species elicit significantly varied pro-inflammatory cytokine responses (78). Chitin is also expressed by a wide variety of respiratory allergens such as house dust mite (HDM), cockroach, and many fungal species including Aspergillus fumigatus (9, 79). Chitinous particles derived from these organisms may be considered as carriers associated with protein allergens that act as cargo (9, 10). For example, the hyphal cell wall of A. fumigatus contains chitin as a structural element and bears the major allergens Asp f 1 and Asp f 2. Engagement of innate receptors by chitin-bearing particles can promote the cellular entry to promote the processing of these allergenic proteins. Targeting the chitinous portion of these particles to prevent engagement of these innate receptors would also prevent entry of many different types of inflammatory elements associated with allergens, such as LPS, β-glucan, and proteases aside from allergenic proteins (80). Our laboratory has demonstrated that anti-GlcNAc antibodies bind chitin particles and germinated A. fumigatus and significantly decrease their uptake by alveolar macrophages and APCs in the lung. Additionally, neonatal exposure of mice to S. pyogenes results in increased levels of serum anti-GlcNAc antibody and an attenuation of A. fumigatus-induced allergic disease. These studies further demonstrated that intravenously (i.v.) delivered anti-GlcNAc antibodies suppressed A. fumigatus-induced airway disease development (9). Serum antibodies against GlcNAc occur naturally, and may vary depending on the microbiome, blood type, as well as the history of infection (12). Accordingly, therapeutic manipulation of anti-GlcNAc antibody levels may protect against bacterial and protozoan infection as well as respiratory and food allergies associated with chitin-bearing organisms. Anti-Sialyl-Lacto-N-Tetraose Antibodies are Protective Against Invasive Aspergillosis and Decrease Carriage of and Infection with Streptococcus agalactiae Aside from GlcNAc expression in its cell wall, germinated A. fumigatus also expresses sialyl-lacto-N-tetraose and α-1,3-glucan (9), an epitope described later in this review. Sialyl-lacto-N-tetraose is also expressed by Group B streptococci type 1b (44) and is a component of human milk oligosaccharides (HMOs) (81). Apart from causing allergic airway disease, A. fumigatus is the leading cause of invasive aspergillosis (I.A.) (82). I.A. results from an opportunistic infection, and is common among immunocompromised individuals (82) and Aspergillus is commonly isolated from deep tissue wounds of soldiers wounded in combat (83, 84). Current I.A. treatments have a limited success rate (85), and an efficacious vaccine for preventing I.A. has not yet been developed (86). Although some fungal polysaccharides can induce protective antibodies, they do so very poorly. This could be a result of multiple factors, including: suppression of T and B cell activity (87), variation in epitopes expressed between hyphal and yeast states, as well as fungus-induced proteosomal degradation of antibodies (88). Since the generation of antibodies to fungal epitopes is difficult, we hypothesized that generating antibodies to a bacterial epitope that is also expressed by fungi would stimulate the production of immunoprotective antibodies. Mice vaccinated with a Group B Streptococci 1b strain containing sialyl-lacto-N-tetraose were protected in a model of disseminated aspergillosis (55). Additionally, i.v., infusion of antibodies against sialyl-lacto-N-tetraose (clone SMB19) or use of SMB19 B cell transgenic mice expressing high levels of endogenous antibodies against sialyl-lacto-N-tetraose without deliberate bacterial immunization resulted in protection. Interestingly, J558 B cell transgenic mice with a high frequency of B cells specific for α-1,3-glucan and increased levels of antibody against α-1,3-glucan, which is also a major component of the A. fumigatus hyphal cell wall, were not protected against disseminated aspergillosis (55). Compared with antibodies against α-1,3-glucan, the SMB19 antibody is unique because it preferentially binds the tip of the germinating hyphae (24, 55). Calcium pumps and channels regulate the growing tip of the hyphae, and dysregulation of these calcium gradients may inhibit hyphal growth (89). The hyphal tip of many other yeast and fungal species such as Candida albicans also contains sialyl-lacto-N-tetraose (Figure 2). Therefore, targeting the hyphal tip may be an effective therapeutic for preventing many different types of invasive fungal diseases. Group B Streptococci 1b polysaccharide conjugate vaccine, provides protection against Group B Streptococci 1b infection (90). Group B Streptococci can cause severe infection among newborns, pregnant women, and the elderly (91). Currently, women testing positive for Group B Streptococci during pregnancy receive antibiotics during labor to prevent maternal and neonatal Group B Streptococcal infections. To generate long-lasting immunity and to circumvent antibiotic use, glycoconjugated Group B streptococcal vaccines have been developed (90). Group B Streptococcal vaccine generation was predated by the observation that maternal antibodies against the Group B streptococcus capsular polysaccharide correlated with lower neonatal susceptibility to Group B Streptococcal disease (92). In clinical trials, pregnant women immunized with the glycoconjugate Group B Streptococcus vaccine produced type-specific antibodies that were also detectable in the newborn cord blood and in the infants for up to 2 months of age. In these studies, Group B Streptococcus carriage rates were also lower among adult females receiving the vaccine (90). Collectively, the effectiveness of the Group B Streptococcal polysaccharide conjugate vaccine for inducing human antibodies that both protect against Group B Streptococci infections and also react with multiple fungi have motivated our efforts to repurpose Group B Streptococcus-conjugate vaccines for protection against IA development in immunocompromised individuals. In addition to being expressed on bacteria and fungi, lacto-N-tetraose and sialyl-lacto-N-tetraose are highly abundant oligosaccharide found in breast milk (81, 93–96). These HMOs can be advantageous by promoting growth of many beneficial Bifidobacterium species (97, 98) or as decoys for pathogenic diarrhea-causing organisms that use glycan-mediated attachment mechanisms for accessing the host immune system (95). However, HMOs may also promote pathogenicity as suggested by their ability to increase Staphylococcus epidermidis and S. aureus growth in vitro (99). Both of these staphylococcal organisms cause mastitis (100, 101), contaminate breast milk, and cause infections among infants (102–104). The exact mechanism by which the HMOs modulate the host immune system, apart from benefiting the gut microbiome (105), has not yet been determined. The presence of lacto-N-tetraose in breast milk and the infant digestive system is of interest because Group B Streptococcus 1b-vaccinated women and their children had antibodies against sialyl-lacto-N-tetraose (90). Although these antibodies can potentially react with an HMO epitope that is highly expressed in the mammary glands of pregnant females and digestive tracts of neonates, the vaccine did not have related adverse effects, and was shown to be safe for use in women of childbearing age, including those who were pregnant (106, 107). This further supports observations that antibodies against shared epitopes can reside in the body naturally without causing autoimmune manifestation. α-1,3-Glucan Expressing Organisms, Non-Canonical T cell-Independent B cell Responses, and Protection Against Cockroach Allergy α-1,3-glucan is a linear α-1,3-linked homopolymer of glucose, and antibodies against α-1,3-glucan are naturally occurring (24). In mice, B cells specific for α-1,3-glucan are enriched within the marginal zone and B1b cell populations (108). Anti-α-1,3-glucan antibodies are detectable in human plasma (unpublished observations); however, unlike the other antibodies discussed in this review, our understanding of the levels and characteristics of human antibodies against α-1,3-glucan is poor. Fungi, including yeast, are among the best documented expressers of the α-1,3-glucan epitope (74). Molecules expressing these epitopes can function as virulence factors when expressed by yeast (109) and fungal cell walls (74), or oral plaque-forming bacteria (110). However, commensal enteric organisms from mice can also express this molecule without any known pathogenic consequence (111). In fungi, such as Aspergillus nidulans, α-1,3-glucan accumulates during vegetative cell growth, and is used as an endogenous carbon source during sexual development (112). Whereas in yeast such as Cryptococcus neoformans, α-1,3-glucan is involved in anchoring the capsule to the cell wall (109). In some Histoplasma capsulatum strains, the α-1,3-glucan cell wall polymer may act as a shield for β-glucan, thereby evading immune clearance via recognition by Dectin-1, a mammalian β-glucan receptor (113), however a receptor for α-1,3-glucan has not yet been identified. Although antibodies against α-1,3-glucan were not protective in a mouse model of invasive aspergillosis (55), it is possible that they could protect against non-invasive fungal and yeast infections similarly to antibodies against β-1,3-glucan, which have been shown to inhibit growth of multiple fungi species (114). Bacterial species expressing α-1,3-glucan include oral streptococci (110, 115), cyst-forming bacteria (116), and some enteric organisms isolated from mice (111). Planktonic Streptococcus mutans do not express α-1,3-glucan; however, when these organisms begin producing biofilms, a series of glucosyltransferases (GTFs) synthesize α-1,3-glucan from sucrose (117). α-1,3-glucan provides structural stability for these biofilms, and is critical for S. mutans attachment, aggregation, as well as accumulation on tooth surfaces (118). It has yet to be determined whether oral antibodies against α-1,3-glucan could specifically prevent cariogenic biofilm formation. Additionally, biofilms formed by other bacterial organisms or yeasts may also contain α-1,3-glucan. Aside from S. mutans, cyst-forming bacteria such as Azotobacter vinelandii and some commensal enteric organisms such as Enterobacter cloacae and Serratia liquefaciens can also express α-1,3-glucan under non biofilm-forming conditions (111, 116). We and others have used α-1,3-glucan to study the immune response to T cell-independent antigens (108, 119–123). Textbook descriptions of T-independent responses state that these immune reactions do not result in the formation of memory; however, using the α-1,3-glucan epitope-bearing strain of Enterobacter (MK7), we demonstrated that α-1,3-glucan-specific IgM-secreting cells contribute to polysaccharide-specific memory antibody responses, which are maintained long-term (108, 124). Aside from bacteria and fungi, German cockroach (Blattella germanica) allergen also contains detectable α-1,3-glucan epitopes. These epitopes are expressed in the insect’s muscle and exoskeleton, and binding of anti-α-1,3-glucan is not exclusively due to microorganism contamination. Cockroach fecal pellets contain both allergenic protein Bla g 2 as well as α-1,3-glucan, and we suggest that particles bearing α-1,3-glucan epitopes can act as carriers of allergenic Bla g 2 proteins (Patel and Kearney manuscript submitted). Natural antibody based therapeutics for treating cockroach allergy are attractive because children that are skin-prick test positive for cockroach allergens are more likely to have difficult-to-control asthma (125, 126) and visit the emergency room due to an asthmatic event (127) than children that are not allergic to cockroaches. Treatment of neonatal, but not adult, mice with purified α-1,3-glucan or an α-1,3-glucan-expressing MK7 Enterobacter strain resulted in suppressed development of cockroach allergy during adult life. α-1,3-glucan-specific IgA-secreting cells were identified in the lungs of mice immunized with MK7 as neonates, but not as adults. By generating mice that are unable to make IgA responses to α-1,3-glucan, we confirmed that these B cells are responsible for protection against cockroach allergy (Patel and Kearney manuscript submitted). It has been shown that neutralization of bacteria and viral particles by antigen-specific IgA at mucosal sites is critical for preventing some diseases (128, 129). Additionally, this mechanism may also be important for suppressing allergic disease and selection of B cell clones capable of producing IgA during neonatal life may also be crucial in this process. Considering these, and other observations, we suggest that early exposure to α-1,3-glucan, in the form of a probiotic or natural colonization, may be sufficient to protect against mycosis and suppress fungal or cockroach allergy development for the lifetime of the individual. Autoreactive B Cell Stimulation and Maintenance As we have described in this review, B cells and their antibody products with the potential to react with self glycans exist naturally and there are multiple mechanisms that maintain self-reactive B cells within the repertoire. These antibodies against self antigens are not limited to those mentioned in this review, and also include antibodies against major and minor blood group antigens as well as antibodies against the phospholipid epitope phosphorylcholine (PC). One major unanswered question is: what mechanisms govern the maintenance of these potentially autoreactive B cell within the adult B cell repertoire? The levels of antibodies in human serum that react with self glycans, such as GlcNAc, are much lower compared to those that react with non-self ligands such as Gal-(α-1,3)-Gal (130) or anti-Neu5Gc (131). A possible explanation for this is that a tolerance mechanism may be involved in maintaining self-reactive B cells within the B cell repertoire. For example, engagement of CD22 or SiglecG by sialic acid-bearing glycans dampens B cell signaling and can even cause B cell apoptosis (17). However, glycan engagement by the BCR along with additional signals through TLR or NOD receptors, such as following encounter with a pathogenic organism, could break these tolerance-inducing mechanisms, allowing for autoreactive antibody production (132). Retention of B cells with reactivity for both self and microbial antigens may be evolutionarily advantageous such it drives the development and maintenance of a complete B cell repertoire (unpublished observations). In another suggested mechanism, epitopes similar among microorganisms and host may be sequestered in the host such that they are not normally exposed to B cell receptors at a level sufficient to initiate B cell activation. For example, GlcNAc subunits of mammalian glycans are usually capped by mannose or highly charged sialic acids, which prevent BCR receptor engagement (12). These antigens, which are normally sequestered intracellularly, may be exposed only as part of the apoptotic program or released during necrotic cell death. Normally, immune disposal mechanisms clear these antigens rapidly and limit their ability to drive immune reactions. (133). However, chronic inflammation and cancer induce states of glycan remodeling (4), during which these epitopes could possibly be exposed. Our last suggested mechanism for the maintenance of autoreactive B cells in the repertoire involves B cell clonal anergy. Many self-reactive B cells are found in innate-like B1 and marginal zone B cell subsets in an activated state (134). Their heightened activation state most likely results from BCR engagement by self antigen. Studies of the regulatory mechanism controlling B cell responses following exposure to self antigen revealed that a proportion of B cells, but not T cells, exist in a state of anergy instead of being deleted. In this particular study, B cell unresponsiveness was manifest as a continuum with the highest indication of BCR engagement apparent in MZ B cells (18), consistent with our previous findings that the MZ B cell population is one of the main reservoirs of B cells that can respond to self- and microbe-associated epitopes (134). Much of the dogma related to the tolerant and anergic state of B cells was originally generated with B cell receptor transgenic or knock-in mice (135). The BCRs in many, but not all, of these models were derived from B cells initially isolated from mice that were repeatedly immunized with T cell-dependent antigens or from autoimmune mice that made somatically hypermutated and isotype-switched pathogenic antibodies. Therefore, forced expression of transgenes among B cells in these models involves the expression of highly mutated high-affinity receptors that may not be reflective of those that would exist in a normal primary repertoire, where B cells would express mostly germline immunoglobulin genes. Thus, B cells expressing self-reactive BCRs may normally co-exist peacefully in a system with self-reactive antigens due to various mechanisms such as SiglecG/CD22 engagement, self-antigen sequestration, anergy, or even receptor editing (136). However, these tolerance mechanisms can be breeched upon engagement of microorganism-associated epitopes that co-engage TLRs, enabling B cells to respond rapidly to the corresponding environmental organisms. Most likely, frequencies of these autoreactive B cell clones fluctuate throughout the lifetime of an individual depending on acquisition of and alterations in the microbiome, bacterial infection, and tissue injury. Conclusions Glycans can exist in different states, such as attached to lipids or proteins, and can be reshaped by post-translational modification or metabolic or catabolic processes. Glycan epitopes are extensively distributed over various organisms across kingdoms (Table 1, Figure 1). Some of these glycans mediate primary biological functions such as self/non-self distinction and apoptotic cell removal. It is clear that some bacteria and allergens can display epitopes that are also expressed on host glycans. The bacterial expression may serve as an effort to evade host immune responses. However, antibodies reactive with these shared glycans are common in mammals, where they mediate both homeostasis within the host and provide defense against pathogenic organisms and allergens. Production of antibodies against these glycans can vary with age, and antibody reactivity against glycans expressed commonly by both the host and these environmental organisms can be involved not only in protection against infectious diseases but also allergic diseases. Despite the association of some of these glycans with host molecules, anti-self antibodies exist naturally, and self-reactive B cells exist in equilibrium within the normal B cell repertoire as the result of highly regulatory and complex dynamics. Figure 1 Overlapping expression of conserved glycan epitopes shared among mammalian cells, bacteria, fungi, and allergens and mechanisms by which autoreactive B cells are maintained in the B cell repertoire (A) Streptococcus pyogenes (Group A streptococcus) expresses GlcNAc, which is also expressed by Aspergillus fumigatus in the form of chitin. (B) Sialyl-lacto-N-tetraose is found on A. fumigatus, Streptococcus agalactiae (Group B streptococcus), and in breast milk, which promotes the growth of commensal gut organisms. (C) Some commensal enteric organisms express α-1,3-glucan, which is also expressed by German cockroach (Blattella germanica) allergen. (D) Autoreactive B cells can be maintained within the normal B cell repertoire by a variety of mechanisms including: signaling through CD22 or Siglec G, obtaining an anergic state, sequestration away from antigen, and receptor editing. Figure 2 Candida albicans (CA) hyphae express β-1,3-glucan and sialyl-lacto-N-tetraose epitopes The yeast form of CA was grown at 37°C on glass slides in RPMI 1640 medium for 2.5 hrs. (A) Cells were washed, fixed in ethanol at −20°C, then stained with monoclonal antibodies against β-1,3-glucan (green) or sialyl-lacto-N-tetraose (red) along with DAPI (blue). (B) Phase-contrast view of the same field; Scale bar= 20µm Table 1 Glycans Shared Between Microorganisms and Mammals Epitope Expression Immunological Outcomes of Glycan-Binding Antibodies N-acetyl-D-Glucosamine (GlcNAc) Mammalian: Apoptotic cells Post-translational modification Bacterial: Staphlococcus aureus Streptococcus pneumoniae Streptococcus pyogenes Mycobacterium tuberculosis Fungal/Yeast: Aspergillus fumigatus Helminth: Heligmosomoides polygyrus Strongyloides stercoralis Allergens: Aspergillus fumigatus Dermatophagoides pteronyssinus (dust mite) Blattella germanica (cockroach) Other chitin-containing organisms (e.g., shellfish) Protection against: Bacterial infections Allergic disease Promote apoptotic cell clearance May modulate fungal infections Sialyl-lacto-N-tetraose Mammalian: Human breast milk Bacterial: Streptococcus agalactiae type 1b Fungal/Yeast: Aspergillus fumigatus Candida albicans Protection against: Bacterial infections Invasive fungal infections May modulate: Allergic disease α-1,3-glucan Mammalian: None to date Bacterial: Streptococcus mutans biofilms Select Enterobacter cloacae Select Serratia liquefaciens Azotobacter vinelandii Fungal/Yeast: Aspergillus fumigatus Aspergillus nidulans Cryptococcus neoformans Histoplasma capsulatum Allergens: Blattella germanica (German cockroach) Protection against: Some invasive fungal diseases Allergic disease May modulate: Biofilm formation 1 This work was supported by the National Institutes of Health (NIH) Grants AI14782-37 and AI100005-05 and T32 AI00705 as well as the American Asthma Foundation. References 1 Marth JD Grewal PK Mammalian glycosylation in immunity Nature reviews. Immunology 2008 8 874 887 2 Ohtsubo K Marth JD Glycosylation in cellular mechanisms of health and disease Cell 2006 126 855 867 16959566 3 Freeze HH Genetic defects in the human glycome Nature reviews. Genetics 2006 7 537 551 4 Dube DH Bertozzi CR Glycans in cancer and inflammation--potential for therapeutics and diagnostics Nature reviews. Drug discovery 2005 4 477 488 15931257 5 Gagneux P Varki A Evolutionary considerations in relating oligosaccharide diversity to biological function Glycobiology 1999 9 747 755 10406840 6 Springer SA Gagneux P Glycan evolution in response to collaboration, conflict, and constraint The Journal of biological chemistry 2013 288 6904 6911 23329843 7 Dube DH Champasa K Wang B Chemical tools to discover and target bacterial glycoproteins Chemical communications (Cambridge, England) 2011 47 87 101 8 Tra VN Dube DH Glycans in pathogenic bacteria--potential for targeted covalent therapeutics and imaging agents Chemical communications (Cambridge, England) 2014 50 4659 4673 9 Kin NW Stefanov EK Dizon BL Kearney JF Antibodies generated against conserved antigens expressed by bacteria and allergen-bearing fungi suppress airway disease Journal of immunology 2012 189 2246 2256 10 Patel PS Kearney JF Neonatal Exposure to Pneumococcal Phosphorylcholine Modulates the Development of House Dust Mite Allergy during Adult Life Journal of immunology 2015 11 Bovin NV Natural antibodies to glycans Biochemistry. Biokhimiia 2013 78 786 797 24010841 12 New JS King RG Kearney JF Manipulation of the glycan-specific natural antibody repertoire for immunotherapy Immunological reviews 2016 270 32 50 26864103 13 Lutz HU Bogdanova A Mechanisms tagging senescent red blood cells for clearance in healthy humans Frontiers in physiology 2013 4 387 24399969 14 Binder CJ Naturally occurring IgM antibodies to oxidation-specific epitopes Advances in experimental medicine and biology 2012 750 2 13 22903662 15 Bouhlal H Kaveri S Multi-faceted role of naturally occurring autoantibodies in fighting pathogens Advances in experimental medicine and biology 2012 750 100 113 22903669 16 Gronwall C Vas J Silverman GJ Protective Roles of Natural IgM Antibodies Frontiers in immunology 2012 3 66 22566947 17 Chaouchi N Vazquez A Galanaud P Leprince C B cell antigen receptor-mediated apoptosis. Importance of accessory molecules CD19 and CD22, and of surface IgM cross-linking Journal of immunology 1995 154 3096 3104 18 Zikherman J Parameswaran R Weiss A Endogenous antigen tunes the responsiveness of naive B cells but not T cells Nature 2012 489 160 164 22902503 19 Zhou ZH Tzioufas AG Notkins AL Properties and function of polyreactive antibodies and polyreactive antigen-binding B cells Journal of autoimmunity 2007 29 219 228 17888628 20 Sedykh MA Buneva VN Nevinsky GA Polyreactivity of natural antibodies: exchange by HL-fragments Biochemistry. Biokhimiia 2013 78 1305 1320 24460965 21 Jones DD DeIulio GA Winslow GM Antigen-driven induction of polyreactive IgM during intracellular bacterial infection Journal of immunology 2012 189 1440 1447 22 Crouzier R Martin T Pasquali JL Heavy chain variable region, light chain variable region, and heavy chain CDR3 influences on the mono- and polyreactivity and on the affinity of human monoclonal rheumatoid factors Journal of immunology 1995 154 4526 4535 23 Martin T Crouzier R Weber JC Kipps TJ Pasquali JL Structure-function studies on a polyreactive (natural) autoantibody. Polyreactivity is dependent on somatically generated sequences in the third complementarity-determining region of the antibody heavy chain Journal of immunology 1994 152 5988 5996 24 Kearney JF Patel P Stefanov EK King RG Natural Antibody Repertoires: Development and Functional Role in Inhibiting Allergic Airway Disease Annual review of immunology 2015 25 Shackelford PG Nelson SJ Palma AT Nahm MH Human antibodies to group A streptococcal carbohydrate. Ontogeny, subclass restriction, and clonal diversity Journal of immunology 1988 140 3200 3205 26 Cusick MF Libbey JE Fujinami RS Molecular mimicry as a mechanism of autoimmune disease Clinical reviews in allergy & immunology 2012 42 102 111 22095454 27 Fujinami RS von Herrath MG Christen U Whitton JL Molecular mimicry, bystander activation, or viral persistence: infections and autoimmune disease Clinical microbiology reviews 2006 19 80 94 16418524 28 Albert LJ Inman RD Molecular mimicry and autoimmunity The New England journal of medicine 1999 341 2068 2074 10615080 29 Oldstone MB Molecular mimicry and immune-mediated diseases FASEB journal : official publication of the Federation of American Societies for Experimental Biology 1998 12 1255 1265 9761770 30 Guilherme L Kalil J Rheumatic fever and rheumatic heart disease: cellular mechanisms leading autoimmune reactivity and disease Journal of clinical immunology 2010 30 17 23 19802690 31 Galvin JE Hemric ME Ward K Cunningham MW Cytotoxic mAb from rheumatic carditis recognizes heart valves and laminin The Journal of clinical investigation 2000 106 217 224 10903337 32 Cunningham MW Streptococcus and rheumatic fever Current opinion in rheumatology 2012 24 408 416 22617826 33 Greenspan NS Davie JM Serologic and topographic characterization of idiotopes on murine monoclonal anti-streptococcal group A carbohydrate antibodies Journal of immunology 1985 134 1065 1072 34 Zabriskie JB Freimer EH An immunological relationship between the group. A streptococcus and mammalian muscle J Exp Med 1966 124 661 678 5922288 35 Dale JB Beachey EH Epitopes of streptococcal M proteins shared with cardiac myosin J Exp Med 1985 162 583 591 2410530 36 Huber SA Cunningham MW Streptococcal M protein peptide with similarity to myosin induces CD4+ T cell-dependent myocarditis in MRL/++ mice and induces partial tolerance against coxsakieviral myocarditis Journal of immunology 1996 156 3528 3534 37 Cunningham MW T cell mimicry in inflammatory heart disease Molecular immunology 2004 40 1121 1127 15036918 38 van Sorge NM Cole JN Kuipers K Henningham A Aziz RK Kasirer-Friede A Lin L Berends ET Davies MR Dougan G Zhang F Dahesh S Shaw L Gin J Cunningham M Merriman JA Hutter J Lepenies B Rooijakkers SH Malley R Walker MJ Shattil SJ Schlievert PM Choudhury B Nizet V The classical lancefield antigen of group a Streptococcus is a virulence determinant with implications for vaccine design Cell host & microbe 2014 15 729 740 24922575 39 Dimachkie MM Barohn RJ Guillain-Barre syndrome and variants Neurologic clinics 2013 31 491 510 23642721 40 Hughes RA Cornblath DR Guillain-Barre syndrome Lancet 2005 366 1653 1666 16271648 41 Sonnino S Mauri L Chigorno V Prinetti A Gangliosides as components of lipid membrane domains Glycobiology 2007 17 1r 13r 42 Seyfried TN el-Abbadi M Roy ML Ganglioside distribution in murine neural tumors Molecular and chemical neuropathology / sponsored by the International Society for Neurochemistry and the World Federation of Neurology and research groups on neurochemistry and cerebrospinal fluid 1992 17 147 167 43 Pincus SH Moran E Maresh G Jennings HJ Pritchard DG Egan ML Blixt O Fine specificity and cross-reactions of monoclonal antibodies to group B streptococcal capsular polysaccharide type III Vaccine 2012 30 4849 4858 22634296 44 Pritchard DG Gray BM Egan ML Murine monoclonal antibodies to type Ib polysaccharide of group B streptococci bind to human milk oligosaccharides Infection and immunity 1992 60 1598 1602 1548081 45 Sharma MB Chaudhry R Tabassum I Ahmed NH Sahu JK Dhawan B Kalra V The presence of Mycoplasma pneumoniae infection and GM1 ganglioside antibodies in Guillain-Barre syndrome Journal of infection in developing countries 2011 5 459 464 21727645 46 Ang CW Jacobs BC Laman JD The Guillain-Barre syndrome: a true case of molecular mimicry Trends in immunology 2004 25 61 66 15102364 47 Nachamkin I Allos BM Ho T Campylobacter species and Guillain-Barre syndrome Clinical microbiology reviews 1998 11 555 567 9665983 48 Yu RK Usuki S Ariga T Ganglioside molecular mimicry and its pathological roles in Guillain-Barre syndrome and related diseases Infection and immunity 2006 74 6517 6527 16966405 49 Yuki N Tagawa Y Handa S Autoantibodies to peripheral nerve glycosphingolipids SPG, SLPG, and SGPG in Guillain-Barre syndrome and chronic inflammatory demyelinating polyneuropathy Journal of neuroimmunology 1996 70 1 6 8862128 50 McCarthy N Giesecke J Incidence of Guillain-Barre syndrome following infection with Campylobacter jejuni American journal of epidemiology 2001 153 610 614 11257070 51 Khalili-Shirazi A Gregson N Gray I Rees J Winer J Hughes R Antiganglioside antibodies in Guillain-Barre syndrome after a recent cytomegalovirus infection Journal of neurology, neurosurgery, and psychiatry 1999 66 376 379 52 Taheraghdam A Pourkhanjar P Talebi M Bonyadi M Pashapour A Sharifipour E Rikhtegar R Correlations between cytomegalovirus, Epstein-Barr virus, anti-ganglioside antibodies, electrodiagnostic findings and functional status in Guillain-Barre syndrome Iranian journal of neurology 2014 13 7 12 24800041 53 Cieslewicz MJ Chaffin D Glusman G Kasper D Madan A Rodrigues S Fahey J Wessels MR Rubens CE Structural and genetic diversity of group B streptococcus capsular polysaccharides Infection and immunity 2005 73 3096 3103 15845517 54 Ebel F Schmitt E Peter-Katalinic J Kniep B Muhlradt PF Gangliosides: differentiation markers for murine T helper lymphocyte subpopulations TH1 and TH2 Biochemistry 1992 31 12190 12197 1457415 55 Wharton RE Stefanov EK King RG Kearney JF Antibodies generated against Streptococci protect in a mouse model of disseminated aspergillosis Journal of immunology 2015 194 4387 4396 56 Hakomori SI Handa K GM3 and cancer Glycoconjugate journal 2015 32 1 8 25613425 57 Nakakuma H Horikawa K Kawaguchi T Hidaka M Nagakura S Hirai S Kageshita T Ono T Kagimoto T Iwamori M Common phenotypic expression of gangliosides GM3 and GD3 in normal human tissues and neoplastic skin lesions Japanese journal of clinical oncology 1992 22 308 312 1469793 58 Hersey P Jamal O Henderson C Zardawi I D’Alessandro G Expression of the gangliosides GM3, GD3 and GD2 in tissue sections of normal skin, naevi, primary and metastatic melanoma International journal of cancer. Journal international du cancer 1988 41 336 343 3346097 59 Dohi T Nores G Hakomori S An IgG3 monoclonal antibody established after immunization with GM3 lactone: immunochemical specificity and inhibition of melanoma cell growth in vitro and in vivo Cancer research 1988 48 5680 5685 3167827 60 Cywes-Bentley C Skurnik D Zaidi T Roux D Deoliveira RB Garrett WS Lu X O’Malley J Kinzel K Zaidi T Rey A Perrin C Fichorova RN Kayatani AK Maira-Litran T Gening ML Tsvetkov YE Nifantiev NE Bakaletz LO Pelton SI Golenbock DT Pier GB Antibody to a conserved antigenic target is protective against diverse prokaryotic and eukaryotic pathogens Proceedings of the National Academy of Sciences of the United States of America 2013 110 E2209 E2218 23716675 61 van Die I Cummings RD Glycan gimmickry by parasitic helminths: a strategy for modulating the host immune response? Glycobiology 2010 20 2 12 19748975 62 Chapot-Chartier MP Kulakauskas S Cell wall structure and function in lactic acid bacteria Microbial cell factories 13 Suppl 2014 1 S9 63 Roberts WS Strominger JL Soll D Biosynthesis of the peptidoglycan of bacterial cell walls. VI. Incorporation of L-threonine into interpeptide bridges in Micrococcus roseus The Journal of biological chemistry 1968 243 749 756 5638591 64 Wade WF O’Toole GA Antibodies and immune effectors: shaping Gram-negative bacterial phenotypes Trends in microbiology 2010 18 234 239 20359893 65 Beuth J Ko HL Schroten H Solter J Uhlenbruck G Pulverer G Lectin mediated adhesion of Streptococcus pneumoniae and its specific inhibition in vitro and in vivo Zentralblatt fur Bakteriologie, Mikrobiologie, und Hygiene. Series A, Medical microbiology, infectious diseases, virology, parasitology 1987 265 160 168 66 Sabharwal H Michon F Nelson D Dong W Fuchs K Manjarrez RC Sarkar A Uitz C Viteri-Jackson A Suarez RS Blake M Zabriskie JB Group A streptococcus (GAS) carbohydrate as an immunogen for protection against GAS infection The Journal of infectious diseases 2006 193 129 135 16323141 67 Salvadori LG Blake MS McCarty M Tai JY Zabriskie JB Group A streptococcus-liposome ELISA antibody titers to group A polysaccharide and opsonophagocytic capabilities of the antibodies The Journal of infectious diseases 1995 171 593 600 7876606 68 Kabanova A Margarit I Berti F Romano MR Grandi G Bensi G Chiarot E Proietti D Swennen E Cappelletti E Fontani P Casini D Adamo R Pinto V Skibinski D Capo S Buffi G Gallotta M Christ WJ Campbell AS Pena J Seeberger PH Rappuoli R Costantino P Evaluation of a Group A Streptococcus synthetic oligosaccharide as vaccine candidate Vaccine 2010 29 104 114 20870056 69 Steer AC Dale JB Carapetis JR Progress toward a global group a streptococcal vaccine The Pediatric infectious disease journal 2013 32 180 182 23328823 70 Maira-Litran T Kropec A Goldmann D Pier GB Biologic properties and vaccine potential of the staphylococcal poly-N-acetyl glucosamine surface polysaccharide Vaccine 2004 22 872 879 15040940 71 Lehar SM Pillow T Xu M Staben L Kajihara KK Vandlen R DePalatis L Raab H Hazenbos WL Morisaki JH Kim J Park S Darwish M Lee BC Hernandez H Loyet KM Lupardus P Fong R Yan D Chalouni C Luis E Khalfin Y Plise E Cheong J Lyssikatos JP Strandh M Koefoed K Andersen PS Flygare JA Wah Tan M Brown EJ Mariathasan S Novel antibody-antibiotic conjugate eliminates intracellular S. aureus Nature 2015 527 323 328 26536114 72 Asensio JL Canada FJ Siebert HC Laynez J Poveda A Nieto PM Soedjanaamadja UM Gabius HJ Jimenez-Barbero J Structural basis for chitin recognition by defense proteins: GlcNAc residues are bound in a multivalent fashion by extended binding sites in hevein domains Chemistry & biology 2000 7 529 543 10903932 73 Lenardon MD Munro CA Gow NA Chitin synthesis and fungal pathogenesis Current opinion in microbiology 2010 13 416 423 20561815 74 Latge JP Tasting the fungal cell wall Cellular microbiology 2010 12 863 872 20482553 75 Merzendorfer H Zimoch L Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases The Journal of experimental biology 2003 206 4393 4412 14610026 76 Beckerman AP de Roij J Dennis SR Little TJ A shared mechanism of defense against predators and parasites: chitin regulation and its implications for life-history theory Ecology and evolution 2013 3 5119 5126 24455141 77 Foglesong MA Walker DH Jr Puffer JS Markovetz AJ Ultrastructal morphology of some prokaryotic microorganisms associated with the hindgut of cockroaches Journal of bacteriology 1975 123 336 345 166980 78 Alvarez FJ The effect of chitin size, shape, source and purification method on immune recognition Molecules (Basel, Switzerland) 2014 19 4433 4451 79 Lee CG Chitin, chitinases and chitinase-like proteins in allergic inflammation and tissue remodeling Yonsei medical journal 2009 50 22 30 19259344 80 Chai LY Vonk AG Kullberg BJ Verweij PE Verschueren I van der Meer JW Joosten LA Latge JP Netea MG Aspergillus fumigatus cell wall components differentially modulate host TLR2 and TLR4 responses Microbes and infection / Institut Pasteur 2011 13 151 159 81 De Leoz ML Gaerlan SC Strum JS Dimapasoc LM Mirmiran M Tancredi DJ Smilowitz JT Kalanetra KM Mills DA German JB Lebrilla CB Underwood MA Lacto-N-tetraose, fucosylation, and secretor status are highly variable in human milk oligosaccharides from women delivering preterm Journal of proteome research 2012 11 4662 4672 22900748 82 Latge JP Aspergillus fumigatus and aspergillosis Clinical microbiology reviews 1999 12 310 350 10194462 83 Warkentien T Rodriguez C Lloyd B Wells J Weintrob A Dunne JR Ganesan A Li P Bradley W Gaskins LJ Seillier-Moiseiwitsch F Murray CK Millar EV Keenan B Paolino K Fleming M Hospenthal DR Wortmann GW Landrum ML Kortepeter MG Tribble DR Invasive mold infections following combat-related injuries Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2012 55 1441 1449 23042971 84 Paolino KM Henry JA Hospenthal DR Wortmann GW Hartzell JD Invasive fungal infections following combat-related injury Military medicine 2012 177 681 685 22730844 85 Karthaus M Prophylaxis and treatment of invasive aspergillosis with voriconazole, posaconazole and caspofungin: review of the literature European journal of medical research 2011 16 145 152 21486728 86 Cassone A Casadevall A Recent progress in vaccines against fungal diseases Current opinion in microbiology 2012 15 427 433 22564747 87 Ponton J Omaetxebarria MJ Elguezabal N Alvarez M Moragues MD Immunoreactivity of the fungal cell wall Medical mycology 39 Suppl 2001 1 101 110 88 Casadevall A Antibody immunity and invasive fungal infections Infection and immunity 1995 63 4211 4218 7591049 89 Brand A Gow NA Mechanisms of hypha orientation of fungi Current opinion in microbiology 2009 12 350 357 19546023 90 Chen VL Avci FY Kasper DL A maternal vaccine against group B Streptococcus: past, present, and future Vaccine 31 Suppl 2013 4 D13 D19 91 Schuchat A Epidemiology of group B streptococcal disease in the United States: shifting paradigms Clinical microbiology reviews 1998 11 497 513 9665980 92 Baker CJ Carey VJ Rench MA Edwards MS Hillier SL Kasper DL Platt R Maternal antibody at delivery protects neonates from early onset group B streptococcal disease The Journal of infectious diseases 2014 209 781 788 24133184 93 Smilowitz JT Lebrilla CB Mills DA German JB Freeman SL Breast milk oligosaccharides: structure-function relationships in the neonate Annual review of nutrition 2014 34 143 169 94 Coppa GV Pierani P Zampini L Bruni S Carloni I Gabrielli O Characterization of oligosaccharides in milk and feces of breast-fed infants by high-performance anion-exchange chromatography Advances in experimental medicine and biology 2001 501 307 314 11787695 95 Bode L Human milk oligosaccharides: every baby needs a sugar mama Glycobiology 2012 22 1147 1162 22513036 96 Chaturvedi P Warren CD Altaye M Morrow AL Ruiz-Palacios G Pickering LK Newburg DS Fucosylated human milk oligosaccharides vary between individuals and over the course of lactation Glycobiology 2001 11 365 372 11425797 97 LoCascio RG Ninonuevo MR Freeman SL Sela DA Grimm R Lebrilla CB Mills DA German JB Glycoprofiling of bifidobacterial consumption of human milk oligosaccharides demonstrates strain specific, preferential consumption of small chain glycans secreted in early human lactation Journal of agricultural and food chemistry 2007 55 8914 8919 17915960 98 Gauhe A Gyorgy P Hoover JR Kuhn R Rose CS Ruelius HW Zilliken F Bifidus factor. IV. Preparations obtained from human milk Archives of biochemistry and biophysics 1954 48 214 224 13125592 99 Hunt KM Preuss J Nissan C Davlin CA Williams JE Shafii B Richardson AD McGuire MK Bode L McGuire MA Human milk oligosaccharides promote the growth of staphylococci Applied and environmental microbiology 2012 78 4763 4770 22562995 100 Kulakov AA Shkoda SM Astashov EI Protasenko P Petrov VI [Lactation mastitis: problems and perspectives] Khirurgiia 2004 36 38 101 Delgado S Arroyo R Jimenez E Marin ML del Campo R Fernandez L Rodriguez JM Staphylococcus epidermidis strains isolated from breast milk of women suffering infectious mastitis: potential virulence traits and resistance to antibiotics BMC microbiology 2009 9 82 19422689 102 Behari P Englund J Alcasid G Garcia-Houchins S Weber SG Transmission of methicillin-resistant Staphylococcus aureus to preterm infants through breast milk Infection control and hospital epidemiology 2004 25 778 780 15484804 103 Ballot DE Nana T Sriruttan C Cooper PA Bacterial bloodstream infections in neonates in a developing country ISRN pediatrics 2012 2012 508512 22919509 104 Cheung GY Otto M Understanding the significance of Staphylococcus epidermidis bacteremia in babies and children Current opinion in infectious diseases 2010 23 208 216 20179594 105 Charbonneau MR O’Donnell D Blanton LV Totten SM Davis JC Barratt MJ Cheng J Guruge J Talcott M Bain JR Muehlbauer MJ Ilkayeva O Wu C Struckmeyer T Barile D Mangani C Jorgensen J Fan YM Maleta K Dewey KG Ashorn P Newgard CB Lebrilla C Mills DA Gordon JI Sialylated Milk Oligosaccharides Promote Microbiota-Dependent Growth in Models of Infant Undernutrition Cell 2016 164 859 871 26898329 106 Baker CJ Edwards MS Group B streptococcal conjugate vaccines Archives of disease in childhood 2003 88 375 378 12716700 107 Kasper DL Paoletti LC Wessels MR Guttormsen HK Carey VJ Jennings HJ Baker CJ Immune response to type III group B streptococcal polysaccharide-tetanus toxoid conjugate vaccine The Journal of clinical investigation 1996 98 2308 2314 8941648 108 Foote JB Kearney JF Generation of B cell memory to the bacterial polysaccharide alpha-1,3 dextran Journal of immunology 2009 183 6359 6368 109 Reese AJ Doering TL Cell wall alpha-1,3-glucan is required to anchor the Cryptococcus neoformans capsule Molecular microbiology 2003 50 1401 1409 14622425 110 Bowen WH Koo H Biology of Streptococcus mutans-derived glucosyltransferases: role in extracellular matrix formation of cariogenic biofilms Caries research 2011 45 69 86 111 Kearney JF McCarthy MT Stohrer R Benjamin WH Jr Briles DE Induction of germ-line anti-alpha 1–3 dextran antibody responses in mice by members of the Enterobacteriaceae family Journal of immunology 1985 135 3468 3472 112 Zonneveld BJ Morphogenesis in Aspergillus nidulans. The significance of a alpha-1, 3-glucan of the cell wall and alpha-1, 3-glucanase for cleistothecium development Biochimica et biophysica acta 1972 273 174 187 4556771 113 Rappleye CA Eissenberg LG Goldman WE Histoplasma capsulatum alpha-(1,3)-glucan blocks innate immune recognition by the beta-glucan receptor Proceedings of the National Academy of Sciences of the United States of America 2007 104 1366 1370 17227865 114 Casadevall A Pirofski LA Immunoglobulins in defense, pathogenesis, and therapy of fungal diseases Cell host & microbe 2012 11 447 456 22607798 115 Chen L Ren Z Zhou X Zeng J Zou J Li Y Inhibition of Streptococcus mutans biofilm formation, extracellular polysaccharide production, and virulence by an oxazole derivative Applied microbiology and biotechnology 2015 116 Sutherland IW Mackenzie CL Glucan common to the microcyst walls of cyst-forming bacteria Journal of bacteriology 1977 129 599 605 402353 117 Ooshima T Matsumura M Hoshino T Kawabata S Sobue S Fujiwara T Contributions of three glycosyltransferases to sucrose-dependent adherence of Streptococcus mutans Journal of dental research 2001 80 1672 1677 11597030 118 Lynch DJ Fountain TL Mazurkiewicz JE Banas JA Glucan-binding proteins are essential for shaping Streptococcus mutans biofilm architecture FEMS microbiology letters 2007 268 158 165 17214736 119 Stohrer R Kearney JF Fine idiotype analysis of B cell precursors in the T-dependent and T-independent responses to alpha 1–3 dextran in BALB/c mice J Exp Med 1983 158 2081 2094 6196436 120 Stohrer R Kearney J Ontogeny of B cell precursors responding to alpha 1- greater than 3 dextran in BALB/c mice Journal of immunology 1984 133 2323 2326 121 Blomberg B Geckeler WR Weigert M Genetics of the antibody response to dextran in mice Science 1972 177 178 180 4114394 122 Froscher BG Klinman NR Strain-specific silencing of a predominant antidextran clonotype family J Exp Med 1985 162 1620 1633 2414388 123 Juy D Cazenave PA Igh restriction of the anti-alpha (1–3) dextran response: polyclonal B cell activators induce the synthesis of anti-alpha (1–3) dextran antibodies in lymphocytes from Igha mice only Journal of immunology 1985 135 1239 1244 124 Foote JB Mahmoud TI Vale AM Kearney JF Long-term maintenance of polysaccharide-specific antibodies by IgM-secreting cells Journal of immunology 2012 188 57 67 125 Eggleston PA Arruda LK Ecology and elimination of cockroaches and allergens in the home The Journal of allergy and clinical immunology 2001 107 S422 S429 11242603 126 Sohn MH Kim KE The cockroach and allergic diseases Allergy, asthma & immunology research 2012 4 264 269 127 Gao P Sensitization to cockroach allergen: immune regulation and genetic determinants Clinical & developmental immunology 2012 2012 563760 22272212 128 Zhang Y Pacheco S Acuna CL Switzer KC Wang Y Gilmore X Harriman GR Mbawuike IN Immunoglobulin A-deficient mice exhibit altered T helper 1-type immune responses but retain mucosal immunity to influenza virus Immunology 2002 105 286 294 11918690 129 Macpherson AJ McCoy KD Johansen FE Brandtzaeg P The immune geography of IgA induction and function Mucosal immunology 2008 1 11 22 19079156 130 Teranishi K Manez R Awwad M Cooper DK Anti-Gal alpha 1-3Gal IgM and IgG antibody levels in sera of humans and old world non-human primates Xenotransplantation 2002 9 148 154 11897007 131 Padler-Karavani V Yu H Cao H Chokhawala H Karp F Varki N Chen X Varki A Diversity in specificity, abundance, and composition of anti-Neu5Gc antibodies in normal humans: potential implications for disease Glycobiology 2008 18 818 830 18669916 132 Jellusova J Nitschke L Regulation of B cell functions by the sialic acid-binding receptors siglec-G and CD22 Frontiers in immunology 2011 2 96 22566885 133 Fishelson Z Attali G Mevorach D Complement and apoptosis Molecular immunology 2001 38 207 219 11532282 134 Martin F Kearney JF Marginal-zone B cells Nature reviews. Immunology 2002 2 323 335 135 Cambier JC Gauld SB Merrell KT Vilen BJ B-cell anergy: from transgenic models to naturally occurring anergic B cells? Nature reviews. Immunology 2007 7 633 643 136 Sukumar S Schlissel MS Receptor editing as a mechanism of B cell tolerance Journal of immunology 2011 186 1301 1302

PMC005xxxxxx/PMC5119655.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101631050 42499 EcoSal Plus EcoSal Plus EcoSal Plus 2324-6200 27735785 5119655 10.1128/ecosalplus.ESP-0020-2015 NIHMS787922 Article The Mosaic Type IV Secretion Systems Christie Peter J. Department of Microbiology and Molecular Genetics, The University of Texas Medical School at Houston, Houston, TX 77030 Correspondence: Peter J. Christie,: Peter.J. Christie@uth.tmc.edu 6 8 2016 10 2016 22 11 2016 7 1 10.1128/ecosalplus.ESP-0020-2015This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Escherichia coli and other Gram-negative and -positive bacteria employ type IV secretion systems (T4SSs) to translocate DNA and protein substrates, generally by contact-dependent mechanisms, to other cells. The T4SSs functionally encompass two major subfamilies, the conjugation systems and the effector translocators. The conjugation systems are responsible for interbacterial transfer of antibiotic resistance genes, virulence determinants, and genes encoding other traits of potential benefit to the bacterial host. The effector translocators are used by many Gram-negative pathogens for delivery of potentially hundreds of virulence proteins termed effectors to eukaryotic cells during infection. In E. coli and other species of Enterobacteriaceae, T4SSs identified to date function exclusively in conjugative DNA transfer. In these species, the plasmid-encoded systems can be classified as the P, F, and I types. The P-type systems are the simplest in terms of subunit composition and architecture, and members of this subfamily share features in common with the paradigmatic Agrobacterium tumefaciens VirB/VirD4 T4SS. This review will summarize our current knowledge of the E. coli systems and the A. tumefaciens P-type system, with emphasis on the structural diversity of the T4SSs. Ancestral P-, F-, and I-type systems were adapted throughout evolution to yield the extant effector translocators, and information about well-characterized effector translocators also is included to further illustrate the adaptive and mosaic nature of these highly versatile machines. INTRODUCTION Secretion of DNA and protein macromolecules across the Gram-negative cell envelope requires elaboration of dedicated translocation systems and the coupling in space and time of substrate recruitment, processing, and translocation reactions. Translocation of substrates to another cell by contact-dependent mechanisms adds more complexity through the requisite establishment of productive cell-cell junctions. The type IV secretion systems (T4SSs) deliver various macromolecular substrates, including DNA and monomeric and multimeric proteins, to a diversity of bacterial, fungal, plant, and human cell types (1, 2). Some T4SSs also function as contact-independent exporters or importers of substrates to or from the extracellular milieu. This functional versatility makes the T4SSs unique among the known bacterial translocation systems. In Escherichia coli, the known T4SSs are restricted in their substrate repertoires to mobile genetic elements (MGEs), although these systems also are capable of translocating certain proteins associated with the DNA transfer process. Detailed investigations of conjugation machines have established the importance of T4SSs over evolutionary time in the shaping of bacterial genomes and, on a more immediate time scale, for dissemination of virulence and antibiotic resistance genes in clinical settings (3, 4). These studies also have contributed greatly to our current understanding of T4SS machine architecture, mechanism of action, and evolution (3, 5, 6, 7). The T4SSs are functionally grouped as (i) conjugation systems, (ii) effector translocators, or (iii) contact-independent DNA/protein exchange systems (1). The conjugation systems are the largest subfamily, present in nearly all bacterial species and some archaeal species (5). These systems are specifically employed for dissemination of the mobile elements that encode them, and they also can mediate transfer of some genetically unlinked, non-self-transmissible mobile elements. The effector translocators, so far shown to function only in Gram-negative pathogens or symbionts, deliver effector proteins to eukaryotic cells (8, 9, 10). The translocated substrates disrupt host cell physiological processes, enabling bacterial colonization and spread. The contact-independent exchange systems, consisting of only a few members, function in release of DNA or protein substrates to the milieu or, alternatively, uptake of exogenous DNA (11, 12, 13). In E. coli, several conjugation systems have been characterized in detail (see Fig. 1). These systems elaborate two surface structures, a cell-envelope-spanning translocation channel and an extracellular pilus (6, 7). A classification scheme that will be used in this review groups the E. coli conjugation systems by the type of conjugative pilus produced. Accordingly, F-type systems elaborate long, flexible pili that dynamically extend and retract (14, 15, 16). These pili support equally efficient mating in liquid or solid surfaces. The P-type systems, by contrast, elaborate short, rigid pili for which no evidence exists of dynamic extension and retraction. Well-characterized conjugation systems encoded by E. coli plasmids RP4, R388, and pKM101, as well as the Agrobacterium tumefaciens VirB/VirD4 system, are P-type systems (17). A third group of plasmids represented by R64 and ColIb-P9 encode two types of pilus structures: a thick, rigid conjugative pilus and a long, flexible type IV pilus. Despite their name, type IV pili are ancestrally unrelated to conjugative pili produced by T4SSs and instead are phylogenetically and functionally related to type II secretion systems (T2SSs) (18). Nevertheless, type IV pili do play a role in enhancing conjugative transfer of these plasmids, especially in liquid media, by virtue of their ability to extend and retract, reminiscent of F pili. The T4SSs encoded by this group of plasmids are designated as I type, and they have a number of physical and functional features that distinguish them from the F- and P-type systems (19, 20). A major aim of this review is to summarize structure-function relationships of paradigmatic F-, P-, and I-type conjugation systems of E. coli and the VirB/VirD4 system of A. tumefaciens. On the basis of detailed phylogenetic studies, it is postulated that the present-day effector translocators evolved from P-, F-, and I-type conjugation systems (5). In general, this process involved the appropriation by ancestral conjugation systems of structural motifs from unknown ancestries that enabled diversification of substrate repertoires and the capacity to deliver substrates to eukaryotic host cells. A second goal of this review is to compare well-characterized conjugation and effector translocator systems to shed light on this evolutionary process and to illustrate the striking structural diversity and functional versatility of this fascinating translocator superfamily. OVERVIEW OF T4SS ARCHITECTURE AND TRANSLOCATION PATHWAY The paradigmatic A. tumefaciens VirB/VirD4 T4SS provides a unifying genetic nomenclature for a discussion of the T4SSs (Fig. 1) (21). The T4SSs of Gram-negative bacteria can be viewed as compilations of four distinct proteins or subassemblies: (i) the VirD4 type IV coupling protein (T4CP); (ii) the inner membrane complex (IMC) composed of VirB4 ATPase, VirB11 ATPase (when present), polytopic VirB3 and VirB6, and bitopic VirB8; (iii) the outer membrane complex (OMC) composed of outer membrane-associated VirB7 and VirB9, and a cell-envelope-spanning subunit VirB10; (iv) the conjugative pilus composed of VirB2 pilin, a proteolytic fragment of the VirB1 transglycosylase (VirB1*) and the pilus-tip protein VirB5 (Fig. 2A) (6, 7). The VirB/VirD4 T4SS and related P-type systems in E. coli are among the simplest T4SSs functioning in Gram-negative species in terms of subunit composition and machine architecture. Other P-type systems as well as the F- and I-type systems encode homologs or orthologs of most or all of the VirB subunits, but also possess additional domains, subunits, or protein subcomplexes that endow these systems with specialized functions and confer greater structural complexity (9, 14, 22). Two recent lines of investigation have shaped our current view of T4SS architecture and the substrate translocation pathway (Fig. 2). Although investigations commencing in the 1940s established the importance of so-called mating pair formation (Mpf) proteins for conjugative DNA transfer, it was not until the early 2000s that the first definitive evidence was obtained that these proteins assemble as a mating channel. By use of a ChIP-based, formaldehyde (FA)-crosslinking assay termed transfer DNA immunoprecipitation (TrIP), a translocation pathway was described for the A. tumefaciens transfer-DNA (T-DNA) substrate through the VirB/ VirD4 T4SS (23). With the TrIP assay, FA-cross-linkable substrate contacts were detected with the VirD4 T4CP, VirB11 ATPase, inner membrane proteins VirB6 and VirB8, and periplasmic/outer membrane subunits VirB2 and VirB9 (Fig. 2B). These subunits were thus postulated to constitute the cell-envelope-spanning “mating” channel. Other components of the T4SS that did not form FA cross-links with the T-DNA substrates were proposed to function as structural scaffolds or stabilizing elements for the channel (23). Second, structures of T4SS subunits or soluble domains and of larger subassemblies have been solved by X-ray crystallography or electron microscopy. Most notably, structures recently were presented for the OMC and the IMC/OMC subassemblies of P-type systems encoded by E. coli plasmids pKM101 and R388 (24, 25, 26). The R388-encoded IMC/OMC subassembly is the largest T4SS structure solved to date and is composed of homologs of the A. tumefaciens VirB3 through VirB10 subunits, lacking only homologs of VirD4, VirB11, VirB1, and VirB2 (26). This structure, here referred to as R388 T4SS3–10, is the current architectural blueprint for the T4SSs (Fig. 2C) and is described in more detail below. By integrating results of the in vivo TrIP and ultrastructural studies, a current model for translocation of substrates through a P-type T4SS postulates that substrates first dock with the VirD4 T4CP and then are transferred to VirB11 for further processing. All three ATPases, VirD4, VirB11, and VirB4, then coordinate substrate transfer across the inner membrane via a channel composed minimally of VirB6 and VirB8. The substrate is then delivered into a channel, composed of VirB2 and VirB9, that is housed within the OMC for translocation across the periplasm and outer membrane (Fig. 2) (7, 23, 27). The following sections describe this P-type translocation pathway and interesting structural or mechanistic variations among the F- and I-type systems. SUBSTRATES AND SUBSTRATE RECOGNITION In the initiating steps of type IV secretion, DNA and protein substrates are processed and recruited to cognate T4SSs for translocation. The early DNA substrate-processing reactions have been extensively reviewed elsewhere (3, 28). In brief, these are conserved reactions in most bacteria, carried out by proteins termed DNA transfer and replication (Dtr) factors. Dtr proteins bind specific origin-of-transfer (oriT) sequences associated with mobile elements, resulting in formation of the catalytically active relaxosome. One Dtr protein, the relaxase, cleaves the DNA strand destined for transfer (the T strand) at the nic site through a DNA phosphodiesterase reaction that results in covalent linkage of a catalytic Tyr residue with the 5′ end of the cleaved DNA. Accessory Dtr factors play critical roles in guiding the relaxase to the oriT sequence and facilitating the nicking reaction on supercoiled DNA substrates. Upon nicking, the relaxase serves to pilot the covalently associated T strand through the conjugation or “mating” channel. The early DNA-processing reactions are spatially positioned at or near the entrance to the conjugation channel, specifically mediated by contacts formed between relaxosome components and the VirD4-like T4CP (28, 29, 30). Accordingly, the T4CP functions as the receptor and docking site for cognate DNA substrates. This receptor activity, however, is not restricted to DNA substrates. T4CPs are associated with nearly all effector translocator systems, where they also function as receptors for protein substrates (2, 31). DNA Substrate Recognition Signals The nature of DNA substrate-T4CP contacts is not yet fully defined, but studies have identified motifs associated with relaxases that are required for DNA transfer. These motifs are postulated to specify the docking of DNA substrates with cognate T4CPs and are designated as translocation signals (TSs). Among the relaxases characterized to date, two types of TSs have been identified. The F plasmid-encoded TraI relaxase, for example, is a large protein with relaxase and helicase domains in its two halves. TSs termed TSA and TSB were mapped, respectively, in the two halves of the protein (32). Both TSs have a consensus sequence (G[E/D]R[L/M]R[V/F]T), and a structure of the TSA region of TraI was solved recently by X-ray crystallography (33). It consists of three domains, each structurally similar to SF1B helicase family domains. A putative T4CP interaction surface was mapped to one of the helicase domains through mutational analysis. Other relaxases, including R388-encoded TrwC and RSF1010-encoded MobA, carry TSs similar to those identified in TraI, suggestive of a common mode of relaxase binding to cognate T4CPs (34, 35, 36). Relaxases alternatively have C-terminal signals that can contribute to DNA substrate-T4CP docking. In A. tumefaciens, for example, the VirD2 relaxase carries such a TS, which is composed of a high proportion of positively charged residues. This motif is critical for the VirD2-VirD4 T4CP interaction and for transfer of the T-DNA substrate through the VirB/VirD4 T4SS to plant cells (see below). Interestingly, the VirB/VirD4 T4SS also translocates several protein substrates to plant cells, and these substrates also carry C-terminal, positively charged TSs (37, 38). Finally, relaxases can carry a combination of internal and C-terminal TSs for docking with cognate receptors. In some cases, such relaxases have been shown to employ these signals to mediate transfer of DNA substrates through distinct T4SSs. RSF1010 plasmids, for example, are promiscuous in the sense that they can translocate through many different T4SSs. RSF1010’s MobA relaxase carries internal TSs required for RSF1010 transfer through P-type T4SSs in E. coli. However, the relaxase also carries a C-terminal, positively charged TS that promotes RSF1010 transfer through the A. tumefaciens VirB/VirD4 T4SS (34, 36, 38). Similarly, the TrwC relaxase employs internal TSs for substrate transfer through the R388-encoded T4SS, and a C-terminal motif for substrate passage through a Bartonella henselae VirB/ VirD4 T4SS (35). Relaxases are only partly responsible for specifying the docking of DNA substrates with cognate T4CPs. Other Dtr accessory factors can confer substrate specificity, as best documented for the F-type systems. For F plasmids, substrate docking relies on the formation of highly specific contacts between the accessory factor TraM and the TraD receptor. In fact, the TraM-TraD interaction is the only structural interface solved to date by X-ray crystallography, as described in further detail in the next section. T4CPs and the DNA Substrate-Docking Reaction The T4CPs are so named because they link secretion substrates with cognate T4SS channels (23, 39, 40, 41). They typically are composed of an N-terminal transmembrane domain (NTD) joined to a large cytoplasmic moiety that consists of a nucleotide binding domain (NBD) and an α-helical bundle termed the all-alpha-domain (AAD) (Fig. 3A, B). An X-ray structure showed that a soluble fragment of the TrwB T4CP assembles as a globular homohexamer of ~110 Å in diameter and 90 Å in height, with a ~20-Å-wide channel in the center. This channel forms an 8-Å-wide constriction at the cytoplasmic pole of the molecule. The NTD extends from the opposite end of the globular assembly, as shown by electron microscopy, giving rise to an overall F1-F0-like ball-stem structure (Fig. 3A) (41, 42). These findings, coupled with evidence that T4CPs exhibit sequence and structural similarities with the FtsK and SpoIIIE DNA translocases, prompted a model in which the T4CP serves as the translocase for delivery of DNA substrates across the inner membrane (See Fig. 2, route 1) (43). Two features of the TrwB structure point to a role for the AAD in DNA substrate docking: (i) the AAD is positioned at the cytoplasmic pole of the TrwB homohexamer (Fig. 3A) and (ii) the AAD structurally resembles the DNA binding domain of XerD recombinase (41). Mutational analyses confirmed the functional importance of AADs associated with TrwB as well as a homolog encoded by plasmid RP4 (44, 45). A recent study further showed that the AADs of the A. tumefaciens VirD4 T4CP and of a T4CP encoded by Enterococcus faecalis pCF10 are essential for translocation of cognate DNA substrates in vivo (46). Purified forms of these domains bind DNA substrates and, more importantly, display specificity for cognate relaxases and other Dtr factors (46). As mentioned above, the A. tumefaciens VirD2 relaxase possesses a C-terminal TS required for T-DNA transfer to plants (38, 47, 48, 49). A VirD2 mutant deleted of this TS fails to bind VirD4’s AAD in vitro, suggesting that a VirD2 TS-VirD4 AAD interaction constitutes a basis for docking of the T-DNA substrate with the VirB/VirD4 channel in A. tumefaciens (49). T4CPs also can carry C-terminal domains (CTDs) of variable lengths and sequence compositions (Fig. 3B) (2). Contributions of these CTDs to substrate transfer are unspecified, with one notable exception. For F-type systems, the CTD of the TraD receptor plays a critical role in docking of the F plasmid substrate (Fig. 3C) (50). An acidic motif at the extreme C terminus of the CTD was shown by X-ray crystallography to mediate specific contacts with the Dtr accessory factor TraM (51, 52). Mutational analyses further confirmed that the TraD hexamer-TraM interaction forms the basis of a highly specific relaxosome-T4CP interaction in vivo (52, 53). This interaction both mediates highly efficient F plasmid transfer and blocks access of the RSF1010 plasmid substrate to the T4CP, resulting in strong inhibition of RSF1010 plasmid transfer (50). TraD’s CTD thus functions as a specificity checkpoint by ensuring efficient docking of the cognate F plasmid substrate at the exclusion of alternative substrates. At this juncture, therefore, the available data suggest that T4CPs recruit cognate DNA substrates through interactions mediated by AADs and, when present, CTDs. Further work is needed to define substrate-T4CP interactions at higher resolutions and identify factors controlling the dynamics of these contacts. The “coupling” function of T4CPs is only partly defined by its receptor activities. T4CPs also interact with cognate T4SS channel subunits to mediate delivery of docked substrates through the channel. The NTDs of T4CPs are required for these contacts, as shown by two-hybrid screens and mutational analyses (54, 55). These domains are implicated in formation of contacts with VirB10-like subunits (54, 55, 56), and it has been proposed that such contacts are responsible for the coupling of T4CPs with cognate channels. On purely stoichiometric grounds, however, it is difficult to envision how a T4CP hexamer (s) forms specific contacts with VirB10 subunits, which, according to the R388 T4SS3–10 structure, exist in 14 copies in the translocation channel (Fig. 2) (26, 41). Another complicating aspect of this interaction is that T4CPs have been shown to undergo profound changes both in conformation and oligomeric state in response to substrate docking and ATP binding signals (57, 58; see below). Accordingly, it seems more reasonable that, rather than forming stable, high-affinity contacts with T4SS channel subunits, T4CPs associate dynamically with T4SSs, possibly in response to activation by intracellular signals. That the T4CP is not a fixed, structural component of the T4SS is further supported by the fact that, in its absence, the T4SS is fully functional in its ability to elaborate conjugative pili (2, 22). STRUCTURE AND FUNCTION OF TYPE IV MACHINES The T4CPs deliver secretion substrates to the translocation channel composed of the VirB homologs. As mentioned above, the R388 T4SS3–10 structure currently represents the architectural blueprint for the T4SS superfamily (Fig. 2C). It is dominated by two large subassemblies, the IMC and OMC, which are connected by a narrow stalk. The following sections summarize current structure-function information about these subassemblies and the respective components, with emphasis on their contributions to the dynamics of substrate transfer across the inner and outer membranes. Contributions of ATPases to Substrate Trafficking In the R388 T4SS3–10 structure, two hexamers of the VirB4-like ATPase form the “legs” of the IMC substructure, but the VirD4- and VirB11-like ATPases were absent (Fig. 2). The ATPase interaction network is thus not yet described in molecular detail, nor is the specific contribution(s) of ATP energy to the overall process of substrate transfer. There is, however, compelling evidence for the P-type systems that all three ATPases coordinate their activities to drive substrate transfer across the inner membrane (54, 59). Interestingly, while the P- and I-type conjugation systems employ homologs of the VirD4, VirB4, and VirB11 ATPases (19, 20, 60, 61), in striking contrast, the F-type systems as well as conjugation systems functioning in Gram-positive species employ only homologs VirD4 and VirB4 (8, 22). The role of the T4CP as a substrate receptor is described above. Importantly, receptor activity does not require the ATP hydrolysis by the T4CP, suggesting that this catalytic activity is required for subsequent steps of the transfer process. One such postulated function is that the T4CP itself catalyzes the delivery of secretion substrates across the inner membrane (Fig. 2, route 1). According to this model, the DNA substrate binds the T4CP via its AAD and then enters the central lumen of the hexamer. The substrate is then pumped through the lumen by rounds of ATP hydrolysis across the membrane (3). At this time, the only support for this model is that the T4CP is phylogenetically and structurally similar to the DNA translocases SpoIIIE and FtsK (43). It is also difficult to reconcile this model with results of the TrIP studies showing that all three ATPases, VirD4, VirB4, and VirB11, are required for delivery of the DNA substrate to the inner membrane subunits VirB6 and VirB8 (23). Thus, alternative models have postulated that the three ATPases form an energy center that coordinates early-stage DNA-processing reactions as well as delivery of the DNA transfer intermediate across the membrane (Fig. 2, routes 2 and 3). Specific contributions of the VirB-like ATPases to substrate transfer have not been defined, but results of mutational analyses as well as structural information about these subunits have supplied important clues regarding their functions. For the VirB11-like ATPases, results of the TrIP studies showed that VirB11 interacts with the DNA substrate only in cells producing the VirD4 receptor, implying that VirD4 delivers the substrate to VirB11 (23). Mutational studies further supplied evidence that VirB11 also engages with effector protein substrates (62). Structural studies of VirB11 homologs revealed that these proteins belong to a secretion ATPase superfamily, whose other members include GspE subunits associated with T2SSs and type IV pilus assembly systems and InvC associated with a Salmonella enterica type III secretion system (T3SS) (63–65). These ATPases undergo large nucleotide-dependent conformational changes thought to be important for providing the mechanical leverage necessary for driving machine assembly processes or substrate unfolding. Indeed, studies confirmed that InvC catalyzes both the release of bound secretion chaperones and unfolding of effector proteins in an ATP-dependent manner (66). By analogy, VirB11 subunits might catalyze the unfolding of relaxases and effector proteins prior to translocation through the T4SS. The VirB4 ATPases are large (>70 kDa) proteins that bind tightly or integrally with the cytoplasmic membrane. Interestingly, these proteins are phylogenetically related to and exhibit structural similarities with T4CPs (5, 67). VirB4-like subunits have been shown to form multiple contacts with other IMC subunits, as well as with the VirB10 subunit (54, 59, 68). In A. tumefaciens, cross-linking of the T-DNA substrate with VirB4 was not detected by TrIP, but production of VirB4 was necessary for detectable substrate cross-linking with the VirB6 and VirB8 channel components (23). In more recent studies, evidence was presented for DNA substrate binding in vivo by a VirB4 homolog associated with a E. faecalis T4SS (69) and for DNA binding in vitro by VirB4 homologs associated with the E. coli pKM101- and R388-encoded T4SSs (70, 71). These findings raise the possibility that VirB4 plays a more direct role in mediating the transfer of DNA substrates across the inner membrane than originally envisioned (23). Accordingly, it can be proposed that, upon docking of the DNA substrate with the VirD4-like receptor, the substrate is delivered to the VirB11 ATPase for further processing. In turn, VirB11 might deliver the substrate to the VirB4 ATPase for transfer to or across the inner membrane (Fig. 2, routes 2 and 3). The Inner Membrane Complex and Its Contribution to Substrate Transfer across the Inner Membrane In early studies, the IMC was envisioned as a subassembly consisting of the ATPases and one or a few copies each of the membrane proteins VirB3, VirB6, and VirB8 (17). Structural definition of the R388 T4SS3–10 subassembly, however, now establishes that the IMC is considerably larger than previously thought (Fig. 2C) (26). This substructure presents as an arch of 255 Å in width sitting on top of the two VirB4 hexamer barrel structures each with dimensions of 105 Å in width and 135 Å in length. The arch, composed of 12 copies each of VirB8 and VirB3, and 24 copies of VirB6, sits in the inner membrane. Intriguingly, the arch also is composed of 12 copies of VirB5, which also localizes at the tip of the conjugative pilus (26). Finally, the IMC is composed of 14 copies of VirB10, which also is a major structural component of the OMC (see below). In view of known topologies of these subunits, the IMCs of P-type systems can be estimated to consist of at least 200 transmembrane helices. VirB6 channel activity? Results of the TrIP studies in A. tumefaciens favored a translocation pathway whereby the three ATPases coordinate delivery of DNA substrates to a membrane channel composed of VirB6 and VirB8 (Fig. 2, route 3) (23, 54). Various VirB6 mutations also were found to selectively block formation of T-DNA contacts with VirB8, suggesting that VirB6 and VirB8 act sequentially in the substrate transfer pathway (Fig. 2B) (72). Other mutations in VirB6 selectively inhibited formation of substrate contacts with the VirB2 and VirB9 subunits, further indicating that VirB6 also might form contacts with OMC components necessary for passage of the DNA substrate through the distal portion of the channel (see below) (72). VirB6 subunits minimally are ~300 residues with 5 or more membrane-spanning domains and a large central, periplasmic domain (2, 72). In view of the recent T4SS3–10 structure, it is intriguing to consider how 24 copies of VirB6 might be configured as part of the translocation channel. One possibility warranting further investigation is that VirB6 oligomerizes to form more than one functional channel across the inner membrane. This could, for example, enable the simultaneous passage of multiple substrates through a given T4SS in response to environmental cues. For conjugation machines mediating the transfer of a mobile genetic element to a bacterial recipient, such redundancy of function might not seem necessary. However, as noted earlier, conjugation systems translocate not only DNA substrates but also certain proteins, e.g., primases, single-stranded DNA binding proteins (SSBs), that function in establishment of the mobile element upon transfer to recipient cells (73, 74). Successful transfer of a DNA substrate therefore could rely on the coordinated and simultaneous transfer of DNA-metabolizing proteins, sometimes in many copies as in the case of SSBs, through a given T4SS. It is also noteworthy that effector translocator systems typically deliver multiple substrates to eukaryotic target cells. While this could be achieved through reiterative rounds of transfer through a single channel or different T4SS machines, a potentially more efficient mechanism would employ multiple channels within the framework of a single T4SS machine for the simultaneous export of multiple substrates. This could explain how, in Legionella pneumophila, the Dot/Icm system is able to rapidly deliver as many as several hundred effectors to eukaryotic cells during an infection (75, 76). With these considerations in mind, the various translocation routes depicted in Fig. 2 might not be mutually exclusive. For example, the T4CP or VirB4-like ATPase might suffice to translocate DNA substrates (routes 1 and 2), while the VirB6/VirB8 channel mediates protein trafficking (route 3). Even for DNA transfer, a model was proposed that the T4CP functions as the translocase for the DNA component of a DNA substrate and the VirB6/ 8 subunits mediate transfer of the relaxase bound to the 5′ end of the DNA substrate (3). Such a model is intriguing as it invokes two distinct translocation pathways for the delivery of a single DNA substrate, the relaxase-T-strand transfer intermediate, across the inner membrane. It also has been envisioned that large periplasmic loops of VirB6 monomers might assemble to form an entry portal for DNA substrates delivered to the periplasm by the T4CP ATPase. This proposal was based on findings in A. tumefaciens that mutations in a central periplasmic loop of VirB6 blocked formation of formaldehyde crosslinks with the DNA substrate (72). In fact, the existence of an entry portal in the periplasm for feeding substrates into the channel housed by the OMC is appealing in view of evidence that some effector translocator systems recruit and export substrates that are first delivered to the periplasm via the general secretory pathway (GSP; see below). VirB8, an IMC-OMC connector? The VirB8 subunits are bitopic proteins with a short cytoplasmic N-terminal domain, a membrane-spanning domain, and a large C-terminal periplasmic domain (77). VirB8 subunits are signatures of T4SSs carried by Gram-negative and -positive species (8), and recent work has established that their periplasmic domains adopt a common structural fold (78, 79, 80, 81, 82). This fold presents as large extended β-sheets with five α-helices, giving rise to an overall globular fold. VirB8 subunits pack as dimers or trimers in the crystal structures, and results of mutational analyses suggest that oligomerization is physiologically relevant. Indeed, the dimer interface has served as a target for a high-throughput screen that resulted in the identification of small-molecule inhibitors of T4SSs (83). Recently, crystal structures were solved for the periplasmic domains of VirB8-like DotI and TraM associated with the I-type L. pneumophila Dot/Icm and plasmid R64 T4SSs (82). Interestingly, the VirB8-like domain of DotI forms a stack of two octameric rings in the crystal structure, and results of mutational analyses suggest that the contacts between the two octameric rings are biologically important in vivo. The octameric structure distinguishes DotI of the I-type systems from P-type homologs, which are predicted to assemble as a dodecameric ring in the IMC (26, 82). Nevertheless, in view of the earlier TrIP data generated with the A. tumefaciens system, it is reasonable to propose that these subunits assemble as ring-like structures at the IMC/OMC junction for conveyance of substrates from the IMC into the channel housed by the OMC. The Outer Membrane Complex and Its Contribution to Substrate Transfer across the Outer Membrane In A. tumefaciens, the outer membrane lipoprotein VirB7 and VirB9 interact with bitopic inner membrane protein VirB10, forming a subassembly originally termed the core complex (see Fig. 2A) (17, 21). This complex, renamed the OMC, is intrinsically stable and stabilizing for most of the other VirB subunits (17). A. tumefaciens ring-shaped OMC complexes were visualized by transmission electron microscopy (TEM) (84), and corresponding OMCs associated with P-type T4SSs encoded by E. coli plasmids pKM101 and R388 have been structurally solved by X-ray crystallography, cryoelectron microscopy, and TEM (24, 25, 26). The pKM101-encoded OMC is composed of 14 copies each of the VirB7, VirB9, and VirB10 homologs (24, 25). This is a large, 1.05 MDa barrel-shaped structure of 185 Å in width and length. It is composed of two layers (I and O layers) forming a double-walled structure. The I layer, composed of the N-terminal domains of VirB9 and VirB10, is anchored in the inner membrane via the N-terminal transmembrane domain of VirB10. The N-terminal domains of VirB10 form a ring of 55 Å in diameter when the OMC is produced in the absence of the IMC. The O layer, composed of VirB7 and domains of VirB9 and VirB10, has a main body and cap that is thought to span the outer membrane. The cap, with a narrow hole of 10 Å in diameter, consists of 14 copies of a domain of VirB10 termed the antennae projection (AP). The OMC thus forms a large barrel with openings at both ends (24, 25). VirB10-like proteins are highly unusual bacterial membrane proteins in extending across the entire cell envelope (24, 85). This feature lends itself well to roles for these subunits not only as structural scaffolds for the T4SS (27) but also as transducers of activating signals from within or outside the cell across the T4SS (see below). In the R388 T4SS3–10 structure, the OMC is linked to the IMC by a thin structure termed the stalk (Fig. 2B) (26). The composition of the stalk is not defined but must consist in part of the 14 copies of VirB10’s N-proximal region since the N terminus of VirB10 extends across the inner membrane. The biological importance of the stalk, or even its existence in the context of the T4SS when embedded in the cell envelope of an intact cell, is not yet established. However, a gap between the IMC and OMC junction may enable access of substrates delivered into the periplasm via the GSP to the OMC for translocation across the outer membrane. The interior chamber of the OMC is sufficiently large to house a translocation channel consisting of VirB2 and VirB9, which were shown by TrIP to cross-link with the translocating T-DNA substrate (24, 25, 86). Interestingly, VirB10 did not detectably interact with the DNA substrate (23), even though VirB10’s AP presumably forms the outer membrane-spanning pore (24, 25). OMC structures solved to date lack VirB2 pilin homologs, and it is conceivable that VirB2-like subunits assemble as part of the channel at the distal end of the OMC to add structural complexity to the outer membrane pore (Fig. 2A) (27). The Translocation Route With an emerging knowledge of T4SS architecture, it is interesting to return to the question of how substrates are routed through the T4SS across the cell envelope. Translocation systems of Gram-negative bacteria deliver their cargoes either in one or in two steps. One-step pathways employ a single channel that spans the entire cell envelope, and two-step pathways employ an inner membrane translocase and a second system that recruits the substrate from the periplasm for delivery across the outer membrane (see reference 18). Most T4SSs are thought to employ a one-step route, as depicted in Fig. 2, routes 2 and 3. For conjugation systems, such a pathway is highly likely given the prevalence of nucleases in the periplasm. Indeed, in uninterrupted matings over periods of many minutes, E. coli Hfr strains can translocate single-stranded forms of nearly entire chromosomes through F-type T4SSs without apparent degradation. A one-step pathway does not exclude the use of a T4CP for delivery of DNA cargoes across the inner membrane (Fig. 2, route 1), however, because the physical disposition of the T4CP relative to the T4SS3–10 subassembly is not known presently. Thus, the T4CP could fit within the body of the IMC, possibly even substituting for one or both of the VirB4 “legs,” to enable DNA delivery through a T4CP-channel contact without substrate exposure to the periplasm (see Fig. 2). It is simplest to envision that a one-step pathway also is employed for translocation of protein substrates, but this is not invariably the case, because at least two systems lack a T4CP and instead rely on the GSP or another inner membrane translocase for delivery of the protein substrate to the periplasm. The Bordetella pertussis Ptl system was the first described T4SS utilizing a two-step pathway. The sole substrate of the Ptl system is the pertussis toxin (PT) of the A/B family of toxins. The A and B subunits of PT carry canonical sec signal sequences for secretion to the periplasm by the GSP. Once in the periplasm, the PT subunits assemble as the holotoxin, which is then recruited and exported by the Ptl system to the extracellular milieu (11). The Ptl system resembles other P-type systems in subunit composition and most likely in overall architecture (see Fig. 1 and Fig. 2) (87). With the R388 T4SS3–10 subassembly as a structural reference (Fig. 2C), it is enticing to suggest that the holotoxin enters the Ptl T4SS through a portal positioned at the IMC/OMC junction. The VirB T4SSs elaborated by Brucella spp., also likely employ a two-step translocation route for delivery of effectors into eukaryotic host cells. Interestingly, the Brucella T4SSs also lack associated T4CPs, yet only some effectors carry canonical sec signal sequences, while others do not. Precisely how the latter are translocated across the inner membrane remains an intriguing question for further study (88, 89, 90). Signal-Mediated Activation of T4SSs The T4SSs generally translocate substrates only upon establishment of direct contact with target cells. This indicates that the channels are gated and activated by propagation of a signal from recipient to donor cells. Studies in the 1970s of the F transfer system supplied strong experimental support for such a signal, generated upon contact of the F pilus with a recipient cell (91). This signal is transduced to the interior of the donor cell where it stimulates early F plasmid-processing reactions. More recent work supplied further evidence for signal communication from the F pilus to the cell interior (92). The F-like R1-encoded pilus is the receptor for bacteriophage R17, which enters cells through undefined mechanisms involving pilus retraction or penetration of the pilus lumen. It was discovered that R17 binds and enters only cells in which the TraD T4CP has already engaged with the F relaxosome. Phage R17 binding appears to supply a requisite extracellular signal, possibly by mimicking recipient cell contact, which together with the DNA substrate-T4CP docking signal activates the channel to allow phage uptake. Studies in A. tumefaciens also have provided evidence for signal transduction along the VirB/VirD4 T4SS. In this system, VirB10 was shown to undergo a conformational switch in response to sensing of two intracellular signals mediated through the VirD4 T4CP. The first signal is conveyed through a two-step reaction involving the docking of the T-DNA substrate with VirD4 and its subsequent transfer to the VirB11 ATPase (49). The second signal corresponds to the ATP hydrolysis activities of VirD4 and VirB11 (93). Integration of these signals results in a conformational change in the structural scaffold VirB10, as shown by a change in protease susceptibility. This conformational change is necessary for passage of the DNA substrate through the distal portion of the VirB channel composed of the VirB2 pilin and VirB9 subunits, as shown with the TrIP assay (49, 93). The nature of the conformational change is not yet known, but a role for VirB10 in the coupling of intra-cellular signals to gating of an outer membrane pore is suggested by isolation of a mutation near the VirB10 AP pore that “locks” VirB10 in the energized conformation and allows for release of secretion substrates to the cell surface independently of target cell contact (94). Therefore, as shown for a number of small-molecule and macromolecular transport systems, the VirB/VirD4 T4SS appears to be activated by a combination of ligand and energy signals. It is tempting to suggest that intra- and extracellular signals generally activate T4SS channels through the envelope-spanning VirD4-VirB10 transduction network. For type IV secretion, signal activation would open channels enabling substrate passage to target cells. In a reverse reaction, exogenous DNA or phage binding to pilus or other T4SS receptors might activate channels for uptake of nucleic acids across the cell envelope. MODULATION OF DONOR-TARGET CELL INTERACTIONS BY ADAPTED FORMS OF CHANNEL SUBUNITS Besides their functions as structural components of the translocation channel, several T4SS subunits have been adapted for novel purposes through acquisition of novel domains or motifs. Most notably, the IMC component VirB6 and the OMC subunits VirB7 and VirB10 of several T4SSs carry domains of demonstrated or postulated importance for modulation of donor cell contacts with bacterial or eukaryotic target cells. The next sections describe some of these specialized functions. Extended VirB6s VirB6 subunits are minimally ~300 kDa and have at least 5 or 6 membrane-spanning helices (2, 72). A subset additionally carries a large hydrophilic domain of functional importance. These subunits were designated “extended VirB6,” and they are found in conjugation systems and many effector translocator systems (2). For the F-type T4SSs, the polytopic motif of VirB6-like TraG subunits is followed by a large ~600 residue C-terminal domain (Fig. 4) (22). This domain is extensively α-helical in its predicted secondary structure, and has been suggested to functionally substitute for VirB8, on the basis of its periplasmic location and the absence of a VirB8 ortholog in the F-type systems (14, 22). Interestingly, TraG also is involved in entry exclusion, a process that blocks redundant DNA transfer between identical donor cells. When a donor cell contacts another donor cell, the C-terminal domain of TraG is thought to extend across the outer membranes of the paired cells to form a specific contact with TraS, a protein also encoded by F-carrying donor cells (Fig. 4) (95, 96). This contact signals a nonproductive donor-donor cell junction, and blocks DNA transfer. Similar findings were reported for homologs of TraG and TraS encoded by the SXT ICE (integrative and conjugative element) of Vibrio cholera (97). Precisely how the TraG-TraS interaction blocks DNA transfer is not yet known, but might involve transduction of a signal across the donor-donor cell junction resulting in a conformational change in the T4SS that blocks channel function (Fig. 4). The Rickettsia spp. P-type T4SSs have multiple copies of variant forms of “extended-VirB6” subunits (98, 99). They range in size between ~600 and 1,400 residues and the polytopic VirB6 motif can be located in one or both terminal regions or centrally. There is evidence for surface display of VirB6 domains in Wolbachia, supporting the notion that such exported domains are involved in establishment of endosymbiotic relationships (Fig. 4) (100). Whether these domains protrude through the OMC chamber as depicted in Fig. 4, or are proteolytically cleaved from the rest of the protein prior to export to the cell surface is presently not known. I-type systems also code for “extended-VirB6” subunits. E. coli plasmid R64 encodes TraY, a 745-residue protein with an unusual hydropathy profile (19). The N- and C-terminal thirds of the protein are highly hydrophobic with between 4 and 6 TM motifs, and the central third is hydrophilic and predicted to reside in the periplasm. In the related L. pneumophila Dot/Icm system, the TraY ortholog is DotA. DotA is ~300 residues larger than TraY, and possesses the same general hydropathy profile with multiple N- and C-terminal TM domains flanking a central hydrophilic domain. Strikingly, DotA is released in a Dot/Icm-dependent manner to the extracellular milieu where it forms ring-shaped complexes (Fig. 4) (101). In contrast to R64-encoded TraY, DotA has an N-terminal signal sequence, leading to a proposal that it is secreted across the membrane by the GSP (101). If true, DotA would be delivered to the cell surface in two steps: first, across the inner membrane by the GSP and, second, across the outer membrane by the Dot/Icm T4SS (Fig. 4). No function has been ascribed to the extracellular form of DotA, and no follow-up studies have defined its export route in further detail. VirB7 Adaptations VirB7 subunits are typically small lipoproteins that play an important role in stabilizing the OMC at the outer membrane. However, larger forms of VirB7-like lipoproteins functioning in P- and I-type systems have novel features, including surface variable regions, as shown for Helicobacter pylori CagT (102, 103), or N0 domains as shown for Xanthomonas citri VirB7 and L. pneumophila DotD (Fig. 5) (104, 105). Studies have shown that surface-variable CagT is required for CagA translocation and pilus biogenesis (106, 107), and also might contribute to immune evasion by H. pylori (102). By contrast, No domains are structural modules thought to serve a variety of functions relating to transport processes. They are features of transport-associated proteins, including TonB, secretins of the types II and III secretion systems and related type IV pilus systems, bacteriophages, and type VI secretion systems (see reference 104). The N0 domains of X. citri VirB7 and L. pneumophila DotD are envisioned to form additional rings at the base of the OMC that contribute to channel gating or contacts with the IMC (104, 105). VirB10 Adaptations The VirB10-like subunits are among the most sequence-and structurally variable subunits of the T4SSs. In the Cag T4SS of H. pylori, only a small C-terminal region of CagY is similar to VirB10, whereas a large central region is composed of multiple repeats (Fig. 5) (108). This central region is surface displayed and associates with a pilus structure (109, 110). Furthermore, during infection this region undergoes extensive genetic rearrangements that disrupt or activate the Cag T4SS (108, 110). Through host immunedriven recombination, CagY is postulated to function as a sensor of the host immune response and, in turn, regulate Cag T4SS function to maximize persistent infection (110). A second example of an interesting but as yet unexplored OMC adaptation is found in the I-type systems. In E. coli, plasmid R64 codes for TraO, which closely resembles VirB10 in size and predicted overall structure (19). In striking contrast, in the related L. pneumophila Dot/Icm system, “VirB10-like” DotG is over 1,000 residues and only the extreme C-terminal region resembles VirB10 (111, 112). DotG possesses central variable domains consisting in part of multiple sets of pentapeptide repeats (111). Although this region of DotG has been postulated to reside in the periplasm, it is noteworthy that DotG homologs from various Legionella species carry structural motifs of known bacterial effector proteins, as shown by Phyre 2 modeling (Fig. 5) (113). Most strikingly, L. pneumophila DotG (accession no. AF026534.1) has structural folds highly similar to PipB2, a T3SS effector of Salmonella enterica serovar Typhimurium (Phyre2 modeling: c2leza; 99.2% confidence; 18% sequence identity) During Salmonella infection, secreted PipB2 participates in reorganization of late endosome/lysosome compartments by linking kinesin-1 onto the Salmonella-containing vacuole (SCV) membrane. The pentapeptide motif is required to efficiently recruit kinesin-1, whereas the N-terminal domain suffices for type III translocation and association with SCVs (114, 115). A C-proximal pentapeptide motif of DotG (accession no. AF026534.1) also has a structural fold similar to the Salmonella T3SS effector SopA (Phyre2 modeling: c2qzaA, 99.7% confidence; 12% sequence identity), which structurally mimics eukaryotic ubiquitin ligase (Fig. 5) (116). The central regions of DotG subunits from different L. pneumophila species are highly variable, and a search among 11 other DotG/IcmE subunits identified structural folds similar to other effector domains that were mainly embedded within pentapeptide motifs (P. J. Christie, unpublished findings). Thus, it can be proposed that DotG subunits of Dot/Icm T4SSs might elaborate effector domains that, reminiscent of passenger domains of type V autotransporters (117), are either tethered to the L. pneumophila cell surface or released by proteolysis for delivery into eukaryotic cells. CONJUGATIVE PILI AND PILUS BIOGENESIS All conjugation systems and probably most effector translocator systems of Gram-negative bacteria elaborate pili or other surface appendages. Although there is compelling evidence that these structures promote attachment of donor cells to potential bacterial or eukaryotic target cells, the contributions of these organelles to the process of substrate transfer is somewhat controversial. On the one hand, studies of E. coli F-type pili showed that donor cells can inefficiently deliver the F plasmid to distantly located recipient cells, suggestive of a role for extended F pili as conduits for the DNA substrate (118, 119). On the other hand, E. coli and A. tumefaciens mutants lacking detectable pili can still efficiently transfer DNA substrates, establishing that extended pili are not obligatory features of T4SS channels (61, 62, 120, 121). In natural settings, it is likely these surface organelles enhance DNA transfer efficiencies, not as conduits for substrates, but rather through their ability to promote aggregation of donor and recipient cells and development of robust biofilms on biotic and abiotic surfaces. Through these adhesive activities, conjugative pili promote cellular contacts favoring formation of mating junctions that resist shear forces and other environmental perturbations. E. coli F pili are presently the best characterized group of conjugative pili. These pili are typically long, ranging from 1 to 20 µm in length, and flexible (14). Among their most noteworthy properties, they stochastically and dynamically extend and retract. This is thought to enable donor cells to scan the local environment for viable recipient cells as a means of enhancing F plasmid transfer efficiencies, particularly in low-cell-density, aqueous environments (14, 15, 16). In contrast to the F pili, the P pili are thicker, more rigid, and typically much shorter, although length measurements are complicated by the fact that isolated pili are often broken (122, 123, 124). These pili do not appear to undergo cycles of extension and retraction, but instead accumulate in the milieu, either through breakage or an active sloughing mechanism. They tend to aggregate as a mesh of polymers, which might promote nonspecific clumping of donor and recipient cells resulting in formation of productive mating pairs. The biogenesis of pili elaborated by E. coli conjugation systems has been a subject of study for over 50 years. The general requirements for pilus assembly are the same as those for elaboration of the translocation channel, with two exceptions. First, the T4CP is not required for pilus production (14). Second, the VirB1 transglycosylase is important for pilus assembly but not a functional translocation channel (60, 125). In fact, the contribution of VirB1 to pilus assembly appears not to be associated with its cell wall hydrolase activity but rather with the export of a proteolytic fragment termed VirB1* to the cell surface. At this location, VirB1* interacts with VirB2 pilin and VirB5 and, by an undefined mechanism, promotes pilus polymerization (126). The biogenesis of conjugative pili can be divided into early- and later-stage reactions, as summarized below. Early-Stage Pilus Assembly Reactions F and P pili are synthesized as pro-proteins with unusually long (~30 to 50 residues) leader peptides that are cleaved upon insertion into the inner membrane (127). The F plasmid-encoded TraA pro-pilin inserts into the membrane by a mechanism dependent on the proton motive force, but not the GSP (128). The F plasmidencoded membrane protein TraQ also is required for correct orientation and stabilization of TraA pro-pilin in the membrane (129). Membrane-embedded pro-pilins are processed further by proteases and other posttranslational modifications. Following signal sequence cleavage by LepB, F pili are acetylated at their N termini, although this modification is not required for F pilus assembly (14). Pilin subunits of P pili undergo a novel head-to-tail cyclization required for stabilization of the pilin in the membrane and for pilus assembly (130, 131). Processed forms of the conjugative pilins typically are composed of alternating stretches of hydrophilic and hydrophobic residues along the length of the protein, giving a characteristic membrane topology such that the N and C termini are in the periplasm and a small hydrophilic domain of only a few residues is in the cytoplasm (128, 132, 133, 134). Approximately ~100,000 monomers of F pilin assemble in the membrane for use in building the pilus upon receipt of an unknown signal (135). F pili assemble by sequential addition of pilin monomers to the base of the growing pilus (15). These pili are composed of a single type of pilin, although the composition and nature of the pilus tip is currently unknown. F pili are hollow cylinders of ~9 nm in diameter with an axial lumen of ~3 nm. Recent studies showed that a single F pilus is composed of distinct 1-start and 4-start helical symmetries, a property thought to enable these pili to withstand considerable extension and contraction forces as might be encountered upon engagement of the pilus with recipient cells (136). E. coli cells typically possess 1 to 5 F pili per cell, although variant F systems can encode up to 20 pili per cell. The pili are randomly distributed around the cell, and it is thought that there are more pilus assembly sites on the cell surface than extended pili (14). The structures of P pili have not yet been solved. They are difficult to visualize when attached to the cell surface, but indirect detection through decoration with pilus-specific phage or green fluorescent protein labeling of pilus assembly proteins yielded estimates as high as 20 pilus assembly sites per cell (14, 137). The mechanism by which pilins are extracted from the hydrophobic membrane environment was evaluated by monitoring accessibility of Cys residues engineered along the length of A. tumefaciens VirB2 to a membrane-impermeable thiol-reactive reagent. This study defined the inner membrane topology of VirB2 and also presented evidence that the VirB4 ATPase, with a contribution by the VirB11 ATPase, catalyzes extraction of the pilin from the membrane (134). Consistent with these findings, VirB4 also stabilizes subunits required for pilus assembly and is essential for interaction of VirB2 with another pilus-associated protein, VirB5 (68). VirB5 associates with the tip of pili elaborated by the A. tumefaciens VirB/VirD4 system, and the VirB2-VirB5 interaction is thought to be essential for pilus nucleation (138). VirB4 thus is postulated to function in dislocation of mature pilins from the inner membrane to mediate formation of the VirB2-VirB5 nucleation complex for subsequent pilus polymerization (134). F-type systems encode VirB5-like TraE subunits, but no evidence exists for a pilus tip association. Rather, these subunits are thought to form part of the T4SS in the periplasm, which, interestingly, agrees with recent evidence for association of VirB5-like TrwJ with the IMC of the R388 T4SS3–10 structure (22, 26). VirB5 subunits might have multiple functions in relation to assembly of the IMC as well as the conjugative pilus. Later-Stage Pilus Assembly and F Pilus Dynamics The membrane pool of pilin monomers is recruited upon receipt of an unknown signal(s) to build the pilus, presumably within or on top of the T4SS. In view of the R388 T4SS3–10 structure showing that two VirB4 hexamers are located at the base of the IMC (Fig. 2) (26), it is reasonable to suggest that VirB4 subunits catalyze extraction of pilin monomers from the membrane and then feed them into the chamber of the OMC. Once in the OMC chamber, pilins would undergo polymerization to build the conjugative pilus. The central chamber of the pKM101-encoded OMC is approximately 100 Å in diameter, sufficiently large to house the conjugative pilus, but the diameter of the outer membrane cap is at most only 32 Å and clearly not large enough to accommodate the pilus. These observations suggest at least two alternative models for pilus assembly. First, the pilus assembles from an IMC platform, and, as it extends through the OMC chamber, it induces profound structural changes in the distal portion of OMC. Second, pilin monomers are delivered through the OMC chamber where polymerization initiates on an outer membrane platform formed by VirB10’s AP (cap) domain (see Fig. 2C). At this time, it is not possible to discriminate between these models. As mentioned above, the F pili are so far the only group of conjugative pili for which a dynamic mode of extension and retraction has been unequivocally established. F pilus assembly requires a putative TraV/TraK/TraB OMC, as well as IMC subunits VirB3-like TraL, VirB4-like ATPase TraC, and VirB6-like TraG (14). These systems code for a unique set of proteins, TraF, -H, -U, -W and TrbB, -I (Fig. 5) (16, 139, 140). These proteins form an interaction network distinct from the presumptive TraV/K/B OMC (141). The TraF/H/U/W-TrbB/I complex is thought to regulate the dynamics of F pilus extension/retraction, although the underlying mechanism is not yet defined. TraF, -H, -U, and -W localize at the outer membrane (14). TrbI is of special interest as a bitopic inner membrane protein that contributes specifically to the process of pilus retraction through contacts with the outer membrane-associated TraF/H/U/W protein complex (Fig. 5) (142). F pilus extension and retraction has been visualized by use of fluorescently labeled phage R17, which binds along the sides of F pili. Intriguingly, F pili were shown to stochastically extend and retract regardless of the presence of recipient cells in the vicinity (15). Pilus polymerization and retraction requires ~5 min for completion, and new pili are elaborated prior to retraction of older pili, suggesting that initiation and retraction events are randomly timed and not coordinated. These studies also presented evidence supporting earlier work showing that F pili rotate along their longitudinal axis during retraction, causing a bound recipient cell to spin as it is drawn closer to the donor cell (15). Such a rotary motion of flexible pili during extension and retraction is thought to allow pili to “sweep” a large volume around the donor cell in order to enhance the probability of a productive encounter with a recipient cell, which would be of particular benefit for cells growing in liquid environments (16). MATING PAIR FORMATION AND THE MATING JUNCTION In E. coli, conjugative pili first form loose contacts with potential recipients. At this stage, mating by F- and P-type plasmids can be disrupted by physical means or treatment with 0.01% SDS. Donor-recipient cell contacts soon stabilize and become difficult to break apart (143, 144, 145, 146). F-type systems encode two proteins, TraN and TraG, that specify this function (Fig. 5). These proteins, termed mating pair stabilization proteins, are unique to the F-type systems and promote formation of tight mating junctions (145). TraN family proteins are large (~600 to 1200 residues) cysteine-rich proteins thought to function as adhesins for formation of stable donor-recipient cell-mating pairs. TraN also was shown to coordinate its functions through interactions with the TraV subunit of the putative TraV/ K/B OMC (145). TraN binds both OmpA and LPS on recipient cells, which likely explains the basis for F-specific mating-pair stabilization. This was evidenced by the identification of mutations in recipients that disrupt formation of tight mating junctions in genes encoding OmpA and constituents of the lipopolysaccharide (LPS) biosynthesis pathway (145, 147). The second mating-pair stabilization protein, TraG, was discussed above as a VirB6-like protein with an α-helical C terminus that can extend or be delivered into recipient cells where it binds the inner membrane protein TraS to block DNA transfer in nonproductive donor-donor pairs (96). In matings with F-minus recipients, however, the C-terminal domain of TraG is thought to alternatively form contacts with another, unidentified inner membrane protein to stabilize the mating junction (96). TraG also has been implicated as a binding partner of TraN (148) and, through this interaction, promote formation of mating pairs (Fig. 5). The architecture of conjugative mating junctions is presently not defined in E. coli or any other species. E. coli donor cells carrying plasmid RP4 form mating junctions at poles and along the lengths of cells, as shown by electron microscopy. Junctions at cell poles were estimated to be ~150 to 200 nm in length, whereas those along the cell body extended up to 1,500 nm in length or up to half the lengths of the cells (143). Intriguingly, no structures resembling the R388 T4SS3–10 or the OMC or IMC subassemblies, or any other structures, have been detected at the mating junction (149). This suggests the intriguing possibility that intact channels form only transiently upon channel activation by substrate-docking signals and establishment of donor-recipient cell contacts. Discerning the assembly mechanism and architecture of the mating junction is further complicated by evidence that E. coli conjugation systems are exceedingly promiscuous in their capacity to deliver cargos to a wide range of Gram-negative and -positive bacteria. In fact, this promiscuity extends to eukaryotic cells, as both F-and P-type systems have been shown to deliver substrates to yeast and even human cells (150, 151, 152, 153). CONCLUSIONS AND FUTURE DIRECTIONS In recent years, mechanistic and structural features of E. coli conjugation systems have been described in unprecedented detail. Most noteworthy, the recent ultrastructural studies have generated an architectural blueprint of a nearly entire T4SS that, when combined with results of the earlier TrIP studies, enhances our understanding of the substrate transfer route. Despite this exciting progress, long-term studies of E. coli conjugation systems have described many features of the F-, P-, and I-type systems whose underlying mechanisms remain poorly understood. Answers to the following “big picture” questions surrounding type IV secretion will require sustained efforts and implementation of creative state-of-the-art technologies and approaches. What is the architecture of the translocation channel and the pathway(s) by which substrates are delivered through the channel? In this review, three possible routes across the inner membrane were described involving the T4CP, VirB ATPases, and integral membrane components of the IMC. Refined studies are needed to discriminate between these possibilities and to define specific contributions of the ATPases and ATP energy consumption to the early-stage translocation reactions. We also now have an OMC structure in atomic detail and, based on this structure and results of the TrIP studies, it is proposed that the translocation channel responsible for conveying substrates to the cell surface is located within the OMC. Visualization of this channel and the nature of its interaction with the IMC channel is critical for answering the larger question of whether T4SSs employ one- or two-step translocation pathways. Donor and recipient cells are known to form tight mating junctions, but the architecture of this junction remains poorly defined. Indeed, the mechanisms by which T4SSs establish productive target cell contacts and then presumably dissociate after mating are perhaps the greatest mysteries surrounding type IV secretion. What are the physical and functional relationships of the conjugative pilus, the translocation channel, and the mating junction? With recent advances in superresolution and cryoelectron microscopy, it should be feasible using the highly efficient F- and P-type systems in E. coli to visualize mating junction formation in real time and the junction itself at near atomic resolution. Equally important is the goal of visualizing the T4SS within the mating junction and specific T4SS machine contacts with the recipient cell envelope. Of broader biological interest, since E. coli also can deliver DNA substrates to yeast and human cells, we should be able to use this model bacterium to define the architecture of junctions formed during interkingdom mating. Intracellular and extracellular signals are known to activate T4SSs, but what are the molecular and structural details of signal activation? Also of interest, conjugative DNA transfer must be coordinated with DNA replication and cell division to avoid replication of a mobile element from occurring at the same time as transfer. A stabilization system (Stb) has been identified as one mechanism functioning to reconcile the maintenance mode with the propagation mode a mobile element (154). What other types of molecular signals and pathways coordinate these potentially conflicting processes? Finally, while there is clear evidence for the structural diversification of P-, F-, and I-type systems during evolution, what novel functions do these structural acquisitions specify? An especially interesting avenue of research for the coming years is to decipher how T4SS machine adaptations and host-derived or other environmental signals confer spatiotemporal control over substrate translocation. The E. coli conjugation systems remain the best subjects for refined structurefunction studies of T4SSs. Yet, a full understanding of T4SS diversity also will require detailed investigations of effector translocators with particular attention to specialized features acquired during the evolution of these machines. I thank members of the Christie laboratory for helpful discussions. This work was supported by NIH R01GM48476 and R21AI105454. Figure 1 Conservation of type IV secretion genes among P-, F-, and I-type T4SSs (P type) The A. tumefaciens VirB/VirD4 reference system with subunit enzymatic functions and associations with inner membrane complex (IMC), outer membrane complex (OMC), or pilus. PG Hydrolase, peptidoglycan hydrolase; T4CP, type IV coupling protein. (F type) Genes related to the virB/virD4 genes are color coded. Genes encoding functions required for F pilus/retraction are shaded in dark gray, and for mating pair stabilization (Mps) or surface exclusion in light gray (VirB6-like TraG also functions in mating pair stabilization). Uppercase letters are tra genes, lowercase letters are trb genes. (I type) virB/virD4-like genes are color coded. Genes unique to the I-type systems encoding inner membrane proteins are in beige; genes encoding subunits that functionally interact with the DotL T4CP are in light-shaded purple. P- and I-type systems employ three ATPases related to VirD4, VirB4, and VirB11; F-type systems employ only homologs of VirD4 and VirB4. Unless otherwise indicated, the systems shown are functional in E. coli. Figure 2 Architecture of a prototypical P-type T4SS The E. coli R388 T4SS3–10 structure (26), presented schematically using the A. tumefaciens VirB/VirD4 reference nomenclature and as visualized by transmission electron microscopy. (A) The T4SS is composed of an extracellular pilus, an outer membrane complex (OMC), and inner membrane complex (IMC), the VirD4-like T4CP (substrate receptor) and, for most systems, a VirB11 ATPase. Substructures of the T4SS: The conjugative pilus was not part of the solved T4SS3–10 structure, but is postulated to associate with the structure as depicted. In the A. tumefaciens VirB/VirD4 T4SS, the pilus is composed of a fragment of the VirB1 transglycosylase (VirB1*), VirB2 pilin, and VirB5 tip protein (126, 138). The OMC is composed of the VirB7, VirB9, and VirB10 subunits in copy numbers listed in parentheses (26). The VirB1 transglycosylase is required for pilus assembly but not for elaboration of the translocation channel. The IMC is composed of the VirB5, VirB8, VirB6, VirB3, and VirB4 subunits with copy numbers listed in parentheses (26). Other ATPases, including VirD4 and VirB11, were not part of the solved T4SS3–10 structure, but are postulated to associate with the IMC as depicted. Color-coding of the subunits matches that for the corresponding virB/virD4 genes in Fig. 1. (B) The A. tumefaciens subunits shown to form formaldehyde cross-links with the T-DNA substrate during transfer through the VirB/VirD4 T4SS with the TrIP assay. Red arrows denote the proposed translocation pathway (23). (C) The R3883–10 structure solved by transmission electron microscopy, reproduced with permission by reference 26. The OMC is postulated to house the translocation channel that extends through the periplasm and across the outer membrane. The OMC is connected to the IMC by a narrow stalk. Two hexamers of VirB4 extend into the cytoplasm and are postulated to establish contacts with the VirD4 T4CP and the VirB11 ATPase. (A and C) For both the schematic and solved T4SS3–10 structure, three possible routes of substrate translocation across the inner membrane are depicted in red dashed lines. A single route is postulated, in a red dashed line, through the OMC for substrate passage to the cell exterior (see text for details). IM, inner membrane; P, periplasm; OM, outer membrane. Figure 3 Domain architecture of type IV coupling proteins (T4CPs) (A) The structural prototype of the T4CPs, R388-encoded TrwB, is a hexamer with a stem composed of the N-terminal transmembrane domains (NTDs) of the 6 protomers and a ball composed of the nucleotide binding domains (NBDs) and all-alpha-domains (AADs). (Left) A space-filling model of the TrwB hexamer showing the Connolly surface (red, negatively charged; blue, positively charged) and central channel connecting the cytoplasm to the periplasm. IM, inner membrane. (Right) A ribbon diagram of the NBD (multicolored) and AAD (green). In the assembled hexamer, the AAD sits at the entrance of the hexamer lumen. Images reproduced with permission by reference 43. (B) Schematics display the domain architectures of P-types TrwB and A. tumefaciens VirD4 and F-type TraD; numbers represent domain junctions with residue numbers relative to the start codon. The AAD is boxed. Known or predicted properties are listed above each domain. TrwB lacks a C-terminal domain (CTD), but such domains are carried by other VirD4 homologs including F-type TraD. (C) The TraD T4CP in the inner membrane with the various domains listed. The extreme C terminus of the CTD (residues in parentheses) is negatively charged and forms specific contacts with C-terminal α-helical domains (purple cylinders) of the TraM Dtr accessory protein. TraM’s N-terminal ribbon-helix-helix domain (RHH; purple dots) mediates binding to sbmA sites located within the F plasmid oriT sequence. The TraD-TraM interaction specifies F plasmid transfer through the F-encoded T4SS. IM, inner membrane. Figure 4 Adaptations of the IMC: VirB6-like subunits Polytopic VirB6 subunits with lengths of ~300 residues are components of the IMC. Many T4SSs have larger forms of these subunits, designated extended-VirB6. These forms of VirB6 are composed of a polytopic VirB6-like domain and a hydrophilic domain as large as ~1,000 residues. (Left) F-type F plasmids of E. coli encode large VirB6-like TraG subunits. These subunits participate in entry exclusion, which prevents nonproductive plasmid transfer between donor cells. TraG’s C-terminal domain extends or is delivered via the T4SS across the donor-donor cell junction, where it binds the entry exclusion protein TraS located in the inner membrane of the paired donor. This interaction might transduce a signal resulting in a conformational change in the T4SS that blocks nonproductive F plasmid transfer to other F-carrying cells (96). (Middle) Rickettsial P-type T4SSs are composed of 4 or more large VirB6 subunits whose hydrophilic domains are unrelated in sequence and which might be surface-displayed for target cell binding or immune evasion (98, 99, 100). (Right) The I-type Dot/Icm system of L. pneumophila secretes the highly hydrophobic DotA (~1,000 residues) to the milieu where it assembles as large ring-shaped complexes. DotA has a canonical signal sequence, which is thought to mediate DotA delivery through the General Secretory Pathway (GSP) across the inner membrane. In the periplasm, DotA is then recruited to the Dot/Icm T4SS for delivery across the outer membrane to the cell exterior (101). Figure 5 Adaptations of the OMC: VirB7- and VirB10-like subunits The outer membrane complexes (OMCs) of paradigmatic T4SSs are composed of the small (~5-kDa) lipoprotein VirB7, outer membrane-associated VirB9, and envelope-spanning VirB10 (~50-kDa). Many T4SSs have larger forms of the VirB7- and VirB10-like subunits that carry surface-displayed domains of functional importance. (Left) H. pylori P-type T4SSs carry larger forms of VirB7-like CagT and VirB10-like CagY with repeat domains that are exposed on the cell surface. TM, transmembrane domain. (Middle) L. pneumophila I-type T4SSs also carry larger forms of the VirB7- and VirB10-like proteins. VirB7-like DotD has an N0 domain that is thought to form an extra ring of structural importance for the OMC. DotG has an N-terminal transmembrane domain (TM), a C-terminal VirB10 structural fold (20), and internal structural folds similar to that of Salmonella T3SS effector PipB2 and the Salmonella effector SopA (114, 116), as determined by Phyre2 structural modeling (113). It is proposed that these domains of DotG protrude through the OMC or are proteolytically released from the VirB10 scaffold domain for delivery into the eukaryotic cell where they exert effector activities. (Right) E. coli F-type T4SSs carry novel subcomplexes that may or may not be physically associated with the T4SS at the cell surface. The TraW/U/F/B/F/TrbI subunits mediate F pilus extension and retraction, TraN and VirB10-like TraG stabilize mating pairs and the TraT lipoprotein prevents redundant F plasmid transfer among donor cells through surface exclusion (see reference 22). Conflicts of interest: The author declares no conflicts. REFERENCES 1 Cascales E Christie PJ The versatile bacterial type IV secretion systems Nat Rev Microbiol 2003 1 137 149 15035043 2 Alvarez-Martinez CE Christie PJ Biological diversity of prokaryotic type IV secretion systems Microbiol Mol Biol Rev 2009 73 775 808 19946141 3 Cabezón E Ripoll-Rozada J Peña A de la Cruz F Arechaga I Towards an integrated model of bacterial conjugation FEMS Microbiol Rev 2015 39 81 95 25154632 4 Juhas M Horizontal gene transfer in human pathogens Crit Rev Microbiol 2015 41 101 108 23862575 5 Guglielmini J de la Cruz F Rocha EP Evolution of conjugation and type IV secretion systems Mol Biol Evol 2013 30 315 331 22977114 6 Christie PJ Whitaker N González-Rivera C Mechanism and structure of the bacterial type IV secretion systems Biochim Biophys Acta 2014 1843 1578 1591 24389247 7 Chandran Darbari V Waksman G Structural biology of bacterial type IV secretion systems Annu Rev Biochem 2015 84 603 629 26034891 8 Bhatty M Laverde Gomez JA Christie PJ The expanding bacterial type IV secretion lexicon Res Microbiol 2013 164 620 639 23542405 9 Kubori T Nagai H The Type IVB secretion system: an enigmatic chimera Curr Opin Microbiol 2016 29 22 29 26529574 10 Asrat S Davis KM Isberg RR Modulation of the host innate immune and inflammatory response by translocated bacterial proteins Cell Microbiol 2015 17 785 795 25850689 11 Locht C Coutte L Mielcarek N The ins and outs of pertussis toxin FEBS J 2011 278 4668 4682 21740523 12 Ramsey ME Woodhams KL Dillard JP The gonococcal genetic island and type IV secretion in the pathogenic Neisseria Front Microbiol 2011 2 61 21833316 13 Stingl K Müller S Scheidgen-Kleyboldt G Clausen M Maier B Composite system mediates two-step DNA uptake into Helicobacter pylori Proc Natl Acad Sci USA 2010 107 1184 1189 20080542 14 Arutyunov D Frost LS F conjugation: back to the beginning Plasmid 2013 70 18 32 23632276 15 Clarke M Maddera L Harris RL Silverman PM F-pili dynamics by live-cell imaging Proc Natl Acad Sci USA 2008 105 17978 17981 19004777 16 Silverman PM Clarke MB New insights into F-pilus structure, dynamics, and function Integr Biol Camb 2010 2 25 31 20473409 17 Christie PJ Atmakuri K Krishnamoorthy V Jakubowski S Cascales E Biogenesis, architecture, and function of bacterial type IV secretion systems Annu Rev Microbiol 2005 59 451 485 16153176 18 Costa TR Felisberto-Rodrigues C Meir A Prevost MS Redzej A Trokter M Waksman G Secretion systems in Gram-negative bacteria: structural and mechanistic insights Nat Rev Microbiol 2015 13 343 359 25978706 19 Sampei G Furuya N Tachibana K Saitou Y Suzuki T Mizobuchi K Komano T Complete genome sequence of the incompatibility group I1 plasmid R64 Plasmid 2010 64 92 103 20594994 20 Nagai H Kubori T Type IVB secretion systems of Legionella and other Gram-negative bacteria Front Microbiol 2011 2 136 21743810 21 Christie PJ Type IV secretion: the Agrobacterium VirB/ D4 and related conjugation systems Biochim Biophys Acta 2004 1694 219 234 15546668 22 Lawley TD Klimke WA Gubbins MJ Frost LS F factor conjugation is a true type IV secretion system FEMS Microbiol Lett 2003 224 1 15 12855161 23 Cascales E Christie PJ Definition of a bacterial type IV secretion pathway for a DNA substrate Science 2004 304 1170 1173 15155952 24 Fronzes R Schäfer E Wang L Saibil HR Orlova EV Waksman G Structure of a type IV secretion system core complex Science 2009 323 266 268 19131631 25 Chandran V Fronzes R Duquerroy S Cronin N Navaza J Waksman G Structure of the outer membrane complex of a type IV secretion system Nature 2009 462 1011 1015 19946264 26 Low HH Gubellini F Rivera-Calzada A Braun N Connery S Dujeancourt A Lu F Redzej A Fronzes R Orlova EV Waksman G Structure of a type IV secretion system Nature 2014 508 550 553 24670658 27 Trokter M Felisberto-Rodrigues C Christie PJ Waksman G Recent advances in the structural and molecular biology of type IV secretion systems Curr Opin Struct Biol 2014 27 16 23 24709394 28 Zechner EL Lang S Schildbach JF Assembly and mechanisms of bacterial type IV secretion machines Philos Trans R Soc Lond B Biol Sci 2012 367 1073 1087 22411979 29 Mihajlovic S Lang S Sut MV Strohmaier H Gruber CJ Koraimann G Cabezón E Moncalián G de la Cruz F Zechner EL Plasmid R1 conjugative DNA processing is regulated at the coupling protein interface J Bacteriol 2009 191 6877 6887 19767437 30 Sut MV Mihajlovic S Lang S Gruber CJ Zechner EL Protein and DNA effectors control the TraI conjugative helicase of plasmid R1 J Bacteriol 2009 191 6888 6899 19767439 31 Vincent CD Friedman JR Jeong KC Sutherland MC Vogel JP Identification of the DotL coupling protein subcomplex of the Legionella Dot/Icm type IV secretion system Mol Microbiol 2012 85 378 391 22694730 32 Lang S Gruber K Mihajlovic S Arnold R Gruber CJ Steinlechner S Jehl MA Rattei T Fröhlich KU Zechner EL Molecular recognition determinants for type IV secretion of diverse families of conjugative relaxases Mol Microbiol 2010 78 1539 1555 21143323 33 Redzej A Ilangovan A Lang S Gruber CJ Topf M Zangger K Zechner EL Waksman G Structure of a translocation signal domain mediating conjugative transfer by type IV secretion systems Mol Microbiol 2013 89 324 333 23710762 34 Parker C Meyer RJ The R1162 relaxase/primase contains two, type IV transport signals that require the small plasmid protein MobB Mol Microbiol 2007 66 252 261 17880426 35 Alperi A Larrea D Fernández-González E Dehio C Zechner EL Llosa M A translocation motif in relaxase TrwC specifically affects recruitment by its conjugative type IV secretion system J Bacteriol 2013 195 4999 5006 23995644 36 Meyer R Mapping type IV secretion signals on the primase encoded by the broad-host-range plasmid R1162 (RSF1010) J Bacteriol 2015 197 3245 3254 26381189 37 Vergunst AC Schrammeijer B den Dulk-Ras A de Vlaam CM Regensburg-Tuïnk TJ Hooykaas PJ VirB/D4-dependent protein translocation from Agrobacterium into plant cells Science 2000 290 979 982 11062129 38 Vergunst AC van Lier MC den Dulk-Ras A Stüve TA Ouwehand A Hooykaas PJ Positive charge is an important feature of the C-terminal transport signal of the VirB/D4-translocated proteins of Agrobacterium Proc Natl Acad Sci USA 2005 102 832 837 15644442 39 Cabezón E Sastre JI de la Cruz F Genetic evidence of a coupling role for the TraG protein family in bacterial conjugation Mol Gen Genet 1997 254 400 406 9180693 40 Schröder G Lanka E TraG-like proteins of type IV secretion systems: functional dissection of the multiple activities of TraG (RP4) and TrwB (R388) J Bacteriol 2003 185 4371 4381 12867445 41 Gomis-Rüth FX Moncalián G Pérez-Luque R González A Cabezón E de la Cruz F Coll M The bacterial conjugation protein TrwB resembles ring helicases and F1-ATPase Nature 2001 409 637 641 11214325 42 Hormaeche I Alkorta I Moro F Valpuesta JM Goni FM De La Cruz F Purification and properties of TrwB, a hexameric, ATP-binding integral membrane protein essential for R388 plasmid conjugation J Biol Chem 2002 277 46456 46462 12244053 43 Gomis-Rüth FX Coll M Structure of TrwB, a gatekeeper in bacterial conjugation Int J Biochem Cell Biol 2001 33 839 843 11461827 44 Schröder G Krause S Zechner EL Traxler B Yeo HJ Lurz R Waksman G Lanka E TraG-like proteins of DNA transfer systems and of the Helicobacter pylori type IV secretion system: inner membrane gate for exported substrates? J Bacteriol 2002 184 2767 2779 11976307 45 de Paz HD Larrea D Zunzunegui S Dehio C de la Cruz F Llosa M Functional dissection of the conjugative coupling protein TrwB J Bacteriol 2010 192 2655 2669 20363945 46 Whitaker N Chen Y Jakubowski SJ Sarkar MK Li F Christie PJ The all-alpha domains of coupling proteins from the Agrobacterium tumefaciens VirB/VirD4 and Enterococcus faecalis pCF10-encoded type IV secretion systems confer specificity to binding of cognate DNA substrates J Bacteriol 2015 197 2335 2349 25939830 47 van Kregten M Lindhout BI Hooykaas PJ van der Zaal BJ Agrobacterium-mediated T-DNA transfer and integration by minimal VirD2 consisting of the relaxase domain and a type IV secretion system translocation signal Mol Plant Microbe Interact 2009 22 1356 1365 19810805 48 Atmakuri K Cascales E Burton OT Banta LM Christie PJ Agrobacterium ParA/MinD-like VirC1 spatially coordinates early conjugative DNA transfer reactions EMBO J 2007 26 2540 2551 17505518 49 Cascales E Atmakuri K Sarkar MK Christie PJ DNA substrate-induced activation of the Agrobacterium VirB/VirD4 type IV secretion system J Bacteriol 2013 195 2691 2704 23564169 50 Sastre JI Cabezón E de la Cruz F The carboxyl terminus of protein TraD adds specificity and efficiency to F-plasmid conjugative transfer J Bacteriol 1998 180 6039 6042 9811665 51 Beranek A Zettl M Lorenzoni K Schauer A Manhart M Koraimann G Thirty-eight C-terminal amino acids of the coupling protein TraD of the F-like conjugative resistance plasmid R1 are required and sufficient to confer binding to the substrate selector protein TraM J Bacteriol 2004 186 6999 7006 15466052 52 Lu J Wong JJ Edwards RA Manchak J Frost LS Glover JN Structural basis of specific TraD-TraM recognition during F plasmid-mediated bacterial conjugation Mol Microbiol 2008 70 89 99 18717787 53 Lu J Frost LS Mutations in the C-terminal region of TraM provide evidence for in vivo TraM-TraD interactions during F-plasmid conjugation J Bacteriol 2005 187 4767 4773 15995191 54 Atmakuri K Cascales E Christie PJ Energetic components VirD4, VirB11 and VirB4 mediate early DNA transfer reactions required for bacterial type IV secretion Mol Microbiol 2004 54 1199 1211 15554962 55 Llosa M Zunzunegui S de la Cruz F Conjugative coupling proteins interact with cognate and heterologous VirB10-like proteins while exhibiting specificity for cognate relaxosomes Proc Natl Acad Sci USA 2003 100 10465 10470 12925737 56 de Paz HD Sangari FJ Bolland S García-Lobo JM Dehio C de la Cruz F Llosa M Functional interactions between type IV secretion systems involved in DNA transfer and virulence Microbiology 2005 151 3505 3516 16272374 57 Tato I Zunzunegui S de la Cruz F Cabezon E TrwB, the coupling protein involved in DNA transport during bacterial conjugation, is a DNA-dependent ATPase Proc Natl Acad Sci USA 2005 102 8156 8161 15919815 58 Tato I Matilla I Arechaga I Zunzunegui S de la Cruz F Cabezon E The ATPase activity of the DNA transporter TrwB is modulated by protein TrwA: implications for a common assembly mechanism of DNA translocating motors J Biol Chem 2007 282 25569 25576 17599913 59 Ripoll-Rozada J Zunzunegui S de la Cruz F Arechaga I Cabezón E Functional interactions of VirB11 traffic ATPases with VirB4 and VirD4 molecular motors in type IV secretion systems J Bacteriol 2013 195 4195 4201 23852869 60 Berger BR Christie PJ Genetic complementation analysis of the Agrobacterium tumefaciens virB operon: virB2 through virB11 are essential virulence genes J Bacteriol 1994 176 3646 3660 8206843 61 Haase J Lurz R Grahn AM Bamford DH Lanka E Bacterial conjugation mediated by plasmid RP4: RSF1010 mobilization, donor-specific phage propagation, and pilus production require the same Tra2 core components of a proposed DNA transport complex J Bacteriol 1995 177 4779 4791 7642506 62 Sagulenko E Sagulenko V Chen J Christie PJ Role of Agrobacterium VirB11 ATPase in T-pilus assembly and substrate selection J Bacteriol 2001 183 5813 5825 11566978 63 Savvides SN Yeo HJ Beck MR Blaesing F Lurz R Lanka E Buhrdorf R Fischer W Haas R Waksman G VirB11 ATPases are dynamic hexameric assemblies: new insights into bacterial type IV secretion EMBO J 2003 22 1969 1980 12727865 64 Hare S Bayliss R Baron C Waksman G A large domain swap in the VirB11 ATPase of Brucella suis leaves the hexameric assembly intact J Mol Biol 2006 360 56 66 16730027 65 Savvides SN Secretion superfamily ATPases swing big Structure 2007 15 255 257 17355860 66 Akeda Y Galán JE Chaperone release and unfolding of substrates in type III secretion Nature 2005 437 911 915 16208377 67 Middleton R Sjölander K Krishnamurthy N Foley J Zambryski P Predicted hexameric structure of the Agrobacterium VirB4 C terminus suggests VirB4 acts as a docking site during type IV secretion Proc Natl Acad Sci USA 2005 102 1685 1690 15668378 68 Yuan Q Carle A Gao C Sivanesan D Aly KA Höppner C Krall L Domke N Baron C Identification of the VirB4-VirB8-VirB5-VirB2 pilus assembly sequence of type IV secretion systems J Biol Chem 2005 280 26349 26359 15901731 69 Li F Alvarez-Martinez C Chen Y Choi KJ Yeo HJ Christie PJ Enterococcus faecalis PrgJ, a VirB4-like ATPase, mediates pCF10 conjugative transfer through substrate binding J Bacteriol 2012 194 4041 4051 22636769 70 Peña A Matilla I Martín-Benito J Valpuesta JM Carrascosa JL de la Cruz F Cabezón E Arechaga I The hexameric structure of a conjugative VirB4 protein ATPase provides new insights for a functional and phylogenetic relationship with DNA translocases J Biol Chem 2012 287 39925 39932 23035111 71 Durand E Oomen C Waksman G Biochemical dissection of the ATPase TraB, the VirB4 homologue of the Escherichia coli pKM101 conjugation machinery J Bacteriol 2010 192 2315 2323 20172994 72 Jakubowski SJ Krishnamoorthy V Cascales E Christie PJ Agrobacterium tumefaciens VirB6 domains direct the ordered export of a DNA substrate through a type IV secretion system J Mol Biol 2004 341 961 977 15328612 73 Rees CE Wilkins BM Transfer of Tra proteins into the recipient cell during bacterial conjugation mediated by plasmid ColIb-P9 J Bacteriol 1989 171 3152 3157 2656642 74 Rees CED Wilkins BM Protein transfer into the recipient cell during bacterial conjugation: studies with F and RP4 Mol Microbiol 1990 4 1199 1205 2172695 75 Ninio S Roy CR Effector proteins translocated by Legionella pneumophila: strength in numbers Trends Microbiol 2007 15 372 380 17632005 76 Zhu W Banga S Tan Y Zheng C Stephenson R Gately J Luo ZQ Comprehensive identification of protein substrates of the Dot/Icm type IV transporter of Legionella pneumophila PLoS One 2011 6 e17638 21408005 77 Kumar RB Das A Functional analysis of the Agrobacterium tumefaciens T-DNA transport pore protein VirB8 J Bacteriol 2001 183 3636 3641 11371528 78 Terradot L Bayliss R Oomen C Leonard GA Baron C Waksman G Structures of two core subunits of the bacterial type IV secretion system, VirB8 from Brucella suis and ComB10 from Helicobacter pylori Proc Natl Acad Sci USA 2005 102 4596 4601 15764702 79 Bailey S Ward D Middleton R Grossmann JG Zambryski PC Agrobacterium tumefaciens VirB8 structure reveals potential protein-protein interaction sites Proc Natl Acad Sci USA 2006 103 2582 2587 16481621 80 Porter CJ Bantwal R Bannam TL Rosado CJ Pearce MC Adams V Lyras D Whisstock JC Rood JI The conjugation protein TcpC from Clostridium perfringens is structurally related to the type IV secretion system protein VirB8 from Gram-negative bacteria Mol Microbiol 2012 83 275 288 22150951 81 Goessweiner-Mohr N Grumet L Arends K Pavkov-Keller T Gruber CC Gruber K Birner-Gruenberger R Kropec-Huebner A Huebner J Grohmann E Keller W The 2.5 Å structure of the Enterococcus conjugation protein TraM resembles VirB8 type IV secretion proteins J Biol Chem 2013 288 2018 2028 23188825 82 Kuroda T Kubori T Thanh Bui X Hyakutake A Uchida Y Imada K Nagai H Molecular and structural analysis of Legionella DotI gives insights into an inner membrane complex essential for type IV secretion Sci Rep 2015 5 10912 26039110 83 Paschos A den Hartigh A Smith MA Atluri VL Sivanesan D Tsolis RM Baron C An in vivo high-throughput screening approach targeting the type IV secretion system component VirB8 identified inhibitors of Brucella abortus 2308 proliferation Infect Immun 2011 79 1033 1043 21173315 84 Sarkar MK Husnain SI Jakubowski SJ Christie PJ Isolation of bacterial type IV machine subassemblies Methods Mol Biol 2013 966 187 204 23299736 85 Jakubowski SJ Kerr JE Garza I Krishnamoorthy V Bayliss R Waksman G Christie PJ Agrobacterium VirB10 domain requirements for type IV secretion and T pilus biogenesis Mol Microbiol 2009 71 779 794 19054325 86 Fronzes R Christie PJ Waksman G The structural biology of type IV secretion systems Nat Rev Microbiol 2009 7 703 714 19756009 87 Winans SC Burns DL Christie PJ Adaptation of a conjugal transfer system for the export of pathogenic macromolecules Trends Microbiol 1996 4 64 68 8820569 88 de Jong MF Sun YH den Hartigh AB van Dijl JM Tsolis RM Identification of VceA and VceC, two members of the VjbR regulon that are translocated into macrophages by the Brucella type IV secretion system Mol Microbiol 2008 70 1378 1396 19019140 89 Myeni S Child R Ng TW Kupko JJ III Wehrly TD Porcella SF Knodler LA Celli J Brucella modulates secretory trafficking via multiple type IV secretion effector proteins PLoS Pathog 2013 9 e1003556 23950720 90 Ke Y Wang Y Li W Chen Z Type IV secretion system of Brucella spp. and its effectors Front Cell Infect Microbiol 2015 5 72 26528442 91 Ou JT Mating signal and DNA penetration deficiency in conjugation between male Escherichia coli and minicells Proc Natl Acad Sci USA 1975 72 3721 3725 1059160 92 Lang S Kirchberger PC Gruber CJ Redzej A Raffl S Zellnig G Zangger K Zechner EL An activation domain of plasmid R1 TraI protein delineates stages of gene transfer initiation Mol Microbiol 2011 82 1071 1085 22066957 93 Cascales E Christie PJ Agrobacterium VirB10, an ATP energy sensor required for type IV secretion Proc Natl Acad Sci USA 2004 101 17228 17233 15569944 94 Banta LM Kerr JE Cascales E Giuliano ME Bailey ME McKay C Chandran V Waksman G Christie PJ An Agrobacterium VirB10 mutation conferring a type IV secretion system gating defect J Bacteriol 2011 193 2566 2574 21421757 95 Anthony KG Klimke WA Manchak J Frost LS Comparison of proteins involved in pilus synthesis and mating pair stabilization from the related plasmids F and R100-1: insights into the mechanism of conjugation J Bacteriol 1999 181 5149 5159 10464182 96 Audette GF Manchak J Beatty P Klimke WA Frost LS Entry exclusion in F-like plasmids requires intact TraG in the donor that recognizes its cognate TraS in the recipient Microbiology 2007 153 442 451 17259615 97 Marrero J Waldor MK Determinants of entry exclusion within Eex and TraG are cytoplasmic J Bacteriol 2007 189 6469 6473 17573467 98 Gillespie JJ Ammerman NC Dreher-Lesnick SM Rahman MS Worley MJ Setubal JC Sobral BS Azad AF An anomalous type IV secretion system in Rickettsia is evolutionarily conserved PLoS One 2009 4 e4833 19279686 99 Gillespie JJ Brayton KA Williams KP Diaz MA Brown WC Azad AF Sobral BW Phylogenomics reveals a diverse Rickettsiales type IV secretion system Infect Immun 2010 78 1809 1823 20176788 100 Rancès E Voronin D Tran-Van V Mavingui P Genetic and functional characterization of the type IV secretion system in Wolbachia J Bacteriol 2008 190 5020 5030 18502862 101 Nagai H Roy CR The DotA protein from Legionella pneumophila is secreted by a novel process that requires the Dot/Icm transporter EMBO J 2001 20 5962 5970 11689436 102 Fischer W Assembly and molecular mode of action of the Helicobacter pylori Cag type IV secretion apparatus FEBS J 2011 278 1203 1212 21352490 103 Terradot L Waksman G Architecture of the Helicobacter pylori Cag-type IV secretion system FEBS J 2011 278 1213 1222 21352491 104 Nakano N Kubori T Kinoshita M Imada K Nagai H Crystal structure of Legionella DotD: insights into the relationship between type IVB and type II/III secretion systems PLoS Pathog 2010 6 e1001129 20949065 105 Souza DP Andrade MO Alvarez-Martinez CE Arantes GM Farah CS Salinas RK A component of the Xanthomonadaceae type IV secretion system combines a VirB7 motif with a N0 domain found in outer membrane transport proteins PLoS Pathog 2011 7 e1002031 21589901 106 Ding H Zeng H Huang L Dong Y Duan Y Mao X Guo G Zou Q Helicobacter pylori chaperone-like protein CagT plays an essential role in the translocation of CagA into host cells J Micro-biol Biotechnol 2012 22 1343 1349 107 Johnson EM Gaddy JA Voss BJ Hennig EE Cover TL Genes required for assembly of pili associated with the Helicobacter pylori cag type IV secretion system Infect Immun 2014 82 3457 3470 24891108 108 Aras RA Fischer W Perez-Perez GI Crosatti M Ando T Haas R Blaser MJ Plasticity of repetitive DNA sequences within a bacterial (Type IV) secretion system component J Exp Med 2003 198 1349 1360 14581606 109 Rohde M Püls J Buhrdorf R Fischer W Haas R A novel sheathed surface organelle of the Helicobacter pylori cag type IV secretion system Mol Microbiol 2003 49 219 234 12823823 110 Barrozo RM Cooke CL Hansen LM Lam AM Gaddy JA Johnson EM Cariaga TA Suarez G Peek RM Jr Cover TL Solnick JV Functional plasticity in the type IV secretion system of Helicobacter pylori PLoS Pathog 2013 9 e1003189 23468628 111 Segal G Purcell M Shuman HA Host cell killing and bacterial conjugation require overlapping sets of genes within a 22-kb region of the Legionella pneumophila genome Proc Natl Acad Sci USA 1998 95 1669 1674 9465074 112 Vogel JP Andrews HL Wong SK Isberg RR Conjugative transfer by the virulence system of Legionella pneumophila Science 1998 279 873 876 9452389 113 Kelley LA Sternberg MJ Protein structure prediction on the Web: a case study using the Phyre server Nat Protoc 2009 4 363 371 19247286 114 Henry T Couillault C Rockenfeller P Boucrot E Dumont A Schroeder N Hermant A Knodler LA Lecine P Steele-Mortimer O Borg JP Gorvel JP Méresse S The Salmonella effector protein PipB2 is a linker for kinesin-1 Proc Natl Acad Sci USA 2006 103 13497 13502 16938850 115 Baisón-Olmo F Cardenal-Muñoz E Ramos-Morales F PipB2 is a substrate of the Salmonella pathogenicity island 1-encoded type III secretion system Biochem Biophys Res Commun 2012 423 240 246 22640733 116 Diao J Zhang Y Huibregtse JM Zhou D Chen J Crystal structure of SopA, a Salmonella effector protein mimicking a eukaryotic ubiquitin ligase Nat Struct Mol Biol 2008 15 65 70 18066077 117 Bernstein HD Type V secretion: the autotransporter and two-partner secretion pathways Ecosal Plus 2010 4 118 Harrington LC Rogerson AC The F pilus of Escherichia coli appears to support stable DNA transfer in the absence of wall-to-wall contact between cells J Bacteriol 1990 172 7263 7264 1979324 119 Babic A Lindner AB Vulic M Stewart EJ Radman M Direct visualization of horizontal gene transfer Science 2008 319 1533 1536 18339941 120 Lawley TD Gordon GS Wright A Taylor DE Bacterial conjugative transfer: visualization of successful mating pairs and plasmid establishment in live Escherichia coli Mol Microbiol 2002 44 947 956 12010490 121 Jakubowski SJ Cascales E Krishnamoorthy V Christie PJ Agrobacterium tumefaciens VirB9, an outer-membrane-associated component of a type IV secretion system, regulates substrate selection and T-pilus biogenesis J Bacteriol 2005 187 3486 3495 15866936 122 Bradley DE Morphological and serological relationships of conjugative pili Plasmid 1980 4 155 169 6152840 123 Bradley DE Taylor DE Cohen DR Specification of surface mating systems among conjugative drug resistance plasmids in Escherichia coli K-12 J Bacteriol 1980 143 1466 1470 6106013 124 Paranchych W Frost LS The physiology and biochemistry of pili Adv Microb Physiol 1988 29 53 114 2898203 125 Bayer M Eferl R Zellnig G Teferle K Dijkstra A Koraimann G Högenauer G Gene 19 of plasmid R1 is required for both efficient conjugative DNA transfer and bacteriophage R17 infection J Bacteriol 1995 177 4279 4288 7543471 126 Zupan J Hackworth CA Aguilar J Ward D Zambryski P VirB1* promotes T-pilus formation in the vir-Type IV secretion system of Agrobacterium tumefaciens J Bacteriol 2007 189 6551 6563 17631630 127 Majdalani N Moore D Maneewannakul S Ippen-Ihler K Role of the propilin leader peptide in the maturation of F pilin J Bacteriol 1996 178 3748 3754 8682776 128 Majdalani N Ippen-Ihler K Membrane insertion of the F-pilin subunit is Sec independent but requires leader peptidase B and the proton motive force J Bacteriol 1996 178 3742 3747 8682775 129 Maneewannakul K Maneewannakul S Ippen-Ihler K Synthesis of F pilin J Bacteriol 1993 175 1384 1391 8095257 130 Eisenbrandt R Kalkum M Lai EM Lurz R Kado CI Lanka E Conjugative pili of IncP plasmids, and the Ti plasmid T pilus are composed of cyclic subunits J Biol Chem 1999 274 22548 22555 10428832 131 Eisenbrandt R Kalkum M Lurz R Lanka E Maturation of IncP pilin precursors resembles the catalytic Dyad-like mechanism of leader peptidases J Bacteriol 2000 182 6751 6761 11073921 132 Paiva WD Grossman T Silverman PM Characterization of F-pilin as an inner membrane component of Escherichia coli K12 J Biol Chem 1992 267 26191 26197 1464628 133 Paiva WD Silverman PM Effects of F-encoded components and F-pilin domains on the synthesis and membrane insertion of TraA-’PhoA fusion proteins Mol Microbiol 1996 19 1277 1286 8730869 134 Kerr JE Christie PJ Evidence for VirB4-mediated dislocation of membrane-integrated VirB2 pilin during biogenesis of the Agrobacterium VirB/VirD4 type IV secretion system J Bacteriol 2010 192 4923 4934 20656905 135 Frost LS Schaechter M Conjugation, bacterial Desk Encyclopedia of Microbiology 2009 2nd San Diego, CA Academic Press 294 308 136 Wang YA Yu X Silverman PM Harris RL Egelman EH The structure of F-pili J Mol Biol 2009 385 22 29 18992755 137 Gilmour MW Lawley TD Rooker MM Newnham PJ Taylor DE Cellular location and temperature-dependent assembly of IncHI1 plasmid R27-encoded TrhC-associated conjugative transfer protein complexes Mol Microbiol 2001 42 705 715 11722736 138 Aly KA Baron C The VirB5 protein localizes to the T-pilus tips in Agrobacterium tumefaciens Microbiology 2007 153 3766 3775 17975085 139 Harris RL Silverman PM Tra proteins characteristic of F-like type IV secretion systems constitute an interaction group by yeast two-hybrid analysis J Bacteriol 2004 186 5480 5485 15292150 140 Arutyunov D Arenson B Manchak J Frost LS F plasmid TraF and TraH are components of an outer membrane complex involved in conjugation J Bacteriol 2010 192 1730 1734 20081027 141 Harris RL Hombs V Silverman PM Evidence that F-plasmid proteins TraV, TraK and TraB assemble into an envelope-spanning structure in Escherichia coli Mol Microbiol 2001 42 757 766 11722740 142 Maneewannakul S Maneewannakul K Ippen-Ihler K Characterization, localization, and sequence of F transfer region products: the pilus assembly gene product TraW and a new product, TrbI J Bacteriol 1992 174 5567 5574 1355084 143 Dürrenberger MB Villiger W Bächi T Conjugational junctions: morphology of specific contacts in conjugating Escherichia coli bacteria J Struct Biol 1991 107 146 156 1807350 144 Achtman M Kennedy N Skurray R Cell-cell interactions in conjugating Escherichia coli: role of traT protein in surface exclusion Proc Natl Acad Sci USA 1977 74 5104 5108 337311 145 Klimke WA Rypien CD Klinger B Kennedy RA Rodriguez-Maillard JM Frost LS The mating pair stabilization protein, TraN, of the F plasmid is an outer-membrane protein with two regions that are important for its function in conjugation Microbiology 2005 151 3527 3540 16272376 146 Low B Wood TH A quick and efficient method for interruption of bacterial conjugation Genet Res 1965 6 300 303 14345913 147 Anthony KG Sherburne C Sherburne R Frost LS The role of the pilus in recipient cell recognition during bacterial conjugation mediated by F-like plasmids Mol Microbiol 1994 13 939 953 7854127 148 Firth N Skurray R Characterization of the F plasmid bifunctional conjugation gene, traG Mol Gen Genet 1992 232 145 153 1348105 149 Samuels AL Lanka E Davies JE Conjugative junctions in RP4-mediated mating of Escherichia coli J Bacteriol 2000 182 2709 2715 10781537 150 Heinemann JA Sprague GFJ Jr Bacterial conjugative plasmids mobilize DNA transfer between bacteria and yeast Nature 1989 340 205 209 2666856 151 Bates S Cashmore AM Wilkins BM IncP plasmids are unusually effective in mediating conjugation of Escherichia coli and Saccharomyces cerevisiae: involvement of the tra2 mating system J Bacteriol 1998 180 6538 6543 9851996 152 Waters VL Conjugation between bacterial and mammalian cells Nat Genet 2001 29 375 376 11726922 153 Silby MW Ferguson GC Billington C Heinemann JA Localization of the plasmid-encoded proteins TraI and MobA in eukaryotic cells Plasmid 2007 57 118 130 17084894 154 Guynet C Cuevas A Moncalián G de la Cruz F The stb operon balances the requirements for vegetative stability and conjugative transfer of plasmid R388 PLoS Genet 2011 7 e1002073 21625564

PMC005xxxxxx/PMC5119750.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8804207 1862 Schizophr Res Schizophr. Res. Schizophrenia research 0920-9964 1573-2509 26260078 5119750 10.1016/j.schres.2015.07.026 NIHMS714436 Article Molecular evidence for decreased synaptic efficacy in the postmortem olfactory bulb of individuals with schizophrenia Egbujo Chijioke a Sinclair Duncan a Borgmann-Winter Karin ab Arnold Steven E a Turetsky Bruce a Hahn Chang-Gyu a# a Department of Psychiatry, University of Pennsylvania, Philadelphia, PA b Children's Hospital of Philadelphia, Philadelphia, PA # Corresponding author: Chang-Gyu Hahn, Neuropsychiatric Signaling Program, Center for Neurobiology and Behavior, University of Pennsylvania, 125 S 31st St, Philadelphia, PA, 19104 USA; phone 610 520 9131, fax 215-573-2041 7 10 2016 07 8 2015 10 2015 22 11 2016 168 1-2 554562 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Multiple lines of evidence suggest altered synaptic plasticity/connectivity as a pathophysiologic mechanism for various symptom domains of schizophrenia. Olfactory dysfunction, an endophenotype of schizophrenia, reflects altered activity of the olfactory circuitry, which conveys signals from olfactory receptor neurons to the olfactory cortex via synaptic connections in the glomeruli of the olfactory bulb. The olfactory system begins with intranasal olfactory receptor neuron axons synapsing with mitral and tufted cells in the glomeruli of the olfactory bulb, which then convey signals directly to the olfactory cortex. We hypothesized that olfactory dysfunction in schizophrenia is associated with dysregulation of synaptic efficacy in the glomeruli of the olfactory bulb. To test this, we employed semi-quantitative immunohistochemistry to examine the olfactory bulbs of 13 postmortem samples from schizophrenia and their matched control pairs for glomerular expression of 5 pre- and postsynaptic proteins that are involved in the integrity and function of synapses. In the glomeruli of schizophrenia cases compared to their matched controls, we found significant decreases in three presynaptic proteins which play crucial roles in vesicular glutamate transport- synapsin IIa (−18.05%, p= 0.019), synaptophysin (−24.08% p=0.0016) and SNAP-25 (−23.9%, p=0.046). Two postsynaptic proteins important for spine formation and glutamatergic signaling were also decreased- spinophilin (−17.40%, p=0.042) and PSD-95 (−34.06%, p=0.015). These findings provide molecular evidence for decreased efficacy of synapses within the olfactory bulb, which may represent a synaptic mechanism underlying olfactory dysfunction in schizophrenia. schizophrenia synapse olfactory bulb synaptophysin SNAP-25 PSD-95 1. Introduction Multiple lines of evidence implicate synaptic alterations in schizophrenia. A number of genes and pathways critical for synaptic function have been associated with schizophrenia in genetic studies (Fromer et al., 2014; Kirov et al., 2012; Schizophrenia Working Group of the Psychiatric Genomics, 2014). In postmortem studies, altered expression of genes and proteins involved in synaptic structure or function, such as synaptophysin, synapsin II, GAP-43, SNAP-25, Cdc42 and Duo, have been observed in the dorsolateral prefrontal cortex (DLPFC), temporal lobe, hippocampus and amygdala in schizophrenia (Eastwood et al., 1995; Eastwood et al., 2000; Eastwood and Harrison, 1995; Glantz and Lewis, 1997; Hill et al., 2006; Ide and Lewis, 2010; Karson et al., 1999; Sawada et al., 2002; Tan et al., 2014; Tcherepanov and Sokolov, 1997; Thompson et al., 1998; Varea et al., 2012; Webster et al., 2001). Furthermore, brain imaging studies indicate altered structural and functional connectivity in cortico-limbic and meso-limbic circuitry in schizophrenia (Allen et al., 2012; Alonso-Solis et al., 2014; Amad et al., 2014; Genzel et al., 2014; Whitfield-Gabrieli et al., 2009; Woodward et al., 2012), some of which may be modifiable by antipsychotic medications (Sarpal et al., 2014). As ultra-structural correlates of synaptic connectivity, decreased dendritic spine density observed in the layer 3 of DLPFC (Glantz and Lewis, 2000; Rosoklija et al., 2000) and auditory cortex (Sweet et al., 2009) further supports impaired synaptic connectivity in schizophrenia. The neural circuitry underlying olfaction may serve as a useful system for studying synaptic efficacy and connectivity in schizophrenia. The essential circuitry consists of the olfactory epithelium (OE) – olfactory bulb (OB) – pyriform cortex. In the OE, odorants bind to odorant receptors generating action potentials in glutamatergic olfactory receptor neurons (ORNs), which synapse with mitral and tufted cells in the glomeruli of the OB (Gottfried, 2010). Within the glomeruli, OB neurons are modulated by relay cells, and axons of OB neurons then travel to the forebrain via the olfactory nerve. Some of these neurons project to the olfactory tubercle, while most of the fibers terminate at the pyriform cortex, where they form synapses with pyramidal neurons that in turn terminate in other parts of the forebrain and limbic system. Neural transmission between the OE and olfactory cortex therefore relies on the relatively simple OE – OB synaptic connection, which is highly concentrated in the olfactory glomeruli. As such, the connectivity/strength in glomerular synapses can affect sensitivity and cognition of olfactory signals (Gottfried, 2010), which can offer clues to synaptic dysregulations in patients with schizophrenia. Increasing evidence suggests that olfactory dysfunction is an endophenotype of schizophrenia (Turetsky et al., 2008; Turetsky et al., 2003a). Individuals with schizophrenia and those at high risk exhibit deficits in multiple domains of olfactory function including deficits in odor identification (Moberg et al., 1997), odor detection threshold sensitivity (Rupp et al., 2005), odor discrimination (Rupp et al., 2005) and odor memory and odor hedonic judgments (Moberg et al., 2014; Moberg and Turetsky, 2003; Nguyen et al., 2011; Turetsky et al., 2009; Turetsky et al., 2003b). Interestingly, a lesser degree of olfactory dysfunction also may be found among patients’ family members (Moberg et al., 2014) but see (Compton and Chien, 2008; Kamath et al., 2013), suggesting a heritability of this phenotype. Together, olfactory dysfunction may be a heritable trait that co-segregates with the illness and thus meets the criteria for an endophenotype. Structural and functional alterations in the olfactory circuitry have been reported in schizophrenia (Arnold et al., 2001; Kamath et al., 2011; Nguyen et al., 2011; Rioux et al., 2004). Olfactory event related potentials that are measured in response to odorant stimulation were found to be decreased in patients with schizophrenia compared to controls (Turetsky, 2003a) which indicates an overall decrease in odorant induced neural transmission. Such changes could be in part due to structural changes in key regions of the circuit, as evidenced by decreased size of the OB (Turetsky et al., 2003a; Turetsky et al., 2000) and olfactory sulci (Nguyen et al., 2011; Takahashi et al., 2013a; Takahashi et al., 2013b; Takahashi et al., 2014) in schizophrenia patients and individuals at high risk. In addition, histologic examination of the postmortem OE showed increased ratios of immature vs. mature (GAP-43 immunoreactive vs. OMP immunoreactive) neurons (Arnold et al., 2001), suggesting that altered neuronal maturation may impact olfactory neuron function in schizophrenia patients. The olfactory glomerulus is a specialized structure in the OB where the axons of the olfactory receptor neurons and dendrites of the olfactory nerves synapse. As such, it permits an opportunity to examine synaptic efficacy with relative ease compared to other regions of the brain. We hypothesized that there would be a decrease in key synaptic proteins in the OB of subjects with schizophrenia, which could mirror similar alterations in other areas of the brain. The results of our study indicate that key presynaptic and postsynaptic proteins are strikingly decreased in the OB of schizophrenia cases compared to their matched controls. 2. Method 2.1 Postmortem samples Olfactory bulbs were removed at autopsy from 13 prospectively assessed elderly subjects with schizophrenia and 13 non-psychiatric control subjects carefully matched for age, postmortem interval (PMI), gender and fixative. No significant differences in age at death or PMI were found between schizophrenia and control groups. Subjects with schizophrenia had been diagnosed prospectively according to DSM-III-R/DSM-IV criteria based on medical history, interview with caregivers and clinical examination of the patient. Subjects with schizophrenia had been elderly participants in a prospective clinicopathological studies program with post mortem tissues stored with the University of Pennsylvania brain bank. Written informed consent for antemortem evaluation and autopsy in the event of death were obtained from next of kin according to approved institutional review board protocols. The antipsychotic exposures of individuals with schizophrenia were converted to chlorpromazine equivalent doses using established methods (Davis, 1974; Woods, 2003). Brain tissues from non-neuropsychiatric elderly controls were obtained through the University of Pennsylvania's Center for Neurodegenerative Disease Research. While none of these control subjects had undergone antemortem assessments, a review of their clinical histories found no evidence of any neurological or psychiatric illness. 2.2 Immunohistochemistry Olfactory bulbs were fixed in formalin (3.7% formaldehyde in 100 mM Tris) or 70% ethanol for 24 h, paraffin embedded, and cut into 10 um thick sections, and then mounted on poly-l-lysine coated slides. Immunohistochemistry experiments were conducted according to previously described procedures (Talbot et al., 2012), to assess the expression of the synaptic proteins. Briefly, sections were deparaffinized, rehydrated and then incubated for 20 min in 3% H2O2 in methanol at room temperature to remove endogenous peroxidase activity. Antigen retrieval was performed, involving boiling for 10 min in 1 mM EDTA, 0.1 M Tris, pH 8.0. After two washes in TTB (0.1M Tris Buffer with 0.01% Triton X-100 pH 7.6) sections were incubated in TTB containing 10% normal horse serum (NS) for 60 min at room temperature. The sections were then incubated overnight at 4°C in primary antibodies diluted in TTB containing 10% NS. Primary antibodies used were a mouse monoclonal anti synapsin-IIa antibody (1:2000, cat #610667, BD Biosciences, San Jose, CA, USA), a rabbit polyclonal anti–synaptophysin antibody (1:2000, cat # 180130, Invitrogen, Waltham, MA, USA), a mouse monoclonal anti-SNAP-25 antibody (1:400, H-1 [sc-376713], Santa Cruz, Santa Cruz, CA, USA), a rabbit polyclonal anti-spinophilin antibody (1:1200, ABA5669, Millipore, Billerica, MA, USA), a mouse monoclonal anti-PSD-95 antibody (1:200, clone K28/43, UC Davis/NIH NeuroMab Facility, Davis, CA, USA), a mouse monoclonal anti- GFAP antibody (1:100,000, MAB360 Millipore, Billerica, MA, USA) and a mouse monoclonal anti-β-actin antibody (1:1000, cat # A2228 Sigma-Aldrich, St. Louis, MO, USA). Slides were then rinsed three times with TTB and incubated in secondary antibody (1:500, Vector Labs, Burlingame, CA, USA) in 10% NS for 1h. After being rinsed two times with TTB, they were incubated for 1 h in Vectastain ABC reagent (1:500, Vector Labs) in TBS at room temperature. The slides were rinsed, and then incubated with the chromogen diaminobenzidine (DAB in 0.1 M Tris) and 0.03% H2O2 for 10 min. Sections were then rinsed, dehydrated and coverslipped with Cytoseal (Thermo-Fisher Scientific, Waltham, MA, USA). For PSD-95 quantification, only formaldehyde fixed samples (pairs 1-11) could be used. One section per subject was used in each run. For each protein including actin and GFAP, all cases were processed in a single, precisely timed run. Sections incubated without primary antibody or with an IgG of the same species than the primary antibody were run in parallel and served as negative controls. The concentration of the primary antibodies was based on manufacturers’ recommendations and our titration studies. 2.3 Quantification and analysis To determine the optical density (OD) of the DAB/AB reaction product in the glomeruli of the olfactory bulbs, we used customized Image pro 7.2 software (Media Cybernetics, Rockville, MD, USA) run on a desktop computer attached to a Leitz DMRB microscope (Leica, Wetzlar, Germany) and an Evolution OE camera (Media Cybernetics, Rockville, MD, USA). All the images were captured at the same scope, with the same illumination and settings for each antibody. For each subject a randomly selected slide section of the olfactory bulb was used, the entire section of the bulb was captured and all glomeruli contained in it were delineated. For fifteen subjects (6 cases, 9 controls), both left and right olfactory bulbs were sectioned and analyzed, while for eleven subjects (7 cases, 4 controls) a single bulb was sectioned and analyzed. Glomeruli were readily identified as areas of increased density of pre- and post-synaptic protein immunostaining within the glomerular layer of the bulb. The gray value for each glomerulus within each section was recorded and converted to an OD value, from which was subtracted the background OD (obtained from non-stained portion of the bulb), yielding the mean OD of each protein across all glomeruli in the section (Balu et al., 2012; Rioux et al., 2005; Talbot et al., 2004). All background-corrected mean glomerular OD data were approximately normally distributed (skewness between −1 and 1). Diagnostic groups were compared using paired two-tailed T-tests, and the relationship between continuous variables and glomerular OD was determined by Pearson's correlations. 3. Results 3.1 Presynaptic protein expression in OB glomeruli In schizophrenia cases and controls, we investigated the expression levels of three important markers of presynaptic function. Synapsin-IIa is a synaptic vesicle phosphoprotein that is abundantly expressed within the glomeruli, but found much less elsewhere in the olfactory bulb (Figure 1A). There was a significant decrease in the mean expression levels of synapsin-IIa within glomeruli in schizophrenia cases compared to matched control subjects. (−18.1%, p= 0.019; Figure 1B). Synaptophysin is a presynaptic protein localized to synaptic vesicle membranes (Wiedenmann and Franke, 1985). Synaptophysin was strongly expressed in the glomeruli of the bulbs, but much less expressed in the peri-glomerular area (Figure 1C). There was a significant decrease in the glomerular levels of synaptophysin in schizophrenia cases compared to normal controls (−24.1% p=0.0016; Figure 1D). SNAP-25 (Soluble NSF Attachment Protein) is a SNARE protein which plays a major role in the fusion of synaptic vesicles with the presynaptic membrane for neurotransmitter release (Washbourne et al., 2002). We observed that SNAP-25 is highly expressed in the olfactory bulb glomeruli (Figure 1E). There was a significant decrease in mean glomerular SNAP-25 expression levels in schizophrenia cases compared to matched controls. (−23.9%, p=0.046; Figure 1F). 3.2 Postsynaptic protein expression in OB glomeruli The glomerular expression of two postsynaptic proteins was also quantified. Firstly, spinophilin is a postsynaptic marker which is highly expressed in dendritic spines (Muly et al., 2004). Spinophilin immunostaining was evident in various regions of the bulb but more so in the glomeruli (Figure 2A). There was a significant decrease in spinophilin expression within the glomeruli in schizophrenia cases compared to matched controls. (−17.4%, p=0.042; Figure 2B). PSD-95 is a core scaffolding protein located within the postsynaptic density of excitatory synapses (Kim and Sheng, 2004). PSD-95 was highly expressed in the olfactory bulb glomeruli, but sparsely expressed in the peri-glomerular area (Figure 2C). We observed a significant decrease in the expression levels of PSD-95 in the glomeruli of schizophrenia subjects when compared to matched controls (−34.1%, p=0.015; Figure 2D). To determine whether non-specific or possibly artifactual differences between groups may account for group differences in synaptic proteins, β-actin and GFAP were quantified as internal controls. Both GFAP and β-actin was not highly concentrated in glomeruli (Figure 3E), so glomerular and peri-glomerular areas were quantified together. There were no differences between schizophrenia and matched control subjects in glomerular and peri-glomerular GFAP and β-actin expression ([GFAP] p=0.41, β-actin [p=0.64]; Figure 3F for not shown and for). Overall, the glomerular expression of each synaptic protein investigated in this study was significantly decreased in subjects with schizophrenia when compared to controls (Figure 3). We found no significant correlations between expression levels of the synaptic proteins studied with age or PMI within control and schizophrenia groups individually, or both groups combined. Expression levels also were not correlated with age of onset, duration of illness or antipsychotic dosage within the schizophrenia group. 4. Discussion Given multiple reports of abnormal synaptic density and protein expression in various cortical and limbic brain regions in schizophrenia (Glantz and Lewis, 1997, 2000; Hill et al., 2006; Karson et al., 1999; Rosoklija et al., 2000; Sweet et al., 2009; Thompson et al., 1998), we hypothesized that altered synaptic connectivity in schizophrenia would be reflected in dysregulation of key synaptic proteins in the olfactory glomeruli of schizophrenia cases. Consistent with this hypothesis, we observed a significant decrease in the expression levels of pre- and post-synaptic proteins synapsin-IIa, synaptophysin, SNAP-25, spinophilin and PSD-95 in the glomeruli of the olfactory bulb in schizophrenia, but no decrease in GFAP or β-actin. These findings may reflect a decrease in synaptic connectivity within the glomeruli, which could contribute to the olfactory dysfunction observed in schizophrenia. In this study, levels of three presynaptic proteins (synapsin-IIa, synaptophysin and SNAP-25) and two postsynaptic proteins (spinophilin and PSD-95) were decreased in the glomeruli of individuals with schizophrenia. In presynaptic terminals, these three proteins are involved in multiple stages of vesicular neurotransmitter release, including reserve pool regulation [synapsin-IIa; (Gitler et al., 2008)], vesicle fusion [SNAP-25; (Washbourne et al., 2002)] and endocytosis [synaptophysin; (Kwon and Chapman, 2011)]. Postsynaptically, PSD-95 stabilizes the synapse and modulates synaptic long-term potentiation (Nagura et al., 2012) while spinophilin regulates spine morphology and plasticity (Feng et al., 2000; Sarrouilhe et al., 2006). While deficits in a subset of these proteins may have suggested dysregulation of specific genes or pathways, the concurrent decreases of all five proteins are likely to reflect overall changes in synaptic integrity, such as decreased number of synapses or diminished synaptic strength throughout olfactory glomeruli in schizophrenia. The spatial resolution of our methods did not allow us to distinguish between these two possibilities. The origin of such overall changes in the olfactory synapses is unclear at present. It is possible that dysregulations either in the presynaptic or postsynaptic segments are primary, and the other segment is affected via trans-synaptic communication. It is also possible that fundamental neurobiological changes affect each segment separately and together disrupt the synaptic ultrastructure. In this regard, our previous observations of altered neuronal lineage in the postmortem OE may suggest a role of presynaptic alterations. In the OE of schizophrenia cases, we found decreased ratios of OMP immunoreactive (IR) cells with respect to GAP-43 IR cells, each representing mature and immature olfactory sensory neurons (OSNs) respectively (Arnold et al., 2001). Given that presynaptic molecules in the OB arise from OSNs, decreased proportions of mature neurons could account for decreased expression of presynaptic molecules in the OB, which in turn may compromise the integrity of postsynaptic segment. Two previous studies have investigated synaptic protein expression in the OB in schizophrenia. One study reported decreased MAP2 expression in OB glomeruli in schizophrenia (Rioux et al., 2004), suggesting a possible decrease in glomerular dendritic arborization which would be consistent with our findings. However, in a follow-up study, the same group did not identify differences in pre- and post-synaptic glomerular proteins OMP, GAP-43, NCAM, calbindin, NFM/HP, β-tubulin III or TrkB, and although synaptophysin was decreased by about 13% it was not statistically significant (Rioux et al., 2005). Further work is required to determine to what extent pre- and postsynaptic deficits in OB glomeruli are a consistent feature of schizophrenia, and to clarify whether deficits are specific to key cellular processes such as vesicular transport, which was the predominant focus of this study. Our observations of decreased expression of synaptic proteins are consistent with similar findings in various postmortem brain regions in schizophrenia, many of which demonstrated brain region- or cell layer-specificity. Decreased levels of synapsin I and pan synapsin proteins in the hippocampus (Browning et al., 1993; Vawter et al., 2002), and synapsin III in the DLPFC (Porton and Wetsel, 2007), have been reported in schizophrenia. Decreased SNAP-25 protein has also been described in the PFC (BA10) (Karson et al., 1999; Thompson et al., 2003) and hippocampus (Fatemi et al., 2001; Young et al., 1998), while fewer spinophilin-positive puncta have been described in the auditory cortex in schizophrenia (Sweet et al., 2009). Decreased PSD-95 in the anterior cingulate cortex in schizophrenia (Funk et al., 2009), but increased PSD-95 in the dorsomedial thalamus (Clinton et al., 2006), have been described. Finally, decreased synaptophysin has been reported in the PFC (BA9, 10, 46) (Glantz and Lewis, 1997; Karson et al., 1999; Rao et al., 2013), thalamus (Landen et al., 1999), hippocampus (Davidsson et al., 1999; Vawter et al., 1999) and cingulate cortex (Davidsson et al., 1999). These broad array of findings, which reveal decreases in the same proteins quantified in our study, support that altered synaptic connectivity in multiple brain regions may be a hallmark of schizophrenia. Although many of these findings have been replicated, other studies have failed to find differences in other brain regions in implicated in schizophrenia (Barksdale et al., 2014; Gray et al., 2010; Halim et al., 2003; Vawter et al., 2002; Young et al., 1998). One reason for the divergence of these findings may be that, while widespread, synaptic deficits may still be localized to discrete brain regions and cell types within them. The glomerulus of the OB represents a subcellular domain which has a concentration of synaptic inputs and can be extremely accurately demarcated, suggesting that it may be a useful brain region for further exploration of synaptic abnormalities in schizophrenia. Given that all the individuals with schizophrenia had been on varying doses of antipsychotics for several years, it is important to consider the impact of antipsychotics on our findings. Consistent with previous findings with spinophilin in schizophrenia (Sweet et al., 2009), we did not observe a correlation of antipsychotic dosage with levels of pre- or post-synaptic molecules. This analysis, however, has inherent limitations because although coefficients used in converting different antipsychotics to their chlorpromazine equivalents have been empirically derived and are based on expert consensus, they do not the take into account the distinct effects of individual drugs or the effects of polypharmacy. In addition, the doses used for conversion were the dosages that the patients were on at the time of death, and may not accurately reflect cumulative lifetime exposure. This however remains an area for further study to evaluate the long term effects of antipsychotics on the olfactory circuit given that significant neurogenesis takes place in this circuit as OSNs are constantly regenerated in adult life. An additional possible confound in this study is smoking, which is more common among individuals with schizophrenia than normal controls (Dickerson et al., 2013). Smoking may impair olfaction [(Katotomichelakis et al., 2007; Schriever et al., 2013)] and could lead to decreased olfactory bulb volume (Schriever et al., 2013). In previous studies, olfactory bulb volume decreases in schizophrenia cases and first degree relatives were significant when covarying for smoking status and volume (Moberg et al., 2014; Turetsky et al., 2003a; Turetsky et al., 2000), suggesting deficits in schizophrenia independent of any effects of smoking. 5. Conclusion We report decreased levels of pre- and postsynaptic proteins in the OB glomeruli of individuals with schizophrenia which is consistent with the hypothesis that abnormal synaptic efficacy is a feature of the illness. Future studies will need to integrate findings from the OB, OE and olfactory cortex to shed further light on the origins of synaptic and circuit abnormalities that underlie schizophrenia. Acknowledgements We thank Konrad Talbot for providing technical expertise and advice in the design and execution of the project, Hala Kazi for technical assistance, and John Robinson, Theresa Schuck and the PENN Brain Bank for assistance with tissues and demographic data. Role of Funding Source CE is supported by an ITMAT CTSA KL2 Grant, DS by an NHMRC (Australia) CJ Martin Fellowship (APP1072878), and CGH by NIH grants MH890140 and MH10028395. Figure 1 Cytoarchitecture and connections of the olfactory bulb. Figure 2 Expression of presynaptic proteins in the glomeruli of the olfactory bulb from individuals with schizophrenia and controls. A, C, E) Immunostaining in the entire olfactory bulb and in glomeruli (inset) of a representative matched pair (pair 9) for (A) synapsin IIa, (C) synaptophysin and (E) SNAP-25; B, D, F) comparison of protein expression between individuals with schizophrenia and controls within each matched pair for (B) synapsin IIa, (D) synaptophysin and (F) SNAP-25. SCZ- schizophrenia. Inset scale bars indicate 150μm. Figure 3 Expression of postsynaptic proteins and GFAP in olfactory bulb glomeruli from individuals with schizophrenia and controls. A, C, E) Immunostaining in the entire olfactory bulb and in glomeruli (inset) of a representative matched pair (pair 9) for (A) spinophilin, (C) PSD-95 and (E) GFAP; B, D, F) comparison of protein expression between individuals with schizophrenia and controls within each matched pair for (B) spinophilin, (D) PSD-95 and (F) GFAP. SCZ- schizophrenia. Inset scale bars indicate 150μm. Figure 4 Differences in glomerular expression of pre- and post-synaptic proteins in the olfactory bulb between individuals with schizophrenia and controls. For GFAP and β-actin, mean ODs included glomerular and peri-glomerular areas. SCZ- schizophrenia. * p<0.05, ** p<0.005, N.S.- non-significant. Error bars represent standard error of the mean. Table 1 Demographic and clinical information for individuals whose olfactory bulbs were used in this study. Pair # Diagnosis Age (years) Age at diagnosis (years) Sex Race PMI (hr) Tissue age (years) Antipsychotic dose / day Chlopromazine dose equivalance Cause of death 1 Schizophrenia 90 35 Female African American 7.5 10 Risperidone 4mg 400 Hypertensive vasculopathy 2 Schizophrenia 86 24 Female Caucasian 7.5 12 Unknown Unknown Resp failure 2nd to MRSA 3 Schizophrenia 83 19 Female Caucasian 7.5 11 Risperidone 3mg 300 Pneumonia complications 4 Schizophrenia 90 20 Female Caucasian 16 18 Thioridazine dose unknown Unknown Resp failure 5 Schizophrenia 82 42 Female Caucasian 12 12 Clozapine 250mg 250 Resp failure 2nd to anemia, DM2, COPD, CVD 6 Schizophrenia 81 19 Male Caucasian 9 9 Olanzapine 7.5mg 225 Resp failure 2nd to vasculopathy and CVA 7 Schizophrenia 84 Unknown Male Caucasian 20 3 Risperidone 6mg 600 Resp failure 2nd to COPD, cardiomyopathy, anemia 8 Schizophrenia 81 19 Male Caucasian 5 11 Thioridazine 450mg 400 Post fall & head injury 9 Schizophrenia 88 24 Female Caucasian 7 14 Thioridazine 25mg 25 Chronic Renal Failure 10 Schizophrenia 39 16 Female African American 17.5 15 Risperidone 8mg 800 Complications of HTN, cardiomyopathy, CHF, 11 Schizophrenia 72 Unknown Male Caucasian 26 1 Unknown Unknown Unknown 12 Schizophrenia 76 21 Female Caucasian 10 15 Thiothixene 10mg 200 Metastatic cecal CA, anemia 13 Schizophrenia 70 17 Female Caucasian 17.5 12 Unknown Unknown Resp failure 2nd to COPD 1 Control 92 - Female Caucasian 5 11 - Endometrial CA 2 Control 84 - Female Caucasian 3 3 - CAD/HTN, Leukemia, breast CA, DM 3 Control 75 - Female Caucasian 15 3 - Cardiac arrest 2nd to aortic stenosis 4 Control 97 - Female Caucasian 15 3 - CAD, HTN, heart block 5 Control 78 - Female Caucasian 19 2 - Post-op complications following GI bleed. 6 Control 72 - Female Caucasian 13.5 2 - Acute renal failure 2nd to AA dissection 7 Control 82 - Male Caucasian 18 6 - Cardiac arrest 8 Control 80 - Male Caucasian 10.5 7 - Myocardial Infarction 9 Control 82 - Female Caucasian 5 3 - Unknown 10 Control 46 - Female Multiracial 12 2 - Fulminant hepatic & multi organ failure 11 Control 70 - Male Caucasian 36 1 - Complications of DM & HTN 12 Control 72 - Female Caucasian 4 23 - Unknown 13 Control 71 - Female Caucasian 10.5 8 - Abdominal aortic aneurysm PMI- postmortem interval, Resp- respiratory, 2nd- secondary, MRSA- multidrug resistant staphylococcus aureus, PNA- pneumonia, COPD- chronic obstructive pulmonary disease, CVD- cardiovascular disease, HTN- hypertension, CHF- congestive heart failure, CA- cancer, CAD- coronary artery disease, DM- diabetes mellitus, GI- gastrointestinal, AA- aortic arch. Table 2 Summary of demographic information. Schizophrenia Control Age 78.6 (± 13.4) 77.5 (± 12.3) PMI 12.5 (± 6.3) 12.4 (± 9.0) Tissue age 11 (± 4.7) 5.7 (± 6.0) Data represented as group mean (± standard deviation). PMI- postmortem interval. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Contributions CE, KBW, SEA, BT, and CGH designed the study together. CE, DS and KBW together conducted the experiments, analyzed and interpreted the data. CE, DS, SEA, BT and CGH wrote the manuscript. Conflict of Interest The authors declare no conflict of interest. References Allen P Luigjes J Howes OD Egerton A Hirao K Valli I Kambeitz J Fusar-Poli P Broome M McGuire P Transition to psychosis associated with prefrontal and subcortical dysfunction in ultra high-risk individuals. Schizophr Bull 2012 38 6 1268 1276 22290265 Alonso-Solis A Vives-Gilabert Y Grasa E Portella MJ Rabella M Sauras RB Roldan A Nunez-Marin F Gomez-Anson B Perez V Alvarez E Corripio I Resting-state functional connectivity alterations in the default network of schizophrenia patients with persistent auditory verbal hallucinations. Schizophr Res 2014 Amad A Cachia A Gorwood P Pins D Delmaire C Rolland B Mondino M Thomas P Jardri R The multimodal connectivity of the hippocampal complex in auditory and visual hallucinations. Mol Psychiatry 2014 19 2 184 191 23318999 Arnold SE Han LY Moberg PJ Turetsky BI Gur RE Trojanowski JQ Hahn CG Dysregulation of olfactory receptor neuron lineage in schizophrenia. Arch Gen Psychiatry 2001 58 9 829 835 11545665 Balu DT Carlson GC Talbot K Kazi H Hill-Smith TE Easton RM Birnbaum MJ Lucki I Akt1 deficiency in schizophrenia and impairment of hippocampal plasticity and function. Hippocampus 2012 22 2 230 240 21049487 Barksdale KA Lahti AC Roberts RC Synaptic proteins in the postmortem anterior cingulate cortex in schizophrenia: relationship to treatment and treatment response. Neuropsychopharmacology 2014 39 9 2095 2103 24603856 Browning MD Dudek EM Rapier JL Leonard S Freedman R Significant reductions in synapsin but not synaptophysin specific activity in the brains of some schizophrenics. Biol Psychiatry 1993 34 8 529 535 8274580 Clinton SM Haroutunian V Meador-Woodruff JH Up-regulation of NMDA receptor subunit and post-synaptic density protein expression in the thalamus of elderly patients with schizophrenia. J Neurochem 2006 98 4 1114 1125 16762023 Compton MT Chien VH No association between psychometrically-determined schizotypy and olfactory identification ability in first-degree relatives of patients with schizophrenia and non-psychiatric controls. Schizophr Res 2008 100 1-3 216 223 18077137 Davidsson P Gottfries J Bogdanovic N Ekman R Karlsson I Gottfries CG Blennow K The synaptic-vesicle-specific proteins rab3a and synaptophysin are reduced in thalamus and related cortical brain regions in schizophrenic brains. Schizophr Res 1999 40 1 23 29 10541003 Davis JM Dose equivalence of the antipsychotic drugs. J Psychiatr Res 1974 11 65 69 4156792 Dickerson F Stallings CR Origoni AE Vaughan C Khushalani S Schroeder J Yolken RH Cigarette smoking among persons with schizophrenia or bipolar disorder in routine clinical settings, 1999-2011. Psychiatr Serv 2013 64 1 44 50 23280457 Eastwood SL Burnet PW Harrison PJ Altered synaptophysin expression as a marker of synaptic pathology in schizophrenia. Neuroscience 1995 66 2 309 319 7477874 Eastwood SL Cairns NJ Harrison PJ Synaptophysin gene expression in schizophrenia. Investigation of synaptic pathology in the cerebral cortex. Br J Psychiatry 2000 176 236 242 10755070 Eastwood SL Harrison PJ Decreased synaptophysin in the medial temporal lobe in schizophrenia demonstrated using immunoautoradiography. Neuroscience 1995 69 2 339 343 8552231 Fatemi SH Earle JA Stary JM Lee S Sedgewick J Altered levels of the synaptosomal associated protein SNAP-25 in hippocampus of subjects with mood disorders and schizophrenia. Neuroreport 2001 12 15 3257 3262 11711867 Feng J Yan Z Ferreira A Tomizawa K Liauw JA Zhuo M Allen PB Ouimet CC Greengard P Spinophilin regulates the formation and function of dendritic spines. Proc Natl Acad Sci U S A 2000 97 16 9287 9292 10922077 Fromer M Pocklington AJ Kavanagh DH Williams HJ Dwyer S Gormley P Georgieva L Rees E Palta P Ruderfer DM Carrera N Humphreys I Johnson JS Roussos P Barker DD Banks E Milanova V Grant SG Hannon E Rose SA Chambert K Mahajan M Scolnick EM Moran JL Kirov G Palotie A McCarroll SA Holmans P Sklar P Owen MJ Purcell SM O'Donovan MC De novo mutations in schizophrenia implicate synaptic networks. Nature 2014 506 7487 179 184 24463507 Funk AJ Rumbaugh G Harotunian V McCullumsmith RE Meador-Woodruff JH Decreased expression of NMDA receptor-associated proteins in frontal cortex of elderly patients with schizophrenia. Neuroreport 2009 20 11 1019 1022 19483657 Genzel L Dresler M Cornu M Jager E Konrad B Adamczyk M Friess E Steiger A Czisch M Goya-Maldonado R Medial Prefrontal-Hippocampal Connectivity and Motor Memory Consolidation in Depression and Schizophrenia. Biol Psychiatry 2014 Gitler D Cheng Q Greengard P Augustine GJ Synapsin IIa controls the reserve pool of glutamatergic synaptic vesicles. J Neurosci 2008 28 43 10835 10843 18945891 Glantz LA Lewis DA Reduction of synaptophysin immunoreactivity in the prefrontal cortex of subjects with schizophrenia. Regional and diagnostic specificity. Archives of general psychiatry 1997 54 10 943 952 9337775 Glantz LA Lewis DA Decreased dendritic spine density on prefrontal cortical pyramidal neurons in schizophrenia. Arch Gen Psychiatry 2000 57 1 65 73 10632234 Gottfried JA Central mechanisms of odour object perception. Nat Rev Neurosci 2010 11 9 628 641 20700142 Gray LJ Dean B Kronsbein HC Robinson PJ Scarr E Region and diagnosis-specific changes in synaptic proteins in schizophrenia and bipolar I disorder. Psychiatry Res 2010 178 2 374 380 20488553 Halim ND Weickert CS McClintock BW Hyde TM Weinberger DR Kleinman JE Lipska BK Presynaptic proteins in the prefrontal cortex of patients with schizophrenia and rats with abnormal prefrontal development. Molecular psychiatry 2003 8 9 797 810 12931207 Hill JJ Hashimoto T Lewis DA Molecular mechanisms contributing to dendritic spine alterations in the prefrontal cortex of subjects with schizophrenia. Mol Psychiatry 2006 11 6 557 566 16402129 Ide M Lewis DA Altered cortical CDC42 signaling pathways in schizophrenia: implications for dendritic spine deficits. Biol Psychiatry 2010 68 1 25 32 20385374 Kamath V Moberg PJ Kohler CG Gur RE Turetsky BI Odor hedonic capacity and anhedonia in schizophrenia and unaffected first-degree relatives of schizophrenia patients. Schizophr Bull 2013 39 1 59 67 21616912 Kamath V Turetsky BI Moberg PJ Identification of pleasant, neutral, and unpleasant odors in schizophrenia. Psychiatry research 2011 187 1-2 30 35 21239063 Karson CN Mrak RE Schluterman KO Sturner WQ Sheng JG Griffin WS Alterations in synaptic proteins and their encoding mRNAs in prefrontal cortex in schizophrenia: a possible neurochemical basis for ‘hypofrontality’. Mol Psychiatry 1999 4 1 39 45 10089007 Katotomichelakis M Balatsouras D Tripsianis G Davris S Maroudias N Danielides V Simopoulos C The effect of smoking on the olfactory function. Rhinology 2007 45 4 273 280 18085020 Kim E Sheng M PDZ domain proteins of synapses. Nat Rev Neurosci 2004 5 10 771 781 15378037 Kirov G Pocklington AJ Holmans P Ivanov D Ikeda M Ruderfer D Moran J Chambert K Toncheva D Georgieva L Grozeva D Fjodorova M Wollerton R Rees E Nikolov I van de Lagemaat LN Bayes A Fernandez E Olason PI Bottcher Y Komiyama NH Collins MO Choudhary J Stefansson K Stefansson H Grant SG Purcell S Sklar P O'Donovan MC Owen MJ De novo CNV analysis implicates specific abnormalities of postsynaptic signalling complexes in the pathogenesis of schizophrenia. Mol Psychiatry 2012 17 2 142 153 22083728 Kwon SE Chapman ER Synaptophysin regulates the kinetics of synaptic vesicle endocytosis in central neurons. Neuron 2011 70 5 847 854 21658579 Landen M Davidsson P Gottfries CG Grenfeldt B Stridsberg M Blennow K Reduction of the small synaptic vesicle protein synaptophysin but not the large dense core chromogranins in the left thalamus of subjects with schizophrenia. Biol Psychiatry 1999 46 12 1698 1702 10624552 Moberg PJ Doty RL Turetsky BI Arnold SE Mahr RN Gur RC Bilker W Gur RE Olfactory identification deficits in schizophrenia: correlation with duration of illness. American Journal of Psychiatry 1997 154 7 1016 1018 9210756 Moberg PJ Kamath V Marchetto DM Calkins ME Doty RL Hahn CG Borgmann-Winter KE Kohler CG Gur RE Turetsky BI Meta-analysis of olfactory function in schizophrenia, first-degree family members, and youths at-risk for psychosis. Schizophr Bull 2014 40 1 50 59 23641047 Moberg PJ Turetsky BI Scent of a disorder: olfactory functioning in schizophrenia. Curr Psychiatry Rep 2003 5 4 311 319 12857535 Muly EC Smith Y Allen P Greengard P Subcellular distribution of spinophilin immunolabeling in primate prefrontal cortex: localization to and within dendritic spines. J Comp Neurol 2004 469 2 185 197 14694533 Nagura H Ishikawa Y Kobayashi K Takao K Tanaka T Nishikawa K Tamura H Shiosaka S Suzuki H Miyakawa T Fujiyoshi Y Doi T Impaired synaptic clustering of postsynaptic density proteins and altered signal transmission in hippocampal neurons, and disrupted learning behavior in PDZ1 and PDZ2 ligand binding-deficient PSD-95 knockin mice. Molecular brain 2012 5 43 23268962 Nguyen AD Pelavin PE Shenton ME Chilakamarri P McCarley RW Nestor PG Levitt JJ Olfactory sulcal depth and olfactory bulb volume in patients with schizophrenia: an MRI study. Brain Imaging Behav 2011 5 4 252 261 21728040 Porton B Wetsel WC Reduction of synapsin III in the prefrontal cortex of individuals with schizophrenia. Schizophr Res 2007 94 1-3 366 370 17540541 Rao JS Kim HW Harry GJ Rapoport SI Reese EA Increased neuroinflammatory and arachidonic acid cascade markers, and reduced synaptic proteins, in the postmortem frontal cortex from schizophrenia patients. Schizophr Res 2013 147 1 24 31 23566496 Rioux L Gelber EI Parand L Kazi HA Yeh J Wintering R Bilker W Arnold SE Characterization of olfactory bulb glomeruli in schizophrenia. Schizophr Res 2005 77 2-3 229 239 15946825 Rioux L Ruscheinsky D Arnold SE Microtubule-associated protein MAP2 expression in olfactory bulb in schizophrenia. Psychiatry Res 2004 128 1 1 7 15450909 Rosoklija G Toomayan G Ellis SP Keilp J Mann JJ Latov N Hays AP Dwork AJ Structural abnormalities of subicular dendrites in subjects with schizophrenia and mood disorders: preliminary findings. Arch Gen Psychiatry 2000 57 4 349 356 10768696 Rupp CI Fleischhacker WW Kemmler G Oberbauer H Scholtz AW Wanko C Hinterhuber H Various bilateral olfactory deficits in male patients with schizophrenia. Schizophrenia bulletin 2005 31 1 155 165 15888433 Sarpal DK Robinson DG Lencz T Argyelan M Ikuta T Karlsgodt K Gallego JA Kane JM Szeszko PR Malhotra AK Antipsychotic Treatment and Functional Connectivity of the Striatum in First-Episode Schizophrenia. JAMA psychiatry 2014 Sarrouilhe D di Tommaso A Métayé T Ladeveze V Spinophilin: from partners to functions. Biochimie 2006 88 9 1099 1113 16737766 Sawada K Young CE Barr AM Longworth K Takahashi S Arango V Mann JJ Dwork AJ Falkai P Phillips AG Honer WG Altered immunoreactivity of complexin protein in prefrontal cortex in severe mental illness. Mol Psychiatry 2002 7 5 484 492 12082566 Schizophrenia Working Group of the Psychiatric Genomics, C. Biological insights from 108 schizophrenia-associated genetic loci. Nature 2014 511 7510 421 427 25056061 Schriever VA Reither N Gerber J Iannilli E Hummel T Olfactory bulb volume in smokers. Exp Brain Res 2013 225 2 153 157 23212471 Sweet RA Henteleff RA Zhang W Sampson AR Lewis DA Reduced dendritic spine density in auditory cortex of subjects with schizophrenia. Neuropsychopharmacology 2009 34 2 374 389 18463626 Takahashi T Nakamura Y Nakamura K Ikeda E Furuichi A Kido M Kawasaki Y Noguchi K Seto H Suzuki M Altered depth of the olfactory sulcus in first-episode schizophrenia. Progress in neuro-psychopharmacology & biological psychiatry 2013a 40 167 172 23063493 Takahashi T Nakamura Y Nakamura K Nishiyama S Ikeda E Furuichi A Kido M Noguchi K Suzuki M Altered depth of the olfactory sulcus in subjects at risk of psychosis. Schizophr Res 2013b 149 1-3 186 187 23830683 Takahashi T Wood SJ Yung AR Nelson B Lin A Yucel M Phillips LJ Nakamura Y Suzuki M Brewer WJ Proffitt TM McGorry PD Velakoulis D Pantelis C Altered depth of the olfactory sulcus in ultra high-risk individuals and patients with psychotic disorders. Schizophr Res 2014 153 1-3 18 24 24530137 Talbot K Eidem WL Tinsley CL Benson MA Thompson EW Smith RJ Hahn CG Siegel SJ Trojanowski JQ Gur RE Blake DJ Arnold SE Dysbindin-1 is reduced in intrinsic, glutamatergic terminals of the hippocampal formation in schizophrenia. J Clin Invest 2004 113 9 1353 1363 15124027 Talbot K Wang HY Kazi H Han LY Bakshi KP Stucky A Fuino RL Kawaguchi KR Samoyedny AJ Wilson RS Arvanitakis Z Schneider JA Wolf BA Bennett DA Trojanowski JQ Arnold SE Demonstrated brain insulin resistance in Alzheimer's disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. The Journal of clinical investigation 2012 122 4 1316 1338 22476197 Tan ML Dyck BA Gabriele J Daya RP Thomas N Sookram C Basu D Ferro MA Chong VZ Mishra RK Synapsin II gene expression in the dorsolateral prefrontal cortex of brain specimens from patients with schizophrenia and bipolar disorder: effect of lifetime intake of antipsychotic drugs. Pharmacogenomics J 2014 14 1 63 69 23529008 Tcherepanov AA Sokolov BP Age-related abnormalities in expression of mRNAs encoding synapsin 1A, synapsin 1B, and synaptophysin in the temporal cortex of schizophrenics. J Neurosci Res 1997 49 5 639 644 9302085 Thompson PM Egbufoama S Vawter MP SNAP-25 reduction in the hippocampus of patients with schizophrenia. Progress in neuro-psychopharmacology & biological psychiatry 2003 27 3 411 417 12691775 Thompson PM Sower AC Perrone-Bizzozero NI Altered levels of the synaptosomal associated protein SNAP-25 in schizophrenia. Biol Psychiatry 1998 43 4 239 243 9513732 Turetsky B Moberg PJ Owzar K Johnson SC Doty RL Gur RE Physiologic impairment of olfactory stimulus processing in schizophrenia. Biol Psychiatry 2003 53 403 411 12614993 Turetsky BI Hahn CG Borgmann-Winter K Moberg PJ Scents and nonsense: olfactory dysfunction in schizophrenia. Schizophr Bull 2009 35 6 1117 1131 19793796 Turetsky BI Kohler CG Gur RE Moberg PJ Olfactory physiological impairment in first-degree relatives of schizophrenia patients. Schizophrenia research 2008 102 1-3 220 229 18457935 Turetsky BI Moberg PJ Arnold SE Doty RL Gur RE Low olfactory bulb volume in first-degree relatives of patients with schizophrenia. Am J Psychiatry 2003a 160 4 703 708 12668359 Turetsky BI Moberg PJ Roalf DR Arnold SE Gur RE Decrements in volume of anterior ventromedial temporal lobe and olfactory dysfunction in schizophrenia. Arch Gen Psychiatry 2003b 60 12 1193 1200 14662551 Turetsky BI Moberg PJ Yousem DM Doty RL Arnold SE Gur RE Reduced olfactory bulb volume in patients with schizophrenia. Am J Psychiatry 2000 157 5 828 830 10784482 Varea E Guirado R Gilabert-Juan J Marti U Castillo-Gomez E Blasco-Ibanez JM Crespo C Nacher J Expression of PSA-NCAM and synaptic proteins in the amygdala of psychiatric disorder patients. J Psychiatr Res 2012 46 2 189 197 22099865 Vawter MP Howard AL Hyde TM Kleinman JE Freed WJ Alterations of hippocampal secreted N-CAM in bipolar disorder and synaptophysin in schizophrenia. Mol Psychiatry 1999 4 5 467 475 10523820 Vawter MP Thatcher L Usen N Hyde TM Kleinman JE Freed WJ Reduction of synapsin in the hippocampus of patients with bipolar disorder and schizophrenia. Mol Psychiatry 2002 7 6 571 578 12140780 Washbourne P Thompson PM Carta M Costa ET Mathews JR Lopez-Bendito G Molnar Z Becher MW Valenzuela CF Partridge LD Wilson MC Genetic ablation of the t-SNARE SNAP-25 distinguishes mechanisms of neuroexocytosis. Nat Neurosci 2002 5 1 19 26 11753414 Webster MJ Shannon Weickert C Herman MM Hyde TM Kleinman JE Synaptophysin and GAP-43 mRNA levels in the hippocampus of subjects with schizophrenia. Schizophr Res 2001 49 1-2 89 98 11343868 Whitfield-Gabrieli S Thermenos HW Milanovic S Tsuang MT Faraone SV McCarley RW Shenton ME Green AI Nieto-Castanon A LaViolette P Wojcik J Gabrieli JD Seidman LJ Hyperactivity and hyperconnectivity of the default network in schizophrenia and in first-degree relatives of persons with schizophrenia. Proc Natl Acad Sci U S A 2009 106 4 1279 1284 19164577 Wiedenmann B Franke WW Identification and localization of synaptophysin, an integral membrane glycoprotein of Mr 38,000 characteristic of presynaptic vesicles. Cell 1985 41 3 1017 1028 3924408 Woods SW Chlorpromazine equivalent doses for the newer atypical antipsychotics. J Clin Psychiatry 2003 64 6 663 667 12823080 Woodward ND Karbasforoushan H Heckers S Thalamocortical dysconnectivity in schizophrenia. Am J Psychiatry 2012 169 10 1092 1099 23032387 Young CE Arima K Xie J Hu L Beach TG Falkai P Honer WG SNAP-25 deficit and hippocampal connectivity in schizophrenia. Cereb Cortex 1998 8 3 261 268 9617921

PMC005xxxxxx/PMC5119761.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8805635 27439 Immunol Allergy Clin North Am Immunol Allergy Clin North Am Immunology and allergy clinics of North America 0889-8561 1557-8607 27712762 5119761 10.1016/j.iac.2016.06.004 NIHMS795547 Article AERD as an Endotype of Chronic Rhinosinusitis Stevens Whitney W. MD, PhD 1 Schleimer Robert P. PhD 12 1 Division of Allergy-Immunology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL USA 2 Department of Otolaryngology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA Corresponding author: Dr. Robert P. Schleimer, Division of Allergy-Immunology, Northwestern University Feinberg School of Medicine, 240 E. Huron St., McGaw Room M-318, Chicago, IL, 60611, USA, Tel.: 312-503-0076, Fax: 312-503-0078, rpschleimer@northwestern.edu 22 6 2016 13 9 2016 11 2016 01 11 2017 36 4 669680 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Synopsis Aspirin-Exacerbated Respiratory Disease (AERD) and Chronic Rhinosinusitis with Nasal Polyps (CRSwNP) are both diseases characterized by the presence of chronic sinonasal inflammation and nasal polyps. Both diseases are associated with asthma; AERD by definition and CRSwNP by common comorbidity (~50%). Therefore, the most prominent clinical feature distinguishing patients with AERD from those with CRSwNP remains the development of an upper or lower respiratory tract reaction following the ingestion of a COX-1 inhibitor. However, even in the absence of COX-1 inhibitors, there are other notable clinical and pathophysiological differences between AERD and CRSwNP patients. Patients with AERD on average have worse upper respiratory disease with increased sinonasal symptoms, mucosal inflammation and requirements for revision sinus surgery when compared to patients with CRSwNP. While no single genetic factor has been identified in either CRSwNP or AERD pathogenesis to date, many studies have evaluated whether there are differences in the underlying cellular and molecular mechanisms that account for the observed clinical variations. Studies from several laboratories have discovered important differences in the metabolism of arachidonic acid, including increased activity of the 5-lipoxygenase pathway and decreased levels of the anti-inflammatory prostanoid PGE2. Clear evidence for activation of platelets in AERD also distinguishes it from CRSwNP. Nasal polyp tissues from both AERD and CRSwNP are characterized by type-2 inflammation but there are significantly increased levels of eosinophil and mast cell degranulation products in AERD sinonasal tissue, and recent evidence suggesting that spontaneous activation of eosinophils, basophils and mast cells occurs. Chronic Rhinosinusitis Chronic Rhinosinusitis with Nasal Polyps CRSwNP Aspirin-Exacerbated Respiratory Disease AERD Samter’s Disease Samter’s Triad Introduction Chronic Rhinosinusitis with Nasal Polyps (CRSwNP) and Aspirin-Exacerbated Respiratory Disease (AERD) are both conditions characterized by the presence of chronic sinonasal inflammation and nasal polyps. Numerous groups have labored to describe the clinical features of these diseases as well as investigate the underlying mechanisms that could be driving their overlapping but distinct pathophysiologic mechanisms. However, there are few head-to-head studies directly comparing the sinonasal inflammatory environment of CRSwNP with that of AERD. As a result, the relationship between CRSwNP and AERD remains incompletely defined. Classification To further understand the relationship between CRSwNP and AERD, it is important to review the clinical criteria needed to make the diagnosis of each condition. Per the consensus guidelines, patients with CRSwNP must present with greater than 12 weeks of rhinorrhea, facial pressure or pain, nasal congestion, and/or a reduction in sense of smell 1, 2. Additionally, there must be objective evidence of nasal polyps and mucosal disease visualized upon sinus CT imaging or nasal endoscopy. CRSwNP can exist without medical comorbidities but more often is observed with other chronic conditions such as asthma, hay fever, cystic fibrosis, or eosinophilic granulomatosis with polyangiitis. The clinical diagnosis of AERD is based upon evolving criteria. In 1922, Widal published the first study describing a patent with asthma, nasal polyps, and sensitivity to aspirin. Later, Samter and Beers described a cohort of aspirin-sensitive patients of whom 85% had only respiratory symptoms and 51% had nasal polyps 3. In this report, it was noted “that every patient does not necessarily present with every potential component of the syndrome”. Furthermore, later studies suggested that AERD patients first developed upper respiratory tract symptoms which progressed to involve the lower respiratory tract and then lastly acquired the aspirin intolerance 4. In 2001, the term “aspirin-exacerbated respiratory disease” was termed by Stevenson and colleagues to emphasize that these patients had underlying respiratory disease that was worsened but not induced by aspirin 5. Additionally, over the course of time, it was also discovered that all medications that inhibit the cyclooxygenase-1 (COX-1) enzyme, not just aspirin, could elicit upper and lower respiratory tract symptoms in AERD patients. Given this history, it is not surprising that different terminologies are still used to define the same or related clinical phenotypes (e.g. Samter’s Disease, Samter’s Triad, AERD, NSAID-exacerbated respiratory disease (NERD), aspirin-intolerant asthma). However, even when the same terminology is used, there can be variations in the clinical features of the study cohort. For example, AERD has been investigated in patients with asthma and aspirin intolerance as well as in patients with asthma, aspirin intolerance, and CRSwNP. How the presence (or absence) of sinonasal inflammation might influence the overall disease pathology is not known. As a result, it is not clear whether these two groups are distinct subsets or rather part of a disease continuum. Unless otherwise specified, this review will define AERD as the presence of the triad of CRSwNP, asthma, and worsening of upper and lower respiratory tract symptoms following the ingestion of COX-1 inhibitors. Epidemiology By definition, all patients with AERD have CRSwNP. However, not all patients with CRSwNP have AERD. It is estimated that only ~10% of patients with nasal polyps and ~9% of patients with chronic rhinosinusitis have AERD 6. However, the true prevalence of AERD in the general population or among patients with both chronic rhinosinusitis and nasal polyps (CRSwNP) is still unknown. This is partly due to the lack of epidemiological studies evaluating patients with CRSwNP in large primary care populations. As an additional layer of complexity, both AERD and CRSwNP are associated with asthma. Nearly all AERD patients and ~26–63% of CRSwNP patients have asthma 1, 7, 8. In contrast, only ~7–10% of asthmatics have AERD but as many as 90% of asthmatics may have radiographic evidence of sinonasal inflammation, with or without nasal polyps 1, 6, 9, 10. Interestingly, the prevalence of AERD can increase to ~15% among severe asthmatics 6, 9 and a significant association between severe asthma and enhanced sinonasal inflammation has been reported 11. However, the true prevalence of AERD among patients with both CRSwNP and asthma is not known and, as a result, further investigation is needed. Clinical Features The respiratory reaction following ingestion of a COX-1 inhibitor is the most prominent and definitive clinical feature distinguishing patients with AERD from those with CRSwNP. In AERD patients, this reaction is associated with distinct pathophysiological changes within the sinonasal mucosa. Consequently, most AERD patients typically avoid taking COX-1 inhibitors of their own accord. This results in the presence of a population of patients with CRSwNP and asthma who have aspirin sensitivity that does not appear in their health records. Thus, diagnosis of AERD by examination of medical record charts can overlook a significant number of AERD patients. It has been reported that as many as 15% of asthmatic patients previously unaware of having a COX-1 inhibitor hypersensitivity had a positive reaction to aspirin during an oral challenge 4. Additionally, in separate smaller studies of patients with both CRSwNP and asthma, 20–42% had positive aspirin challenges despite being previously unaware of any COX-1 hypersensitivity 12, 13. In contrast, it has been estimated that as many as 15% of patients with a clinical history consistent with the triad of AERD actually do not react to aspirin upon clinical challenge and instead have only CRSwNP with asthma 4, 13, 14. Taken together, these studies illustrate the limitations in distinguishing CRSwNP and AERD by clinical history alone and highlight the importance of confirming the diagnosis by oral aspirin challenge. There are other less definitive clinical features that can help to distinguish between AERD and CRSwNP. Both CRSwNP and AERD develop in adulthood, with the average age of onset being approximately 42 and 34 years respectively 1, 14. CRSwNP is more common in men while AERD is more common in women 1, 4, 14. Interestingly, a more recent study examining patients who had undergone sinus surgery at a tertiary care hospital suggests that among patients with CRSwNP, women had more severe disease than men as determined by radiographic evidence of enhanced sinus inflammation, need for revision surgery, and use of oral corticosteroids at the time of surgery 15. This is consistent with a large European cohort study that found that women with aspirin-induced asthma have more progressive upper and lower respiratory tract disease than men 4. Clinically, patients with CRSwNP or with AERD both present with symptoms of nasal congestion, rhinorrhea, sinus pressure/pain, and/or hyposmia. Patients with AERD were shown to subjectively report more severe sinus symptoms than patients with CRSwNP 16. Additionally, AERD patients typically have significantly worse endoscopic sinus scores as well as radiographic evidence of sinonasal inflammation compared to CRSwNP patients 16–18. Finally, AERD patients have greater risk of symptom recurrence following surgery and are more likely to undergo revision surgeries sooner and more frequently than those without AERD 19–21. The average number of revision surgeries in AERD patients has been reported to range from 2.6 to 10 4, 14, 21. In contrast, one study reported that patients with CRS on average underwent 1.8 total sinus surgeries but the number of revision surgeries in CRSwNP patients specifically is not well-established 22. In regards to lower respiratory tract disease, CRSwNP and AERD are both associated with asthma that is more severe than that in patients with asthma alone. Conversely, in patients with chronic rhinosinusitis undergoing surgery at a tertiary care facility, asthmatic patients had worse sinus scores than non-asthmatic patients 23. Additionally, as mentioned previously, asthma severity increased with enhanced radiologic evidence of sinus severity 11. In a separate study of 201 AERD patients, 45% had uncontrolled asthma 24. Additionally, Morales and colleagues observed that AERD patients had a 60% increase in risk of severe asthma and 80% increase in emergency room visits compared to patients with aspirin-tolerant asthma 9. AERD patients had lower mean bronchodilator FEV1, more severe asthma by physician assessment, and higher intubation rates than patients without AERD 25. More recently, studies have begun to further define asthma characteristics, as there are variations within the population in regards to clinical features and overall disease severity. For example, the Severe Asthma Research Program performed a cluster analysis of over 700 patients with persistent asthma and identified 5 distinct asthma subsets 26. In particular, one of these cohorts (group 5) was characterized by having the most severe asthma, lowest baseline lung function, and frequent systemic corticosteroid use 26. Additionally, almost half of these patients reported a history of sinus surgery, 26 again suggesting an association between asthma severity and chronic sinus disease. A similar cluster analysis was recently performed in a cohort of AERD patients, albeit only ~81% reported a history of having nasal polyps. In this analysis, 4 distinct subgroups of AERD were defined predominantly based upon levels of asthma severity ranging from mild to severe 27. Interestingly, those patients with significant upper respiratory symptoms (group 1) had the highest levels of peripheral blood eosinophils and urinary leukotrienes among all the subsets examined 27. It is thus clear that patients with NP and AERD have worse asthma than patients with asthma and no NP, and that patients with NP and asthma have worse NP than patients with NP and no asthma. These two diseases each appear to worsen the other. A causal relationship between atopy and either CRSwNP or AERD is still not established and appears unlikely. It has been estimated that 51–86% of patients with CRSwNP 8, 22 and 33–66% with AERD 4, 14 are sensitized to at least one aeroallergen. However, it remains unclear whether allergic sensitization is associated with increased sinus disease severity 8, 22, 23, 28, 29. In summary, AERD, on average, is characterized by more severe upper airway disease than CRSwNP and more significant lower airway disease than asthma. As a result of the severity of their disease, patients with AERD place a disproportionate financial burden on the healthcare system compared to patients with either CRSwNP or asthma alone 30. Given the heterogeneity in past AERD study populations, future investigations are needed to more extensively characterize patients with the clinical triad of AERD and directly compare them to patients with CRSwNP alone as well as with CRSwNP and asthma together. Such studies would also allow for a better understanding of the roles that gender, atopy, asthma, and intolerance to COX-1 inhibitors play in AERD versus CRSwNP. Genetics Since both CRSwNP and AERD develop in adulthood, neither a simple Mendelian pattern of inheritance nor a single genetic defect has been associated with either disease to date 31, 32. In support of this, cohort studies found that only 1% of American and 6% of European AERD patients reported a family member also having AERD 4, 14. Additionally, there have been several studies evaluating whether various specific gene polymorphisms are linked to AERD, although not all AERD patients evaluated in these analyses had CRSwNP (see discussion about definitions above) 32–35. Taken together, present data do not suggest a strong genetic component to CRSwNP or AERD but additional studies are clearly needed to definitively evaluate the role of genetic inheritance in CRSwNP and AERD. As another means to search for potential gene expression differences between AERD and CRSwNP, Stankovic and colleagues examined and compared transcriptional gene profiles of nasal polyps extracted from patients with CRSwNP, AERD, and healthy controls through a microarray analysis 36. While specific gene signatures could distinguish AERD as well as CRSwNP nasal polyps from healthy sinonasal tissue, there were no significant differences in gene expression profiles between AERD and CRSwNP nasal polyps themselves. Interestingly, when compared to healthy controls, CRSwNP nasal polyps were specifically characterized by the up-regulation of met proto-oncogene and protein phosphatase 1 regulatory subunit 9B as well as by the down-regulation of prolactin-induced protein (PIP) and zinc alpha2-glycoprotein. In contrast, up-regulation of periostin gene expression was most characteristic of AERD polyps compared to healthy controls 36. The role these genes play in disease pathogenesis is uncertain but work from our laboratory has also confirmed a significant overlap between gene signature profiles of AERD and CRSwNP polyps (Stevens et al., unpublished data). It is also possible that epigenetic changes and/or differential levels of various proteins or non-protein inflammatory mediators could account for the clinical differences between the two conditions 37. Alterations in gene expression could also be occurring in the bone marrow, in primary or secondary lymphoid tissues or in other sites in AERD such that they are not detectable in NP tissue. Alternatively, unspecified external factors might be needed to induce the enhanced AERD phenotype observed in certain genetically susceptible patients with nasal polyps. As such, chronic viral infections, active smoking, and exposure to environmental tobacco smoke during childhood have all been hypothesized to play a role in the development of AERD 38, 39. Disease Pathophysiology Epithelial barrier dysfunction, exposures to pathogenic and non-pathogenic bacteria, and dysregulation of the host innate and adaptive immune responses are all thought to be important in CRSwNP pathophysiology. Unfortunately, less is known about the specific cellular and molecular mechanisms contributing to AERD pathogenesis in particular, especially at baseline when COX-1 inhibitors are not present. For example, the role of epithelial cells and the microbiome in AERD has not been as extensively characterized as it has in CRSwNP 40–42. As a result, it is possible that specific functional and physical defects in sinonasal epithelial cells as well as alterations in the microbiome of the sinonasal cavity could distinguish AERD and CRSwNP clinical phenotypes. Clearly, further investigations are warranted. Host Immune Response There has been an increasing focus on exploring how the host immune response could contribute to the chronic sinonasal inflammation seen in CRSwNP and AERD. One of the hallmarks of CRSwNP and AERD nasal polyps is an enhanced tissue eosinophilia compared to healthy sinonasal tissue 43. Additionally, levels of the eosinophil granule protein, eosinophil cationic protein (ECP), were also significantly elevated in both subsets of nasal polyps compared to controls as well as in AERD nasal polyps compared to CRSwNP 43–45. In contrast, recent studies found no significant difference in the number of eosinophils counted in hematoxylin and eosin stained slides within nasal polyp sections from AERD and CRSwNP nasal polyps 45, 46. Taken together, these observations lead to the hypothesis that eosinophils are more highly activated in AERD compared to CRSwNP, thus releasing their granule contents (causing more ECP to be detected in AERD) but in the process losing their ability to be detected by traditional histologic staining. It should be noted that not all patients with CRSwNP have enhanced tissue eosinophilia. In particular, Asian patients either living in Asia or in the United States were found to have significantly less eosinophils in their nasal polyps when compared to CRSwNP patients of European descent 44, 47. These observations would in turn suggest that Asian patients, given their relative paucity of tissue eosinophils, might be less likely to have AERD. This is supported by one small study from China that found the prevalence of AERD to be 0.57% among patients evaluated with CRSwNP 48, which is must lower than the 9–10% estimated in a meta-analysis of patients predominantly of European descent 6. Second generation Asians with CRSwNP in the US also had less atopy and less comorbidity with asthma than non Asian Americans. One interpretation of these findings is that the new generation of type 2 targeting biologicals may have reduced efficacy in patients of Asian descent. In addition to eosinophils, CRSwNP nasal polyps are also traditionally characterized as having an increased number of innate type-2 lymphoid cells (ILC2), mast cells, basophils, and neutrophils when compared to healthy sinonasal tissue 46, 49–51. Walford and colleagues reported an overall correlation between the number of eosinophils and ILC2s in nasal polyps but they did not specifically compare ILC2 numbers in AERD versus CRSwNP nasal polyps 50. Similar to eosinophils, mast cell numbers did not significantly differ between CRSwNP and AERD nasal polyps 41, 52. However, mast cells are a major source of prostaglandin D2 (PGD2) and this inflammatory mediator has been shown to be significantly elevated in AERD compared to CRSwNP patients, especially after aspirin challenge 52, 53. Finally, basophil numbers were significantly reduced in AERD versus CRSwNP nasal polyps 46. It remains unclear exactly how eosinophils, basophils, mast cells, and ILC2s each contribute to sinonasal inflammation and the clinical symptoms of either CRSwNP or AERD. Furthermore, given the elevated levels of PGD2 and ECP despite a lack of significant elevation in number of mast cells or eosinophils, it is tempting to speculate that these innate immune cells are more activated in AERD than CRSwNP. However, additional studies are needed to further evaluate this hypothesis. Even less is known regarding adaptive immune cells in AERD. T cells are elevated in nasal polyps of CRSwNP patients compared to healthy controls 43, 54. Additionally, B cells and plasma cells are also elevated in CRSwNP nasal polyps and thought to be locally activated within this tissue to produce increased amounts of IgG, IgA, IgM and IgE 43, 55–57. Some of the antibodies found in CRSwNP nasal polyps are against self-proteins (e.g. nuclear antigens and cytokines) but it is unclear if similar levels of autoantibodies exist in AERD patients 58, 59. It has been estimated that as many as 64% of CRSwNP and 87% of AERD patients are colonized with Staphylococcus aureus with a subset of these patients producing IgE against staphylococcal superantigens 60. Some studies report significantly increased levels of IgE to staphylococcal superantigens in AERD nasal polyps compared to CRSwNP 61, 62 while a more recent study found no difference 63. Taken together, additional studies are needed to further define the relationship between IgE to staphylococcal superantigens in AERD and CRSwNP nasal polyps. However, these specific antibodies may still contribute by an unknown mechanism to the chronic sinonasal inflammation observed in both diseases. In addition, Staphylococcus aureus enterotoxins can activate large numbers of T cells, and there is a body of literature suggesting that T cell activation mediated by these toxins plays a role in CRSwNP 64–66. Arachidonic Acid Metabolites AERD is classically characterized by a systemic dysregulation in arachidonic acid metabolism but a full description of this pathway is beyond the scope of this review. There have been some studies that have specifically focused on examination of prostaglandin and leukotriene mediators in the sinonasal tissue of AERD compared to CRSwNP patients. For example, levels of Prostaglandin E2 (PGE2), along with its receptor EP2, are decreased in AERD versus CRSwNP nasal polyps 67–70. Given the anti-inflammatory properties of PGE2, its reduction in AERD could in turn contribute to the enhanced sinonasal inflammation observed. Laidlaw and colleagues recently reported that the number of platelet-adherent leukocytes was significantly increased in AERD compared to CRSwNP nasal polyps 71. Furthermore, these platelets could convert leukocyte-derived Leukotriene (LT) A4 into LTC4 and thus contribute to the elevated levels of cysteinyl leukotrienes that are classically seen in AERD nasal polyps compared to CRSwNP or healthy controls 67, 71. Furthermore, the cysteinyl leukotriene receptor, CysLT1, is also elevated in AERD versus CRSwNP 68, 72, 73. Interestingly, in CRSwNP, cysteinyl leukotrienes (LTC4, LTD4, and LTE4) were also significantly elevated in sinonasal tissue when compared to healthy mucosae 69, 74. How these mediators all contribute to different clinical phenotypes in AERD versus CRSwNP is still not entirely understood. Other Inflammatory Mediators In addition to arachidonic acid metabolites, there have been numerous studies characterizing the baseline inflammatory milieu of nasal polyps from CRSwNP and AERD patients. Given the association with type-2 innate immune cells (e.g. mast cells, eosinophils, and basophils), it is not surprising that there are elevations in levels of traditional type-2 inflammatory mediators in AERD and also in CRSwNP nasal polyps when compared to healthy sinonasal tissue (e.g. IL-5, IL-4, IL-13, Eotaxin-1 Eotaxin-2) 43, 4448, 45, 75, 76. However, when directly comparing AERD nasal polyps with CRSwNP nasal polyps, there have been conflicting results. Some reports suggest that protein levels of IL-5 and the type-1 cytokine, IFNγ, are elevated in AERD compared to CRSwNP polyps 67, 77. More recently, a separate study evaluated the expression of 19 different inflammatory mediators in AERD and CRSwNP polyps. Surprisingly, there were no differences in levels of type-2 cytokines (IL-4, IL-5, IL-13, Eotaxin-1, Eotaxin-2, and MCP-4) or type-1 cytokines (IL-6, IFNγ) between these diseases 45, although levels of GM-CSF and MCP-1 were significantly elevated in AERD nasal polyps 45. It is unclear why there are discrepancies among studies but possible explanations could include differences in the tissues collected, techniques utilized to measure protein levels, the use of corticosteroids or other drugs within the study populations etc. In summary, it appears that both CRSwNP and AERD are characterized by a type-2 inflammatory milieu with increased levels of both type-2 cytokines and eosinophils when compared to healthy sinonasal tissue but not necessarily when compared to each other. There is a marked increase in levels of certain mediators released by activated innate immune cells (e.g. ECP and PGD2) in AERD compared to CRSwNP. While the underlying mechanisms that contribute to the increase in inflammatory cell degranulation products is unknown, it is worth exploring whether the abnormalities in arachidonic acid metabolism or other factors mediate the exaggerated activation of these cells in AERD. Summary CRSwNP and AERD are both important clinical diseases associated with increased socioeconomic burden and decreased quality of life. The most prominent clinical feature distinguishing patients with AERD from those with CRSwNP remains the development of an upper or lower respiratory tract reaction following the ingestion of a COX-1 inhibitor. However, even in the absence of COX-1 inhibitors, there are other clinical and pathophysiological differences observed between AERD and CRSwNP patients. Those with AERD on average have worse upper respiratory disease with increased sinonasal symptoms, mucosal inflammation and requirements for revision sinus surgery when compared to patients with CRSwNP. While no single genetic factor has been identified in either CRSwNP or AERD pathogenesis to date, studies characterizing differences in the underlying cellular and molecular mechanisms that could account for the observed clinical variations are progressing. Both AERD and CRSwNP nasal polyps are now known to be characterized by a type-2 inflammatory environment but there appear to be significantly increased levels of eosinophil and mast cell degranulation occurring in AERD sinonasal tissue. In addition, AERD, unlike CRSwNP, is also characterized by the systemic dysregulation of arachidonic acid metabolism and activation of platelet function. The underlying mechanisms that contribute to these observations are still not fully known and additional studies are needed to further define both CRSwNP and AERD pathogenesis. Funding: This work was supported by Chronic Rhinosinusitis Integrative Studies Program (U19-AI106683) and by the National Institutes of Health grants T32 AI083216, R37 HLO68546, RO1 HL0788860 and R01 AI104733. Key Points Clinically, patients with AERD can be distinguished from those with CRSwNP and asthma by the development of respiratory symptoms following the ingestion of a COX-1 inhibitor. However, clinical history alone may not always be sufficient to confirm the diagnosis of AERD. In the absence of COX-1 inhibitors, AERD patients on average still have worse upper and lower respiratory tract disease than those patients with CRSwNP with or without asthma. Nasal polyps from AERD and CRSwNP patients are both defined by a predominant type-2 inflammatory environment but there appears to be significantly increased levels of eosinophil and mast cell degranulation occurring in AERD. Mechanistically, AERD, unlike CRSwNP, is characterized by platelet activation as well as a dysregulation in arachidonic acid metabolism. Disclosures: Whitney Stevens has no financial conflicts of interest. Robert Schleimer has served as a consultant with several pharmaceutical companies with interest in CRS, including Astra-Zeneca, Genentech, GSK, Intersect ENT, Merck, Regeneron and Sanofi. Dr. Schleimer is a founder, shareholder and advisor for Allakos. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Fokkens WJ Lund VJ Mullol J Bachert C Alobid I Baroody F European Position Paper on Rhinosinusitis and Nasal Polyps 2012 Rhinol Suppl 2012 3 p preceding table of contents 1 298 2 Peters AT Spector S Hsu J Hamilos DL Baroody FM Chandra RK Diagnosis and management of rhinosinusitis: a practice parameter update Ann Allergy Asthma Immunol 2014 113 347 85 25256029 3 Samter M Beers RF Jr Intolerance to aspirin. Clinical studies and consideration of its pathogenesis Ann Intern Med 1968 68 975 83 5646829 4 Szczeklik A Nizankowska E Duplaga M Natural history of aspirin-induced asthma. AIANE Investigators. European Network on Aspirin-Induced Asthma Eur Respir J 2000 16 432 6 11028656 5 Stevenson DD Sanchez-Borges M Szczeklik A Classification of allergic and pseudoallergic reactions to drugs that inhibit cyclooxygenase enzymes Ann Allergy Asthma Immunol 2001 87 177 80 11570612 6 Rajan JP Wineinger NE Stevenson DD White AA Prevalence of aspirin-exacerbated respiratory disease among asthmatic patients: A meta-analysis of the literature J Allergy Clin Immunol 2015 135 676 81 e1 25282015 7 Promsopa C Kansara S Citardi MJ Fakhri S Porter P Luong A Prevalence of confirmed asthma varies in chronic rhinosinusitis subtypes Int Forum Allergy Rhinol 2015 8 Tan BK Zirkle W Chandra RK Lin D Conley DB Peters AT Atopic profile of patients failing medical therapy for chronic rhinosinusitis Int Forum Allergy Rhinol 2011 1 88 94 21731824 9 Morales DR Guthrie B Lipworth BJ Jackson C Donnan PT Santiago VH NSAID-exacerbated respiratory disease: a meta-analysis evaluating prevalence, mean provocative dose of aspirin and increased asthma morbidity Allergy 2015 70 828 35 25855099 10 Joe SA Thakkar K Chronic rhinosinusitis and asthma Otolaryngol Clin North Am 2008 41 297 309 vi 18328369 11 Lin DC Chandra RK Tan BK Zirkle W Conley DB Grammer LC Association between severity of asthma and degree of chronic rhinosinusitis Am J Rhinol Allergy 2011 25 205 8 21819754 12 Delaney JC The diagnosis of aspirin idiosyncrasy by analgesic challenge Clin Allergy 1976 6 177 81 1277441 13 Dursun AB Woessner KA Simon RA Karasoy D Stevenson DD Predicting outcomes of oral aspirin challenges in patients with asthma, nasal polyps, and chronic sinusitis Ann Allergy Asthma Immunol 2008 100 420 5 18517072 14 Berges-Gimeno MP Simon RA Stevenson DD The natural history and clinical characteristics of aspirin-exacerbated respiratory disease Ann Allergy Asthma Immunol 2002 89 474 8 12452205 15 Stevens WW Peters AT Suh L Norton JE Kern RC Conley DB A retrospective, cross-sectional study reveals that women with CRSwNP have more severe disease than men Immun Inflamm Dis 2015 3 14 22 25866636 16 Jang DW Comer BT Lachanas VA Kountakis SE Aspirin sensitivity does not compromise quality-of-life outcomes in patients with Samter’s triad Laryngoscope 2014 124 34 7 23712910 17 Awad OG Lee JH Fasano MB Graham SM Sinonasal outcomes after endoscopic sinus surgery in asthmatic patients with nasal polyps: a difference between aspirin-tolerant and aspirin-induced asthma? Laryngoscope 2008 118 1282 6 18475212 18 Robinson JL Griest S James KE Smith TL Impact of aspirin intolerance on outcomes of sinus surgery Laryngoscope 2007 117 825 30 17473677 19 Young J Frenkiel S Tewfik MA Mouadeb DA Long-term outcome analysis of endoscopic sinus surgery for chronic sinusitis Am J Rhinol 2007 21 743 7 18201458 20 Yip J Yao CM Lee JM State of the art: a systematic review of the surgical management of aspirin exacerbated respiratory disease Am J Rhinol Allergy 2014 28 493 501 25514486 21 Kim JE Kountakis SE The prevalence of Samter’s triad in patients undergoing functional endoscopic sinus surgery Ear Nose Throat J 2007 86 396 9 17702319 22 Batra PS Tong L Citardi MJ Analysis of comorbidities and objective parameters in refractory chronic rhinosinusitis Laryngoscope 2013 123 Suppl 7 S1 11 23 Pearlman AN Chandra RK Chang D Conley DB Tripathi-Peters A Grammer LC Relationships between severity of chronic rhinosinusitis and nasal polyposis, asthma, and atopy Am J Rhinol Allergy 2009 23 145 8 19401038 24 Bochenek G Szafraniec K Kuschill-Dziurda J Nizankowska-Mogilnicka E Factors associated with asthma control in patients with aspirin-exacerbated respiratory disease Respir Med 2015 109 588 95 25820158 25 Lee JH Haselkorn T Borish L Rasouliyan L Chipps BE Wenzel SE Risk factors associated with persistent airflow limitation in severe or difficult-to-treat asthma: insights from the TENOR study Chest 2007 132 1882 9 18079222 26 Moore WC Meyers DA Wenzel SE Teague WG Li H Li X Identification of asthma phenotypes using cluster analysis in the Severe Asthma Research Program Am J Respir Crit Care Med 2010 181 315 23 19892860 27 Bochenek G Kuschill-Dziurda J Szafraniec K Plutecka H Szczeklik A Nizankowska-Mogilnicka E Certain subphenotypes of aspirin-exacerbated respiratory disease distinguished by latent class analysis J Allergy Clin Immunol 2014 133 98 103 e1 6 23993879 28 Robinson S Douglas R Wormald PJ The relationship between atopy and chronic rhinosinusitis Am J Rhinol 2006 20 625 8 17181106 29 Ramadan HH Fornelli R Ortiz AO Rodman S Correlation of allergy and severity of sinus disease Am J Rhinol 1999 13 345 7 10582111 30 Chang JE White A Simon RA Stevenson DD Aspirin-exacerbated respiratory disease: burden of disease Allergy Asthma Proc 2012 33 117 21 22525387 31 Hsu J Avila PC Kern RC Hayes MG Schleimer RP Pinto JM Genetics of chronic rhinosinusitis: state of the field and directions forward J Allergy Clin Immunol 2013 131 977 93 93 e1 5 23540616 32 Park SM Park JS Park HS Park CS Unraveling the genetic basis of aspirin hypersensitivity in asthma beyond arachidonate pathways Allergy Asthma Immunol Res 2013 5 258 76 24003382 33 Kurosawa M Yukawa T Hozawa S Mochizuki H Recent advance in investigation of gene polymorphisms in Japanese patients with aspirin-exacerbated respiratory disease Allergol Immunopathol (Madr) 2015 43 92 100 25224359 34 Kim SH Choi H Yoon MG Ye YM Park HS Dipeptidyl-peptidase 10 as a genetic biomarker for the aspirin-exacerbated respiratory disease phenotype Ann Allergy Asthma Immunol 2015 114 208 13 25592153 35 Chang HS Park JS Shin HR Park BL Shin HD Park CS Association analysis of FABP1 gene polymorphisms with aspirin-exacerbated respiratory disease in asthma Exp Lung Res 2014 40 485 94 25338211 36 Stankovic KM Goldsztein H Reh DD Platt MP Metson R Gene expression profiling of nasal polyps associated with chronic sinusitis and aspirin-sensitive asthma Laryngoscope 2008 118 881 9 18391768 37 Cheong HS Park SM Kim MO Park JS Lee JY Byun JY Genome-wide methylation profile of nasal polyps: relation to aspirin hypersensitivity in asthmatics Allergy 2011 66 637 44 21121930 38 Szczeklik A Aspirin-induced asthma as a viral disease Clin Allergy 1988 18 15 20 3127083 39 Chang JE Ding D Martin-Lazaro J White A Stevenson DD Smoking, environmental tobacco smoke, and aspirin-exacerbated respiratory disease Ann Allergy Asthma Immunol 2012 108 14 9 22192959 40 Mahdavinia M Keshavarzian A Tobin MC Landay AL Schleimer RP A comprehensive review of the nasal microbiome in chronic rhinosinusitis (CRS) Clin Exp Allergy 2016 46 21 41 26510171 41 Stevens WW Lee RJ Schleimer RP Cohen NA Chronic rhinosinusitis pathogenesis J Allergy Clin Immunol 2015 136 1442 53 26654193 42 Lou H Meng Y Piao Y Zhang N Bachert C Wang C Cellular phenotyping of chronic rhinosinusitis with nasal polyps Rhinology 2016 43 Van Zele T Claeys S Gevaert P Van Maele G Holtappels G Van Cauwenberge P Differentiation of chronic sinus diseases by measurement of inflammatory mediators Allergy 2006 61 1280 9 17002703 44 Zhang N Van Zele T Perez-Novo C Van Bruaene N Holtappels G DeRuyck N Different types of T-effector cells orchestrate mucosal inflammation in chronic sinus disease J Allergy Clin Immunol 2008 122 961 8 18804271 45 Stevens WW Ocampo CJ Berdnikovs S Sakashita M Mahdavinia M Suh L Cytokines in Chronic Rhinosinusitis: Role in Eosinophilia and Aspirin Exacerbated Respiratory Disease Am J Respir Crit Care Med 2015 46 Mahdavinia M Carter RG Ocampo CJ Stevens W Kato A Tan BK Basophils are elevated in nasal polyps of patients with chronic rhinosinusitis without aspirin sensitivity J Allergy Clin Immunol 2014 133 1759 63 24636088 47 Mahdavinia M Suh LA Carter RG Stevens WW Norton JE Kato A Increased noneosinophilic nasal polyps in chronic rhinosinusitis in US second-generation Asians suggest genetic regulation of eosinophilia J Allergy Clin Immunol 2015 135 576 9 25312761 48 Fan Y Feng S Xia W Qu L Li X Chen S Aspirin-exacerbated respiratory disease in China: a cohort investigation and literature review Am J Rhinol Allergy 2012 26 e20 2 22391072 49 Ho J Bailey M Zaunders J Mrad N Sacks R Sewell W Group 2 innate lymphoid cells (ILC2s) are increased in chronic rhinosinusitis with nasal polyps or eosinophilia Clin Exp Allergy 2015 45 394 403 25429730 50 Walford HH Lund SJ Baum RE White AA Bergeron CM Husseman J Increased ILC2s in the eosinophilic nasal polyp endotype are associated with corticosteroid responsiveness Clin Immunol 2014 155 126 35 25236785 51 Takabayashi T Kato A Peters AT Suh LA Carter R Norton J Glandular mast cells with distinct phenotype are highly elevated in chronic rhinosinusitis with nasal polyps J Allergy Clin Immunol 2012 130 410 20 e5 22534535 52 Buchheit KM Cahill KN Katz HR Murphy KC Feng C Lee-Sarwar K Thymic stromal lymphopoietin controls prostaglandin D generation in patients with aspirin-exacerbated respiratory disease J Allergy Clin Immunol 2015 53 Cahill KN Bensko JC Boyce JA Laidlaw TM Prostaglandin D(2): a dominant mediator of aspirin-exacerbated respiratory disease J Allergy Clin Immunol 2015 135 245 52 25218285 54 Derycke L Eyerich S Van Crombruggen K Perez-Novo C Holtappels G Deruyck N Mixed T helper cell signatures in chronic rhinosinusitis with and without polyps PLoS One 2014 9 e97581 24911279 55 Van Zele T Gevaert P Holtappels G van Cauwenberge P Bachert C Local immunoglobulin production in nasal polyposis is modulated by superantigens Clin Exp Allergy 2007 37 1840 7 17941912 56 Hulse KE Norton JE Suh L Zhong Q Mahdavinia M Simon P Chronic rhinosinusitis with nasal polyps is characterized by B-cell inflammation and EBV-induced protein 2 expression J Allergy Clin Immunol 2013 131 1075 83 83 e1 7 23473835 57 Kato A Peters A Suh L Carter R Harris KE Chandra R Evidence of a role for B cell-activating factor of the TNF family in the pathogenesis of chronic rhinosinusitis with nasal polyps J Allergy Clin Immunol 2008 121 1385 92 92 e1 2 18410958 58 Tan BK Li QZ Suh L Kato A Conley DB Chandra RK Evidence for intranasal antinuclear autoantibodies in patients with chronic rhinosinusitis with nasal polyps J Allergy Clin Immunol 2011 128 1198 206 e1 21996343 59 Tsybikov NN Egorova EV Kuznik BI Fefelova EV Magen E Anticytokine autoantibodies in chronic rhinosinusitis Allergy Asthma Proc 2015 36 473 80 26534753 60 Van Zele T Gevaert P Watelet JB Claeys G Holtappels G Claeys C Staphylococcus aureus colonization and IgE antibody formation to enterotoxins is increased in nasal polyposis J Allergy Clin Immunol 2004 114 981 3 15480349 61 Perez-Novo CA Kowalski ML Kuna P Ptasinska A Holtappels G van Cauwenberge P Aspirin sensitivity and IgE antibodies to Staphylococcus aureus enterotoxins in nasal polyposis: studies on the relationship Int Arch Allergy Immunol 2004 133 255 60 14976394 62 Suh YJ Yoon SH Sampson AP Kim HJ Kim SH Nahm DH Specific immunoglobulin E for staphylococcal enterotoxins in nasal polyps from patients with aspirin-intolerant asthma Clin Exp Allergy 2004 34 1270 5 15298569 63 Yoo HS Shin YS Liu JN Kim MA Park HS Clinical significance of immunoglobulin E responses to staphylococcal superantigens in patients with aspirin-exacerbated respiratory disease Int Arch Allergy Immunol 2013 162 340 5 24193355 64 Conley DB Tripathi A Seiberling KA Schleimer RP Suh LA Harris K Superantigens and chronic rhinosinusitis: skewing of T-cell receptor V beta-distributions in polyp-derived CD4+ and CD8+ T cells Am J Rhinol 2006 20 534 9 17063750 65 Huvenne W Hellings PW Bachert C Role of staphylococcal superantigens in airway disease Int Arch Allergy Immunol 2013 161 304 14 23689556 66 Bachert C Zhang N Patou J van Zele T Gevaert P Role of staphylococcal superantigens in upper airway disease Curr Opin Allergy Clin Immunol 2008 8 34 8 18188015 67 Perez-Novo CA Watelet JB Claeys C Van Cauwenberge P Bachert C Prostaglandin, leukotriene, and lipoxin balance in chronic rhinosinusitis with and without nasal polyposis J Allergy Clin Immunol 2005 115 1189 96 15940133 68 Adamusiak AM Stasikowska-Kanicka O Lewandowska-Polak A Danilewicz M Wagrowska-Danilewicz M Jankowski A Expression of arachidonate metabolism enzymes and receptors in nasal polyps of aspirin-hypersensitive asthmatics Int Arch Allergy Immunol 2012 157 354 62 22123288 69 Yoshimura T Yoshikawa M Otori N Haruna S Moriyama H Correlation between the prostaglandin D(2)/E(2) ratio in nasal polyps and the recalcitrant pathophysiology of chronic rhinosinusitis associated with bronchial asthma Allergol Int 2008 57 429 36 18797183 70 Machado-Carvalho L Torres R Perez-Gonzalez M Alobid I Mullol J Pujols L Altered expression and signalling of EP2 receptor in nasal polyps of AERD patients: role in inflammation and remodelling Rhinology 2016 71 Laidlaw TM Kidder MS Bhattacharyya N Xing W Shen S Milne GL Cysteinyl leukotriene overproduction in aspirin-exacerbated respiratory disease is driven by platelet-adherent leukocytes Blood 2012 119 3790 8 22262771 72 Sousa AR Parikh A Scadding G Corrigan CJ Lee TH Leukotriene-receptor expression on nasal mucosal inflammatory cells in aspirin-sensitive rhinosinusitis N Engl J Med 2002 347 1493 9 12421891 73 Corrigan C Mallett K Ying S Roberts D Parikh A Scadding G Expression of the cysteinyl leukotriene receptors cysLT(1) and cysLT(2) in aspirin-sensitive and aspirin-tolerant chronic rhinosinusitis J Allergy Clin Immunol 2005 115 316 22 15696087 74 Perez-Novo CA Claeys C Van Cauwenberge P Bachert C Expression of eicosanoid receptors subtypes and eosinophilic inflammation: implication on chronic rhinosinusitis Respir Res 2006 7 75 16689996 75 Van Bruaene N Perez-Novo CA Basinski TM Van Zele T Holtappels G De Ruyck N T-cell regulation in chronic paranasal sinus disease J Allergy Clin Immunol 2008 121 1435 41 41 e1 3 18423831 76 Steinke JW Payne SC Borish L Interleukin-4 in the Generation of the AERD Phenotype: Implications for Molecular Mechanisms Driving Therapeutic Benefit of Aspirin Desensitization J Allergy (Cairo) 2012 2012 182090 22262978 77 Steinke JW Liu L Huyett P Negri J Payne SC Borish L Prominent role of IFN-gamma in patients with aspirin-exacerbated respiratory disease J Allergy Clin Immunol 2013 132 856 65 e1 3 23806637

PMC005xxxxxx/PMC5119898.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8704771 1572 Alzheimer Dis Assoc Disord Alzheimer Dis Assoc Disord Alzheimer disease and associated disorders 0893-0341 1546-4156 27819841 5119898 10.1097/WAD.0000000000000162 NIHMS798771 Article Recruiting U.S. Chinese Elders into Clinical Research for Dementia Li Clara Ph.D. 1 Neugroschl Judith M.D. 1 Umpierre Mari Ph.D., LCSW. 1 Martin Jane Ph.D. 1 Huang QiYing MSW. 1 Zeng Xiaoyi B.A. 1 Cai Dongming M.D., Ph.D. 1 Sano Mary Ph.D. 12 1 Icahn School of Medicine at Mount Sinai, New York, NY 2 James J. Peters VA Medical Center, Bronx, NY Corresponding Author: Clara Li, Ph.D., Department of Psychiatry Alzheimer’s Disease Research Center, Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, Box 1230, New York, NY 10029-6574, Tel: 212.241.6500 ext. 88786, Fax: 212.996.0987, clara.li@mssm.edu 7 7 2016 Oct-Dec 2016 01 10 2017 30 4 345347 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Purpose This study described and evaluated the rapid recruitment of elderly Chinese into clinical research at the Mount Sinai Alzheimer’s Disease Research Center (MSADRC). Design and Methods Methods of publicizing the study included lectures to local senior centers/churches and publications in local Chinese newspapers. The amount of time and success of these methods were evaluated. A “go to them” model of evaluation was employed to enable participants to complete the study visit at locations where they were comfortable. Results From January to December 2015, we recruited 98 participants aged ≥ 65 who primarily speak Mandarin/Cantonese and reside in New York. The mean age and years of education was 73.93±6.34 and 12.79±4.58, respectively. The majority of participants were female (65.3%) and primarily Mandarin speaking (53.1%). Of all enrollees, 54.1% were recruited from community lectures, 29.6% through newspapers, 10.2% through word of mouth, and 6.1% from our clinical services. 40.8% of participants underwent evaluations at the MSADRC, 44.9% at local senior centers/churches, and 14.3% at home. Implications Given that the majority of our participants had low English proficiency, the use of bilingual recruiters probably allowed us to overcome the language barrier, facilitating recruitment. Our “go to them” model of evaluation is another important factor contributing to our successful recruitment. elderly Chinese Americans clinical study dementia aging recruitment minority Chinese Americans are one of the fastest growing populations among the elders in the United States (U.S.).1 Many elderly Chinese are at a high risk for developing cognitive loss and dementia, probably due to the high prevalence of vascular risk factors,2 low level of education,3 and limited access to healthcare services.4 However, recruitment of elderly Chinese Americans into research remains to be a challenge for a variety of reasons, such as misconceptions that memory loss is an inevitable part of aging, lack of trust in research and/or researchers, and social stigma associated with dementia.5 Additionally, many elderly Chinese Americans are fearful of going to a medical center and/or reluctant to leave their neighborhood for healthcare. This manuscript describes the experience with recruitment of U.S. Chinese elders into aging and dementia research at the Mount Sinai Alzheimer’s Disease Research Center (MSADRC) in New York City. Design and Methods Chinese Outreach Program Our Chinese Outreach Program was staffed by a bilingual Ph.D. level neuropsychologist (CL) at 70% effort and a bilingual research coordinator (QH) at 50% effort. The bilingual staff translated recruitment flyers, brochures, and consent forms from English into Simplified and Traditional Chinese. The Chinese recruitment flyers and brochures were distributed in local senior centers and churches following community talks. All study materials were approved by the local Intuitional Review Board (IRB). PowerPoint lectures on dementia and cognitive aging were given at four local churches and six senior centers frequented by elderly Chinese. The lectures were conducted in Cantonese and simultaneously translated into Mandarin. These lectures varied in length from 30 minutes to one hour and ended with information about research opportunities at the MSADRC. At the end of the lectures, audiences were given the opportunity to sign up to participate in the aging and dementia study at the MSADRC. A newspaper announcement about the lecture was printed in a local Chinese newspaper and paid for by the church where the lecture took place. The contexts of the lecture and research opportunities at the MSADRC were published as a column in a local Chinese newspaper at no cost following an interview with the newspaper journalist who attended the announced lecture. Subjects Our outreach effort targeted individuals who were 65 years of age or older, primarily Chinese speaking, New York residents, and able to complete cognitive assessments. Study Procedures An IRB approved Chinese version of the MSADRC consent form was signed by participants or study partners as appropriate prior to study participation. The study involved a 3-hour in-person clinical evaluation modeled by the MSADRC, which includes medical, functional, and cognitive assessments. An optional blood draw for DNA banking and genetic analysis was included as part of the standard dementia evaluation. The “go to them” model of evaluation was used, which allows participants to complete the study procedures in locations that were comfortable and convenient for them, such as participants’ own home, local senior centers/churches, and the MSADRC. Upon completion of the evaluation, all participants were assigned a research diagnosis of normal cognition, Mild Cognitive Impairment (MCI), and dementia using the diagnostic criteria used by the MSADRC. A brief report, written in both Chinese and English, was mailed to each research participant detailing results of the cognitive testing along with research diagnosis and clinical recommendations. Results Through our Chinese outreach program, we successfully enrolled 98 participants between January and December 2015. In the one-year period, community lectures generated 65 potential interests and 54 enrollees. Newspaper exposures, on the other hand, produced 66 potential interests and 29 enrollees. Table 1 shows details in recruitment efforts and outcomes from the two recruitment methods. Figure 1 shows the lag time between outreach and recruitment. Of all the enrollees, 54.1% were recruited from community talks; 29.6% from newspapers; 10.2% through word of mouth; and 6.1% from our clinical services (e.g. social workers and physician referrals). The average age at initial interviews was 73.93±6.34 years. The sample cohort has a relatively high level of education (12.79± 4.58 years). Many of the research subjects have been living in the U.S. for over three decades. Most were right-hand predominant (92.9%), married (65.3%), females (65.3%), and Mandarin speaking (53.1%). 40.8% of participants were evaluated at MSADRC, 44.9% at their senior centers/church, and 14.3% at home. 87.5% of the participants and 59.3% of the informants reported concerns about memory functioning and/or thinking abilities of the research participants. The informants were spouses (43.9%), children (33.7%), other family member (6.1%), friends (9.2%), and health aides (4.1%). 81.6% of participants agreed to have their blood drawn. Discussion Our Chinese outreach program, established to engage elderly Chinese Americans into research, demonstrates that direct face to face outreach with bilingual staff members is the most successful recruitment technique. A majority of our elderly Chinese Americans enrolled in that year were a result of this face to face approach. While newspaper exposures generated a greater number of potential interests than community lectures, the lectures (a face to face methodology) produced greater enrollment. In addition, community talks appeared to be less time consuming and yielded a higher success rate in recruiting the targeted population compared to newspaper announcements, suggesting that direct face to face outreach with bilingual staff members was a more successful form of recruitment. Overall, our experience provides a template for successful recruitment of elderly Chinese Americans, and perhaps recruitment of other immigrant elderly populations, into aging and dementia research. First, we find that U.S. elderly Chinese participants, like many older adults, have concerns about their memory functions, and are willing to participate in clinical research. Most participants agreed to DNA banking and genetic analysis, suggesting that other biomarker studies may be of interest. A factor that appears to increase willingness to participate is participants’ desires to understand their levels of cognitive health. Most of our participants reported experiencing a decline in their memory and were motivated to learn more about their cognitive functions. Most of our participants do not speak English well, and the use of bilingual recruiters allowed us to overcome the language barrier that would prevent access to cognitive assessment. In addition, our graph illustrates that recruitment efforts takes time to show effect. It took several talks before enrollments occurred and there was a sharp increase in enrollment following each outreach event but then recruitment reached a plateau and needed another outreach to boost enrollment. Further, nearly 60% of our evaluations were conducted at the participants’ homes and senior centers/churches, indicating our “go to them” model of evaluation was another essential element to our successful recruitments. Data from the current study indicates that our participants were largely in their 70s, high school graduates, residents in the U.S. for over three decades, females, married, right-handed, born in China, and Mandarin speaking. The results help to identify specific sub-groups of participants who are more likely to participate in research and those who may require more attention for improving future recruitments in clinical research for elderly Chinese Americans. While our data showed that it is possible to recruit the minority population into aging and dementia research, our cohort may not represent the average population of elderly Chinese Americans in New York. In addition, our sample has a relatively high level of education. Previous studies have reported that low education may threaten the validity of diagnostic testing.6 Systematic collection of clinical data will help to establish better norms. Future studies focusing on recruitment of participants with low levels of education are needed. Source of Support: NIH National Institute on Aging grant (P50AG05138) Figure 1 Graph 1: Lag Time between Outreach and Recruitment Community talks and newspaper announcements were successful in recruiting participants but their effectiveness began to fade towards the end of the year. Table 1 Outreach Recruitment Efforts and Outcomes Community talks provided an opportunity for information sharing pertaining to study project and resulted in less one-on-one calls to explain study proceudre and yielded more eligible participants compared to newspaper annoucements. Recruitment Methods Community Lectures Newspaper Announcements Total exposures 249 38,000 Total potential interests 65 66 Total enrollments 54 29 Bilingual staff efforts Giving community lectures (11 hours) Speaking with potential participants (6 hours) Scheduling participants for study visits (5 hours) Creating power point slides (5 hours) Translating brochures/flyers (6 hours) Contacting local senior centers/churches (3 hours) An interview with a newspaper journalist (1 hour) Speaking with potential participants (36 hours) Scheduling participants for study visits (4 hours) Total staff hours 36 hours 41 hours Staff hour(s) spent to enroll one participant 0.67 hour 1.41 hours Total costs $600 $0 References 1 Asian/Pacific American Heritage Month EaSAW USDoC Facts for Features U.S. Census Bureau 2011 5 Washington, DC USA 2011 2 He J Iosif AM Lee DY Brain structure and cerebrovascular risk in cognitively impaired patients: Shanghai Community Brain Health Initiative-pilot phase Archives of neurology 2010 10 67 10 1231 1237 20937951 3 Salmon DP Jin H Zhang M Grant I Yu E Neuropsychological assessment of Chinese elderly in the Shanghai Dementia Survey The Clinical Neuropsychologist 1995 9 159 168 4 Nguyen D Bornheimer LA Mental health service use types among Asian Americans with a psychiatric disorder: considerations of culture and need The journal of behavioral health services & research 2014 10 41 4 520 528 24402440 5 Chao SZ Lai NB Tse MM Recruitment of Chinese American elders into dementia research: the UCSF ADRC experience The Gerontologist 2011 6 51 Suppl 1 S125 S133 21565814 6 Lezak M Neuropsychological assessment 1995 New York Oxford University Press

PMC005xxxxxx/PMC5119922.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8303331 22425 Nutr Res Nutr Res Nutrition research (New York, N.Y.) 0271-5317 1879-0739 27865347 5119922 10.1016/j.nutres.2016.08.001 NIHMS808774 Article Gestational food restriction decreases placental IL10 expression and markers of autophagy and ER stress in murine intrauterine growth restriction Chu Alison a1 Thamotharan Shanthie a2 Ganguly Amit a3 Wadehra Madhuri mwadehra@mednet.ucla.edu b Pellegrini Matteo matteop@mcdb.ucla.edu c Devaskar Sherin U. a4* a David Geffen School of Medicine at UCLA, Department of Pediatrics, Division of Neonatology & Developmental Biology, Neonatal Research Center of the UCLA Children's Discovery and Innovation Institute. 10833 Le Conte Avenue, MDCC 22-402, Los Angeles, CA, 90095, USA b David Geffen School of Medicine at UCLA, Department of Pathology, 4525 MacDonald Research Laboratories, Los Angeles, CA 90095, USA c David Geffen School of Medicine at UCLA, Department of Molecular, Cell, and Developmental Biology, 3000 Terasaki Life Sciences Building, 610 Charles Young Drive East, Los Angeles, California 90095, USA Corresponding author: Sherin U. Devaskar, David Geffen School of Medicine at UCLA, Department of Pediatrics, Division of Neonatology, 10833 Le Conte Avenue, MDCC 22-402, Los Angeles, CA 90095; (t) 310.825.9357, (f) 310.267.0154 1 alisonchu@mednet.ucla.edu 2 sthamotharan@mednet.ucla.edu 3 aganguly@mednet.ucla.edu 4 sdevaskar@mednet.ucla.edu 11 8 2016 5 8 2016 10 2016 01 10 2017 36 10 10551067 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Intrauterine growth restriction (IUGR) affects up to 10% of pregnancies and often results in short- and long- term sequelae for offspring. The mechanisms underlying IUGR are poorly understood, but it is known that healthy placentation is essential for nutrient provision to fuel fetal growth, and is regulated by immunologic inputs. We hypothesized that in pregnancy, maternal food restriction (FR) resulting in IUGR would decrease the overall immunotolerant milieu in the placenta, leading to increased cellular stress and death. Our specific objectives were to evaluate: (1) key cytokines (e.g. IL10) that regulate maternal-fetal tolerance, (2) cellular processes (autophagy and ER stress) that are immunologically-mediated and important for cellular survival and functioning, and (3) evaluate the resulting IUGR phenotype and placental histopathology. After subjecting pregnant mice to mild and moderate FR from gestational day (GD) 10 to 19, we collected placentas and embryos at GD19. We examined RNA sequencing data to identify immunologic pathways affected in IUGR-associated placentas and validated mRNA expression changes of genes important in cellular integrity. We also evaluated histopathologic changes in vascular and trophoblastic structures as well as protein expression changes in autophagy, ER stress, and apoptosis in the mouse placentas. Several differentially expressed genes were identified in FR compared to control mice, including a considerable subset that regulates immune tolerance, inflammation and cellular integrity. In summary, maternal food restriction decreases anti-inflammatory effect of IL10 and suppresses placental autophagic and ER stress responses, despite evidence of dysregulated vascular and trophoblast structures leading to IUGR. mouse placenta vasculature intrauterine growth restriction autophagy ER stress Interleukin-10 1. Introduction Intrauterine growth restriction (IUGR) is a poorly understood complication of pregnancy, affecting up to 10% of pregnancies. As the causes of IUGR are myriad, IUGR is thought to encompass a broad spectrum of diseases that results from poor nutrient provision to the fetus. This adverse intrauterine environment may, in some cases, result from disordered placentation or poor maternal health and nutrition [1,2]. It is critical to understand the contribution of maternal diet towards the pathophysiology of IUGR, as IUGR not only places offspring of these pregnancies at increased risk of short-term morbidity, but also at increased risk for adult-onset cardiovascular and metabolic diseases [3-6]. The placenta plays an important role in IUGR as it is the essential organ for transportation of nutrients from maternal blood to the fetus. Studies of placental gene expression have shown that the placenta is able to sense and respond to changes in the immediate environment (e.g. maternal diet) in ways that directly alter the transport of nutrients to the fetus [7-9]. The transfer of nutrients is largely mediated by blood flow and transcellular transport, via placental vasculature and across trophoblast cells. The reported placental changes seen in human IUGR include a degenerated syncytiotrophoblastic lining, fibrin deposition, hypovascular/avascular villi and large areas of infarction [10]. These findings likely represent changes that occur in response to longstanding nutrient and oxygen insufficiency in IUGR. These changes further exacerbate poor blood flow to the fetus, resulting in poor in-utero growth of the fetus. The development and maintenance of placental structure and cellular function are dependent upon immune modulation with several pathways communicating to allow immune tolerance at the maternal-fetal interface. A growing body of evidence suggests that pathology in pregnancy disorders such as preeclampsia, recurrent miscarriage and IUGR can be related to altered immune tolerance. Interleukin-10 (IL10), an anti-inflammatory cytokine, is a key immunologic signal that is altered in these pregnancy states. It normally acts as an anti-inflammatory pleiotropic regulator in pregnancy via suppression of inflammatory cytokines, inhibition of antigen expression via major histocompatibility complex (MHC) class II expression, protection against vascular dysfunction and inflammation, and modulation of endoplasmic reticulum (ER) stress and autophagy [11] (Figure 1). Our specific research objectives were to ascertain the effect of maternal food restriction resulting in placental insufficiency and IUGR on key cytokines in pregnancy affecting immune tolerance, e.g. IL10, and on downstream immunologic pathways regulating cellular maintenance, e.g. autophagy and ER stress. As autophagy and ER stress are processes that are essential for cellular functioning in the face of pathologic stimuli such as nutrient deprivation and hypoxia, we hypothesized that maternal food restriction during pregnancy would result in IUGR and concomitantly decrease “anti-inflammatory” signaling in the placenta and increase cellular stress and death via alterations in autophagy and ER stress. To do this, we exposed pregnant mice to varying degrees (25% and 50% restriction of daily food intake, by weight) of maternal food restriction (FR) during the second half of gestation, and assessed the effect of maternal FR on placenta and embryo weight. FR during gestational days (GD) 10-19 was chosen to mimic the chronicity of human IUGR, which more commonly manifests in the second half of pregnancy. This time period was also chosen because placental function and gene expression dramatically affect fetal growth patterns during this portion of pregnancy. To initially undertake a global approach, we analyzed RNA sequencing data generated from whole placental samples to identify key genes involved in immunomodulatory pathways that were affected in our animal model of IUGR. We identified candidate genes not previously characterized in placental pathology per se, but known to be important in autophagy, ER stress, and vascular inflammation in other human diseases. We then validated altered mRNA expression of these specific genes by real time PCR. We employed a combination of western immunoblotting as well as immunohistochemistry to evaluate whether tissue-level and/or protein expression of IL10 and markers of autophagy and ER stress pathways reflected the same directional changes as seen in the transcriptomic changes in these placentas. We also examined the placentas for histopathologic findings consistent with IUGR, and to localize associated degenerative changes that we would expect to see with altered autophagy and ER stress pathways. Taken together, these experiments further our understanding of how maternal diet specifically affects the immunologic milieu at the maternal-fetal interface and the cellular health of key placental structures for nutrient provision to the fetus. 2. Methods and materials This study was conducted in accordance with established guidelines and all protocols [12] were approved by the Animal Research Committee of the University of California Los Angeles in accordance with the guidelines set by the National Institutes of Health. C57/BL6 mice were housed in 12:12 hour light-dark cycles with ad libitum access to a standard rodent chow diet (Pico Lab Rodent Diet 20, cat# 5053, Lab Diet, St. Louis, MO; major ingredients included: ground corn, soybean meal, wheat middlings, whole wheat, fish meal, dried beet pulp, wheat germ, cane molasses, brewers dried yeast, ground oats, dehydrated alfalfa meal, soybean oil, whey, and calcium carbonate), and the vitamin and mineral premixes. The macronutrient content of the diet was comprised of the following: carbohydrate (62.3% energy), protein (24.5% energy), and fat (13.1 % energy). 2.1. Placental samples At 8 weeks of age, male and female mice were mated overnight, and pregnant mice were identified by the presence of a vaginal plug (designed as gestational day 1). Pregnant females were transferred to individual cages and reared on the same chow diet ad libitum until gestational day 10, at which point they were randomly assigned to three groups: 1) control group - ad libitum access to the standard rodent diet, 2) mild restriction group – restricted by 25% of their daily intake from gestational day 10 to 19, and 3) moderate restriction group – restricted by 50% of their daily intake from gestational day 10 to 19, as previously described (n=4-5 pregnant mothers/group) [7]. On GD 19, animals were euthanized by intraperitoneal injection of 100 mg/kg of phenobarbital. The placentas were separated from the respective fetuses and collected (n=20-25/group). The placentas and fetuses were weighed in a Mettler AB104 precision balance (0.01 mg sensitivity), and then immediately snap-frozen and stored at -80°C until further analysis. In some cases, placentas were fixed in paraffin and then processed for immunohistochemical staining, as described below (section 2.6). 2.2. RNA sequencing For this analysis, only two groups were considered. Whole placental samples were obtained from control and 50% FR groups (n=5/group). Briefly, tissue from single placentas was homogenized and RNA extracted as previously described [7]. Total RNA was quantified using the Qubit RNA assay (Thermofisher Scientific, Waltham, MA). 1,000ng of total RNA was used as starting material for each sample. Library preparation was performed using the Illumina TruSeq RNA Sample Preparation kit, according to the manufacturer's instructions. Libraries were run using 50-bp single-end reads on the HiSeq2000 System (Illumina), and reads were mapped with TopHat, which keeps unique alignments and those with up to two mismatches [13]. The quality of alignments was checked with FastQC and the resulting file was processed through the HTSeq program to create a gene matrix as input for downstream analysis. DESeq [14] was used to calculate differential expression, generating Reads Per Kilobase per Million mapped reads (RPKM) per gene [7]. 2.3. Library analysis Genes were considered differentially expressed between control and the 50% FR group if differentially methylated regions (DMR) located close to those genes had significantly different mean expression levels as determined by Student's t-test, p<0.05. Details on determination of mean methylation levels are included in section 2.3.1 below. Pathway analysis was then conducted to identify genes that were involved in functional immunomodulatory pathways. Specifically, genes previously identified as important in inflammatory, immunologic, vascular, autophagy, or ER stress pathways in human disease were selected for further validation. 2.3.1. Statistical analysis methods used in RNA sequencing and library analysis For each CG site, a t-score was calculated from the t-test of mean difference between the two groups of comparison. Sites with an absolute t-score greater than or equal to 1.5 (approximately top 10%) were deemed candidate DMRs. For each candidate DMR, z-score of average t-score from all CG sites within the region was calculated. If the absolute z-score was greater than threshold, as calculated based upon the false discovery rate of <5% [7], and mean methylation levels in the two groups differed by at least 15%, the region was considered a DMR. Genes that overlapped or had transcription start sites within 5Kbp of the DMR were considered associated genes to the DMR of interest. Genes were considered differentially expressed between control and the 50% FR group if differentially methylated regions (DMR) associated with those genes had mean expression levels that were significantly different between the two groups, by Student's t-test, with unadjusted p-value<0.05. Log2 ratios and adjusted p-values using Bonferroni correction were also calculated. False discovery rate for p value cutoffs was <5% [7]. 2.4. qRT-PCR Total cellular RNA was isolated using the RNeasy mini kit (Qiagen, Valencia, CA. USA). cDNA was generated from 1 μg of total RNA from placental tissue (n=6/group) by reverse transcription (RT) using a Superscript III Reverse Transcriptase kit (Invitrogen, San Diego, CA). Amplification was performed in triplicate using Taqman-based detection on a Step One real-time quantitative PCR thermocycler (Applied Biosystems), as previously described [12]. For IL-10, the IL-10 Taqman Gene expression Assay Mm01288386_m1 was used according to the manufacturer's instructions (Applied Biosystems). For all other genes of interest, a Fam/Tamra probe was used (Eurofins, MWG Operon). Relative gene expression was calculated using the comparative CT method [15] with 18S (Applied Biosystems, #4319413E) expression used as the internal control for normalization. The amplification cycles consisted of: 50°C for 2 minutes, 95°C for 20 seconds, then 45 cycles of 95°C for 1 second (denaturation) and 56-59°C for 20 seconds (annealing), followed by 72°C for 5 minutes (extension). Specific annealing temperatures, exon-spanning primers for amplification, and probe sequences for detection are provided in Table 1. 2.5. Western blotting BAX (Bcl-2 associated X protein) and BCL2 (B-cell lymphoma 2) protein detection was undertaken to evaluate for changes in apoptosis in whole placental samples. BIP (Binding immunoglobulin protein) detection was undertaken by Western blot analysis to evaluate overall ER stress in whole placental samples. Briefly, tissue homogenates from whole placental samples were solubilized in 50mM Tris, pH 6.8, containing 2% SDS. Protein concentration was determined by using the Bio-Rad dye-binding assay. Western blotting was performed as described previously (n=9/group) [16]. Briefly, the solubilized protein homogenates (20 μg) were subjected to electrophoresis on 10% SDS-polyacrylamide gels and transferred to nitrocellulose membranes (Transblot; Bio-Rad, Hercules, CA). The following primary antibodies were used for signal detection: BAX antibody (Cell Signaling Technology, 1:750 dilution), BCL2 antibody (Cell Signaling Technology, 1:750 dilution), and BIP antibody (Cell Signaling Technology, 1:1000 dilution). Anti-vinculin antibody (Sigma-Aldrich, St. Louis, MO, 1:7000) was used to detect endogenous vinculin, which served as an internal control for inter-lane loading variability [17]. For all bands, resultant optical density was ensured as linear to the loading protein concentrations, and the intensity of the protein bands was assessed by densitometry using the Scion Image software program [16]. 2.6. Immunohistochemistry Immunohistochemistry for detection of IL10, markers of autophagy (LC3B, a ubiquitin-like protein involved in the ATG-8-conjugation system, necessary for the formation of the autophagosome), vasculature (CD34+ endothelial cells), and trophoblast cells (cytokeratin staining) was performed. All tissues were rinsed in PBS, fixed in cold 10% neutral buffered formalin for 24 hours and then transferred to cold 70% ethanol. Samples were embedded in paraffin and sectioned at 5 μm thickness. A single section from each placenta was stained with hematoxylin and eosin by the Tissue Procurement Core Laboratory (TPCL) at UCLA. Samples were deparaffinized and dehydrated in alcohol. Antigen retrieval was performed using 0.1 M citrate, pH 6.0, at 95°C for 20 min. The slides were then incubated with antibody or control rabbit pre-immune serum at the same dilution overnight. The antibody signal was visualized using the Vectastain ABC kit (Vector, PK6101). Negative controls (no incubation with primary antibody) were included for each experiment. Anti-IL10 (Abcam, ab33471) was used at a dilution of 1:100; anti-CD34 (Abcam, ab81289) was used at a dilution of 1:100; pan cytokeratin (DAKO, Z0622) was used at a dilution of 1:50; LC3B (Abcam, ab168831) was used at a dilution of 1:50. Antibody staining was detected using the appropriate species Vector ABC kit (Vector Laboratories, PK6106) followed by the DAB substrate [18]. Image Calibration: Staining was captured using a 4×, 10×, 20× or 40× objective (×40 or ×100 or ×200 or ×400 magnification) from at least four non-overlapping microscopic fields by an AxioCam CCD digital camera (Carl Zeiss). Files were saved in a 12-bit tagged image file format (.tif). To perform area analysis, unmodified, uncompressed. tif image files of staining were accessed in Photoshop as previously described [19]. 2.6.1. Statistical analysis of immunohistochemistry data We conducted quantitative analysis of vasculature using ImageJ software. The observer was blinded to experimental group, and using systematic random sampling, at least three non-overlapping regions within the labyrinthine layer from each placental section (40× magnification images) were analyzed. Control, 25% FR, and 50% placentas (n=8/group) were evaluated and standardized for equal pixel size. For vascular quantification, CD34 positively stained cells were identified and contiguous positively stained cells were marked to create an outline of the vascular area. Region of interest analysis was conducted to calculate the summative area of positively outlined areas per region. Because of the pattern of distribution of positive staining, semi-quantitative analysis of IL10 (n=4-7/group) and trophoblast staining (n=8/group) was conducted by a pathologist blinded to phenotypic information. Staining was quantified by grading the overall positive staining of cells with intensities of 0 to 4 akin to a Likert scale (0=below the level of detection, 1=weak, 2=mild, 3=moderate, 4=strong) [20]. Quantitative analysis of LC3B staining was performed by counting the number of positively- and non-staining spongiotrophoblast cells per region of interest, located in the junctional zone. All cell counts were completed in triplicate on randomly selected fields (taken at 40×) per placental section and average cell counts calculated for each mouse placenta (n=3-4/group). 2.7. Statistical Analyses All statistical analyses were conducted in GraphPad Prism software (version 5, GraphPad Software Inc., La Jolla, CA) and data are presented as means ± SEM. Significant differences between three groups were analyzed using one-way analysis of variance (ANOVA) with Fisher's LSD, unless otherwise indicated (for qRT-PCR and western blotting studies). Post-hoc Tukey's multiple comparison testing was used to determine significant differences in pairwise comparisons. When comparisons were made between two groups only, Student's t-test was performed. All p-values are reported as 2-tailed with statistical significance set at <0.05 for all comparisons. Sample sizes were based upon previously published data evaluating placental gene expression using this murine model of maternal FR resulting in IUGR [7]. Power analyses conducted using gene expression data (by qRT-PCR) from that study, which showed mean relative expression=1.0 with SD=0.1, indicate that 5-8 animals per experimental group was sufficient to detect a 25-35% change in gene expression from the control group. These sample sizes achieve 89% power to detect a difference of at least 25-35% using the Hsu (with Best) multiple comparison test at a 0.05 overall significance level. 3. Results 3.1 Placenta and embryo weight Compared to controls (n=20), 25% FR mice (n=25) did not demonstrate a decrease in placental weight. However, 50% FR mice (n=21) demonstrated a significant (p<0.01) decrease in placental weight compared to control and 25% FR mice (control: 0.08±0.002g versus 25% FR: 0.08±0.002g versus 50% FR: 0.06±0.002g). The 50% FR mice displayed a 29% decrease in placental weight compared to the control and a 31% reduction compared to the 25% FR mice. Both the 25% FR mice and 50% FR mice showed a significant (p<0.01) decrease in embryo weight compared to controls (25% FR: 42% decrease; 50% FR: 48% decrease) (control: 1.16±0.02g versus 25% FR: 0.67±0.01g versus 50% FR: 0.61±0.03g). There was no significant difference in embryo weight between 25% and 50% FR mice. 3.2 RNA sequencing pathway analysis in placental tissue There were almost 700 genes that were differentially expressed between 50% FR and control groups, as previously reported [7]. Approximately 15% of these genes have unknown clinical significance in human physiology or disease. Of the remainder, approximately 7% of the genes that had differential RNA expression in the 50% FR group compared to controls have been associated with immunologic function, vascular and endothelial maintenance, inflammation, autophagy, or ER stress (Table 2). 3.3 qRT-PCR In order to extend the results found by RNA sequencing to the 25% FR group, we utilized qRT-PCR to determine gene expression in all groups, with a focus on genes involved in autophagy, ER stress and inflammation in human disease. We validated differences in expression between control, 25% FR and 50% FR groups (n=6/group) for three genes involved in autophagy, Dram1 (control: 1.009±0.059 versus 25% FR: 0.816±0.094 versus 50% FR: 0.301±0.062; p<0.001), Fbxo32 (control: 1.010±0.062 versus 25% FR: 0.708±0.046 versus 50% FR: 0.305±0.046; p<0.001), and Scd1 (control: 1.012±0.067 versus 25%FR: 0.851÷0.114 versus 50% FR: 0.581±0.048; p<0.006). Three genes which are known to be functionally important in the endoplasmic reticulum (ER) stress response, Pdia4, Creld2, and Derl3 were also significantly differentially expressed between groups (control: 1.014±0.075 versus 25% FR: 0.955±0.038 versus 50% FR: 0.493±0.038; p<0.0001), (control: 1.009±0.057 versus 25% FR: 0.885±0.054 versus 50% FR: 0.738±0.089; p<0.04), (control: 1.007±0.073 versus 25% FR: 0.746±0.078 versus 50% FR: 0.479±0.091; p<0.001), respectively. Lastly, two genes involved in vascular inflammation were found to be differentially expressed, Map3k6 (control: 1.030±0.084 versus 25% FR: 1.697±0.120 versus 50% FR: 2.000±0.305; p<0.02) and C1qa (control: 1.022±0.088 versus 25% FR: 0.804±0.096 versus 50% FR: 0.459±0.059; p<0.0008) (Fig 2). In addition, given the key anti-inflammatory role of IL10 in pregnancy, we quantitated its expression and found that Il10 levels in whole placenta decreased in FR compared to control groups (control: 1.113±0.214 versus 25% FR: 0.328±0.058 versus 50% FR: 0.320±0.072; p<0.001). 3.4 Western immunoblotting Our results above suggested that FR results in alterations to autophagy and ER stress pathways. To translate these differences, BAX, BCL2, and BIP protein expression were assessed by Western blot analysis. There was no significant difference in BAX (p=0.19) or BCL2 (p=0.46) protein expression among groups (Fig 3A,3B). However, BIP expression was down regulated in FR groups (control: 97.62±5.94 versus 25% FR: 77.46±4.71 versus 50% FR: 69.21±4.60; p=0.0016) (Fig 3C). As a marker of generalized ER stress, this observation is consistent with our qPCR findings of decreasing ER stress with increasing severity of food restriction. 3.5 Placental histopathology Histopathology was undertaken to evaluate for differences in placental structures among groups. No appreciable differences were seen between groups in the overall architectural organization of the placenta as visualized by H&E staining (Fig 4). IL10 and LC3B immunostaining was performed to evaluate for protein-level expression of inflammatory markers and bulk autophagy. Positive staining for IL10 was seen predominantly within the junctional zone of the placenta (Fig 5A), and overall positive staining for IL10 was decreased with increasing degrees of food restriction (control: 3.60±0.24 versus 25%FR: 1.75±0.25 versus 50% FR: 1.25±0.25; p<0.0001) (Fig 5B). Similarly, positive staining for LC3B was seen predominantly within the junctional zone, and detected in mostly spongiotrophoblast cells, though occasionally seen in giant trophoblast cells (Fig 5C). Overall numbers of spongiotrophoblasts and giant cell trophoblasts (determined by morphology) were not different between groups (p=0.38 and p=0.81, respectively). Average number of LC3B-positively staining spongiotrophoblast cells per section (using region of interest analysis) was significantly different between groups (control: 13.44±2.39 versus 25% FR: 4.89±0.29 versus 50% FR: 5.17±1.18, p=0.0087) (Fig 3D). Evaluation of vascular and trophoblast structures was specifically undertaken, as these structures are essential for nutrient provision from the mother to the fetus. Within the labyrinthine area, FR mice showed a graded decrease in vasculature with increasing severity of FR, as analyzed by CD34 immunohistochemical analysis (Fig 6A). When summative area of vasculature per region of interest was compared, there was a significant decrease in both FR groups compared to controls (control: 35264±3530 versus 25% FR: 23224±1623 versus 50% FR: 20156±2193; p<0.01) (Fig 6B). Overall trophoblast integrity was altered in the FR mice, as demonstrated by pan cytokeratin immunohistochemistry staining (Fig 6C). Sections were graded 0 to 4+ on overall positivity of staining, and there was a significant difference between groups (control: 3.125±0.125 versus 25% FR: 2.000±0.267 versus 50% FR: 1.500±0.267; p<0.01) (Fig 6D). Qualitative review revealed degenerative changes of the cytokeratin-positively staining structures with a decreased percentage of positively stained cells overall in FR groups compared to control. 4. Discussion Our animal study demonstrates that maternal nutrient deprivation during pregnancy affects the immunologic milieu in the placenta, specifically affecting both gene and protein expression of IL10 and associated cellular stress responses to nutrient deprivation via autophagy and ER stress. Our findings are consistent with our hypothesis that IL10 expression is decreased in maternal food-restricted placentas. These findings are in line with human data showing that placental insufficiency uniformly seen in various disorders of pregnancy such as recurrent miscarriages, preeclampsia, and/or intrauterine growth restriction may be due to disordered immune tolerance and/or dysregulation of IL10 [11,21-23]. IL10 is a known key anti-inflammatory cytokine in pregnancy, but has been shown in other inflammatory human disease to also regulate autophagic and ER stress responses. In our model, we provide evidence that bulk measures of autophagy and ER stress are suppressed at the junctional zone of the murine placenta in conditions of maternal food restriction. These findings are contradictory to our hypothesis that IL10 reduction leads to increased cellular death via autophagy and ER stress. Instead, these findings suggest that placental cell populations adapt to chronic nutrient restriction by decreasing cellular autophagic and ER stress responses in order to continue to function and survive to provide nutrients and blood flow to the growing fetus. We identified, by RNA sequencing and pathway analysis, novel genes important in these cell homeostasis regulatory pathways that are differentially affected by maternal food restriction. However, the suppression of autophagy, ER stress, and anti-inflammatory signals localized largely to the junctional zone in maternal FR-associated placentas does not adequately compensate for dysregulation of vascular and trophoblast structures within the murine labyrinth. Autophagy is the process by which damaged organelles and protein aggregates are engulfed into autophagosomes, fused with lysosomes and degraded by hydrolases. This process is essential for cell survival and regulation of energy and nutrient homeostasis. Therefore, we had hypothesized that in our model of maternal food restriction and intrauterine growth restriction, markers of autophagy would be increased. However, we found the opposite – that bulk measures of autophagy are decreased. These contradictory findings are in line with the existing debate on whether placental insufficiency leads to induction or suppression of autophagy as some groups have shown that up-regulation of autophagy is associated with fetal growth restriction and preeclampsia [24-26]. It is generally accepted that both in vivo and in vitro, starvation conditions activate autophagy. However, these responses are highly tissue-specific, and modulated by multiple and complex input signals. The placenta is a unique organ that is responsible for sustenance of the growing fetus, often at the expense of maternal well-being, and therefore, may regulate autophagic processes differently in response to nutrient deprivation. Our results in placentas associated with FR overall suggest that bulk autophagy is decreased, especially in the junctional zone of the murine placenta. We believe that this seemingly contradictory finding may reflect the function of certain autophagy genes to inhibit inflammation, as seen in tissue-specific knock-out animal models of autophagy genes resulting in wide spread tissue over-inflammation [27,28]. In fact, in in vivo cancer models, autophagy inhibition has been demonstrated to favor immunosuppression, leading to increased tumor aggressiveness [29]. If this is also true in pregnancy at the maternal-fetal interface, our results suggests that the autophagic suppression seen in the FR-associated placentas may represent a compensatory mechanism to maintain pregnancy in the face of a pro-inflammatory environment. By RNA sequencing, we identified a number of autophagic genes that may be suppressed in IUGR induced by maternal food restriction. For example, Dram1, or DNA-damage regulated autophagy modulator 1, is a gene regulated as part of the p53 tumor suppressor pathway that encodes a lysosomal membrane protein required for the induction of autophagy via this pathway. In our murine FR-associated placentas, Dram1 expression was significantly decreased compared to controls implying suppressed autophagy via this pathway. This is in contrast to reports by Hung et al, who found increased levels of DRAM in human IUGR placentas [24]. Fbxo32, or F box protein 32 (also known as atrogin-1), deficiency in a knock-out animal model has been associated with premature death of cardiomyocytes via impaired autophagy [30]. In our model, FR led to decreased production of Fbxo32, perhaps affecting autophagy-induced cell death. Scd1, stearoyl-CoA desaturase 1, is a key player in fatty acid biosynthesis. Therefore, its reduction in our FR groups may simply represent poor maternal nutrient supply. However, inhibition of Scd1 has been shown to reduce starvation-induced autophagy [31]. Therefore, the reduced levels of Scd1 seen in FR groups may reflect compensation in these placentas to reduce the induction of autophagy by starvation. ER stress is the disruption of protein folding in the endoplasmic reticulum, resulting in the accumulation of misfolded proteins within the ER. This pathologic process results from disturbances such as ER calcium depletion, nutrient deprivation, oxidative stress, DNA damage, or energy perturbation. When ER stress is persistent and excessive, the unfolded protein response (UPR) initiates apoptosis to eliminate stressed cells and also activates NF-κB and the inflammasome. ER stress has been implicated as a contributor to pathology in human intrauterine growth restriction, likely due to deficient placental perfusion inducing oxidative stress [32-35]. Again, our findings are contradictory to our hypothesis that ER stress increases in the nutritionally inadequate environment of IUGR. In our study, overall protein expression of BIP decreases with increasing severity of maternal FR, with no change in apoptotic markers between groups. This finding may again be related to the unique ability of the placenta to immunologically mediate stress responses in order to maintain cellular viability and function to protect and provide for the developing fetus. IL10 has been found to modulate ER stress in other inflammatory disorders [36-37], but its role in placental ER stress has not been defined. By RNA sequencing, we have discovered 3 novel markers of ER stress that are differentially expressed in placentas of food-restricted mothers. We demonstrate decreased levels of Pdia4 (also known as ERP72), a mediator of the unfolded protein response, in FR groups compared to control mice. We also demonstrated decreased levels of Creld2, a stress-inducible gene that may regulate the ER stress response and progression. Mutations in promoter sites of this gene result in decreased basal activity and responsiveness to ER stress stimuli using luciferase reporter analyses [38]. Lastly, Derl3 was found to have decreased expression in FR placentas. Derl3 encodes an important component of ER-associated degradation. In vitro studies have demonstrated that overexpression of this gene enhances ER stress responses and promotes cellular survival, whereas decreased expression decreases ER response but increased cell death in response to ischemia [39]. Taken together, the decreased expression of these 3 genes would result in decreased functional ER stress activity in the face of nutritional deprivation. Lastly, the protective effects of IL10 on vascular function and inflammation have been observed both in pregnancy models and non- pregnancy states [22-23]. IL10 knockout mice have been shown to exhibit impaired spiral artery remodeling and poor placental angiogenesis, resulting in IUGR, when exposed to environmental toxins. Treatment with recombinant IL10 rescues the pregnancy, seemingly via restoration of endovascular activity [40]. Outside of pregnancy, IL10 has also been implicated as a mediator of vascular protection in hypertension, diabetes and atherosclerosis [41-43]. Therefore, one interesting consequence of disordered IL10 seen in adverse pregnancy outcomes is that it may be involved in vascular remodeling seen in placental insufficiency and the developmental programming of cardiovascular disease seen in the offspring of pregnancies complicated by IUGR, preeclampsia, or prematurity. An interesting candidate gene that was greatly increased in FR placentas is Map3k6 (mitogen-activated protein 3 kinase 6), which mediates angiogenic and tumorigenic effects via VEGF expression [44]. Again, this increase may represent an attempt at compensation for poor placental vascularization secondary to nutritional restriction-induced remodeling, as is evident by our immunohistochemical staining for CD34+ endothelial cells. We also demonstrated decreased levels of C1qa, a polypeptide chain of complement subcomponent C1q. Complement C1q-induced activation of β-catenin signaling has recently been shown to contribute to hypertensive arterial remodeling [45]. Contrary to our expectations, C1qa levels were decreased in our FR placentas, again highlighting the inconsistencies of whether cellular protective mechanisms are impaired or enhanced in response to adverse environments in these pathologic conditions of pregnancy. Our histopathologic findings were consistent with the reported placental findings in human IUGR [46-47], where placental hypovascularity and trophoblast degeneration lead to poor blood flow and nutrient transfer to the fetus [48]. However, it should be noted that despite decreased vascularity and damaged trophoblast layers within the labyrinthine area of FR-associated placentas, placental weight was not changed. It may be that in mild FR, placental mass is relatively maintained at the expense of fetal growth, whereas moderate FR crosses a critical threshold, beyond which placental self-preservation is impaired. Taken together, our histopathologic data suggests that alterations in immune tolerance, and placental compensatory responses to protect cellular health via decreasing autophagic and ER stress responses, may allow for functional compensation at mild levels of FR, but are inadequate at more moderate levels of FR. There are several strengths and limitations to our study. The use of animal models to study human disease is often necessary in diseases like IUGR, where detection occurs late in the course of disease making progression of disease difficult to study. However, careful consideration into the specific animal model chosen is essential, as there are both similarities and differences to human placental structure and pregnancy with any animal model [49,50]. Though both mouse and human placentas are hemochorial, meaning fetally-derived trophoblasts are directly bathed in maternal blood [48], there are significant differences in gestation and litter size, which may explain why human pregnancies demonstrate much greater adaptive capacity in protecting fetal weight. These differences present a major limitation of our study, as the extent to which animal data can be applied to human disease remains under debate. We specifically chose this mouse model of late chronic maternal food restriction as it has been shown to result in IUGR, and it has been well-characterized in its effect of maternal health and hormonal status during pregnancy, nutrient transfer to the fetus, and short- and long-term effects on the offspring [7,48,51]. However, as stated previously, architectural differences in mouse and human placenta present challenges in comparing changes specific to the nutrient exchange layers. To address this, we utilized a discovery-based technique of RNA sequencing to identify patterns of placental gene expression changes that may be altered in growth restriction. This allows for a non-biased evaluation of how specific genes involved in immunologic function and cellular maintenance are affected in IUGR, and identification of potentially novel players in placental insufficiency. Using whole placenta, we found 700 genes that were significantly differentially expressed between groups, and performed pathway analysis, which demonstrated that the pathways of interest were well represented among the differentially expressed genes. However, another limitation of our study was that adjusted p-values from RNA sequencing analysis were nonsignificant in some of our genes of interest involved in immunologic pathways. This limitation may have occurred because of the smaller sample sizes used for RNA sequencing. This sample size was chosen as genome-wide sequencing techniques applied to large groups can be cost-prohibitive, and the main purpose of RNA sequencing in this study was to screen for pathway changes. To address this issue on a gene-specific basis, we validated RNA expression of specific genes of interest by qRT-PCR using larger sample sizes across all three experimental groups. In addition, there is still significant debate and remaining unknowns in the field on the crosstalk between autophagy, ER stress and apoptotic pathways. Our study does not establish cause-and-effect, and further mechanistic studies on these pathways in IUGR are warranted to more definitively link decreased IL10 with autophagy and ER stress. In conclusion, our murine model of chronic, mild to moderate late gestational FR of the pregnant mouse represents an animal model of IUGR with reasonable fidelity to the human condition, based on histopathologic similarities. Consistent with our hypothesis, we found that IL10 expression in the murine placenta is decreased in maternal gestational food restriction. Using RNA sequencing techniques, we have demonstrated that a significant portion of gene expression changes seen on a whole placental level involve immunologic pathways and associated cellular maintenance processes. However, contrary to our hypothesis that maternal FR results in increased autophagy and ER stress, leading to cellular death, we found that bulk measures of these pathways are decreased in maternal FR-associated placentas. We believe that these findings suggest that the placenta serves as a unique interface, that must preserve fetal growth in the face of adverse maternal environment, and that these mechanisms are immunologically mediated. As such, placental compensatory responses appear to be distinct and opposite to the inflammatory, autophagic and ER stresses described in in vitro nutrient deprivation. This work was supported by the UCLA Children's Discovery and Innovation Institute [CDI-SGA-01012016 (AC)], the American Heart Association [15BGIA25710060 (AC)], and the National Institutes of Health [5K12HD034610, HD 081206, HD 41230 (SUD)]. Abbreviations BAX Bcl-2 associated X protein BCL2 B-cell lymphoma BIP Binding immunoglobulin protein ER endoplasmic reticulum GD gestational day FR food restriction IL10 interleukin 10 IUGR intrauterine growth restriction LC3B microtubule-associated protein 1 light chain 3 beta Figure 1 Scheme representing crosstalk between IL10, autophagy, ER stress, and apoptotic pathways. Figure 2 qRT-PCR validation of placental RNA expression of genes involved in vascular inflammation, autophagy and ER stress in mice exposed to maternal FR For each candidate gene, average mRNA expression level from control, 25% FR, and 50% FR groups (n=6/group, based on power calculations for detect a 25% difference in mean expression between groups to achieve 89% power at a 0.05 overall significance level) was calculated by qRT-PCR. Data are represented in graphs as means ± SEM. Asterisks indicate a significant difference between indicated FR group compared to control (p<0.05) by post-hoc Tukey's multiple comparison testing. Reported p-value for each gene is by one-way ANOVA between all three groups. Figure 3 Western immunoblotting of apoptosis and ER stress markers in the placentas of gestational FR-exposed mice Representative blots of BAX (A), BCL2 (B) proteins as markers of apoptosis, and BIP (C) as a marker of ER stress are shown. There were no significant differences between groups for BAX and BCL2 expression. BIP expression however, was decreased in both the 25% and 50% FR groups, compared to controls. Data are represented in graphs as means ± SEM. Asterisk indicates p<0.05 by ANOVA. C=control, 25%=25% FR group, 50%=50% FR group (n=9/group). Figure 4 Hematoxylin & eosin staining of control, mild and moderate FR mouse placentas at 4× magnification Representative murine placental sections stained by H&E from control, mild- and moderate-FR exposed mice (n=8/group). La: labyrinth; JZ: junctional zone; D: decidua. Scale bars represent 500μm. Figure 5 IL10 and LC3B staining in murine placenta after maternal FR (A) Representative sections from each group (control, 25% FR, 50% FR) stained for IL10 demonstrate decreasing positive staining for IL10 (brown; 20× magnification) with increasing severity of maternal FR. Arrows indicate positive staining. Scale bars represent 100 μm. D: decidua; JZ: junctional zone; La: labyrinth. (B) Decreasing histologic grading scores of overall positively of IL10 staining in FR groups compared to controls (n=4-7/group). Data are represented in graphs as means ± SEM. Asterisks indicate p<0.0001 between the indicated group compared to the control group, by post-hoc Tukey's multiple comparison testing. Listed p-value represents significance between groups, by ANOVA. (C) Representative sections from each group (control, 25% FR, 50% FR) stained for LC3B (brown; 20× magnification) demonstrating decreasing numbers of positively-stained spongiotrophoblasts in the junctional zone of FR mouse placenta with increasing severity of FR. Arrows indicate positive staining of spongiotrophoblasts. Scale bars represent 100μm. (D) Decreased average numbers of positively staining spongiotrophoblasts for LC3B per region of interest in FR groups compared to controls (n=3-4/group). Data are represented in graphs as means ± SEM. Asterisk indicates p<0.05 between all groups, by ANOVA. Fig 6 Vascular and trophoblast changes in the murine placenta with maternal FR (A) Representative sections from each group (control, 25% FR, 50% FR) stained for CD34 (brown; 20× magnification) in endothelial cells within the labyrinthine region (20×). Arrows indicate blood vessels. Scale bars represent 100μm. (B) Quantitative analysis of sum of vascular area per region of interest at 40×. Each placental section from all three groups (n=8/group) was analyzed in triplicate. Vascular areas were defined as outlined by contiguous positively stained cells. Data are represented in graphs as means ± SEM. Asterisks indicate p<0.05 (between indicated FR group and control), by post-hoc Tukey's multiple comparison testing. Listed p-value indicates significance between all groups, by ANOVA. (C) Representative placental sections from each group (control, 25% FR, 50% FR) stained for trophoblast cells using cytokeratin (brown, 20× magnification). Qualitatively, degenerative changes (dilated maternal spaces, with a decreased percentage of positive stained cells in the 25% and 50% FR groups compared to controls) are seen in murine trophoblast layers within the labyrinth. Arrows indicate trophoblast lining. Scale bars represent 100μm. (D) Decreased histologic grading scores of overall positivity of cytokeratin stain in FR groups compared to controls (n=8/group). Data are represented in graph as means ± SEM. Asterisks indicate p<0.01 between indicated FR group vs. control, by post-hoc Tukey's multiple comparison testing. Listed p-value indicates significance between all groups, by ANOVA. Table 1 Primer and probe sequences and conditions used for qRT-PCR Gene Annealing temperature Forward primer Reverse primer Probe Dram1 57°C 5′-ACACAGGAACAAC TCCTCCA 5′-AACGGGAGTGCT GAAGTAGC 5′-TCTCTGCATTTCT TGGCGCAGC Fbxo32 57°C 5′-TTCTCAGAGAGGCA GATTCG 5′-GAGAATGTGGCA GTGTTTGC 5′-CCAATCCAGCTGC CCTTTGTCA Scd1 59°C 5′-TTCTTCTCTCACGT GGGTTG 5′-CGGGCTTGTAGTA CCTCCTC 5′-CGCAAACACCCG GCTGTCAA Pdia4 57°C 5′-GGATGCTGCTAACA ACCTGA 5′-CCAGGGAGACTTT CAGGAAC 5′-CAAGTTTCACCAC ACTTTCAGCCCTG Creld2 57°C 5′-TGTGTGGATGTGGA TGAGTG 5′-AGCCGTTGACATT CTCACAG 5′-CATCTCCGTGCAG CGATGGC Derl3 57°C 5′-CTCTTCGTGTTCCG CTACTG 5′-AGGAATCCCAGC AGAGTCAT 5′-AACCCTCCTCCAG CATGCGG Map3k6 56°C 5′-TACAACGCGGATGT AGTGGT 5′-AACAGAGGAGCA CGTTGTTG 5′-TTCTACCACCTCG GCGTGCG C1qa 57°C 5′-GAGCATCCAGTTTG ATCGG 5′-CATCCCTGAGAG GTCTCCAT 5′-ACCACGGAGGCA GGGACACC Table 2 Genes differentially expressed between control and 50% FR groups by RNA sequencing that are involved in vascular, immunological and inflammatory pathways Five placentas from control and five placentas from 50% FR mice were processed for RNA sequencing. This table represents the candidate genes identified as differentially expressed between control and 50% FR groups, by unadjusted p-value<0.05, involved in immunologic, inflammatory and vascular pathways. For each gene listed, average expression level for the control and 50% FR group is reported, as well as log2 fold change, p-value based upon Student's t-test and adjusted p-value using Bonferroni correction between groups. Genes are grouped by functional roles in innate immunity, vascular and endothelial function, inflammation, autophagy and ER stress. Gene Average Expression Level (control) Average Expression Level (50% FR) Log2 Fold Change P-value P-value (adj) Innate Immunity IL21r (interleukin 21 receptor) 1.056 0.346 -1.6 7.73E-08 0.001 IL23r (interleukin 23 receptor) 0.366 0.078 -2.2 9.11E-06 0.042 Marco (macrophage receptor with collagenous structure) 0.019 0.856 5.5 0.0002 0.187 Cebpa (CCAAT/enhancer binding protein, alpha) 6.306 9.708 0.62 0.008 1 Sfrp1 (secreted frizzled-related protein 1) 1.199 0.844 -0.5 0.015 1 Syt11 (synaptotagmin XI) 1.481 1.080 -0.45 0.015 1 Siglec1 (sialic acid binding Ig-like lectin 1, sialoadhesin) 0.660 1.132 0.78 0.033 1 Pyhin1 (pyrin and HIN domain family, member 1) 0.300 0.564 0.91 0.034 1 Isg15 (ISG15 ubiquitin-like modifier) 20.849 34.689 0.73 0.045 1 Vascular and endothelial maintenance Tnfsf15 (tumor necrosis factor superfamily, member 15) 0.072 0.019 -1.9 4.03E-05 0.075 Gm52 (syncytin a) 17.555 32.574 0.89 0.0007 0.339 Wls (wntless homolog) 14.802 9.686 -0.61 0.0007 0.339 Ncf1 (neutrophil cytosolic factor 1) 0.322 0.598 0.89 0.001 0.412 Bmper (BMP-binding endothelial regulator) 6.383 11.646 0.87 0.004 0.737 Map3k6 (mitogen activated protein kinase 6) 1.823 2.862 0.65 0.006 0.981 Lyve1 (lymphatic vessel endothelial hyaluronic receptor 1) 11.950 23.472 0.97 0.012 1 Cldn5 (claudin 5) 7.497 11.338 0.60 0.013 1 Ntsr1 (neurotensin receptor 1) 0.349 0.067 -2.4 0.016 1 C5ar1 (complement component 5a receptor 1) 0.257 0.535 1.1 0.021 1 Kcnk6 (potassium inwardly-rectifying channel, K6) 3.905 2.975 -0.39 0.029 1 Lrg1 (leucine-rich alpha-2 glycoprotein 1) 2.044 3.732 0.87 0.035 1 Ucp2 (uncoupling protein 2, mitochondrial, proton carrier) 11.918 16.672 0.48 0.035 1 Trpv6 (transient receptor potential cation channel, V6) 9.530 5.234 -0.86 0.049 1 Gclm (glutamate-cysteine ligase, modifier subunit) 6.098 8.718 0.52 0.049 1 Inflammation Mmp8 (matrix metallopeptidase 8) 0.198 0.561 1.5 2.77E-05 0.070 Sfxn5 (sideroflexin 5) 0.495 0.325 -0.61 0.008 1 Slfn4 (schlafen 4) 0.433 0.736 0.77 0.010 1 IL18rap (interleukin 18 receptor accessory protein) 0.233 0.144 -0.70 0.011 1 Tnf (tumor necrosis factor) 0.338 0.176 -0.94 0.012 1 Bok (Bcl2-related ovarian killer protein) 51.091 38.383 -0.41 0.025 1 Fpr2 (formyl peptide receptor 2) 0.149 0.354 1.25 0.027 1 Plxnc1 (plexin C1) 0.142 0.238 0.74 0.032 1 Blnk (B-cell linker) 0.485 0.742 0.61 0.043 1 C1qa (complement component 1, q subcomponent, alpha polypeptide) 5.333 10.883 1.03 0.049 1 Autophagy Fbxo32 (f box protein 32) 23.176 13.730 -0.76 0.005 0.808 Scd1 (stearoyl-coenzyme A desaturase 1) 5.818 4.339 -0.42 0.017 1 Dram1 (DNA damage regulated autophagy modulator 1) 10.696 7.794 -0.46 0.009 1 ER stress Creld2 (cysteine-rich with EGF like domains 2) 42.240 27.444 -0.62 0.003 0.724 Derl3 (Der1-like domain family, member 3) 23.324 14.517 -0.68 0.019 1 Pdia4 (protein disulfide isomerase associated 4) 82.834 60.377 -0.46 0.010 1 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. 1 Cross JC Hemberger M Lu Y Nozaki T Whitely K Masutani M Adamson SL Trophoblast functions, angiogenesis and remodeling of the maternal vasculature in the placenta Mol Cell Endocrinol 2002 187 1-2 207 12 11988329 2 Herr F Baal N Widmer-Teske R McKinnon T Zygmunt M How to study placental vascular development? Theriogenology 2010 73 6 817 27 20036417 3 Kanaka-Gantenbein C Fetal origins of adult diabetes Ann N Y Acad Sci 2010 1205 99 105 20840260 4 Thornburg KL O'Tierney PF Louey S Review: the placenta is a programming agent for cardiovascular disease Placenta 2010 31 Suppl S54 9 20149453 5 Ligi I Grandvuillemin I Andres V Dignat-George F Simeoni U Low birth weight infants and the developmental programming of hypertension: a focus on vascular factors Semin Perinatol 2010 34 3 188 192 20494734 6 Luyckx VA Bertram JF Brenner BM Fall C Hoy WE Ozanne SE Vikse BE Effect of fetal and child health on kidney developmental and long-term risk of hypertension and kidney disease Lancet 2013 382 273 83 23727166 7 Chen PY Ganguly A Rubbi L Orozco LD Morselli M Ashraf D Jaroszewicz A Feng S Jacobsen SE Nakano A Devaskar SU Pellegrini M Intrauterine calorie restriction affects placental DNA methylation and gene expression Physiol Genomics 2013 45 14 565 76 23695884 8 Gabory A Ferry L Fajardy I Jouneau L Gothié JD Vigé A Fleur C Mayeur S Gallou-Kabani C Gross MS Attig L Vambergue A Maternal diets trigger sex-specific divergent trajectories of gene expression and epigenetic systems in mouse placenta PLoS One 2012 7 11 e47986 23144842 9 Mao J Zhang X Sieli PT Falduto MT Torres KE Rosenfeld CS Contrasting effects of different maternal diets on sexually dimorphic gene expression in the murine placenta Proc Natl Acad Sci U S A 2010 107 12 5557 62 20212133 10 Biswas S Placental changes in idiopathic intrauterine growth restriction OA Anatomy 2013 1 2 11 11 Cheng SB Sharma S Interleukin-10: a pleitropic regulator in pregnancy Am J Reprod Immunol 2015 73 6 487 500 25269386 12 Thamotharan S Stout D Shin BC Devaskar SU Temporal and spatial distribution of murine placental and brain GLUT3-luciferase transgene as a readout of in vivo transcription Am J Physiol Endocrinol Metab 2013 304 3 E254 66 23193055 13 Trapnell C Pachter L Salzbery SL TopHat: discovering splice junctions with RNA-Seq Bioinformatics 2009 25 1105 11 19289445 14 Anders S Huber W Differential expression analysis for sequence count data Genome Biol 2010 11 10 R106 20979621 15 Livak KJ Schmittgen TD Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method Methods 2001 25 4 402 8 11846609 16 Ganguly A McKnight RA Raychaudhuri S Shin BC Ma Z Moley K Devaskar SU Glucose transporter isoform-3 mutations cause early pregnancy loss and fetal growth restriction Am J Physiol Endocrinol Metab 2007 292 5 E1241 55 17213475 17 Sankar R Thamotharan S Shin D Moley KH Devaskar SU Insulin-responsive glucose transporters-GLUT8 and GLUT4 are expressed in the developing mammalian brain Brain Res Mol Brain Res 2002 107 2 157 65 12425944 18 Wadehra M Natarajan S Seligson DB Williams CJ Hummer AJ Hedvat C Expression of epithelial membrane protein-2 is associated with endometrial adenocarcinoma of unfavorable outcome Cancer 2006 107 90 8 16736513 19 Agley CC Velloso CP Lazarus NR Harridge SD An image analysis method for the precise selection and quantitation of fluorescently labeled cellular constituents: application to the measurement of human muscle cells in culture J Histochem Cytochem 2012 60 6 428 38 22511600 20 Klopfleisch R Multiparametric and semiquantitative scoring systems for the evaluation of mouse model histopathology – a systematic review BMC Vet Res 2013 9 123 23800279 21 Tinsley JH South S Chiasson VL Mitchell BM Interleukin-10 reduces inflammation, endothelial dysfunction, and blood pressure in hypertensive pregnant rats Am J Physiol Regul Integr Comp Physiol 2010 298 3 R713 9 20053959 22 Chatterjee P Chiasson VL Kopriva SE Young KJ Chatterjee V Jones KA Mitchell BM Interleukin 10 deficiecy exacerbates toll-like receptor 3-induced preeclampsia-like symptoms in mice Hypertension 2011 58 3 489 96 21768525 23 Chatterjee P Chiasson VL Seerangan G Tobin RP Kopriva SE Newell-Rogers MK Mitchell BM Cotreatment with interluekin 4 and interleukin 10 modulates immune cells and prevents hypertension in pregnant mice Am J Hypertens 2015 28 1 135 42 24906486 24 Hung TH Chen SF Lo LM Li MJ Yeh YL Hsieh TT Increased autophagy in placentas of intrauterine growth-restricted pregnancies PLoS One 2012 7 7 e40957 22815878 25 Oh SY Choi SJ Kim KH Cho EY Kim JH Roh CR Autophagy-related proteins, LC3 and Beclin-1, in placentas from pregnancies complicated by preeclampsia Reprod Sci 2008 15 9 912 20 19050324 26 Saito S Nakashima A A review of the mechanism for poor placentation in early-onset preeclampsia: the role of autophagy in trophoblast invasion and vascular remodeling J Reprod Immunol 2014 101-102 80 8 23969229 27 Park S Buck MD Desai C Zhang X Loginicheva E Martinez J Freeman ML Saitoh T Akira S Guan JL He YW Blackman MA Autophagy genes enhance murine gammaherpervirus 68 reactivation from latency by preventing virus-induced systemic inflammation Cell Host Microbe 2016 19 1 91 101 26764599 28 Lu Q Yokoyama CC Williams JW Baldrge MT Jin X DesRochers B Bricker T Wilen CB Bagaitkar J Logincheva E Sergushichev A Kreamalmeyer D Homeostatic control of innate lung inflammation by vici syndrome gene Epg5 and additional autophagy genes promotes influenza pathogenesis Cell Host Microbe 2016 19 1 102 13 26764600 29 Ladoire S Enot D Senovilla L Ghiringhelli F Poirier-Colame V Chaba K Semeraro M Chaix M Penault-Liorca F Arnould L Poillot ML Arveux P The presence of Lc3b puncta and hmgb1 expression in malignant cells correlate with the immune infiltrate in breast cancer Autophagy 2016 epub ahead of print 30 Zaglia T Milan G Ruhs A Franzoso M Bertaggia E Pianca N Carpi A Carullo P Pesce P Sacerdoti D Sarais C Catalucci D Atrogin-1 deficiency promotes cardiomyopathy and premature death via impaired autophagy J Clin Invest 2014 124 6 2410 24 24789905 31 Ogasawara Y Itakura E Kono N Mizushima N Arai H Nara A Mizukami T Yamamoto A Stearoyl-CoA desaturase 1 activity is required for autophagosome formation J Biol Chem 2014 289 34 23938 50 25023287 32 Burton GJ Yung HW Cindrova-Davies T Charnock-Jones DS Placental endoplasmic reticulum stress and oxidative stress in the pathophysioloy of unexplained intrauterine growth restriction and early onset preeclampsia Placenta 2009 30 Suppl A S43 8 19081132 33 Kawakami T Yoshimi M Kadota Y Inoue M Sato M Suzuki S Prolonged endoplasmic reticulum stress alters placental morphology and causes low birth weight Toxicol Appl Pharmacol 2014 275 2 134 44 24370435 34 Lian IA Løset M Mundal SB Fenstad MH Johnson MP Eide IP Bjørge L Freed KA Moses EK Austgulen R Increased endoplasmic reticulum stress in decidual tissues from pregnancies complicated by fetal growth restriction with and without pre-eclampsia Placenta 2011 32 11 823 9 21907405 35 Yung HW Calabrese S Hynx D Hemmings BA Cetin I Charnock-Jones DS Burton GJ Evidence of placental translation inhibition and endoplasmic reticulum stress in the etiology of human intrauterine growth restriction Am J Pathol 2008 173 2 451 62 18583310 36 Hasnain SZ Tauro S Das I Tong H Chen AC Jeffery PL McDonald V Florin TH McGuckin MA IL-10 promotes production of intestinal mucus by suppressing protein misfolding and endoplasmic reticulum stress in goblet cells Gastroenterology 2013 144 2 357 368 23123183 37 Shkoda A Ruiz PA Daniel H Kim SC Rogler G Sartor RB Haller D Interleukin-10 blocked endoplasmic reticulum stress in intestinal epithelial cells: impact on chronic inflammation Gastroenterology 2007 132 1 190 207 17241871 38 Oh-hashi K Koga H Ikeda S Shimada K Hirata Y Kiuchi K CRELD2 is a novel endoplasmic reticulum stress-inducible gene Biochem Biophys Res Commun 2009 387 3 504 10 19615339 39 Belmont PG Chen WJ San Pedro MN Thuerauf DJ Gellings Lowe N Gude N Hilton B Wolkowicz R Sussman MA Glembotski CC Roles for endoplasmic reticulum-associated degradation and the novel endoplasmic reticulum stress response gene Derlin-3 in the ischemic heart Circ Res 2010 106 2 307 16 19940266 40 Tewari N Kalkunte S Murray DW Sharma S The water channel aquaporin 1 is a novel molecular target of polychlorinated biphenyls for in utero anomalies J Biol Chem 2009 284 22 15224 32 19332547 41 Gunnett CA Heistad DD Berg DJ Faraci FM IL-10 deficiency increases superoxide and endothelial dysfunction during inflammation Am J Physiol Heart Circ Physiol 2000 279 4 H1555 62 11009441 42 Didion SP Kinzenbaw DA Schrader LI Chu Y Faraci FM Endogenous interleukin-10 inhibits angiotensin II-induced vascular dysfunction Hypertension 2009 54 3 619 24 19620507 43 Kinzenbaw DA Chu Y Peña Silva RA Didion SP Faraci FM Interleukin-10 protects against aging-induced endothelial dysfunction Physiol Rep 2013 1 6 e00149 24400151 44 Eto N Miyagishi M Inagi R Fujita T Nangaku M Mitogen-activated protein 3 kinase 6 mediates angiogenic and tumorigenic effects via vascular endothelial growth factor expression Am J Pathol 2009 174 4 1553 63 19246638 45 Sumida T Naito AT Nomura S Nakagawa A Higo T Hasimoto A Okada K Sakai T Ito M Yamaguchi T Oka T Akazawa H Complement C1q-induced activation of β-catenin signalling causes hypertensive arterial remodelling Nat Commun 2015 6 6241 25716000 46 Biswas S Adhikari A Chattopadhyay JC Ghosh SK Histological changes of placentas associated with intra-uterine growth restriction of fetuses: a case control study Nepal Med Coll J 2012 14 1 18 24 23441489 47 Al-Sahan N Grynspan D von Dadelszen P Gruslin A Maternal floor infarction: management of an underrecognized pathology J Obstet Gynaecol Res 2014 40 1 293 6 24102864 48 Coan PM Vaughan OR Sekita Y Finn SL Burton GJ Constancia M Fowden AL Adaptations in placental phenotype support fetal growth during undernutrition of pregnant mice J Physiol 2010 588 Pt 3 527 38 19948659 49 Dilworth MR Sibley CP Review: transport across the placenta of mice and women Placenta 2013 34 Suppl S34 9 23153501 50 Clark DA The use and misuse of animal analog models of human pregnancy disorders J Reprod Immunol 2014 103 1 8 24725995 51 Ganguly A Collis L Devaskar SU Placental glucose and amino acid transport in calorie-restricted wild-type and Glut3 null heterozygous mice Endocrinology 2012 153 8 3995 4007 22700768

PMC005xxxxxx/PMC5119927.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7702118 4161 Immunol Rev Immunol. Rev. Immunological reviews 0105-2896 1600-065X 27782328 5119927 10.1111/imr.12467 NIHMS802102 Article More than complementing Tolls: Complement–Toll-like receptor synergy and crosstalk in innate immunity and inflammation Hajishengallis George 1 Lambris John D. 2 1 Department of Microbiology, University of Pennsylvania School of Dental Medicine, Philadelphia, PA 19104, USA 2 University of Pennsylvania, Perelman School of Medicine, Department of Pathology and Laboratory Medicine, Philadelphia, PA 19104, USA Correspondence: Dr. George Hajishengallis, University of Pennsylvania, School of Dental Medicine, 240 South 40th Street, Philadelphia, PA 19104-6030, USA; Tel.: 215-898-2091, Fax: 215-898-8385; geoh@upenn.edu 14 7 2016 11 2016 01 11 2017 274 1 233244 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Summary Complement and Toll-like receptors (TLRs) play key roles in the host immune response and are swiftly activated by infection or other types of immunological stress. This review focuses on the capacity of complement and TLRs to engage in signaling crosstalk, ostensibly to coordinate immune and inflammatory responses through synergistic or antagonistic (regulatory) interactions. However, over-activation or dysregulation of either system may lead – often synergistically – to exaggerated inflammation and host tissue injury. Intriguingly, moreover, certain pathogens can manipulate complement-TLR crosstalk pathways in ways that undermine host immunity and favor their persistence. In the setting of polymicrobial inflammatory disease, subversion of complement-TLR crosstalk by keystone pathogens can promote dysbiosis. Knowledge of the molecular mechanisms underlying complement-TLR crosstalk pathways can, therefore, be used productively for tailored therapeutic approaches, such as, to enhance host immunity, mitigate destructive inflammation, or counteract microbial subversion of the host response. Complement TLR crosstalk inflammation immune evasion Introduction In co-evolving with the microbial world, mammalian innate immunity has developed effective sentinel mechanisms to promptly detect and respond to infections. Sentinel cells (e.g., neutrophils, macrophages, and dendritic cells) sense invading pathogens through pattern- recognition receptors (PRRs) and alert downstream innate and/or adaptive mechanisms aiming to eradicate or control the infection (1). A major PRR family is represented by the Toll-like receptors (TLRs), each member of which senses distinct types of conserved microbial structures (‘microbe-associated molecular patterns’; MAMPs), thus endowing the innate response with a degree of specificity (e.g., TLR2 responds to lipoteichoic acid, TLR3 to viral double-stranded RNA, TLR4 to lipopolysaccharide [LPS], TLR5 to flagellin and TLR9 to bacterial CpG DNA) (2, 3). The broad but distinct specificities of the TLRs as well as their ability to form heterotypic multi-receptor complexes and engage distinct intracellular signaling molecules further diversifies their recognition and signaling capacities (4, 5). These attributes of the TLRs (and other PRRs) enable the host to detect almost any type of infection, discriminate between different classes of microbes and hence mount a context-relevant immune response. In addition to sentinel cells, innate immunity also has a humoral arm that includes a heterogeneous group of pattern-recognition molecules (PRMs), such as, collectins (e.g., mannose-binding lectin; MBL), ficolins, pentraxins and the complement component C1q (6, 7). Soluble PRMs can be released either locally by stimulated inflammatory cells or systemically following their production in liver. Although structurally heterogeneous, these molecules share evolutionarily conserved functions, such as microbial opsonization as well as activation and regulation of the complement system (8). Historically established as a cascade of antimicrobial proteins in the blood, complement is now appreciated as a network of interacting fluid-phase and cell surface-associated molecules (PRMs, convertases and other proteases, regulators, and signaling receptors) that trigger, amplify, and regulate immunity and inflammation (9). The complement cascade is triggered by distinct mechanisms (classical, lectin, or alternative) that converge at the third component (C3) and lead to the generation of effectors with diverse functions (e.g., recruitment and activation of inflammatory cells via the C3a and C5a anaphylatoxins that activate specific G-protein-coupled receptors; microbial opsonization through C3b; and direct lysis of susceptible targeted microbes by means of the C5b-9 membrane attack complex) (9). During an infection, complement and TLRs are rapidly activated to provide critical frontline defense and act as key mediators between innate and adaptive immunity (10). Interestingly, several microbial products, including LPS (TLR4 agonist), zymosan (TLR2/6 agonist) and CpG DNA (TLR9 agonist), can activate complement in addition to initiating TLR signaling (11, 12). Therefore, an appropriately coordinated host immune response would necessitate signaling crosstalk between TLR and complement pathways, leading to synergistic or antagonistic interactions. Synergistic pathways can enhance the sensitivity of detection, since even individually weak stimuli can potentially combine to elicit a robust immune response. Conversely, antagonistic pathways can augment the specificity of the host response by controlling it and preventing bystander tissue damage (13). Typical examples for these contrasting functions include the cooperation between TLR2 and the C-type lectin dectin-1 for effective anti-fungal immunity (14) and the homeostatic suppression of TLR-induced pro-inflammatory responses by adenosine receptors (15, 16). This review summarizes recent literature on the biological importance of complement–TLR crosstalk pathways. Such pathways lead to diverse effects raging from reinforcement of innate immunity to exacerbation of pathologic inflammation or, conversely, regulation of unwarranted inflammation, depending on the receptors involved and the cellular context. Moreover, mechanisms that allow the interplay between complement and TLRs can be potentially exploited by certain pathogens to modulate the host response in ways that favor pathogen survival and persistence. Regulation of immune and inflammatory responses by complement-TLR cooperation As alluded to above, complement and TLRs are swiftly co-activated in response to microbial infection, while common microbial molecules (such as LPS and CpG DNA) can act as both TLR ligands and complement activators (9). At the cellular level, signaling crosstalk interactions between complement and TLRs have been shown in several cell types, including monocytes, macrophages, neutrophils, and dendritic cells (17–22). In vivo, the early innate immune response is shaped, to a large extent, by bidirectional crosstalk between the two systems (10). In perhaps the first in vivo systematic study to dissect complement-TLR crosstalk pathways, the authors employed systemic administration of different TLR ligands to mice lacking decay-accelerating factor (DAF), a major membrane-associated complement inhibitor. Specifically, LPS (TLR4), zymosan (TLR2/6), and CpG oligodeoxynucleotide (TLR9) all induced significantly higher tumor necrosis factor (TNF), interleukin-1β (IL-1β), and IL-6 responses compared to the same ligands given to wild-type mice (12). Similarly, mice systemically co-treated with TLR ligands and cobra venom factor, a potent complement activator, elicited remarkably high plasma levels of proinflammatory cytokines, further supporting that complement can amplify inflammation in co-operation with TLR signaling (12). Further work revealed a critical involvement of the anaphylatoxin receptors (C3aR and C5aR1 [CD88]) in the complement-TLR synergism for enhanced production of pro-inflammatory and antimicrobial mediators (12, 23). The signaling pathways involved in complement-TLR crosstalk converge at the level of mitogen-activated protein kinases (MAPK), specifically extracellular signal-regulated kinase-1 (ERK1), ERK2 and JUN N-terminal kinase (JNK), which activate the transcriptional factors nuclear factor-κB (NF-κB) and activator protein-1 (AP-1) (12) (Figure 1). Although this synergy could potentially enhance innate immune defenses against infection, it may also contribute to inflammatory pathology. For instance, complement-TLR synergy may actually account for earlier observations that the inhibition of C5a signaling protects against sepsis induced by high-dose LPS or by cecal ligation and puncture (CLP) peritonitis (24). Moreover, the synergistic complement–TLR interaction seen in DAF-deficient mice might explain, at least in part, why DAF-deficient mice are particularly susceptible to inflammatory and autoimmune diseases (25). Complement-TLR crosstalk synergy has also been demonstrated at mucosal sites. Indeed, in the murine gingival tissue, the concomitant activation of C5aR and TLR2 by local co-injection of specific agonists (C5a and the TLR2 ligand Pam3Cys) induced significantly higher levels of TNF, IL-1β, IL-6, and IL-17A mRNA and protein than activation of each receptor alone (26). In fact, destructive periodontal inflammation appears to depend on synergy between C5aR1 and TLR2, since mice deficient in either C5aR1 or TLR2 are essentially resistant against inflammatory bone loss in the periodontium (27, 28). Consistently, treatment of mice subjected to experimental periodontitis with PMX-53, a C5aR1 antagonist, inhibits periodontal inflammation (TNF, IL-1β, IL-6, and IL-17) and bone loss, regardless of the presence of TLR2 (i.e., inflammatory bone loss can be effectively inhibited by blocking just one of the two crosstalking receptors) (26). However, in other experimental systems, where interactions might be partially synergistic or additive, combined inhibition of complement and PRRs may be more effective than inhibition of each system alone. For instance, in a human whole-blood model, combined inhibition of complement and CD14 was shown to be more effective in blocking E. coli-induced cytokine responses than single inhibition (29, 30). CD14 lacks a transmembrane signaling domain but acts as a critical co-receptor of TLRs (mostly TLR4 and TLR2) (3), although it might also have TLR-independent effects that contribute to inflammation. Another study in the human whole-blood model focused on interactions between complement and TLR9 signaling induced by CpG oligodeoxynucleotides, which are considered as vaccine adjuvants (11). These investigators showed that complement inhibition at C3 suppresses both DNA-backbone-mediated maturation of antigen-presenting cells (upregulation of CD40 and CD83) and DNA-sequence-specific induction of cytokines. Interestingly, a CpG oligodeoxynucleotide (CpG-2006) could trigger the classical and the alternative pathway of complement, which in turn promoted the cellular uptake of CpG-2006 (11). Therefore, the immunostimulatory function of oligodeoxynucleotides such as CpG-2006, seems to be reliant upon the combined activation of complement and TLR9. Although C5aR1 synergizes with TLR2 for IL-17A induction in experimental mouse periodontitis (26), C5aR1 downregulates IL-17A in endotoxic shock in mice (21). It is uncertain whether this difference can be attributed to the different disease models or the different TLRs involved (TLR2 vs. TLR4). Intriguingly, C5aR1 promotes the induction of another IL-17 isoform (IL-17F) in endotoxic shock (31). In this regard, C5aR1 synergizes with TLR4 for IL-17F production in mouse macrophages via a MyD88- phosphatidylinositol-3 kinase (PI3K)-Akt pathway (31), whereas the same C5aR1-TLR4 crosstalk in the same cell type inhibits IL-17A production (21) (Figure 2). According to this study, the major source of IL-17A during endotoxemia in mice was not the ‘usual suspects’ (CD4+ T cells, γδ T cells, or NK cells) but rather CD11b+F4/80+ macrophages (21). Mechanistically, C5a was shown to activate PI3K-Akt and mitogen-activated protein kinase kinases 1/2 (MEK1/2)-ERK1/2 pathways, resulting in C5aR1 (but not C5aR2)-dependent induction of IL-10, which subsequently inhibits production of IL-17A (as well as IL-23) (21) (Figure 2). Because IL-17F has considerably reduced bioactivity as compared to IL-17A (32), C5a appears to shift the IL-17A – IL-17F balance toward the less bioactive molecule to mitigate excessive inflammation in acute conditions. However, it is currently unclear why C5aR1-induced IL-10 inhibits IL-17A preferentially over IL-17F. The relatively recently discovered C5aR2 (also referred to as C5a-like receptor 2; GPR77) functions as an alternative high-affinity receptor for C5a (33). Owing to its inability to productively couple to G proteins, C5aR2 was originally perceived as a non-signaling decoy receptor that could compete with C5aR1 for C5a binding, thereby mitigating C5a-dependent inflammation (34, 35). Consistently, upon pulmonary immune complex injury, C5aR2-deficient mice display increased lung inflammation (as revealed by elevated TNF and IL-6 responses and neutrophil recruitment) compared to wild-type controls, although the authors did not rule out G-protein-independent anti-inflammatory signaling downstream of C5aR2 (36). Subsequent studies indeed showed that C5aR2 might also play active, yet complex and poorly understood, roles in inflammation regulation including crosstalk interactions with TLRs (37–40). In the latter regard, C5aR2-deficient mice exhibited increased survival rates compared with wild-type controls after cecal ligation and puncture-induced sepsis (38). Rather than antagonizing C5aR1, C5aR2 synergizes with C5aR1 to cause sepsis by inducing the expression of the mobility group box 1 (HMGB1) protein (38). Interestingly, the induction of HMGB1 by LPS and C5a, or by LPS alone, is diminished in C5aR2-deficient macrophages. This finding suggests involvement of possible C5aR2–TLR4 crosstalk in the induction of HMGB1 that appears to require mitogen-activated protein MEK1/2, JNK1/2 and PI3K (38). Moreover, C5aR2 was shown to mediate C5a-induced activation of mast cells (41) and to promote atherosclerosis and neointimal plaque formation in apolipoprotein E-deficient mice (42). In contrast to these pro-inflammatory roles by C5aR2, other studies showed that C5aR2 interacts physically with and negatively regulates C5aR1 signaling in neutrophils and macrophages (39, 43), thereby providing a mechanistic basis for its reported anti-inflammatory action (36). In toto, the activities of C5aR2 appear to be dynamic and contextual depending on cell type, tissue, and disease model (44). In macrophages, complement receptor 3 (CR3; CD11b/CD18) can regulate the signaling activity of TLRs that utilize Mal (MyD88-adaptor like; also known as Toll/IL-1R(TIR)-domain-containing adaptor protein; TIRAP) as an adaptor, i.e., TLR2 and TLR4 (45) (Figure 3). Specifically, outside-in signaling by CR3 leads to activation of ADP ribosylation factor 6 (ARF6) and induction of phosphatidylinositol-(4,5)-bisphosphate (PIP2) production by phosphatidylinositol 5-kinase (PI5K), thereby promoting the targeting of Mal to membrane-bound PIP2 through its PIP2-binding domain. Mal can subsequently facilitate the recruitment of MyD88 to either TLR2 or TLR4 for initiation of MyD88-dependent signaling (45) (Figure 3). On the other hand, an independent study showed that CR3 may negatively regulate TLR-mediated inflammatory responses in macrophages by activating Syk and promoting degradation of MyD88 and TRIF via the E3 ubiquitin ligase Cbl-b (46) (Figure 3). Moreover, CR3 activation by a small-molecule allosteric agonist was shown to induce MyD88 degradation in macrophages also downstream of TLR7 and TLR8, thereby inhibiting TLR7/8-induced production of TNF (47). Intriguingly, this regulatory effect of CR3 was abrogated in macrophages expressing a genetic variant of CR3 (specifically a missense polymorphism, R77H, of CD11b) (47), which has been identified as a risk factor in systemic lupus erythematosus (48). It could thus be suggested that this mechanism may contribute to the pathogenesis of systemic lupus erythematosus, where RNA-containing immune complexes can readily trigger TLR7/8-mediated inflammation. Taken together, the above-discussed studies indicate that CR3 can exert both positive and negative regulation of TLR signaling by controlling the localization and/or degradation of TLR adaptors, although the contextual basis of these contrasting effects is not clear. TLR regulation of expression of complement components The previous section discussed several studies showing that complement receptors (e.g., C3aR, C5aR1, and CR3) regulate TLR-dependent responses, such as those induced by LPS (12, 21, 31, 45, 46). Reciprocally, TLR activation induces the expression of complement components, thereby potentially contributing to enhance complement activity in an inflammatory environment (49–52). For example, LPS induces robust production and release of factor B of the alternative pathway in macrophages (a major source of extra-hepatic complement synthesis) through a TLR4-TRIF pathway that leads to JNK and NF-κB activation (49). The same study showed that the double-stranded RNA analog polyI:C (a typical TLR3 agonist) also stimulates factor B production in macrophages via a JNK- and NF-κB-dependent mechanism; however, this pathway was not mediated by TLR3, suggesting the involvement of alternative receptors for polyI:C, such as the cytosolic sensors MDA-5 and RIG-I (49). An independent study showed that polyI:C induces factor B expression also in colonic epithelial cells, albeit via a TLR3-dependent mechanism (50). Importantly, the expression of factor B mRNA and protein is significantly enhanced in colonic biopsies of patients with ulcerative colitis and Crohn's disease as compared to healthy controls (50). Therefore, upon TLR stimulation, innate immune and epithelial cells can locally produce a critical component for alternative complement activation, which can in turn further amplify TLR-mediated responses. Although this positive feedback loop may contribute to host defense, the same mechanism could exacerbate pathology in diverse settings, such as inflammatory bowel disease and ischemia/reperfusion. In the latter condition, TLRs can respond to endogenous ligands released from stressed/ischemic tissues and local production of factor B (e.g., by cardiomyocytes in the context of myocardial infarction) may potentially contribute to complement-mediated injury during ischemia (53). Intestinal ischemia/reperfusion induces the expression of factor B and C3 in the gut of wild-type but not TLR4-deficient mice, which exhibit reduced inflammation and tissue damage (52). Administration of a complement inhibitor, CR2-Crry, during reperfusion ameliorated intestinal tissue damage in wild-type mice but did not further inhibit tissue damage in TLR4-deficient mice (52). These findings suggest that ischemia/reperfusion-induced tissue damage in this model requires a crosstalk involving TLR4 regulation of local production of complement, which in turn amplifies TLR4-mediated inflammation. A more recent study showed that, in addition to polyI:C and LPS, Pam3Cys activation of TLR2 (though not CpG activation of TLR9) also induces factor B production and release in macrophages and cardiac cells (51). Moreover, induction of polymicrobial sepsis by cecal ligation and puncture in mice was shown to increase the levels of factor B (in serum, peritoneal cavity, heart and other organs) in an MyD88-dependent manner, whereas genetic ablation of factor B reduced complement activation during sepsis, attenuated organ injury and improved survival (51). This study lends further support that factor B acts downstream of TLR activation and that bacterial sepsis is largely dependent on complement-TLR crosstalk. Modified low-density lipoprotein (mLDL) regulates the expression and release of C3 in macrophages by acting on TLR4 and liver X receptor (54). Specifically, uptake of mLDL by macrophages results in formation of oxysterols that activate Lliver X receptor-dependent transcription of target genes including C3. Moreover, on the cell surface, mLDL interacts with CD14 and TLR4 leading to induction of MEK1/2-ERK1/2-dependent C3 mRNA expression and NF-κB-dependent C3 protein secretion. Furthermore, subsequent activation of C3 leads to C3a activation of C3aR signaling that promotes mLDL uptake by macrophages, thereby reinforcing this positive regulatory feedback loop (54). As complement, TLRs, and mLDL metabolism are involved in atherosclerosis (55), this mechanism may be a contributing factor to the development of atherosclerotic lesions. Interestingly, TLR signaling suppresses the desensitization of GPCRs by downregulating the expression of G-protein-coupled receptor kinases, which induce GPCR phosphorylation and internalization, thereby potentially prolonging the activation of C3aR and C5aR1 (56). Moreover, TLR-induced cytokines, such as IL-6, promote the expression of C3aR and C5aR (57). In summary, TLRs regulate the expression of complement factors as well as the expression and/or activation of complement receptors, which in turn can amplify or limit TLR-dependent responses. Subversion of innate immunity by pathogen-induced complement-TLR crosstalk Periodontitis is a chronic inflammatory disease of the tooth-supporting tissues (periodontium) that is induced by local dysbiotic polymicrobial communities (58). These communities form on subgingival tooth sites and appear to have evolved collective strategies that enable them to persist in an inflammatory environment (59). A formidable challenge for these bacteria is to evade killing without resorting to immune suppression, as this would inhibit inflammation and hence limit their food supply, which is derived from inflammatory tissue breakdown (60). This selective pressure might be responsible for the development of some highly sophisticated microbial tactics, which represent new paradigms in immune evasion and are reviewed below. Immune subversion by periodontal bacteria Porphyromonas gingivalis, a low-abundance gram-negative bacterium associated with periodontitis, was shown to exert a disproportionately high impact on the dysbiotic transformation of periodontal microbial communities, thereby behaving as a keystone pathogen (61, 62). Specifically, P. gingivalis can subvert the innate host response in ways that alter the numbers and composition of the microbiota, that is, causing dysbiosis (63). The overgrowth of a subset of species, including inflammophilic pathobionts, leads to destructive periodontal inflammation and bone loss (59–61). The manipulation of the host response by P. gingivalis is based, at least in part, on its capacity to instigate subversive crosstalk interactions between complement and TLRs. For instance, P. gingivalis can induce a C5aR1-TLR2 crosstalk in neutrophils to uncouple bacterial immune clearance from inflammation (19) (Figure 4A), which creates a nutritionally favorable environment for the bacteria as they can feed off the inflammatory spoils (e.g., degraded collagen peptides and heme-containing compounds, a source of iron) (60, 64). In addition to stimulating TLR2, P. gingivalis can directly activate C5aR1 (i.e., independently of complement activation) through the action of its gingipains that can locally cleave C5 to generate C5a ligand (27, 65). In both human and mouse neutrophils, the P. gingivalis-instigated C5aR-TLR2 signaling crosstalk triggers ubiquitination and proteasomal degradation of the TLR2 adaptor MyD88, leading to suppression of downstream antimicrobial effects that would otherwise clear this bacterium (19, 66, 67) (Figure 4A). Although MyD88 induces also proinflammatory signaling for NF-κB activation, the nutritionally favorable inflammatory response is not abrogated but instead mediated by an alternative TLR2 adaptor, Mal (MyD88 adaptor-like). In this pathway, Mal activates PI3K which mediates a robust inflammatory response. Indeed, genetic ablation or pharmacological inhibition of Mal or PI3K suppresses the induction of pro-inflammatory cytokines by neutrophils in vitro and in vivo (19). Moreover, P. gingivalis-induced Mal-PI3K signaling inhibits GTPase RhoA-dependent actin polymerization and hence P. gingivalis phagocytosis (19) (Figure 4A). These actions also promote the survival of bystander bacteria that are otherwise susceptible to neutrophil killing (19). Conversely, inhibition of PI3K or any of the two crosstalking receptors, C5aR1 or TLR2, in the periodontium of P. gingivalis-colonized mice promotes the elimination of P. gingivalis, reverses the increase in total microbiota counts induced earlier by P. gingivalis colonization, and blocks periodontal inflammation (19). Therefore, P. gingivalis manipulates neutrophils through distinct mechanisms that collectively promote the survival of the microbial community and the perpetuation of inflammation. P. gingivalis induces a C5aR1-TLR2 crosstalk also in macrophages, which are thereby impaired for intracellular killing of this bacterium (68). However, the signaling mechanisms involved are completely different from those operating in neutrophils. In macrophages, the P. gingivalis C5aR1-TLR2 crosstalk leads to synergistic production of high and sustained levels of cAMP, which suppresses nitric oxide-dependent killing of P. gingivalis (68). Specifically, elevation of cAMP leads to activation of protein kinase A (PKA), which inactivates glycogen synthase kinase-3β (GSK3β) and inhibits the expression of inducible nitric oxide synthase (iNOS), hence reducing the production of nitric oxide, a potent antimicrobial molecule (68) (Figure 4B). The P. gingivalis-induced C5aR1-TLR2 crosstalk additionally regulates cytokine expression in macrophages (27). Specifically, P. gingivalis selectively suppresses TLR2-induced IL-12p70 through a C5aR1-dependent mechanism involving ERK1/2 (Figure 1), whereas the same C5aR1-TLR2 crosstalk upregulates the production of proinflammatory cytokines (IL-1β, IL-6, and TNF), which appear to mediate inflammatory bone loss in a murine model of experimental periodontitis (27). Moreover, the ability of P. gingivalis to manipulate TLR2 activation via the C5a-C5aR1 pathway enables this microbe to inhibit the production of IL-12p70 and secondarily interferon (IFN)γ resulting in enhanced pathogen survival (27). Therefore, overall, P. gingivalis appears to inhibit both IFNγ-dependent priming of macrophages and their nitric oxide-dependent pathway for intracellular killing. The ability of complement to regulate TLR-induced IL-12 is a more general property that includes additional TLRs and IL-12-relates cytokines, such as IL-23. For instance, earlier work has shown that activation of C5aR1 in macrophages inhibits TLR4-induced mRNA expression of IL-12p35, IL-12/IL-23p40, IL-23p19 and IL-27p28, and production of IL-12, IL-23 and IL-27 proteins. The underlying mechanism involves induction of PI3K and ERK1/2 signaling, which in turn inhibit critical transcription factors (the IFN regulatory factors 1 and 8; IRF-1 and -8) that are required for expression of IL-12 family cytokines (12, 69, 70) (Figure 1). Similar but relatively attenuated inhibitory effects were observed after C3aR activation (12, 69). CR3 plays many and diverse roles in immunity and inflammation, including leukocyte transmigration and iC3b-mediated phagocytosis (71). Besides interacting with host molecules (iC3b, fibrinogen, and intercellular adhesion molecule-1 [ICAM-1]), CR3 can also interact with various microbial molecules, such as LPS, Bordetella pertussis filamentous hemagglutinin, Leishmania gp63, and P. gingivalis FimA fimbriae (72–76). In this regard, P. gingivalis FimA fimbriae can induce TLR2 inside-out signaling which transactivates the high-affinity conformation and hence the ligand-binding capacity of CR3 (77, 78). The interactions of CR3 on monocytes or macrophages with P. gingivalis lead to induction of proinflammatory cytokines (TNF, IL-1β, and IL-6) (75, 79) and promotion of ICAM-1-dependent monocyte transmigration across endothelial cell monolayers (80). Intriguingly, the aforementioned TLR2-CR3 crosstalk is exploited by P. gingivalis for a relatively safe entry and persistence in macrophages (81). Indeed, the intracellular survival of P. gingivalis is significantly reduced in CR3-deficient (CD11b−/−) mouse macrophages, suggesting that CR3-mediated phagocytosis of P. gingivalis prevents or ameliorates its killing (81). This finding is in line with observations that CR3 is not linked to vigorous microbicidal mechanisms, in contrast to certain other phagocytic receptors, such as Fcγ receptor III (CD16) (82–85). Indeed, in macrophages, CR3-derived phagosomes do not fuse with lysosomes as readily as CD16-derived phagosomes (86). The relatively mild post-phagocytic events downstream of CR3 are consistent with its role in the phagocytosis of iC3b-opsonized apoptotic cells, which entail minimal ‘danger’ as compared to pathogen infection (87, 88). Accordingly, upon phagocytosis of apoptotic cells, the production of IL-12 in efferocytic macrophages is suppressed (87) (Figure 5). Similarly, direct CR3 binding by P. gingivalis FimA fimbriae mitigates TLR2-induced IL-12 via outside-in signaling that induces ERK1/2-dependent inhibition of IL-12p35 and IL-12/IL-23p40 mRNA expression (89) (Figure 5). Consistent with this mechanism, CR3 blockade in a mouse peritonitis model (induced by i.p. injection of P. gingivalis) promotes IL-12-dependent clearance of P. gingivalis. Moreover, CR3-deficient mice are superior to wild-type controls in controlling P. gingivalis i.p. infection owing to elevated production of IL-12 and, secondarily, IFN-γ, a major activator of intracellular killing (89). Although P. gingivalis can exploit C5aR1 in neutrophils and macrophages to suppress their antimicrobial functions (19, 68), as well as bind CR3 for a safe entry into macrophages (81), the same receptors on dendritic cells do not seem to enhance the intracellular persistence of P. gingivalis (90). In stark contrast, C5aR1 promotes the intracellular killing of P. gingivalis in dendritic cells, whereas CR3 does not function as a phagocytic receptor for P. gingivalis. Similar to C5aR1, C3aR enhances the intracellular killing of P. gingivalis in dendritic cells. In contrast to C5aR1, C5aR2 is associated with increased intracellular survival of P. gingivalis in macrophages, consistent with the notion that C5aR2 can – in a certain context - downregulate the activity of C5aR1 (39, 43). The differential effects of C5aR1 in dendritic cells as compared to macrophages might be attributed to differential regulation of the cAMP response in these two leukocyte types. As outlined above for macrophages, activation of C5aR1 leads to high levels of intracellular cAMP and thus PKA activation, which is critical for inhibiting nitric oxide-dependent killing of P. gingivalis (68). In dendritic cells, on the other hand, C5aR1 suppresses cAMP production and hence the activation of PKA (91). C3aR – which also facilitates intracellular killing of P. gingivalis in dendritic cells (90)– similarly inhibits the cAMP-PKA pathway (92). As both C3aR and C5aR1 activate Gαi protein-mediated signaling, it is not clear why the same receptors have different effects on the cAMP responses in macrophages versus dendritic cells.. However, some insights could be discussed at least at a theoretical level. Following activation of Gαi, the released Giβγ subunits regulate the production of cAMP by adenylate cyclase, either positively or negatively depending upon the specific enzyme isoform (93). The isoforms of adenylate cyclase isoforms that are positively regulated by Giβγ are different from those that are sensitive to the inhibitory action of Gαi (93). Thus, it can be reasoned that dendritic cells and macrophages express distinct isoforms of adenylate cyclase, thus C3aR- or C5aR-induced Gαi signaling has different effects on the regulation of the enzyme isoforms. Another cell type-specific difference is that whereas C5a inhibits P. gingivalis-induced IL-12p70 in macrophages (27), C5a promotes P. gingivalis-induced IL-12p70 in dendritic cells (90). The C5a-induced inhibition of IL-12p70 by P. gingivalis is mediated by ERK1/2 signaling (27), consistent with an earlier report that C5a-induced ERK1/2 signaling inhibits enterobacterial lipopolysaccharide-induced IL-12p70 in macrophages (69). Whereas C5a induces ERK1/2 signaling also in dendritic cells (94), the ERK1/2 pathway in this cell type upregulates, rather than inhibits, IL-12p70 production (95). Despite its ability to transactivate and bind CR3 in macrophages, P. gingivalis fails to utilize CR3 as a phagocytic receptor in dendritic cells (90). The reason for this difference is not understood, although a study has suggested that CR3 cannot be readily transactivated in dendritic cells (96). Therefore, C3aR, C5aR1, and CR3, mediate cell-type-specific effects on how innate leukocytes handle P. gingivalis. Since dendritic cells are not as potent in pathogen destruction as compared to neutrophils or macrophages (97), it appears paradoxical that P. gingivalis can exploit complement receptors in neutrophils and macrophages more efficiently than it does in dendritic cells. However, given the abundance of complement cleavage products in the periodontal pocket (98), it makes sense from an evolutionary perspective that P. gingivalis developed complement-dependent evasion mechanisms against those leukocyte types that are most often encountered in its niche. Indeed, the immediate threat to P. gingivalis in its predominant niche, the periodontal pocket, is represented by neutrophils and secondarily by macrophages, which predominate in the leukocyte infiltrate of the periodontal pocket over other leukocyte types (99). Immune subversion by other pathogens The TLR2–CR3 crosstalk pathway may be exploited by additional pathogens. Mycobacteria and spores of Bacillus anthracis can both induce TLR2 inside-out signaling for transactivating and binding CR3, thereby promoting their uptake via CR3 (100, 101). It is thought that the ability of Mycobacterium tuberculosis to parasitize within macrophages may, in part, be reliant on its capacity to stimulate TLR2-induced CR3 uptake (101). Moreover, CR3-deficient mice display enhanced resistance to infection with B. anthracis spores. The susceptibility of wild-type mice in this model was attributed to enhanced uptake of B. anthracis spores and their carriage by the macrophages to sites of spore germination and bacterial growth (100). Upon opsonization with iC3, Francisella tularensis also uses CR3 for efficient macrophage uptake and the resulting outside-in signaling suppresses TLR2-mediated and MAPK-dependent pro-inflammatory responses, thereby promoting the pathogenesis of F. tularensis infection (102). The crosstalk of CR3 with the TLR system is bidirectional since, as discussed above, CR3 also regulates TLR signaling (45, 46). In line with this notion, a recent study has shown CR3 regulation of TLR8 responses in dendritic cells. Indeed, whereas free HIV-1 induces robust TLR8-dependent inflammatory and anti-viral responses (induction of p38, ERK, and NF-κB pathways and activation of IFN regulatory factors 1 and 7) in immature dendritic cells, iC3b-opsonized HIV interacts with CR3 leading to CR3-TLR8 crosstalk that modulates the host response in a way that enhances viral transcription (103). In addition to CR3, gC1qR, a complement receptor for C1q, also suppresses TLR4-induced IL-12 in human monocytes (104) (Figure 5). This regulatory effect is mediated via PI3K signaling and is selective for IL-12 in that TNF, IL-6, and IL-8 are not impacted. However, this crosstalk appears to be exploited by the hepatitis C virus whose core protein acts as a ligand for gC1qR to inhibit IL-12 production and Th1 immunity (105) (Figure 5). The complement regulatory receptor CD46 also engages in a similar crosstalk with TLR4. Indeed, upon binding C3b dimers, CD46 inhibits LPS-induced IL-12 production in monocytes (106). The measles virus interacts with CD46 and thereby inhibits IL-12 production and cell-mediated immunity (106) (Figure 5). The underlying signaling mechanism is uncertain. However, a post-transcriptional mechanism was implicated in a study with human herpesvirus-6, which similarly uses CD46 as a cellular receptor to suppress TLR4-induced IL-12 (107). Concluding remarks and outlook The literature summarized in this review reveals an intricate interplay between complement and TLRs for regulating the expression and activation of critical components of the two systems, thereby contributing to the coordination of host immune and inflammatory responses. These bidirectional interactions range from antagonistic to synergistic (Figures 1–5) and can therefore enhance host immunity and inflammation to clear infections, or can dampen host responses to ameliorate exaggerated inflammation and tissue damage. In the latter case, future therapies for inflammatory or autoimmune diseases could focus on inhibiting either complement or TLRs, or both systems, depending on tissue or disease context (108–110). However, there may be instances where combined inhibition may have unfavorable outcomes. Indeed, although both complement and TLR2 induce inflammation in the context of renal ischemia/reperfusion (mice deficient in either factor B or TLR2 are protected from ischemic acute kidney injury), mice doubly deficient in factor B and TLR2 develop severe inflammatory tissue injury (111). These data suggest that, in this model, complement and TLR2 may also induce compensatory anti-inflammatory signals, the absence of which in the doubly deficient mice may have detrimental effects. Some crosstalk interactions between complement and TLRs appear to be proactively instigated by pathogens ostensibly to dysregulate or modify the host response in ways that favor their persistence, often with concomitant collateral tissue damage. For instance, by inducing a C5aR1-TLR2 crosstalk, periodontal bacteria can disengage immune bacterial clearance from inflammation (Figure 4), thereby contributing to the persistence of ‘inflammophilic’ communities of pathobionts that exacerbate polymicrobial inflammatory diseases, such as periodontitis (19). In this context, novel and potentially effective approaches may be to interfere with the host signaling circuitry that is exploited for microbial subversion of the immune response. Although this review has focused on innate immunity, complement-TLR interactions also impact on adaptive immunity. An important mechanism in this regard involves signaling crosstalk in antigen-presenting cells between C5aR1 and TLR4 which downregulates the expression of IL-12 family cytokines (IL-12, IL-23, and IL-27) (Figure 1) involved in the regulation of distinct T-cell subsets (Th1, Th2, and Th17) (12, 17, 69, 70, 112). The role of C5aR1 signaling in regulating T cell immunity in co-operation with TLRs is complex and contextual, as it can lead to different outcomes depending on the maturation stage of the antigen-presenting cell (20) or the type of crosstalking TLR (112). Moreover, cell type- and species-specific differences have been noted and reviewed elsewhere (17, 113–115)). The complex – and still incompletely understood – crosstalk interactions of complement with TLRs (and other systems reviewed elsewhere (9)) apparently aim to fine-tune a balance between homeostatic immunity and inflammatory pathology. Future research to further dissect the molecular mechanisms of complement-TLR crosstalk and their contextual nature may contribute to the design of novel approaches to maximize the beneficial and minimize the detrimental aspects of these interactions. The authors are supported by grants from the U.S. National Institutes of Health: DE015254, DE017138, DE021685, and DE024716 (GH); AI003040 and AI068730 (JDL) and the European Community’s Seventh Framework Programme under grant agreement number 602699 (DIREKT) (JDL). Figure 1 Synergistic and antagonistic interactions between complement and TLRs Complement and TLRs are co-activated in response to microbial infection. Complement anaphylatoxin receptor signaling induced by C3a or C5a synergizes with TLR signaling resulting in enhanced activation of MAPKs and transcription factors, such as NF-κB and AP-1, resulting in upregulation of proinflammatory cytokine expression. TLRs are activated by MAMPs, some of which (e.g., LPS and zymosan) can additionally co-activate complement. In contrast, complement can downregulate TLR-induced cytokines of the IL-12 family. Activation of C5aR1 by C5a suppresses TLR-induced mRNA expression of IL-12p35, IL-12/IL-23p40, IL-23p19, and IL-27p28 (hence production of bioactive IL-12, IL-23, and IL-27) in monocytes/macrophages. The underlying signaling mechanism involves induction of PI3K or ERK1/2 signaling, which in turn suppress crucial transcription factors (IRF-1 and IRF-8) that regulate these cytokines. Figure 2 C5aR1 regulation of IL-17 isoforms in LPS-activated macrophages C5aR1 and TLR4 promote the induction of IL-17F in mouse macrophages via a MyD88-PI3K-Akt pathway. On the other hand, C5a-induced activation of C5aR1 activates PI3K-Akt and MEK1/2-ERK1/2 pathways that lead to induction of IL-10, which subsequently inhibits production of IL-17A. Figure 3 TLR–CR3 crosstalk pathways CR3 can regulate the signaling activity of TLRs that utilize Mal as an adaptor, i.e. TLR2 and TLR4. Specifically, outside-in signaling by CR3 leads to activation of ADP ribosylation factor 6 (ARF6) and induction of phosphatidylinositol-(4,5)-bisphosphate (PIP2) production by phosphatidylinositol 5-kinase (PI5K). This in turn promotes the targeting of Mal, which has PIP2-binding domain, to membrane-bound PIP2. Mal in turn facilitates the recruitment of MyD88 to either TLR2 or TLR4 to initiate pro-inflammatory signaling. Moreover, CR3 outside-in signaling stimulates ITAM-coupled activation of the tyrosine kinases Src and Syk. Syk in turn binds and phosphorylates MyD88 and TRIF, which are thereby targeted by the E3 ubiquitin ligase Cbl-b for proteolytic cleavage. Figure 4 P. gingivalis-induced C5aR1-TLR2 crosstalk in neutrophils and macrophages P. gingivalis expresses ligands that activate the TLR2–TLR1 complex (TLR2/1) and enzymes (HRgpA and RgpB gingipains) with C5 convertase-like activity that generate high local concentrations of C5a ligand. The bacterium can thus co-activate C5aR and TLR2 in (A) neutrophils and (B) macrophages. In neutrophils (A), the resulting crosstalk leads to ubiquitination and proteasomal degradation of the TLR2 adaptor MyD88, thereby inhibiting a host-protective antimicrobial response. This proteolytic event requires C5aR1-TLR2-dependent release of TGF-β1, which mediates MyD88 ubiquitination via the E3 ubiquitin ligase Smurf1 (enlarged inset). Moreover, the C5aR1-TLR2 crosstalk activates PI3K, which inhibits phagocytosis through suppression of RhoA GTPase and actin polymerization, while inducing inflammatory cytokine production. In contrast to MyD88, Mal contributes to immune subversion by acting upstream of PI3K. In macrophages (B), P. gingivalis activates C5aR1 and induces intracellular Ca2+ signaling which synergistically enhances the otherwise weak cAMP responses induced by TLR2 activation alone. The resulting activation of the cAMP-dependent protein kinase A (PKA) inhibits NF-κB and glycogen synthase kinase-3β (GSK3β), thereby suppressing inducible nitric oxide synthase (iNOS)-dependent killing of the pathogen in macrophages. Figure 5 Immune evasion via complement-mediated suppression of TLR-induced IL-12 production The crosstalk between the indicated complement receptors (C5aR1, CR3, gC1qR and CD46) and TLRs selectively inhibits the induction of IL-12 in macrophages. Signaling molecules that have been implicated, such as ERK1/2, IRF-1, IRF-8 and PI3K, are shown downstream of the corresponding receptors. A posttranscriptional mechanism might be involved in IL-12 regulation by CD46. Activation of these complement receptors by their natural ligands likely mediates homeostatic functions. However, the same receptors can be activated by bacteria or viruses (see text for details) which can thereby downregulate TLR-induced IL-12 production and hence IFNγ to suppress cell-mediated immunity. HCV, hepatitis C virus; MV, measles virus. The authors declare no conflicting financial interests. REFERENCES 1 Kawai T Akira S Toll-like receptors and their crosstalk with other innate receptors in infection and immunity Immunity 2011 34 637 650 21616434 2 Medzhitov R Recognition of microorganisms and activation of the immune response Nature 2007 449 819 826 17943118 3 Beutler BA TLRs and innate immunity Blood 2009 113 1399 1407 18757776 4 Triantafilou M Triantafilou K Lipopolysaccharide recognition: CD14, TLRs and the LPS-activation cluster Trends Immunol 2002 23 301 304 12072369 5 Brown J Wang H Hajishengallis GN Martin M TLR-signaling networks: an integration of adaptor molecules, kinases, and cross-talk J Dent Res 2011 90 417 427 20940366 6 Bottazzi B Doni A Garlanda C Mantovani A An integrated view of humoral innate immunity: pentraxins as a paradigm Annu Rev Immunol 2010 28 157 183 19968561 7 Holmskov U Thiel S Jensenius JC Collectins and ficolins: humoral lectins of the innate immune defense Annu Rev Immunol 2003 21 547 578 12524383 8 Inforzato A Bottazzi B Garlanda C Valentino S Mantovani A Pentraxins in humoral innate immunity Adv Exp Med Biol 2012 946 1 20 21948359 9 Ricklin D Hajishengallis G Yang K Lambris JD Complement: a key system for immune surveillance and homeostasis Nat Immunol 2010 11 785 797 20720586 10 Hajishengallis G Lambris JD Crosstalk pathways between Toll-like receptors and the complement system Trends Immunol 2010 31 154 163 20153254 11 Mangsbo SM Complement activation by CpG in a human whole blood loop system: Mechanisms and immunomodulatory effects J Immunol 2009 183 6724 6732 19864604 12 Zhang X Regulation of Toll-like receptor-mediated inflammatory response by complement in vivo Blood 2007 110 228 236 17363730 13 Hajishengallis G Lambris JD Microbial manipulation of receptor crosstalk in innate immunity Nat Rev Immunol 2011 11 187 200 21350579 14 Goodridge HS Underhill DM Fungal recognition by TLR2 and dectin-1 Handb Exp Pharmacol 2008 87 109 18071656 15 Ogawa S Molecular determinants of crosstalk between nuclear receptors and toll-like receptors Cell 2005 122 707 721 16143103 16 Lukashev D Ohta A Apasov S Chen JF Sitkovsky M Cutting edge: Physiologic attenuation of proinflammatory transcription by the Gs protein-coupled A2A adenosine receptor in vivo J Immunol 2004 173 21 24 15210754 17 Seow V Inflammatory responses induced by lipopolysaccharide are amplified in primary human monocytes but suppressed in macrophages by complement protein C5a J Immunol 2013 191 4308 4316 24043889 18 Fang C Zhang X Miwa T Song WC Complement promotes the development of inflammatory T-helper 17 cells through synergistic interaction with Toll-like receptor signaling and interleukin-6 production Blood 2009 114 1005 1015 19491392 19 Maekawa T Porphyromonas gingivalis manipulates complement and TLR signaling to uncouple bacterial clearance from inflammation and promote dysbiosis Cell Host Microbe 2014 15 768 778 24922578 20 Zaal A Lissenberg-Thunnissen SN van Schijndel G Wouters D van Ham SM ten Brinke A Crosstalk between Toll like receptors and C5a receptor in human monocyte derived DCs suppress inflammatory cytokine production Immunobiology 2013 218 175 180 22559913 21 Bosmann M Sarma JV Atefi G Zetoune FS Ward PA Evidence for anti-inflammatory effects of C5a on the innate IL-17A/IL-23 axis FASEB J 2011 26 1640 1651 22202675 22 Yamada M Complement C1q regulates LPS-induced cytokine production in bone marrow-derived dendritic cells Eur J Immunol 2004 34 221 230 14971048 23 Lappegård KT Human genetic deficiencies reveal the roles of complement in the inflammatory network: lessons from nature Proc Natl Acad Sci U S A 2009 106 15861 15866 19717455 24 Guo RF Riedemann NC Ward PA Role of C5a-C5aR interaction in sepsis Shock 2004 21 1 7 25 Kim DD Song WC Membrane complement regulatory proteins Clin Immunol 2006 118 127 136 16338172 26 Abe T Local complement-targeted intervention in periodontitis: proof-of-concept using a C5a receptor (CD88) antagonist J Immunol 2012 189 5442 5448 23089394 27 Liang S The C5a receptor impairs IL-12-dependent clearance of Porphyromonas gingivalis and is required for induction of periodontal bone loss J Immunol 2011 186 869 877 21149611 28 Burns E Bachrach G Shapira L Nussbaum G Cutting Edge: TLR2 is required for the innate response to Porphyromonas gingivalis : Activation leads to bacterial persistence and TLR2 deficiency attenuates induced alveolar bone resorption J Immunol 2006 177 8296 8300 17142724 29 Lau C CD14 and complement crosstalk and largely mediate the transcriptional response to Escherichia coli in human whole blood as revealed by DNA microarray PLoS One 2015 10 e0117261 25706641 30 Brekke OL Combined inhibition of complement and CD14 abolish E. coli-induced cytokine-, chemokine- and growth factor-synthesis in human whole blood Mol Immunol 2008 45 3804 3813 18606453 31 Bosmann M MyD88-dependent production of IL-17F is modulated by the anaphylatoxin C5a via the Akt signaling pathway FASEB J 2011 25 4222 4232 21859896 32 Gaffen SL Structure and signalling in the IL-17 receptor family Nat Rev Immunol 2009 9 556 567 19575028 33 Monk PN Scola AM Madala P Fairlie DP Function, structure and therapeutic potential of complement C5a receptors Br J Pharmacol 2007 152 429 448 17603557 34 Okinaga S C5L2, a nonsignaling C5a binding Protein Biochemistry 2003 42 9406 9415 12899627 35 Gao H Evidence for a functional role of the second C5a receptor C5L2 Faseb J 2005 19 1003 1005 15784721 36 Gerard NP An anti-inflammatory function for the complement anaphylatoxin C5a-binding protein, C5L2 J Biol Chem 2005 280 39677 39680 16204243 37 Chen NJ C5L2 is critical for the biological activities of the anaphylatoxins C5a and C3a Nature 2007 446 203 207 17322907 38 Rittirsch D Functional roles for C5a receptors in sepsis Nat Med 2008 14 551 557 18454156 39 Bamberg CE The C5a receptor (C5aR) C5L2 is a modulator of C5aR-mediated signal transduction J Biol Chem 2010 285 7633 7644 20044484 40 Horst SA Itzek A Klos A Beineke A Medina E Differential Contributions of the Complement Anaphylotoxin Receptors C5aR1 and C5aR2 to the Early Innate Immune Response against Staphylococcus aureus Infection Pathogens 2015 4 722 738 26512700 41 Pundir P MacDonald CA Kulka M The Novel Receptor C5aR2 Is Required for C5a-Mediated Human Mast Cell Adhesion, Migration, and Proinflammatory Mediator Production J Immunol 2015 195 2774 2787 26283482 42 Selle J Atheroprotective role of C5ar2 deficiency in apolipoprotein E-deficient mice Thromb Haemost 2015 114 848 858 26084965 43 Croker DE C5a2 can modulate ERK1/2 signaling in macrophages via heteromer formation with C5a1 and beta-arrestin recruitment Immunol Cell Biol 2014 92 631 639 24777312 44 Li R Coulthard LG Wu MC Taylor SM Woodruff TM C5L2: a controversial receptor of complement anaphylatoxin, C5a FASEB J 2013 27 855 864 23239822 45 Kagan JC Medzhitov R Phosphoinositide-mediated adaptor recruitment controls Toll-like receptor signaling Cell 2006 125 943 955 16751103 46 Han C Jin J Xu S Liu H Li N Cao X Integrin CD11b negatively regulates TLR-triggered inflammatory responses by activating Syk and promoting degradation of MyD88 and TRIF via Cbl-b Nat Immunol 2010 11 734 742 20639876 47 Reed JH Complement receptor 3 influences toll-like receptor 7/8-dependent inflammation: implications for autoimmune diseases characterized by antibody reactivity to ribonucleoproteins J Biol Chem 2013 288 9077 9083 23386618 48 MacPherson M Lek HS Prescott A Fagerholm SC A systemic lupus erythematosus-associated R77H substitution in the CD11b chain of the Mac-1 integrin compromises leukocyte adhesion and phagocytosis J Biol Chem 2011 286 17303 17310 21454473 49 Kaczorowski DJ Pivotal Advance: The pattern recognition receptor ligands lipopolysaccharide and polyinosine-polycytidylic acid stimulate factor B synthesis by the macrophage through distinct but overlapping mechanisms J Leukoc Biol 2010 50 Ostvik AE Mucosal toll-like receptor 3-dependent synthesis of complement factor B and systemic complement activation in inflammatory bowel disease Inflamm Bowel Dis 2014 20 995 1003 24739633 51 Zou L Complement factor B is the downstream effector of TLRs and plays an important role in a mouse model of severe sepsis J Immunol 2013 191 5625 5635 24154627 52 Pope MR Hoffman SM Tomlinson S Fleming SD Complement regulates TLR4-mediated inflammatory responses during intestinal ischemia reperfusion Mol Immunol 2010 48 356 364 20800895 53 Singh MV Ca2+/calmodulin-dependent kinase II triggers cell membrane injury by inducing complement factor B gene expression in the mouse heart J Clin Invest 2009 119 986 996 19273909 54 Mogilenko DA Modified low density lipoprotein stimulates complement C3 expression and secretion via liver X receptor and Toll-like receptor 4 activation in human macrophages J Biol Chem 2012 287 5954 5968 22194611 55 Hovland A The complement system and toll-like receptors as integrated players in the pathophysiology of atherosclerosis Atherosclerosis 2015 241 480 494 26086357 56 Fan J Malik AB Toll-like receptor-4 (TLR4) signaling augments chemokine-induced neutrophil migration by modulating cell surface expression of chemokine receptors Nat Med 2003 9 315 321 12592402 57 Rittirsch D Flierl MA Ward PA Harmful molecular mechanisms in sepsis Nat Rev Immunol 2008 8 776 787 18802444 58 Lamont RJ Hajishengallis G Polymicrobial synergy and dysbiosis in inflammatory disease Trends Mol Med 2015 21 172 183 25498392 59 Hajishengallis G Periodontitis: from microbial immune subversion to systemic inflammation Nat Rev Immunol 2015 15 30 44 25534621 60 Hajishengallis G The inflammophilic character of the periodontitis-associated microbiota Mol Oral Microbiol 2014 29 248 257 24976068 61 Hajishengallis G Darveau RP Curtis MA The keystone-pathogen hypothesis Nat Rev Microbiol 2012 10 717 725 22941505 62 Darveau RP The oral microbial consortium's interaction with the periodontal innate defense system DNA Cell Biol 2009 28 389 395 19435427 63 Hajishengallis G Low-abundance biofilm species orchestrates inflammatory periodontal disease through the commensal microbiota and complement Cell Host Microbe 2011 10 497 506 22036469 64 Marsh PD Are dental diseases examples of ecological catastrophes? Microbiology 2003 149 279 294 12624191 65 Popadiak K Potempa J Riesbeck K Blom AM Biphasic effect of gingipains from Porphyromonas gingivalis on the human complement system J Immunol 2007 178 7242 7250 17513773 66 Brzezinska AA Johnson JL Munafo DB Ellis BA Catz SD Signalling mechanisms for Toll-like receptor-activated neutrophil exocytosis: key roles for interleukin-1-receptor-associated kinase-4 and phosphatidylinositol 3-kinase but not Toll/IL-1 receptor (TIR) domain-containing adaptor inducing IFNβ (TRIF) Immunology 2009 127 386 397 19019092 67 Burns E Eliyahu T Uematsu S Akira S Nussbaum G TLR2-dependent inflammatory response to Porphyromonas gingivalis is MyD88 independent, whereas MyD88 is required to clear infection J Immunol 2010 184 1455 1462 20042569 68 Wang M Microbial hijacking of complement-toll-like receptor crosstalk Sci Signal 2010 3 ra11 20159852 69 Hawlisch H Belkaid Y Baelder R Hildeman D Gerard C Kohl J C5a negatively regulates toll-like receptor 4-induced immune responses Immunity 2005 22 415 426 15845447 70 la Sala A Gadina M Kelsall BL G(i)-protein-dependent inhibition of IL-12 production is mediated by activation of the phosphatidylinositol 3-kinase-protein 3 kinase B/Akt pathway and JNK J Immunol 2005 175 2994 2999 16116186 71 Ehlers MRW CR3: a general purpose adhesion-recognition receptor essential for innate immunity Microbes Infect 2000 2 289 294 10758405 72 Diamond MS Garcia-Aguilar J Bickford JK Corbi AL Springer TA The I domain is a major recognition site on the leukocyte integrin Mac-1 (CD11b/CD18) for four distinct adhesion ligands J Cell Biol 1993 120 1031 1043 7679388 73 McGuirk P Mills KH Direct anti-inflammatory effect of a bacterial virulence factor: IL-10-dependent suppression of IL-12 production by filamentous hemagglutinin from Bordetella pertussis Eur J Immunol 2000 30 415 422 10671196 74 Russell DG Wright SD Complement receptor type 3 (CR3) binds to an Arg-Gly-Asp-containing region of the major surface glycoprotein, gp63, of Leishmania promastigotes J Exp Med 1988 168 279 292 3294332 75 Hajishengallis G Ratti P Harokopakis E Peptide mapping of bacterial fimbrial epitopes interacting with pattern recognition receptors J Biol Chem 2005 280 38902 38913 16129673 76 Ingalls RR Arnaout MA Golenbock DT Outside-in signaling by lipopolysaccharide through a tailless integrin J Immunol 1997 159 433 438 9200483 77 Harokopakis E Hajishengallis G Integrin activation by bacterial fimbriae through a pathway involving CD14, Toll-like receptor 2, and phosphatidylinositol-3-kinase Eur J Immunol 2005 35 1201 1210 15739163 78 Hajishengallis G Wang M Liang S Induction of distinct TLR2-mediated proinflammatory and proadhesive signaling pathways in response to Porphyromonas gingivalis fimbriae J Immunol 2009 182 6690 6696 19454663 79 Hajishengallis G Differential interactions of fimbriae and lipopolysaccharide from Porphyromonas gingivalis with the Toll-like receptor 2-centred pattern recognition apparatus Cell Microbiol 2006 8 1557 1570 16984411 80 Harokopakis E Albzreh MH Martin MH Hajishengallis G TLR2 transmodulates monocyte adhesion and transmigration via Rac1- and PI3K-mediated inside-out signaling in response to Porphyromonas gingivalis fimbriae J Immunol 2006 176 7645 7656 16751412 81 Wang M Fimbrial proteins of Porphyromonas gingivalis mediate in vivo virulence and exploit TLR2 and complement receptor 3 to persist in macrophages J Immunol 2007 179 2349 2358 17675496 82 Lowell CA Rewiring phagocytic signal transduction Immunity 2006 24 243 245 16546092 83 Wright SD Silverstein SC Receptors for C3b and C3bi promote phagocytosis but not the release of toxic oxygen from human phagocytes J Exp Med 1983 158 2016 2023 6227677 84 Caron E Hall A Identification of two distinct mechanisms of phagocytosis controlled by different Rho GTPases Science 1998 282 1717 1721 9831565 85 Hellwig SM van Oirschot HF Hazenbos WL van Spriel AB Mooi FR van De Winkel JG Targeting to Fcg receptors, but not CR3 (CD11b/CD18), increases clearance of Bordetella pertussis J Infect Dis 2001 183 871 879 11237803 86 Vieira OV Botelho RJ Grinstein S Phagosome maturation: aging gracefully Biochem J 2002 366 689 704 12061891 87 Kim S Elkon KB Ma X Transcriptional suppression of interleukin-12 gene expression following phagocytosis of apoptotic cells Immunity 2004 21 643 653 15539151 88 Mevorach D Mascarenhas JO Gershov D Elkon KB Complement-dependent clearance of apoptotic cells by human macrophages J Exp Med 1998 188 2313 2320 9858517 89 Hajishengallis G Shakhatreh MA Wang M Liang S Complement receptor 3 blockade promotes IL-12-mediated clearance of Porphyromonas gingivalis and negates its virulence in vivo J Immunol 2007 179 2359 2367 17675497 90 Hajishengallis G Krauss JL Jotwani R Lambris JD Differential capacity for complement receptor-mediated immune evasion by Porphyromonas gingivalis depending on the type of innate leukocyte Mol Oral Microbiol 2016 [Epub ahead of print] 91 Peng Q Dendritic cell function in allostimulation is modulated by C5aR signaling J Immunol 2009 183 6058 6068 19864610 92 Li K Cyclic AMP plays a critical role in C3a-receptor-mediated regulation of dendritic cells in antigen uptake and T-cell stimulation Blood 2008 112 5084 5094 18812470 93 Sunahara RK Taussig R Isoforms of mammalian adenylyl cyclase: Multiplicities of signaling Mol Interv 2002 2 168 184 14993377 94 Weaver DJ Jr C5a receptor-deficient dendritic cells promote induction of Treg and Th17 cells Eur J Immunol 2010 40 710 721 20017191 95 Baruah P C1q enhances IFN-g production by antigen-specific T cells via the CD40 costimulatory pathway on dendritic cells Blood 2009 113 3485 3493 19171874 96 Varga G Active MAC-1 (CD11b/CD18) on DCs inhibits full T-cell activation Blood 2007 109 661 669 17003381 97 Silva MT When two is better than one: macrophages and neutrophils work in concert in innate immunity as complementary and cooperative partners of a myeloid phagocyte system J Leukoc Biol 2010 87 93 106 20052802 98 Hajishengallis G Complement and periodontitis Biochem Pharmacol 2010 80 1992 2001 20599785 99 Delima AJ Van Dyke TE Origin and function of the cellular components in gingival crevice fluid Periodontol 2000 2003 31 55 76 12656996 100 Oliva C Turnbough CL Jr Kearney JF CD14-Mac-1 interactions in Bacillus anthracis spore internalization by macrophages Proc Natl Acad Sci U S A 2009 106 13957 13962 19666536 101 Sendide K Reiner NE Lee JS Bourgoin S Talal A Hmama Z Cross-talk between CD14 and complement receptor 3 promotes phagocytosis of mycobacteria: regulation by phosphatidylinositol 3-kinase and cytohesin-1 J Immunol 2005 174 4210 4219 15778383 102 Dai S Rajaram MV Curry HM Leander R Schlesinger LS Fine tuning inflammation at the front door: macrophage complement receptor 3-mediates phagocytosis and immune suppression for Francisella tularensis PLoS Pathog 2013 9 e1003114 23359218 103 Ellegard R Complement opsonization of HIV-1 results in decreased antiviral and inflammatory responses in immature dendritic cells via CR3 J Immunol 2014 193 4590 4601 25252956 104 Waggoner SN Cruise MW Kassel R Hahn YS gC1q receptor ligation selectively down-regulates human IL-12 production through activation of the phosphoinositide 3-kinase pathway J Immunol 2005 175 4706 4714 16177118 105 Waggoner SN Hall CH Hahn YS HCV core protein interaction with gC1q receptor inhibits Th1 differentiation of CD4+ T cells via suppression of dendritic cell IL-12 production J Leukoc Biol 2007 82 1407 1419 17881511 106 Karp CL Mechanism of suppression of cell-mediated immunity by measles virus Science 1996 273 228 231 8662504 107 Smith A Santoro F Di Lullo G Dagna L Verani A Lusso P Selective suppression of IL-12 production by human herpesvirus 6 Blood 2003 102 2877 2884 12829600 108 Skjeflo EW Combined inhibition of complement and CD14 improved outcome in porcine polymicrobial sepsis Crit Care 2015 19 415 26612199 109 Maekawa T Inhibition of pre-existing natural periodontitis in non-human primates by a locally administered peptide inhibitor of complement C3 J Clin Periodontol 2016 [Epub ahead of print] 110 Mastellos DC Ricklin D Hajishengallis E Hajishengallis G Lambris JD Complement therapeutics in inflammatory diseases: promising drug candidates for C3-targeted intervention Mol Oral Microbiol 2016 31 3 17 26332138 111 Amura CR Renner B Lyubchenko T Faubel S Simonian PL Thurman JM Complement activation and toll-like receptor-2 signaling contribute to cytokine production after renal ischemia/reperfusion Mol Immunol 2012 52 249 257 22750071 112 Bosmann M Haggadone MD Hemmila MR Zetoune FS Sarma JV Ward PA Complement activation product C5a is a selective suppressor of TLR4-induced, but not TLR3-induced, production of IL-27(p28) from macrophages J Immunol 2012 188 5086 5093 22491257 113 Kemper C Kohl J Novel roles for complement receptors in T cell regulation and beyond Mol Immunol 2013 56 181 190 23796748 114 Song WC Crosstalk between complement and toll-like receptors Toxicol Pathol 2012 40 174 182 22109714 115 Merle NS Noe R Halbwachs-Mecarelli L Fremeaux-Bacchi V Roumenina LT Complement System Part II: Role in Immunity Front Immunol 2015 6 257 26074922

PMC005xxxxxx/PMC5120389.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0376331 5212 J Psychiatr Res J Psychiatr Res Journal of psychiatric research 0022-3956 1879-1379 25263276 5120389 10.1016/j.jpsychires.2014.09.007 NIHMS831058 Article Association between mental disorders and subsequent adult onset asthma Alonso Jordi 123 de Jonge Peter 4 Lim Carmen C. W. 5 Aguilar-Gaxiola Sergio 6 Bruffaerts Ronny 7 Caldas-de-Almeida Jose Miguel 8 Liu Zhaorui 9 O'Neill Siobhan 10 Stein Dan J. 11 Viana Maria Carmen 12 Al-Hamzawi Ali Obaid 13 Angermeyer Matthias C. 14 Borges Guilherme 15 Ciutan Marius 16 de Girolamo Giovanni 17 Fiestas Fabian 18 Haro Josep Maria 192021 Hu Chiyi 22 Kessler Ronald C. 23 Lépine Jean Pierre 24 Levinson Daphna 25 Nakamura Yosikazu 26 Posada-Villa Jose 27 Wojtyniak Bogdan J 28 Scott Kate M. 29 1 Health Services Research Unit, IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain 2 CIBER Epidemiología y Salud Pública (CIBERESP), Spain 3 Pompeu Fabra University (UPF), Barcelona, Spain 4 Interdisciplinary Center Psychopathology and Emotion Regulation, University Medical Center, University of Groningen, Groningen, The Netherlands 5 Department of Psychological Medicine, Otago University, Dunedin, New Zealand 6 University of California, Davis, Center for Reducing Health Disparities, School of Medicine, Sacramento, California, USA 7 Universitair Psychiatrisch Centrum - Katholieke Universiteit Leuven (UPC-KUL), Leuven, Belgium 8 Chronic Diseases Research Center (CEDOC) and Department of Mental Health, Faculdade de Ciencias Medicas, Universidade Nova de Lisboa, Lisbon, Portugal 9 Institute of Mental Health, Peking University, Beijing, and People’s Republic of China 10 Bamford Centre for Mental Health and Well-Being, University of Ulster, Derry, Northern Ireland 11 University of Cape Town Department of Psychiatry & Mental Health, Groote Schuur Hospital, Cape Town, South Africa 12 Department of Social Medicine, Federal University of Espírito Santo (UFES), Vitória, Brazil 13 Al-Qadisia University College of Medicine, Diwania, Iraq 14 Center for Public Mental Health, Gosing am Wagram, Austria 15 Division of Epidemiological and Psychosocial Research, National Institute of Psychiatry (Mexico) & Metropolitan Autonomous University, Mexico City, Mexico 16 National School of Public Health and Professional Development, Bucharest, Romania 17 IRCCS Centro S. Giovanni di Dio Fatebenefratelli, Brescia, Italy 18 Evidence Generation for Public Health Research Unit, National Institute of Health, Lima, Peru 19 Parc Sanitari Sant Joan de Déu, Sant Boi de Llobregat, Barcelona, Spain 20 CIBER de Salud Mental (CIBERSAM), Spain 21 University of Barcelona, Barcelona, Spain 22 Shenzhen Institute of Mental Health and Shenzhen Kangning Hospital, Guangdong Province, PR China 23 Department of Health Care Policy, Harvard Medical School, Boston, MA, USA 24 Hôpital Saint-Louis Lariboisière Fernand Widal, INSERM U 705, CNRS UMR 8206, Paris, France 25 Mental Health Services, Ministry of Health, Jerusalem, Israel 26 Department of Public Health, Jichi Medical University, Yakushiji, Shimotsuke-shi, Tochigi-ken, Japan 27 Colegio Mayor de Cundinamarca University, CALLE 28 No. 5B-02, Bogota, DC, Colombia 28 National Institute of Public Health-National Institute of Hygiene and Department-Centre of Monitoring and Analyses of Population Health, Warsaw, Poland 29 Department of Psychological Medicine, Otago University, Dunedin, New Zealand Corresponding author: Jordi Alonso, jalonso@imim.es, Health Services Research Unit, IMIM (Hospital del Mar Research Institute), C. Dr. Aiguader, 88, PRBB building, 08003 BARCELONA, Spain. Phone: (+34) 933 160 760, Fax: (+34) 933 160 797 20 11 2016 16 9 2014 12 2014 23 11 2016 59 179188 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background and objectives Associations between asthma and anxiety and mood disorders are well established, but little is known about their temporal sequence. We examined associations between a wide range of DSM-IV mental disorders with adult onset of asthma and whether observed associations remain after mental comorbidity adjustments. Methods During face-to-face household surveys in community-dwelling adults (n = 52,095) of 19 countries, the WHO Composite International Diagnostic Interview retrospectively assessed lifetime prevalence and age at onset of 16 DSM-IV mental disorders. Asthma was assessed by self-report of physician’s diagnosis together with age of onset. Survival analyses estimated associations between first onset of mental disorders and subsequent adult onset asthma, without and with comorbidity adjustment. Results 1,860 adult onset (21 years+) asthma cases were identified, representing a total of 2,096,486 person-years of follow up. After adjustment for comorbid mental disorders several mental disorders were associated with subsequent adult asthma onset: bipolar (OR=1.8; 95%CI 1.3–2.4), panic (OR=1.4; 95%CI 1.0–2.0), generalized anxiety (OR=1.3; 95%CI 1.1–1.7), specific phobia (OR=1.4; 95%CI 1.2–1.6); post-traumatic stress (OR=1.5; 95%CI 1.1–2.0); binge eating (OR=1.9; 95%CI 1.2–2.9) and alcohol abuse (OR=1.5; 95%CI 1.2–2.0). Mental comorbidity linearly increased the association with adult asthma. The association with subsequent asthma was stronger for mental disorders with an early onset (before age 21). Conclusions A wide range of temporally prior mental disorders are significantly associated with subsequent onset of asthma in adulthood. The extent to which asthma can be avoided or improved among those with early mental disorders deserves study. Asthma Mental Disorders Population Epidemiology Chronic Disease Comorbidity INTRODUCTION Asthma is a major public health problem because a lifetime course and an increasing prevalence (Jenkins et al. 1994; Pearce et al. 2000). An association between asthma and some mental disorders, in particular, anxiety and depression has been shown (Goodwin et al. 2003a; Goodwin et al. 2004; Perna et al. 1997; Shavitt et al. 1992); (Opolski and Wilson, 2005; Toren et al. 2006). While some of the previous evidence was based in small number of countries, recent data have extended similar results to a large number of countries; (Jiang et al. 2013; Wong et al. 2013). Most of the studies showing an association between asthma and mental disorders were cross-sectional in nature, thus limiting their ability to infer the temporal relationship between asthma and mental disorders. Several longitudinal studies suggest that asthma in childhood is followed by some subsequent internalizing mental disorders (Alati et al. 2005; Goodwin et al. 2013; Ramos Olazagasti et al. 2012) and with suicidal ideation and suicide attempts (Goodwin and Eaton, 2005). On the other hand, a number of studies have shown a longitudinal association between psychological distress and atopic disorders, mostly asthma, both in children and adults (Chida et al. 2008); (Sanna et al. 2014). Only very few of these studies included comprehensive diagnostic measures of mental disorders (i.e., based on standard psychiatric diagnostic criteria such as the Diagnostic Statistical Manual (DSM)) for mood and anxiety (Hasler et al. 2005; Wainwright et al. 2007); and eating disorders (Goodwin et al. 2009; Scott et al. 2007; Scott et al. 2008). An additional limitation of previous research is that the influence of mental comorbidity in the association of mental disorders and asthma has not been analyzed in depth. Knowing whether anxiety or depression specifically is associated with asthma (after adjusting for comorbidity with the other) can guide research focused on the mechanisms underlying the association with asthma. We previously reported, based on a large international study including many mental disorders, that there was a concurrent association between 12-month mental disorders and lifetime asthma in many countries, regardless of the important variation in asthma prevalence in these countries (Scott et al. 2008). Associations were similar for anxiety, mood and alcohol abuse disorders. In those analyses we did not assess the effect of mental comorbidity and our focus was on associations between current mental disorders and asthma, rather than on associations between temporally prior mental disorders and subsequent onset of asthma. We therefore undertook analyses that considered the sequential order of the mental disorders and asthma comorbidity and reported that early onset (i.e., before age 21) mental disorders predicted subsequent onset of diagnosed adult onset asthma (i.e., after age 21), even after adjusting for childhood adversities, smoking and other relevant variables (Scott et al. 2008). However in these analyses, we included a limited set of mental disorders and we did not adjust for mental disorder comorbidity. Nor did we examine whether the association held true for mental disorders starting after the age of 21 and subsequent adult asthma. Therefore, we could not determine whether the associations found between early onset and subsequent asthma were reflecting the onset timing of these disorders (i.e., that they occur at critical developmental periods) or were because early disorders are a risk marker for comorbidity. Aims of the Study In this study we analyzed only asthma cases with onset in adulthoods (21+ years of age), as we wanted to test the antecedent model (Scott MK 2009) of mental disorders preceding asthma onset. This model suggests that mental disorders that start early in life and are chronic or recurrent, may have physiological effects akin to chronic stress, leading to Hypothalamus-Pituitary-Adrenal (HPA) dysregulation (Chida Y et al, 2008; Scott KM, 2009). This altered physiological stress response in turn has been associated with immune system dysfunction and increased inflammatory response. These mechanisms could facilitate asthma onset among susceptible individuals, along with the lifestyle risk factors also associated with mental disorders. By selecting only adult-onset asthma cases and conducting survival analyses based on person years only up to asthma diagnosis, this allowed us to investigate this specific temporal sequence from mental disorder to asthma diagnosis, albeit with the limitation of retrospective data. We did not include childhood asthma as it would be difficult for respondents to clearly recall the temporal priority of mental disorders and asthma symptoms. We therefore conducted the present study using a larger dataset from the World Mental Health (WMH) Surveys to examine the associations between first onset of 16 mood, anxiety, impulse control and substance use disorders with subsequent adult onset asthma. Our aims were to examine the influence of mental disorder comorbidity on these associations; to investigate whether the associations vary according to the age of mental disorder or asthma onset; and to assess whether they vary by gender. In this study we analyzed only asthma cases with onset in adulthood, as it helped ensure the temporal order of the association of interest: from prior mental disorders to subsequent asthma. For childhood onset of asthma, which is fairly common, it might be difficult for respondents to clearly recall the temporal priority of mental disorders and asthma symptoms. Thus, we improved accuracy of the estimation of the association between mental disorders and asthma, to the expense of some generalization. METHOD Samples and Procedures This study uses data from 19 of the WMH surveys: Colombia, Mexico, Peru, United States, Shenzhen (China), Japan, New Zealand, Belgium, France, Germany, Italy, the Netherlands, Romania, Spain, Portugal, Israel, Iraq, Northern Ireland, and Poland (see Table 1). A stratified multi-stage clustered area probability sampling strategy was used to select adult respondents (18 years+) in most WMH countries. Most of the surveys were based on nationally representative household samples while Colombia, Mexico and Shenzhen were based on nationally representative household samples in urbanized areas. In most countries, internal subsampling was used to reduce respondent burden and average interview time by dividing the interview into two parts. All respondents completed Part 1, which included the core diagnostic assessment of most mental disorders. All Part 1 respondents who met lifetime criteria for any mental disorder and a probability sample of other respondents were administered Part 2, which assessed physical conditions and collected a range of other information related to survey aims. Part 2 respondents were weighted by the inverse of their probability of selection for Part 2 of the interview to adjust for differential sampling. Analyses in this paper are based on the weighted Part 2 subsample (n= 52,095). Additional weights were used to adjust for differential probabilities of selection within households, to adjust for non-response, and to match the samples to population socio-demographic distributions. Measures taken to ensure interviewer and data accuracy and cross-national consistency are described in detail elsewhere (Kessler and Ustun, 2004). All respondents provided informed consent and procedures for protecting respondents were approved and monitored for compliance by the Institutional Review Boards in each country (Kessler and Ustun, 2004). Measures All WMH instruments were cross-culturally adapted following the standard WHO procedures of translation, back-translation, cognitive debriefing and harmonization to make sure data collected would be comparable across countries. Such procedures are described in detail (Kessler and Ustun, 2008). Mental disorders All surveys used the WMH survey version of the WHO Composite International Diagnostic Interview (CIDI 3.0) (Kessler and Ustun, 2004), a fully structured interview, to assess lifetime history of mental disorders. Disorders were assessed using the definitions and criteria of the DSM-IV (American Psychiatric Association, 2000). The mental disorders adjusted for in this paper include anxiety disorders (panic disorder, agoraphobia without panic, specific phobia, social phobia, post-traumatic stress disorder, generalized anxiety disorder, obsessive compulsive disorder); mood disorders (major depressive episode/dysthymia, bipolar disorders I, II and broad); substance use disorders (alcohol abuse and dependence, drug abuse and dependence); and impulse control disorders (intermittent explosive disorder, bulimia nervosa and binge eating disorder). The selection of these 16 mental disorders was based on their prevalence, including disorders only when there were sufficient numbers of cases to warrant reliable analyses. As a result, for instance anorexia nervosa had to be excluded. We considered early onset mental disorders those with an onset before age of 21 years. CIDI organic exclusion rules were applied in making diagnoses. Clinical reappraisal studies conducted in some of the WMH countries indicate that lifetime diagnoses of anxiety, mood and substance use disorders based on the CIDI have generally good concordance with diagnoses based on blinded clinical interviews (Haro et al. 2006). Asthma In a series of questions adapted from the U.S Health Interview Survey, respondents were asked about the lifetime presence of selected chronic conditions. Respondents were asked: “Did a doctor or other health professional ever tell you that you had any of the following illnesses….asthma?” Clinical guidelines as those issued by the American Thoracic Society, recommend a combination of methods, including medical history, physical examination and respiratory function tests (Goodwin et al. 2003b; Pearce et al. 2000) but such methods are generally not feasible in large epidemiological surveys. It is important to note here that an investigation of the correspondence of self-reported chronic conditions in the US National Health Interview Survey with medical records abstracted in the prior 3 years found that self-reported current asthma to be in fairly good agreement with medical records, although underreported by 20–30% (NCHS, 1994). In our survey the definition of asthma was self-report of a diagnosis of asthma –not simply a self-report of asthma. So it may correspond more closely still to actual medical records. If respondents endorsed this question they were classified as having a history of asthma for these analyses. Respondents were also asked how old they were when they were first diagnosed with asthma. This year is referred to herein as the age of onset of asthma, although it is recognized that the underlying pathology develops over many years and that it is possible to have asthma without being detected. Only adult-onset asthma (onset age 21+) was investigated in this paper. No difference between atopic/non-atopic asthma was made in our study. Statistical Analysis Discrete-time survival analyses (Singer and Willett, 1993) with person-year as the unit of analysis were used to test sequential associations between first onset of mental disorders and subsequent onset of asthma. For these analyses, a person-year data set was created in which each year in the life of each respondent up to and including the age of onset of asthma or their age at interview (whichever came first) was treated as a separate observational record, with the year of asthma onset coded 1 and earlier years coded 0 on a dichotomous outcome variable. Persons who reported asthma onset before age 21 (N=2,186, corresponding to the 54% of all asthma cases in our study) were excluded from analysis. Mental disorder predictors were coded 1 from the year after first onset of each individual mental disorder. This time lag of 1 year in the coding of the predictors ensured that in cases where the first onset of a mental disorder and of asthma occurred in the same year, the mental disorder would not count as a predictor. Only person-years up to the diagnosis of asthma were analyzed so that only mental disorder episodes occurring prior to the onset of asthma were included in the predictor set. Logistic regression analysis was used to analyze these data with the survival coefficients presented as odds ratios, indicating the relative odds of asthma onset in a given year for a person with a prior history of mental disorder compared to a person without that mental disorder (including people without any mental disorder history). It is important to note that, given the retrospective nature of the information collected, results of the above analyses should be considered exploratory in nature. Longitudinal studies will be needed to confirm them. A series of bivariate and multivariate models was developed including the predictor mental disorder plus control variables. Models controlled for person-years, countries, gender, current age, and in the multivariate models, other mental disorders. Bivariate models investigated association of specific mental disorders with subsequent asthma onset. The next model, a multivariate model, estimated the associations of each mental disorder with asthma onset adjusting for mental disorder comorbidity. Our approach was to not control for covariates that could be on the causal pathway between mental disorders and subsequent asthma. However, we recognize that these variables may also confound associations so we re-estimated the multivariate model with adjustment for history of smoking (ever/never/current) and educational attainment. This made virtually no difference to associations (all previously significant associations remained significant and none reduced in magnitude – data available on request) so we report the results from the model unadjusted for smoking and education in this paper. We did not include BMI in the adjusted models, as estimates of current body weight only were available, and including this variable in the analyses would result in a biased estimation of the associations. Our earlier studies of concurrent mental-physical comorbidity in the WMH surveys found that these associations were generally consistent cross-nationally, despite varying prevalence of mental disorder and physical conditions (Lin et al. 2008; Von Korff, 2009). This was particularly the case for the association of mental disorders and adult onset asthma (Scott et al. 2007). All analyses for this paper were therefore run on the pooled cross-national dataset. As the WMH data are both clustered and weighted, the design-based Taylor series linearization (Shah, 1998) implemented in version 10 of the SUDAAN software system (SUDDAN, 1999) was used to estimate standard errors and evaluate the statistical significance of coefficients. RESULTS Sample information The survey characteristics are shown in Table 1 together with information about the number of survey respondents reporting a history of adult-onset asthma (n=1,860). They represented a total of 2,096,486 person-years of follow up. Four surveys had limited their eligible population to less than 65. The overall weighted response rate was 78%, ranging from 45.9% in France to 95.2 in Iraq. The prevalence of adult onset asthma ranged from 0.5% in the Shen Zen survey of People’s Republic of China to 7.4% in New Zealand. The mean age of adult asthma onset was 41 (SE=0.5) with a median age of onset of 38. Lifetime prevalence of mental disorders among those with and without asthma In Table 2 the lifetime prevalence rates of mental disorders are shown for the total sample without adult onset asthma (N= 49,296) and for those participants reporting adult onset asthma (final three columns). 20.8% of the adult onset asthma participants had lifetime MDE or dysthymia, 11.2% had specific phobia, 2% had intermittent explosive disorder, and 8.2%, alcohol abuse. Lifetime prevalence of mood and anxiety disorders among adult onset asthma cases was generally higher than in the total sample without adult onset asthma. It should be noted that the prevalence rates for mental disorders reported in this table represent the history of mental disorders occurring at any age up until the age of interview for each respondent. The analyses that follow include as predictors only those mental disorders occurring prior to the onset of asthma. Associations between temporally prior mental disorders and subsequent asthma onset Table 3 shows the associations between the first onset of temporally prior individual mental disorders and subsequent adult asthma onset. These associations were first investigated in a series of bivariate models (i.e., only one mental disorder considered at a time), but adjusted for age cohort, gender, person-year and country. First section of the table shows that 12 (out of the 16) mental disorders predicted adult onset asthma, with statistically significant ORs in the range of 1.5 to 2.8. Bipolar disorder showed the highest bivariate association (OR= 2.8) followed by binge eating disorder (OR= 2.6). The second section of table 3 shows that many bivariate associations remained significant, although attenuated, after adjusting for lifetime mental disorder comorbidity (that is, adjusting for mental disorders occurring before asthma onset). Dummy variables for all mental disorders were entered simultaneously and the highest multivariate associations observed correspond to binge eating disorder (OR= 1.8), bipolar disorder (OR= 1.8), PTSD (OR= 1.5), and alcohol abuse (OR= 1.5). Of interest, the observed association between major depressive episode/dysthymia became non-significant in the multivariate model. The global Chi square test for the joint effect of all mental disorders was statistically significant (X216= 140.3). That effect was not homogeneous for all mental disorders (X215=35.2). The final section in table 3 shows the coefficients of a multivariate model estimated with dummy predictors for the number of mental disorders without any information about the type of mental disorders, adjusted for age cohort, gender, person-year and country. The joint effect of the number of mental disorders is statistically significant X2=67.6 and a linear trend is observed, with an OR= 1.4 for exactly one mental disorder only to an OR= 3.1 for 5 or more mental disorders. We performed analyses with subgroups of countries (i.e., developed vs developing) and found that none of the interactions are significant indicating that there are no country income group differences in these associations. We also tested whether early onset (before age of 21) mental disorders were more strongly associated with subsequent adult asthma relative to late onset (after age of 21) mental disorders. Although we found that this was the case for most of the mental disorders, this difference between early and late onset became non-significant once all mental disorders were included in the models (results shown, but available on request). This suggests that the stronger association for early onset disorders with asthma is because early onset mental disorders are risk factors or risk markers for lifetime mental disorder comorbidity. Variation over the life-course (timing of asthma diagnoses) When we investigated whether associations between mental disorders and subsequent asthma varied by timing of the asthma diagnosis, we found significant interaction effects with person years for mood disorders (MDE/Dysthymia), and for GAD, Social phobia, and PTSD (Table 4). In all cases, the significant interactions indicate that the association of these mental disorders with adult asthma is stronger when asthma is diagnosed earlier in adulthood relative to later in adulthood. This is an important qualifier of the findings in Table 3 where the multivariate associations between MDE/dysthymia and Social phobia and asthma are not significant, as these results included adult asthma diagnosed throughout the whole adult lifespan. Finally, we assessed whether the association between mental disorders and adult onset asthma was different by gender (Table 5). We did find an interaction effect with gender for the association of two disorders with asthma: specific phobia, which was more strongly associated with subsequent asthma among females (stratified OR for female= 1.5) and for PTSD, which was more strongly associated with asthma among males (stratified OR for male= 2.9). DISCUSSION The results of this study must be interpreted taking into account several limitations. First, our assessment of asthma and mental disorders as well as their age of onset (AOO) is retrospective, which is associated with underestimates and errors (Wells and Horwood, 2004). Nevertheless, there is evidence that retrospective reported age of onset of asthma is reliable (Pattaro et al. 2007) and accurate (Toren et al. 2006). In addition, the instrument used for assessing mental disorders was specifically modified to improve accuracy in reporting age of onset, including decomposition of questions and bounding uncertainty interviewer techniques (Kessler and Ustun, 2004). Moreover, while neuroticism and distress may bias the reporting of disease symptoms (Janssens et al. 2009), they do not bias the reporting of diagnosed conditions (Vassend and Skrondal, 1999). Second, our assessment of asthma is based on self-report rather than on the most recommended strategy of combining medical history, physical examination and respiratory function tests (Douwes et al. 2011; Goodwin et al. 2003b; Pearce et al. 2000). But, as mentioned earlier, in our survey the definition of asthma was self-report of a diagnosis of asthma –not simply a self-report of asthma, a strategy that is associated with a lower bias (Baumeister et al. 2010; Kriegsman et al. 1996). Memory and diagnostic biases might have lead us to inaccurately classify some individuals in relation to their mental disorders and/or asthma status, and misclassification usually results in a bias towards the null, that is, an underestimation of the real association between mental disorders and asthma (Green et al. 2010; Loerbroks et al. 2012). Overall the retrospective nature of the data leads us to consider our results as preliminary evidence to be confirmed by longitudinal studies. Thirdly, diagnosis requires access to health care and there might be differences in access to health care among the countries and regions included in our study, which we have not studied. However the asthma prevalence found is consistent with previous studies (Akinbami et al. 2011; Goodwin et al. 2003b; Loerbroks et al. 2012), if we consider that we are analyzing adult onset asthma cases only, which correspond to a little less than half (46%) of the total asthma cases in our sample. Also a major strength of the WMH-surveys is that it contains a series of population-based samples in which all participants underwent the CIDI interview. The likelihood of mental disorders being diagnosed thus plays no role as every respondent is scored according to DSM criteria. But there may be substantial cultural differences in the likelihood of asthma being diagnosed. It is quite likely, for example, that in developing countries fewer people will get asthma diagnoses and care. This will result in a non-differential misclassification, that is, respondents being classified as non-asthma cases when in fact they are cases (undiagnosed). The effect of this is to bias associations towards the null. Finally, we did not explore asthma severity nor specific asthma symptoms, which are more strongly associated with depression and anxiety disorders than overall asthma (Opolski and Wilson, 2005). Notwithstanding these limitations, we believe our study provides valuable knowledge in several respects. Specifically, we confirm an important association of anxiety disorders and subsequent adult onset asthma (PTSD, panic, specific phobia and GAD) and report an even stronger association for bipolar, binge eating, and alcohol abuse disorders. We provide evidence for the first time that the number of comorbid mental disorders shows an additive association with subsequent asthma. We also show that the association of mental disorders and subsequent adult onset asthma is stronger for early onset mental disorders, and some associations are stronger when asthma occurs earlier rather than later in adulthood. Finally, we determine that for the most part associations between mental disorders and subsequent asthma onset are similar for men and women. This is an interesting finding given the higher prevalence of mood and anxiety disorders, and of adult-onset asthma, among women. The association between anxiety disorders and adult asthma is consistent with previous findings (Deshmukh et al. 2008; Goodwin et al. 2003a; Goodwin et al. 2004; Katon et al. 2004; ten and Petermann, 2000). Several studies have shown an association of adult asthma with panic (Goodwin et al. 2003b); (Favreau et al. 2014), with GAD (Goodwin et al. 2003a; Lavoie et al. 2011) and with PTSD (Katon et al. 2004; Kean et al. 2006; Spitzer et al. 2009; Spitzer et al. 2011), the anxiety disorders which showed the highest association in our study. Less evidence is available about the association of asthma and bipolar disorder (Goodwin et al. 2004) and eating disorders (Moreau et al. 2009; Stevenson, 2003). It is notable that binge eating was the single disorder most strongly associated with adult asthma in our study. This fact, together with a lack of association with bulimia, strongly suggests that obesity may be a causal mechanism. No doubt, this association deserves further research. We found a stronger association between specific phobia and asthma for women, and indeed associations between all anxiety disorders (with the notable exception of PTSD) and asthma were somewhat stronger for women, although this gender difference was only significant for specific phobia. This pattern is consistent with prior research suggesting that women have greater psychobiological reactivity to stress, and a more sensitized hypothalamic-pituitary-adrenal axis than men (Meewisse et al. 2007). In light of this research suggesting greater physiological stress reactivity in women, the stronger association we observed between PTSD and asthma for men is surprising (Olff et al. 2007). It may therefore be a chance finding, or it may reflect complex interactions between gender, traumatic stress and respiratory function that require further investigation to elucidate. A number of mechanisms have been suggested to explain the association between asthma and mental disorders. Among others, the symptomatic nature of asthma, which might be a life-threatening condition, would lead to developing some psychopathological manifestations (Goodwin et al. 2012; Kean et al. 2006); and asthma medication, in particular oral steroids, could cause psychological symptoms among asthma patients (Opolski and Wilson, 2005). These mechanisms assume that asthma would be the antecedent (Scott, 2009) leading to the development of mental disorders. But our study suggests that mental disorders can be the antecedent of subsequent adult-onset asthma, as it had been previously indicated (Chida et al. 2008; Goodwin et al. 2009). Possible pathways for such causation would include alterations in the autonomic nervous (sympathetic-adrenal-medullary -SAM- axis) and/or the neuroendocrine (hypothalamic-pituitary-adrenal –HPA- axis) systems associated to mental disorders, which would increase vulnerability to asthma onset; (Goodwin et al. 2012; Scott, 2009; Wright, 2005). The possibility of a causal role of mental disorders in adult onset asthma is reinforced by our observation that the association was stronger among those individuals whose mental disorder started during their childhood or early youth (early onset). It is also possible that certain behaviors or risk factors which are more frequent among individuals with mental disorders, such as smoking (McLeish et al. 2011) or obesity (Anto et al. 2010), among others, could mediate in the development of asthma in adulthood. It is important to note that the results presented here are adjusted by smoking history and education years. But we could not adjust by obesity due to lack of information regarding obesity prior to asthma onset. Overall our results are compatible with the antecedent model which proposes that mental disorders could, through their association with dysregulated stress physiology and lifestyle risk factors (Chida Y et al, 2008; Scott KM, 2009) contribute to the development of adult-onset asthma. Such mechanisms could occur in concert with others. The evidence of a possible “bidirectional” association (Chida et al. 2008) strongly suggests the existence of mechanisms that are common for both mental disorders and asthma (Van Lieshout and MacQueen, 2012). Inflammation would be one of such mechanisms, as it has been shown to underpin asthma and it has been suggested to have a role in depressive disorders (Dantzer et al. 2008); (Berk et al. 2013). Recently, a bidirectional pathway between depression and comorbid systemic illnesses has been proposed (Iwata et al. 2013; ten and Petermann, 2000), which could imply novel strategies for treating mental disorders. And as stress can affect the autonomous nervous system, its dysregulation could be a mechanism by which stress increases the risk of developing both asthma and emotional problems (Van Lieshout and MacQueen, 2012). Stress, endocrine and immune systems are strongly related and very likely mechanisms in such bidirectional relationship between mental disorders and asthma (Wright, 2005). Shared determinants acting very early in life, maybe in utero, would be also compatible with this bidirectional association. Family and genetic associations between asthma and depression have been described (Van Lieshout and MacQueen, 2012). Childhood adversities are a well-known risk factor for mental disorders, and they have been also found to be risk factors of physical conditions (Scott et al. 2011). More recently we showed that childhood adversities were a risk factor for adult onset asthma (Korkeila et al. 2012), but we also reported that associations between early onset mental disorders and subsequent onset of asthma were independent of childhood adversities. Further research on common pathways for mental disorders and adult onset asthma might bring clues to new therapeutic approaches. We found that early mental disorders (before age of 21) are more strongly associated with subsequent adult onset asthma, but many of these associations became non-significant when the presence of other mental disorders prior to asthma onset was taken into account. We interpret this as the result of earlier onset mental disorders being markers for mental comorbidity, rather than the timing of mental disorders per se having an effect on the risk of asthma. The important role of mental comorbidity is supported in our study by the evidence of a linear effect of mental comorbidity on subsequent asthma and by the fact that the statistical model including only the number of comorbid mental disorders had a better fit predicting adult onset asthma than any of the other more complex models analyzed here. Individuals with mental comorbidity should be considered a group with a higher risk of subsequent asthma. From an etiological perspective, the higher risk of asthma with each additional comorbid mental disorder suggests that pathways towards development of adult onset asthma are not totally overlapping. Our study adds to previous knowledge: a) a large sample of culturally different individuals worldwide; b) a wide range of mental disorders assessed with a face to face diagnostic interview; and c) clear, consistent association between mental disorders and subsequent adult-onset asthma, with adjustment for some obvious confounders. While the cross-sectional nature of the study is a major weakness, our study adds distinctly novel contributions with regard to a) and b) above in particular, with adjustment for some obvious confounders. In conclusion, in this international study we found that there is an association between a wide range of mental disorders and subsequent adult onset asthma, even after adjusting for mental disorder comorbidity. By analyzing only adult onset asthma cases, this helped ensuring the temporal order of the associations under investigation (from prior mental disorder to subsequent asthma). This also means that our results are generalizable only to adult onset asthma. Earlier onset mental disorders were more strongly associated with asthma than later onset disorders; this was explained by mental disorder comorbidity. Some mental disorders were more strongly associated with asthma when the asthma occurred earlier rather than later in adulthood. Associations were for the most part similar for men and women. The study also reveals the importance of comorbid mental disorders, as they increased risk for subsequent asthma. While pathological mechanisms implicated in the development of asthma and/or improving some of their outcomes among those with preexisting mental disorders should be better understood, the possibilities of avoiding asthma by intervening with potential psychological issues deserve further research. Table 1 Characteristics of WMH samples and proportion (and number) with history of adult-onset asthma. Sample Size History of Adult-Onset Asthma Diagnosis1 Country Field Dates Age Range Part 1 Part 2 sub- sample Response Rate (%) Number Unweighted (N) Weighted (%) Americas   Colombia 2003 18–65 4426 2381 87.7 23 0.7   Mexico 2001–2 18–65 5782 2362 76.6 22 0.7   United States 2002–3 18+ 9282 5692 70.9 324 5.6   Peru 2005–6 18–65 3930 1801 90.2 53 2.9 Asia and South Pacific   Japan 2002–6 20+ 4129 1682 55.1 57 2.7   PRC Shenzhen 2006–7 18+ 7132 2475 80.0 20 0.5   New Zealand 2003–4 18+ 12790 7312 73.3 544 7.4 Europe   Belgium 2001–2 18+ 2419 1043 50.6 26 2.4   France 2001–2 18+ 2894 1436 45.9 43 2.4   Germany 2002–3 18+ 3555 1323 57.8 40 3.3   Italy 2001–2 18+ 4712 1779 71.3 56 3.0   The Netherlands 2002–3 18+ 2372 1094 56.4 47 4.0   Spain 2001–2 18+ 5473 2121 78.6 77 3.0   Northern Ireland 2004–7 18+ 4340 1986 68.4 88 4.0   Portugal 2008–9 18+ 3849 2060 57.3 60 2.8   Romania 2005–6 18+ 2357 2357 70.9 83 2.7   Poland 2010–11 18–64 10081 4000 50.4 72 1.3 Middle East   Israel 2002–4 21+ 4859 4859 72.6 188 3.9   Iraq 2006–7 18+ 4332 4332 95.2 37 1.2 Weighted average response rate (%) 78.0 Total sample size 98,714 52,095 1,860 3.4 1 This is the onset of asthma in those 21 years and over. Table 2 Prevalence of mental disorders across WMH samples Types of mental disorders Among the total sample (21+ years) (n= 49296) Among those with adult-onset asthma (n = 1860) n % SE n % SE I. Mood disorders   Major Depressive Episode/Dysthymia 11639 12.9 0.2 615 20.8 1.0   Bipolar Disorder (Broad) 1561 1.8 0.1 98 3.2 0.4 II. Anxiety disorders   Panic Disorder 1670 1.9 0.1 126 4.2 0.5   Generalized Anxiety Disorder 3625 4.3 0.1 270 9.5 0.7   Social Phobia 3646 4.3 0.1 207 7.1 0.6   Specific Phobia 5814 7.2 0.1 354 11.2 0.8   Agoraphobia without Panic 751 0.9 0.0 38 1.2 0.2   Post-Traumatic Stress Disorder 2718 3.4 0.1 221 8.3 0.8   Obsessive Compulsive Disorder 729 1.2 0.1 17 0.4 0.1 III. Impulse-control disorders   Intermittent Explosive Disorder 1523 2.0 0.1 47 2.0 0.4   Binge Eating Disorder 530 0.7 0.0 39 1.5 0.3   Bulimia Nervosa 355 0.4 0.0 27 0.7 0.1 IV. Substance disorders   Alcohol Abuse 4806 7.5 0.2 216 8.2 0.7   Alcohol Dependence with Abuse 1627 2.1 0.1 97 3.3 0.4   Drug Abuse 1746 2.4 0.1 72 2.5 0.3   Drug Dependence with Abuse 678 0.9 0.1 34 1.0 0.2 Table 3 Bivariate and multivariate associations (odds ratios) between DSM-IV mental disorders and the subsequent diagnosis of adult-onset asthma. The WMH Surveys. Bivariate Models1 Multivariate Type Model2 Multivariate Number Model3 OR (95% C.I.) OR (95% C.I.) OR (95% C.I.) I. Mood disorders   Major Depressive Episode/ Dysthymia 1.6* (1.4–1.9) 1.2 (1.0–1.4) - -   Bipolar Disorder (Broad) 2.8* (2.0–3.9) 1.8* (1.3–2.5) - - II. Anxiety disorders   Panic Disorder 2.2* (1.6–3.1) 1.4* (1.0–2.0) - -   Generalized Anxiety Disorder 2.0* (1.6–2.4) 1.3* (1.1–1.7) - -   Social Phobia 1.5* (1.2–1.9) 1.0 (0.8–1.3) - -   Specific Phobia 1.7* (1.4–2.0) 1.3* (1.1–1.6) - -   Agoraphobia without Panic 1.3 (0.9–2.0) 0.8 (0.6–1.3) - -   Post-Traumatic Stress Disorder 2.1* (1.6–2.7) 1.5* (1.1–1.9) - -   Obsessive Compulsive Disorder 0.8 (0.4–1.7) 0.6 (0.3–1.1) - - III. Impulse-control disorders   Intermittent Explosive Disorder 1.5 (1.0–2.2) 1.0 (0.7–1.4) - -   Binge Eating Disorder 2.6* (1.7–3.9) 1.8* (1.2–2.9) - -   Bulimia Nervosa 1.7 (1.0–3.0) 1.0 (0.5–1.8) - - IV. Substance disorders   Alcohol Abuse 2.0* (1.6–2.4) 1.5* (1.1–2.0) - -   Alcohol Dependence with Abuse 2.5* (1.8–3.4) 1.3 (0.9–2.0) - -   Drug Abuse 1.7* (1.2–2.4) 0.9 (0.6–1.4) - -   Drug Dependence with Abuse 1.9* (1.2–3.1) 0.8 (0.5–1.4) - - Joint effect of all types of disorders, χ216 140.3* Difference between types of disorders, χ215 35.2* V. Number of disorders   Exactly 1 disorder - - - - 1.4* (1.2–1.6)   Exactly 2 disorders - - - - 1.6* (1.4–2.0)   Exactly 3 disorders - - - - 2.2* (1.7–2.9)   Exactly 4 disorders - - - - 2.9* (2.0–4.1)   5+ disorders - - - - 3.1* (2.2–4.2) Joint effect of number of disorders, χ25 67.6* * Significant at the 0.05 level, two-tailed test. 1 Bivariate models: each mental disorder type was estimated as a predictor of the physical condition onset in a separate discrete time survival model controlling for age cohorts, gender, person-year and country. 2 Multivariate Type model: the model was estimated with dummy variables for all mental disorders entered simultaneously, including the controls specified above and additionally adjusted for smoking (ever/never/current) and education (number of years). 3 Multivariate Number model: the model was estimated with dummy predictors for number of mental disorders without any information about type of mental disorders, including the controls specified above. Table 4 Interaction between mental disorder and person-year in predicting the subsequent diagnosis of adult-onset asthma. The WMH Surveys. Mental disorder*Person-year interaction1 Type of Mental Disorders OR (95% C.I.) χ21 [p] Major Depressive Episode/Dysthymia 0.97* (0.96–0.98) 49.4* [0.000] Generalized Anxiety Disorder 0.99* (0.97–1.00) 5.3* [0.021] Social Phobia 0.99* (0.98–1.00) 5.5* [0.019] Post-Traumatic Stress Disorder 0.98* (0.97–0.99) 12.1* [0.001] * OR significant at the 0.05 level, 2-sided test 1 A series of multivariate models were estimated. For example, the model for depression included the dummy variables for all mental disorders plus the cross-product term for depression and person-year (as a continuous variable), plus the controls specified for earlier models. Table 5 Multivariate type models for adult-onset asthma, stratified by gender. The WMH Surveys. Stratified Multivariate Type Models Type of Mental Disorders Male Female OR (95% C.I.) OR (95% C.I.) Specific Phobia 0.9 (0.6–1.3) 1.5* (1.2–1.7) Post-Traumatic Stress Disorder 2.9* (1.6–5.4) 1.2 (0.9–1.6) * Significant at the 0.05 level, two-tailed test. Reference List Akinbami LJ Moorman JE Liu X Asthma prevalence, health care use, and mortality: United States, 2005–2009 Natl. Health Stat. Report 2011 1 14 Alati R O'Callaghan M Najman JM Williams GM Bor W Lawlor DA Asthma and internalizing behavior problems in adolescence: a longitudinal study Psychosom. Med 2005 67 462 470 15911911 American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) 2000 Fourth Washington, DC American Psychiatric Association Anto JM Sunyer J Basagana X Garcia-Esteban R Cerveri I de MR Heinrich J Janson C Jarvis D Kogevinas M Kuenzli N Leynaert B Svanes C Wjst M Gislason T Burney P Risk factors of new-onset asthma in adults: a population-based international cohort study Allergy 2010 65 1021 1030 20132157 Baumeister H Kriston L Bengel J Harter M High agreement of self-report and physician-diagnosed somatic conditions yields limited bias in examining mental-physical comorbidity J. Clin. Epidemiol 2010 63 558 565 19959329 Berk M Williams LJ Jacka FN O'Neil A Pasco JA Moylan S Allen NB Stuart AL Hayley AC Byrne ML Maes M So depression is an inflammatory disease, but where does the inflammation come from? BMC. Med 2013 11 200 24228900 Chida Y Hamer M Steptoe A A bidirectional relationship between psychosocial factors and atopic disorders: a systematic review and meta-analysis Psychosom. Med 2008 70 102 116 18158379 Dantzer R O'Connor JC Freund GG Johnson RW Kelley KW From inflammation to sickness and depression: when the immune system subjugates the brain Nat. Rev. Neurosci 2008 9 46 56 18073775 Deshmukh VM Toelle BG Usherwood T O'Grady B Jenkins CR The association of comorbid anxiety and depression with asthma-related quality of life and symptom perception in adults Respirology 2008 13 695 702 18513245 Douwes J Brooks C Pearce N Asthma nervosa: old concept, new insights Eur. Respir. J 2011 37 986 990 21532014 Favreau H Bacon SL Labrecque M Lavoie KL Prospective impact of panic disorder and panic-anxiety on asthma control, health service use, and quality of life in adult patients with asthma over a 4-year follow-up Psychosom. Med 2014 76 147 155 24470131 Goodwin RD Bandiera FC Steinberg D Ortega AN Feldman JM Asthma and mental health among youth: etiology, current knowledge and future directions Expert. Rev. Respir. Med 2012 6 397 406 22971065 Goodwin RD Eaton WW Asthma, suicidal ideation, and suicide attempts: findings from the Baltimore epidemiologic catchment area follow-up Am. J. Public Health 2005 95 717 722 15798135 Goodwin RD Fergusson DM Horwood LJ Asthma and depressive and anxiety disorders among young persons in the community Psychol. Med 2004 34 1465 1474 15724877 Goodwin RD Jacobi F Thefeld W Mental disorders and asthma in the community Arch. Gen. Psychiatry 2003a 60 1125 1130 14609888 Goodwin RD Pine DS Hoven CW Asthma and panic attacks among youth in the community J. Asthma 2003b 40 139 145 12765315 Goodwin RD Robinson M Sly PD McKeague IW Susser ES Zubrick SR Stanley FJ Mattes E Severity and persistence of asthma and mental health: a birth cohort study Psychol. Med 2013 43 1313 1322 23171853 Goodwin RD Sourander A Duarte CS Niemela S Multimaki P Nikolakaros G Helenius H Piha J Kumpulainen K Moilanen I Tamminen T Almqvist F Do mental health problems in childhood predict chronic physical conditions among males in early adulthood? Evidence from a community-based prospective study Psychol. Med 2009 39 301 311 18507873 Green JG McLaughlin KA Berglund PA Gruber MJ Sampson NA Zaslavsky AM Kessler RC Childhood adversities and adult psychiatric disorders in the national comorbidity survey replication I: associations with first onset of DSM-IV disorders Arch. Gen. Psychiatry 2010 67 113 123 20124111 Haro JM Arbabzadeh-Bouchez S Brugha TS de GG Guyer ME Jin R Lepine JP Mazzi F Reneses B Vilagut G Sampson NA Kessler RC Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health surveys Int. J. Methods Psychiatr. Res 2006 15 167 180 17266013 Hasler G Gergen PJ Kleinbaum DG Ajdacic V Gamma A Eich D Rossler W Angst J Asthma and panic in young adults: a 20-year prospective community study Am. J. Respir. Crit Care Med 2005 171 1224 1230 15764721 Iwata M Ota KT Duman RS The inflammasome: pathways linking psychological stress, depression, and systemic illnesses Brain Behav. Immun 2013 31 105 114 23261775 Janssens T Verleden G De PS Van DI Van den Bergh O Inaccurate perception of asthma symptoms: a cognitive-affective framework and implications for asthma treatment Clin. Psychol. Rev 2009 29 317 327 19285771 Jenkins MA Hopper JL Bowes G Carlin JB Flander LB Giles GG Factors in childhood as predictors of asthma in adult life BMJ 1994 309 90 93 8038673 Jiang CQ Loerbroks A Lam KB Bosch JA Thomas GN Zhang WS Cheng KK Lam TH Adab P Mental HEALTH AND asthma in China: the Guangzhou Biobank Cohort Study Int. J. Behav. Med 2013 20 259 264 22297917 Katon WJ Richardson L Lozano P McCauley E The relationship of asthma and anxiety disorders Psychosom. Med 2004 66 349 355 15184694 Kean EM Kelsay K Wamboldt F Wamboldt MZ Posttraumatic stress in adolescents with asthma and their parents J. Am. Acad. Child Adolesc. Psychiatry 2006 45 78 86 16327584 Kessler RC Ustun TB The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders 2008 New York Cambridge University Press Kessler RC Ustun TB The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI) Int. J. Methods Psychiatr. Res 2004 13 93 121 15297906 Korkeila J Lietzen R Sillanmaki LH Rautava P Korkeila K Kivimaki M Koskenvuo M Vahtera J Childhood adversities and adult-onset asthma: a cohort study BMJ Open 2012 2 Kriegsman DM Penninx BW van Eijk JT Boeke AJ Deeg DJ Self-reports and general practitioner information on the presence of chronic diseases in community dwelling elderly. A study on the accuracy of patients' self-reports and on determinants of inaccuracy J. Clin. Epidemiol 1996 49 1407 1417 8970491 Lavoie KL Boudreau M Plourde A Campbell TS Bacon SL Association between generalized anxiety disorder and asthma morbidity Psychosom. Med 2011 73 504 513 21715294 Lin EH Korff MV Alonso J Angermeyer MC Anthony J Bromet E Bruffaerts R Gasquet I de GG Gureje O Haro JM Karam E Lara C Lee S Levinson D Ormel JH Posada-Villa J Scott K Watanabe M Williams D Mental disorders among persons with diabetes--results from the World Mental Health Surveys J. Psychosom. Res 2008 65 571 580 19027447 Loerbroks A Herr RM Subramanian S Bosch JA The association of asthma and wheezing with major depressive episodes: an analysis of 245 727 women and men from 57 countries Int. J. Epidemiol 2012 41 1436 1444 22879363 McLeish AC Cougle JR Zvolensky MJ Asthma and cigarette smoking in a representative sample of adults J. Health Psychol 2011 16 643 652 21346012 Meewisse ML Reitsma JB de Vries GJ Gersons BP Olff M Cortisol and post-traumatic stress disorder in adults: systematic review and meta-analysis Br. J. Psychiatry 2007 191 387 392 17978317 Moreau D Kalaboka S Choquet M Annesi-Maesano I Asthma, obesity, and eating behaviors according to the diagnostic and statistical manual of mental disorders IV in a large population-based sample of adolescents Am. J. Clin. Nutr 2009 89 1292 1298 19321566 NCHS Evaluation of National Health Interview Survey diagnostic reporting Vital Health Stat 1994 2 1 116 Olff M Langeland W Draijer N Gersons BP Gender differences in posttraumatic stress disorder Psychol. Bull 2007 133 183 204 17338596 Opolski M Wilson I Asthma and depression: a pragmatic review of the literature and recommendations for future research Clin. Pract. Epidemiol. Ment. Health 2005 1 18 16185365 Pattaro C Locatelli F Sunyer J de MR Using the age at onset may increase the reliability of longitudinal asthma assessment J. Clin. Epidemiol 2007 60 704 711 17573986 Pearce N Douwes J Beasley R The rise and rise of asthma: a new paradigm for the new millennium? J. Epidemiol. Biostat 2000 5 5 16 10885867 Perna G Bertani A Politi E Colombo G Bellodi L Asthma and panic attacks Biol. Psychiatry 1997 42 625 630 9376459 Ramos Olazagasti MA Shrout PE Yoshikawa H Bird HR Canino GJ The longitudinal relationship between parental reports of asthma and anxiety and depression symptoms among two groups of Puerto Rican youth J. Psychosom. Res 2012 73 283 288 22980534 Sanna L Stuart AL Pasco JA Jacka FN Berk M Maes M O'Neil A Girardi P Williams LJ Atopic disorders and depression: findings from a large, population-based study J. Affect. Disord 2014 155 261 265 24308896 Scott KM Von KM Alonso J Angermeyer MC Benjet C Bruffaerts R de GG Haro JM Kessler RC Kovess V Ono Y Ormel J Posada-Villa J Childhood adversity, early-onset depressive/anxiety disorders, and adult-onset asthma Psychosom. Med 2008 70 1035 1043 18941133 Scott KM Von KM Angermeyer MC Benjet C Bruffaerts R de GG Haro JM Lepine JP Ormel J Posada-Villa J Tachimori H Kessler RC Association of childhood adversities and early-onset mental disorders with adult-onset chronic physical conditions Arch. Gen. Psychiatry 2011 68 838 844 21810647 Scott KM Von KM Ormel J Zhang MY Bruffaerts R Alonso J Kessler RC Tachimori H Karam E Levinson D Bromet EJ Posada-Villa J Gasquet I Angermeyer MC Borges G de GG Herman A Haro JM Mental disorders among adults with asthma: results from the World Mental Health Survey Gen. Hosp. Psychiatry 2007 29 123 133 17336661 Scott K Von Korff MR Scott KM Gureje O The development of mental-physical comorbidity Global Perspectives on Mental-Physical Comorbidity 2009 UK WHO: Cambridge University Press 84 94 Shah BV Armitage P Colton T Linearization methods of variance estimation Encyclopedia of Biostatistics 1998 Chichester John Wiley and Sons 2276 2279 Shavitt RG Gentil V Mandetta R The association of panic/agoraphobia and asthma. Contributing factors and clinical implications Gen. Hosp. Psychiatry 1992 14 420 423 1473713 Singer JD Willett JB It's about time: Using discrete-time survival analysis to study duration and the timing of events Journal of Educational Statistics 1993 18 155 195 Spitzer C Barnow S Volzke H John U Freyberger HJ Grabe HJ Trauma, posttraumatic stress disorder, and physical illness: findings from the general population Psychosom. Med 2009 71 1012 1017 19834051 Spitzer C Koch B Grabe HJ Ewert R Barnow S Felix SB Ittermann T Obst A Volzke H Glaser S Schaper C Association of airflow limitation with trauma exposure and post-traumatic stress disorder Eur. Respir. J 2011 37 1068 1075 20729219 Stevenson J Relationship between behavior and asthma in children with atopic dermatitis Psychosom. Med 2003 65 971 975 14645774 SUDDAN SUDDAN Software for the statistical analysis of correlated data (computer program) Anonymous 1999 North Carolina, USA Research Triangle Park ten TC Petermann F Reviewing asthma and anxiety Respir. Med 2000 94 409 415 10868701 Toren K Palmqvist M Lowhagen O Balder B Tunsater A Self-reported asthma was biased in relation to disease severity while reported year of asthma onset was accurate J. Clin. Epidemiol 2006 59 90 93 16360566 Van Lieshout RJ MacQueen GM Bienenstock J Relations between asthma and psychological distress: an old idea revisited Allergy and the nervous system. Chem Immunol Allergy 2012 98 Basel Karger 1 13 Vassend O Skrondal A The Role of Negative Affectivity in Self assessment of Health: A Structural Equation Approach J. Health Psychol 1999 4 465 482 22021640 Von Korff MR Von Korff MR Scott KM Gureje O Global perspectives on mental-physical comorbidity Global perspectives on mental-physical comorbidity in the WHO World Mental Health Surveys 2009 New York, NY Cambridge University Press 1 11 Wainwright NW Surtees PG Wareham NJ Harrison BD Psychosocial factors and incident asthma hospital admissions in the EPIC-Norfolk cohort study Allergy 2007 62 554 560 17441796 Wells JE Horwood LJ How accurate is recall of key symptoms of depression? A comparison of recall and longitudinal reports Psychol. Med 2004 34 1001 1011 15554571 Wong KO Hunter RB Douwes J Senthilselvan A Asthma and wheezing are associated with depression and anxiety in adults: an analysis from 54 countries Pulm. Med 2013 2013 929028 23577252 Wright RJ Stress and atopic disorders J. Allergy Clin. Immunol 2005 116 1301 1306 16337463

PMC005xxxxxx/PMC5120393.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0376333 5218 J Psychosom Res J Psychosom Res Journal of psychosomatic research 0022-3999 1879-1360 26526305 5120393 10.1016/j.jpsychores.2015.08.005 NIHMS830490 Article Associations between DSM-IV mental disorders and subsequent COPD diagnosis Rapsey Charlene M. a* Lim Carmen C.W. a Al-Hamzawi Ali b Alonso Jordi c Bruffaerts Ronny d Caldas-de-Almeida J.M. e Florescu Silvia fg de Girolamo Giovanni h Hu Chiyi i Kessler Ronald C. j Kovess-Masfety Viviane k Levinson Daphna l Elena Medina-Mora María m Murphy Sam n Ono Yutaka o Piazza Maria p Posada-Villa Jose q ten Have Margreet r Wojtyniak Bogdan g Scott Kate M. a a Department of Psychological Medicine, University of Otago, Dunedin, New Zealand b College of Medicine, Al-Qadisiya University, Diwania governorate, Iraq c IMIM-Institut Hospital del Mar d'Investigacions Mèdiques) CIBER en Epidemiolgía y Salud Pública (CIBERESP) UPF-Pompeu Fabra University, Spain d University Psychiatric Centre, University of Leuven, Campus Gasthuisberg, Belgium e Chronic Diseases Research Center (CEDOC) and Department of Mental Health, Faculdade de Ciências Médicas, Universidade Nova de Lisboa, Portugal f National School of Public Health, Management and Professional Development, Bucharest, Romania g Centre of Monitoring and Analyses of Population Health, National Institute of Public Health-National Institute of Hygiene, Poland h IRCCS Centro S. Giovanni di Dio Fatebenefratelli, Brescia, Italy i Shenzhen Insitute of Mental Health & Shenzhen Kanging Hospital, PRC - Shenzhen j Department of Health Care Policy, Harvard Medical School, Boston, MA, United States k Ecole des Hautes Etudes en Santé Publique (EHESP), EA 4057 Paris Descartes University, Paris, France l Ministry of Health Israel, Mental Health Services, Israel m Nacional Institute of Psychiatry Ramon de la Fuente, Mexico n School of Psychology, University of Ulster, Northern Ireland o Center for Cognitive Behavior Therapy and Research, National Center for Neurology and Psychiatry, Japan p Unit of Analysis and Generation of Evidence for Public Health, Peruvian National Institute of Health, Peru q Colegio Mayor de Cundinamarca University, Colombia r Trimbos-Instituut, Netherlands Institute of Mental Health and Addiction, Netherlands * Corresponding author at: Department of Psychological Medicine, University of Otago, PO Box 913, Dunedin, New Zealand. charlene.rapsey@otago.ac.nz (C.M. Rapsey). 17 11 2016 02 9 2015 11 2015 23 11 2016 79 5 333339 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objectives COPD and mental disorder comorbidity is commonly reported, although findings are limited by substantive weaknesses. Moreover, few studies investigate mental disorder as a risk for COPD onset. This research aims to investigate associations between current (12-month) DSM-IV mental disorders and COPD, associations between temporally prior mental disorders and subsequent COPD diagnosis, and cumulative effect of multiple mental disorders. Methods Data were collected using population surveys of 19 countries (n = 52,095). COPD diagnosis was assessed by self-report of physician's diagnosis. The World Mental Health-Composite International Diagnostic Interview (WMH-CIDI) was used to retrospectively assess lifetime prevalence and age at onset of 16 DSM-IV disorders. Adjusting for age, gender, smoking, education, and country, survival analysis estimated associations between first onset of mental disorder and subsequent COPD diagnosis. Results COPD and several mental disorders were concurrently associated across the 12-month period (ORs 1.5–3.8). When examining associations between temporally prior disorders and COPD, all but two mental disorders were associated with COPD diagnosis (ORs 1.7–3.5). After comorbidity adjustment, depression, generalized anxiety disorder, and alcohol abuse were significantly associated with COPD (ORs 1.6–1.8). There was a substantive cumulative risk of COPD diagnosis following multiple mental disorders experienced over the lifetime. Conclusions: Mental disorder prevalence is higher in those with COPD than those without COPD. Over time, mental disorders are associated with subsequent diagnosis of COPD; further, the risk is cumulative for multiple diagnoses. Attention should be given to the role of mental disorders in the pathogenesis of COPD using prospective study designs. Alcohol abuse Anxiety disorders Comorbidity COPD Depression Introduction Mental disorders are highly prevalent among those with COPD. [1–3] Mental disorder and COPD comorbidity may negatively affect adaptive functioning, quality of life, exercise capacity, treatment adherence, and mortality. [4,5] A review of studies published between 1968 and 2004 reported that in those with COPD, prevalence of depression ranged from 7%-79% and prevalence of anxiety ranged from 10%–100%; [2] a review of studies between 1966 and 2012, which only included studies using a clinical diagnostic tool to assess mental disorder, found rates ranging between 0%–42% and 10%–55% respectively. [1]. The research investigating mental disorder and COPD comorbidity prevalence is limited by multiple methodological weaknesses including use of small samples, clinical populations of patients with COPD, and symptom rating scales rather than diagnostic clinical measures. Symptom rating scales for mental disorders do not differentiate between diagnostic categories, sometimes overlap with COPD, and may describe negative emotional responses to COPD. Thus, it is not surprising that estimates of the prevalence of mental disorder and COPD comorbidity vary widely. Cross-sectional analysis of general population samples reveals elevated risk of mental disorder in those with COPD. [6–8] However, it is difficult to conclude whether there is an association between COPD and some mental disorders but not others; whether there are differences in the relative strength of association between different mental disorders and COPD, or whether there is a non-specific association between COPD and any mental disorder (that is, mental distress generally). Moreover, prospective studies that identify the direction of the association between COPD and mental disorders are needed. Understanding the direction of the association between COPD and mental disorders may help clarify the mechanisms that connect these debilitating conditions. The majority of prospective studies have been conducted with participants with COPD at baseline and have determined that mental disorders increase poor outcomes in those with COPD. [3] However, prospective studies investigating mental disorder as a risk factor for COPD onset are lacking. One study provides evidence toward the hypothesis that mental disorder increases risk for COPD, finding that depression doubled the risk for later COPD diagnosis [9]. The World Mental Health (WMH) Surveys comprise cross-national data drawn from the general populations of a range of developed and developing countries, clinically valid assessment of lifetime prevalence of a wide range of DSM-IV mental disorders, and self-reported physician's diagnosis and year of diagnosis of chronic lung disease (referred to herein as COPD). The surveys are cross-sectional in design but collected information retrospectively on age of diagnosis of mental disorders and of COPD, which allows the use of survival analysis to examine associations between temporally prior mental disorders and the subsequent diagnosis of COPD. The aims of the study were three-fold. First, we estimated concurrent (12 month) associations between lifetime COPD and a wide range of 12 month DSM-IV mental disorders. Second, we investigated associations between first onset of temporally prior mood, anxiety, impulse control, and substance use disorders with subsequent COPD diagnosis, with and without adjustment for mental disorder comorbidity. Third, we investigated whether there was a cumulative effect of multiple mental disorders and subsequent risk of COPD. Adjustment was made for smoking in all analyses. Method Samples and procedures This study uses data from 19 of the WMH surveys (see Table 1). The World Mental Health (WMH) Survey Initiative is a project of the World Health Organization aimed at addressing the global burden of mental disorders. [10,11] A stratified multi-stage clustered area probability sampling strategy was used to select adult respondents (18 years +) in most WMH countries. In most countries, internal subsampling was used to reduce respondent burden and average interview time by dividing the interview into two parts. All respondents completed Part 1 which included the core diagnostic assessment of most mental disorders. All Part 1 respondents who met lifetime criteria for any mental disorder and a probability sample of respondents without mental disorders were administered Part 2 which assessed physical conditions and collected a range of other information related to survey aims. Part 2 respondents were weighted by the inverse of their probability of selection for Part 2 of the interview to adjust for differential sampling, resulting in an unbiased sample. Analyses in this paper are based on the weighted Part 2 subsample (n = 52,095; person years = 2,167,404). Additional weights were used to adjust for differential probabilities of selection within households, to adjust for non-response, and to match the samples to population socio-demographic distributions. Measures taken to ensure data accuracy, cross-national consistency and protection of respondents are described in detail elsewhere. [11,12] All respondents provided informed consent and procedures for protecting respondents were approved and monitored for compliance by the Institutional Review Boards in each country (see [11] for details). Measures Mental disorders All surveys used the WMH survey version of the WHO Composite International Diagnostic Interview (now CIDI 3.0, [12]) a fully structured interview, to assess lifetime history of mental disorders. Retrospective age-of-onset reports were based on a question series designed to avoid the implausible response patterns obtained in using the standard CIDI age-of-onset question. [13] Disorders were assessed using the definitions and criteria of the DSM-IV. The mental disorders adjusted for in this paper include anxiety disorders, mood disorders, substance use disorders, and impulse control disorders. CIDI organic exclusion rules were applied in making diagnoses. Clinical reappraisal studies conducted in some of the WMH countries indicate that lifetime diagnoses of anxiety, mood and substance use disorders based on the CIDI have generally good concordance with diagnoses based on blinded clinical interviews. [14]. COPD status In a series of questions adapted from the U.S Health Interview Survey, [15] respondents were asked about the lifetime presence of selected chronic conditions. Respondents were asked: “Did a doctor or other health professional ever tell you that you had any of the following illnesses… Other chronic lung disease, like COPD or emphysema?” If respondents endorsed this question they were classified as having a history of COPD for these analyses. Asthma was not included in this category because it was asked about in a separate question that preceded this ‘other chronic lung disease’ question. Although it is possible that some respondents with chronic lung diseases other than COPD/emphysema may have endorsed this question we use the term COPD as this is likely to comprise the majority of cases captured by this question. Respondents were also asked how old they were when they were first diagnosed with COPD. Only adult-onset COPD (onsets age 21 +) was investigated in this paper on the assumption that pre-adult onsets would reflect congenital processes, rather than any possible influence of temporally prior mental disorders. Smoking status Smoking was assessed with one item and respondents were classified according to three-levels, current smoker, ex-smoker, or never smoked. Statistical analysis Aim 1 The prevalence of specific mental disorders was estimated separately among respondents with and respondents without a COPD diagnosis. The ORs of the associations between lifetime COPD and specific 12 month DSM-IV mental disorders were calculated in logistic regression equations that adjusted for age, gender, country, smoking, and years of education. Aim 2 Although the data collection was cross-sectional, there was a time element in the data as we asked for the time of onset of mental disorders and of COPD. We analyzed the associations between mental disorder and COPD diagnosis by using this time-related information. Comparable with previous studies [16,17] using these data we used discrete-time survival analyses [19,20] with person-year as the unit of analysis to investigate sequential associations between first onset of mental disorders and the subsequent diagnosis of COPD. For these analyses a person-year data set was created in which each year in the life of each respondent up to and including the age of diagnosis of COPD or their age at interview (whichever came first) was treated as a separate observational record, with the year of COPD diagnosis coded 1 and earlier years coded 0 on a dichotomous outcome variable. As stated earlier, we were interested in adults with a COPD diagnosis over the age of 20, therefore the small number of people who reported COPD diagnosis before age 21 were excluded from the analyses. Mental disorder predictors were coded 1 from the year after first diagnosis of each individual mental disorder. This time lag of 1 year in the coding of the predictors ensured that in cases where the first diagnosis of a mental disorder and of COPD occurred in the same year, the mental disorder would not count as a predictor. Only person-years up to the diagnosis of COPD were analyzed so that only mental disorder episodes occurring prior to the diagnosis of COPD were included in the predictor set. Logistic regression analysis was used to estimate associations with the survival coefficients presented as odds ratios, indicating the relative odds of COPD diagnosis in a given year for a person with a prior history of the specific mental disorder compared to people without that mental disorder and people without any mental disorder history at all. First, a series of single disorder multivariate models were developed including the predictor mental disorder plus control variables to investigate whether each individual mental disorder contributed to an accelerated diagnosis of COPD. Second, a multi disorder model was developed including all mental disorders thereby adjusting for comorbidity to investigate the contribution of each mental disorder over and above the effect of other types of mental disorder. Aim 3 To investigate the cumulative effect of an increasing number of mental disorders, discrete time survival analysis was used to estimate a model that included number of disorders coded as dummy variable (1, 2, 3, 4, 5 +) without information about type of disorder. For example, those with exactly 1 disorder were coded 1 and the rest of the sample was coded 0 in the exactly 1 disorder category. All models control for countries, gender, current age, smoking, education level, and cohort (defined by ages at interview 18–29, 30–44, 45–59, 60 +). Our earlier studies of concurrent mental-physical comorbidity in the WMH surveys found that these associations are generally consistent cross-nationally, despite varying prevalence of mental disorder and physical conditions. [10,21] All analyses for this paper were therefore run on the pooled cross-national dataset. As the WMH data are both clustered and weighted, the design-based Taylor series linearization [22] implemented in version 10 of the SUDAAN software system [23] was used to estimate standard errors and evaluate the statistical significance of coefficients. Results The survey characteristics are shown in Table 1 together with information about the number of survey respondents reporting a history of COPD (n = 790). Prevalence of COPD diagnosis ranged from 0.1 (Mexico) to 3.4 (Israel) with an averaged prevalence across all countries of 1.3%. It should be noted that prevalence of COPD diagnosis will be underestimated in those 4 countries with upper age limits of 65 relative to the other countries with unrestricted upper ages of respondents. Prevalence of COPD diagnosis is also likely to vary considerably according to country differences in health service availability and assessment procedures. The prevalence of COPD reported in this study is lower than other studies, for example in a US study using a similar method for assessing COPD, the prevalence of COPD was 6.0%. [24]. Twelve-month prevalence of mental health disorders by lifetime COPD status and concurrent comorbidity As can be seen in Table 2, across a 12-month period, most mood and anxiety disorders were associated with a 50–220% (OR: 1.5–3.2) increased risk of COPD. The impulse control disorders and most substance use disorders were not associated with significantly elevated risk for COPD, excepting drug abuse with dependence, which was associated with a 280% increased risk of COPD diagnosis. Type and number of mental disorders as predictors of subsequent COPD diagnosis Results of analysis examining associations between temporally prior mental disorders and subsequent COPD are presented in Table 3. In the single disorder models, with no adjustment for comorbid disorders, all but two mental disorders (binge eating disorder and bulimia nervosa) are associated with COPD with ORs between 1.7–3.5. In the multi-disorder model, with adjustments for mental disorder comorbidity, the magnitude of associations was reduced (ORs from 1.6–1.8). Depression/dysthymia, generalized anxiety disorder, and alcohol abuse continued to predict diagnosis of COPD over and above the effect of the other disorders. The global chi square test for the joint effect of all mental disorders was significant (χ162 = 112.9, p < 0.001), however the test for variation in ORs indicates that the associations do not differ significantly in magnitude (χ152 = 17.4, p = 0.298). A cautious interpretation of this latter test result would be that we have found a generalized link between psychopathology and COPD diagnosis, with some suggestion that depression, anxiety, and alcohol abuse may have specific associations with COPD, but these specific relationships would require confirmation in subsequent studies. Table 4 displays the clear dose–response relationship between number of mental disorders experienced over the life course (prior to COPD diagnosis) and an accelerated COPD diagnosis. At each level of number of mental disorders (1, 2, 3, 4, 5 +) a greater proportion of those with COPD experienced mental disorder (18.3%, 9.0%, 6.2%, 1.9%, 3.8%) compared to those without COPD (16.0%, 6.6%, 2.8%, 1.4%, 1.6%). Further, the results from a multivariate model that considered only number of mental disorders (i.e., not including information about type) are presented in the final columns of data in Table 3., with ORs ranging from 1.6 for one mental disorder to 5.8 for 5 + mental disorders. This model was a better fit for the data than the multivariate type model just presented, again reinforcing the idea that it is psychopathology in general that matters most in terms of increasing COPD risk rather than specific types of mental disorders. Additional analyses Multivariate type and number models were developed; other more complex non additive multivariate models were also run, for example including both type and number of mental disorders in the same model, but model fit statistics did not indicate these provided a better fit for the data, so the simpler models are reported here (model fitting statistics available on request). Gender differences were examined by including interaction terms between gender and each mental disorder in the multivariate type model but as results were substantively the same for men and women we report results with adjustment for gender. Minimal changes were observed in the associations between mental disorders and COPD diagnosis after adjustment for nicotine use (data not shown but available on request). Discussion This is the first study evaluating the association of a wide range of DSM-IV mental disorders with COPD diagnosis in an international sample. The main finding were that there was an increased likelihood of COPD and mental disorder comorbidity across a 12-month period; those with mood, anxiety, and drug abuse with dependence were 50–280% more likely to also have a COPD diagnosis, which addressed the first aim of the study. The second aim was to investigate associations between temporally prior mental disorders and subsequent COPD diagnosis; significant associations between most mental disorders and COPD were found. After adjusting for comorbidity, depression, generalized anxiety disorder, and alcohol abuse contributed independently to an increased risk of COPD diagnosis at each time point (year of life). Third, there was a marked dose response for cumulative number of mental disorders experienced over the lifetime and the risk of COPD. The conclusions from this study are constrained by several caveats. COPD status was assessed by self-report of medical diagnosis without verification by clinical assessment and by retrospective report, which may have led to errors in classification and reported age of diagnosis timing. [13] Prevalence of COPD in this study appeared lower than previously reported [24] and rates varied between countries. Moreover, in four countries participation was limited to individuals under the age of 65, which likely excluded the greater proportion of individuals with COPD in those localities. Furthermore, COPD is likely to be markedly under-diagnosed in the general population whereby those who meet criteria for COPD have not received a medical diagnosis. [25] Thus, it is probable that classification errors were made whereby some of those with COPD were classified as not having COPD. Mood has not been found to influence reports of clinician diagnosed conditions [26,27] therefore it is unlikely that current mood disorder moderated self-report of COPD. The result of non-differential, under-reporting would be to weaken possible associations between variables and therefore the findings from this study may be conservative. The age of mental disorder onset distributions from the WMH surveys are consistent across countries and with other research; [28,29] however, some degree of inaccuracy with the precise timing of onset is likely to remain. Mitigating this, the validity of the survival analysis depends more on the temporal sequence of the mental disorders and COPD being correct, and less on accuracy in their precise onset timing. For most individuals their mental disorders and diagnosis of COPD occurred decades apart, thus the temporal sequence is likely to be correct. To further ensure this though, we do not include as predictors any mental disorders that were reported to occur in the same year as their COPD diagnosis. Selective classification errors whereby those most affected by mental disorders and COPD are absent from the sample due to hospitalization or early mortality would also influence associations in the direction of a null finding. Finally, this is a retrospective study and although the information on timing of COPD diagnosis and mental disorder onset allowed the estimation of predictive associations that are indicative of specific temporal associations from mental disorder to COPD, these results are regarded as preliminary and await confirmation from prospective studies. Despite its limitations, this study has a number of unique characteristics that allow for novel contributions to the literature. Controlling for cultural differences was possible through multi-national participation. Selection bias was reduced through surveying the general population. The sample size was sufficient to consider possible associations between COPD and a range of mental disorders, including disorders with lower prevalence rates. Mental disorders were assessed using a clinical instrument allowing for discrimination between diagnostic categories and for adjustment for comorbidity among disorders. Although as noted, prospective studies will be required to confirm the study findings, it is worth considering the potential causal pathways that mental disorders have in common and how these may increase susceptibility to COPD. One direct, biological mechanism that may connect mental disorders with COPD onset is raised inflammatory response and impaired immune regulation. [5] The immune irregularities in depression are complex and there is evidence for both suppression of cellular immunity through HPA axis mediated hyper-secretion of cortisol, but also of immune activation marked by elevated circulating levels of pro-inflammatory cytokines. [30] To the extent that depression does increase inflammation, it may contribute to the development of COPD in conjunction with other COPD pathogenic factors. Factors that contribute to increased inflammatory response that are common to mental disorders, such as disrupted sleep, may also add to the risk of COPD diagnosis. [6,31]. Smoking is another factor common to mental disorders and to COPD, however minimal changes were observed in the associations between mental disorders and COPD diagnosis after adjustment for nicotine use. Consistent with this, adjustment for smoking did not markedly attenuate the relationship between depression and COPD in other studies. [7,9] Goodwin et al. [8] has argued that it is nicotine dependence rather than nicotine use that is explanatory. Thus our limited assessment of nicotine use may have influenced our findings. The current study contributes to accumulating evidence that alcohol is involved in the pathogenesis of COPD, independent of nicotine exposure; alcohol abuse doubled the risk of COPD diagnosis after taking into account smoking and other mental disorders. Interestingly, there was no concurrent association between alcohol use and COPD. Alcohol use may no longer be present in older age when COPD is present; however it may be a risk factor for the development of COPD. Mechanisms that explain the role of heavy alcohol use in impaired lung function include the effects of alcohol and its metabolites and nonalcohol congeners, leading to immunosuppressive effects and compromised mucociliary clearance. [32] Perhaps surprisingly alcohol abuse but not alcohol dependence was associated with an increased risk for COPD. One speculative explanation for this finding may be that those who tolerate alcohol to the extent of developing dependence are genetically less susceptible to alcohol related lung pathology. For example, Asian populations are more likely to have a reduced ability to metabolize alcohol and one of several undesirable results is alcohol triggered asthma. [33]. Although we have discussed possible causal mechanisms thus far, we do not assume these associations are causal as there are also non-causal pathways, common to mental disorder and to COPD, that may explain the association between these conditions such as nutrition and exercise, [5] genetics, [34] and childhood adversities. [35] Moreover, COPD is comorbid with multiple conditions including osteoporosis and coronary heart disease; [5] it may be that diagnosis of other chronic conditions precedes diagnosis of COPD and that the associations we document here reflect in part the associations between mental disorders and these other conditions, or between mental disorders and multiple physical condition comorbidities. Another non-causal pathway may be socio-economic status whereby those with lower SES are more likely to experience mental disorders [36] and are also at greater risk for COPD through several pathways including occupational exposure to lung irritants, exposure to biomass smoke, and poor nutrition and housing. [37] In this study, adjustment was made for education level, which may not have fully controlled for SES. One further conceivable non-causal connection between mental disorders and COPD diagnosis is that seeking medical attention for mental disorders increases the likelihood of receiving a diagnosis of COPD. For example, panic disorder may increase catastrophic interpretation of dyspnea, thereby increasing the chance of COPD assessment. Additionally, patients seeking attention for any mental disorder may have an increased likelihood of a physician noting symptoms of COPD following greater attendance at a medical practice. Inconsistent with this explanation, adjustment for health care use only slightly attenuated the significant association between COPD and depression observed in another study. [9]. In summary this research contributes new evidence to our understanding of the breadth of associations between mental disorders and COPD. This is the first study to identify the relative contribution of different mental disorders to COPD diagnosis and to identify substantive cumulative risk of COPD diagnosis with the experience of multiple mental disorders. The association between COPD and mental disorders is likely to be bidirectional; however research has focused on identifying COPD as a risk factor for mental disorder [3] rather than mental disorder as a risk factor for COPD. This research indicates that future research should prospectively attend to mental disorders as a risk factor for developing COPD. Acknowledgments The authors appreciate the helpful contributions to WMH of Herbert Matschinger, PhD. A complete list of all within-country and cross-national WMH publications can be found at http://www.hcp.med.harvard.edu/wmh/. A complete list of all within-country and cross-national WMH publications can be found at http://www.hcp.med.harvard.edu/wmh/. Funding/support The World Health Organization World Mental Health (WMH) Survey Initiative is supported by the National Institute of Mental Health (NIMH; R01 MH070884), the John D. and Catherine T. MacArthur Foundation, the Pfizer Foundation, the US Public Health Service (R13-MH066849, R01-MH069864, and R01 DA016558), the Fogarty International Center (FIRCA R03-TW006481), the Pan American Health Organization, Eli Lilly and Company, Ortho-McNeil Pharmaceutical, GlaxoSmithKline, and Bristol-Myers Squibb. We thank the staff of the WMH Data Collection and Data Analysis Coordination Centers for assistance with instrumentation, fieldwork, and consultation on data analysis. The Colombian National Study of Mental Health (NSMH) is supported by the Ministry of Social Protection. The ESEMeD project is funded by the European Commission (Contracts QLG5-1999-01042; SANCO 2004123, and EAHC 20081308), the Piedmont Region (Italy)), Fondo de Investigación Sanitaria, Instituto de Salud Carlos III, Spain (FIS 00/0028), Ministerio de Ciencia y Tecnología, Spain (SAF 2000-158-CE), Departament de Salut, Generalitat de Catalunya, Spain, Instituto de Salud Carlos III (CIBER CB06/02/0046, RETICS RD06/0011 REM-TAP), and other local agencies and by an unrestricted educational grant from GlaxoSmithKline. Implementation of the Iraq Mental Health Survey (IMHS) and data entry were carried out by the staff of the Iraqi MOH and MOP with direct support from the Iraqi IMHS team with funding from both the Japanese and European Funds through United Nations Development Group Iraq Trust Fund (UNDG ITF). The Israel National Health Survey is funded by the Ministry of Health with support from the Israel National Institute for Health Policy and Health Services Research and the National Insurance Institute of Israel. The World Mental Health Japan (WMHJ) Survey is supported by the Grant for Research on Psychiatric and Neurological Diseases and Mental Health (H13-SHOGAI-023, H14-TOKUBETSU-026, H16-KOKORO-013) from the Japan Ministry of Health, Labor and Welfare. The Mexican National Comorbidity Survey (MNCS) is supported by The National Institute of Psychiatry Ramon de la Fuente (INPRFMDIES 4280) and by the National Council on Science and Technology (CONACyT-G30544- H), with supplemental support from the PanAmerican Health Organization (PAHO). Te Rau Hinengaro: The New Zealand Mental Health Survey (NZMHS) is supported by the New Zealand Ministry of Health, Alcohol Advisory Council, and the Health Research Council (HRC 11/200). The Northern Ireland Study of Mental Health was funded by the Health & Social Care Research & Development Division of the Public Health Agency. The Chinese World Mental Health Survey Initiative is supported by the Pfizer Foundation. The Shenzhen Mental Health Survey is supported by the Shenzhen Bureau of Health and the Shenzhen Bureau of Science, Technology, and Information. The Peruvian World Mental Health Study was funded by the National Institute of Health of the Ministry of Health of Peru. The Polish project Epidemiology of Mental Health and Access to Care —EZOP Poland was carried out by the Institute of Psychiatry and Neurology in Warsaw in consortium with Department of Psychiatry — Medical University in Wroclaw and National Institute of Public Health-National Institute of Hygiene in Warsaw and in partnership with Psykiatrist Institut Vinderen — Universitet, Oslo. The project was funded by the Norwegian Financial Mechanism and the European Economic Area Mechanism as well as Polish Ministry of Health. No support from pharmaceutical industry neither other commercial sources was received. The Portuguese Mental Health Study was carried out by the Department of Mental Health, Faculty of Medical Sciences, NOVA University of Lisbon, with collaboration of the Portuguese Catholic University, and was funded by Champalimaud Foundation, Gulbenkian Foundation, Foundation for Science and Technology (FCT) and Ministry of Health. The Romania WMH study projects “Policies in Mental Health Area” and “National Study regarding Mental Health and Services Use” were carried out by National School of Public Health & Health Services Management (former National Institute for Research & Development in Health, present National School of Public Health Management & Professional Development, Bucharest), with technical support of Metro Media Transilvania, the National Institute of Statistics — National Centre for Training in Statistics, SC. Cheyenne Services SRL, Statistics Netherlands and were funded by Ministry of Public Health (former Ministry of Health) with supplemental support of Eli Lilly Romania SRL. Additional funding Work on this paper was funded by a grant from the Health Research Council of New Zealand (HRC 11/200) to Kate M. Scott. Financial disclosure Dr. Kessler has been a consultant for Analysis Group, GlaxoSmithKline Inc., Kaiser Permanente, Merck & Co, Inc., Ortho-McNeil Janssen Scientific Affairs, Pfizer Inc., Sanofi-Aventis Groupe, Shire US Inc., SRA International, Inc., Takeda Global Research & Development, Transcept Pharmaceuticals Inc., Wellness and Prevention, Inc., and Wyeth-Ayerst; has served on advisory boards for Eli Lilly & Company, Mindsite, and Wyeth-Ayerst; and has had research support for his epidemiological studies from Analysis Group Inc., Bristol-Myers Squibb, Eli Lilly & Company, EPI-Q, Ortho-McNeil Janssen Scientific Affairs., Pfizer Inc., Sanofi-Aventis Groupe, and Shire US, Inc. He owns stock in Datastat, Inc. Table 1 Characteristics of WMH samples and percent (and number) with history of chronic lung diseases. Country Field dates Age rangea Sample size Response rate (%) History of adult onset chronic lung diseases (21+) Part 1 Part 2 sub-sample N Weighted % SE Mean age of diagnosis SE Americas Colombia 2003 18–65 4426 2381 87.7 3 0.2 0.1 33.7 2.2 Mexico 2001–2 18–65 5782 2362 76.6 7 0.1 0.0 39.6 6.4 United States 2002–3 18+ 9282 5692 70.9 128 2.1 0.3 53.3 1.9 Peru 2005–6 18–65 3930 1801 90.2 4 0.4 0.2 36.8 7.1 Asia and South Pacific Japan 2002–6 20+ 4129 1682 55.1 16 0.9 0.3 55.1 4.2 PRC Shen Zhenb 2006–7 18+ 7132 2475 80.0 20 0.4 0.1 36.2 4.5 New Zealand 2003–4 18+ 12,790 7312 73.3 107 1.4 0.2 54.3 2.3 Europe Belgium 2001–2 18+ 2419 1043 50.6 34 2.4 0.5 51.3 4.2 France 2001–2 18+ 2894 1436 45.9 51 2.9 0.6 41.6 2.2 Germany 2002–3 18+ 3555 1323 57.8 25 1.7 0.4 47.5 3.4 Italy 2001–2 18+ 4712 1779 71.3 37 1.6 0.3 48.9 3.0 The Netherlands 2002–3 18+ 2372 1094 56.4 20 1.1 0.5 50.8 3.7 Spain 2001–2 18+ 5473 2121 78.6 55 1.5 0.3 47.4 2.0 Northern Ireland 2004–7 18+ 4340 1986 68.4 11 0.3 0.1 47.0 3.1 Portugal 2008–9 18+ 3849 2060 57.3 20 0.8 0.2 43.1 3.4 Romania 2005–6 18+ 2357 2357 70.9 31 1.2 0.3 35.8 2.3 Poland 2010–11 18–64 10,081 4000 50.4 26 0.6 0.1 43.5 1.7 Middle East Israel 2002–4 21+ 4859 4859 72.6 177 3.4 0.3 43.4 1.1 Iraq 2006–7 18+ 4332 4332 95.2 18 0.5 0.1 51.9 5.4 Mean age of diagnosis (all countries combined) 47.6 0.8 Prevalence of chronic lung disease (all countries combined) 1.3 0.1 Weighted average response rate (%) 67.4 Total sample size 98,714 52,095 790 a For the purposes of cross-national comparisons we limit the sample to 18 + . b People's Republic of China. Table 2 Prevalence and concurrent associations between 12-month mental disorders and lifetime COPD. Type of 12-month disorder Among those with lifetime COPD Among those without lifetime COPD Concurrent associations between 12 month disorders and lifetime COPD1 % SE % SE OR (95% C.I) Mood disorder Major depressive episode/Dysthymia 8.7 1.5 5.3 0.1 2.1 (1.5–2.9) Bipolar disorder (Broad) 1.8 0.5 1.3 0.1 2.3 (1.2–4.2) Anxiety disorder Panic disorder 2.2 0.6 1.1 0.1 2.5 (1.6–3.9) Generalized anxiety disorder 6.4 1.3 2.0 0.1 2.6 (1.7–4.1) Social phobia 7.0 1.2 2.8 0.1 2.4 (1.7–3.4) Specific phobia 8.5 1.4 5.8 0.1 1.5 (1.1–2.1) Agoraphobia without panic 1.4 0.5 0.5 0.0 3.2 (1.7–5.9) Post-traumatic stress disorder 3.2 0.7 1.7 0.1 2.3 (1.5–3.5) Obsessive compulsive disorder 1.6 0.8 1.2 0.1 2.7 (1.0–7.3) Impulse-control disorder Intermittent explosive disorder 3.0 1.1 2.1 0.1 1.6 (0.8–3.5) Bulimia nervosa – – 0.2 0.0 – – Binge eating disorder 0.7 0.4 0.6 0.1 1.4 (0.5–3.9) Substance use disorder Alcohol abuse 1.7 0.8 1.6 0.1 1.7 (0.8–3.9) Alcohol abuse with dependence 0.9 0.3 0.6 0.0 1.7 (0.8–3.5) Drug abuse – – 0.5 0.0 – – Drug abuse with dependence – – 0.2 0.0 – – “–”: Unstable estimates due to low number of cases(<5). 1 Each 12-month mental disorder type was estimated as a predictor in separate logistic regression model controlling for current age, gender, country, smoking (ever/current/never) and years of education. Table 3 Bivariate and multivariate associations (odds ratios) between DSM-IV mental disorders and the subsequent diagnosis of COPD. Type of disorders Single disorder models1 Comorbid disorders model2 OR (95% C.I.) OR (95% C.I.) I. Mood disorders Major Depressive Episode/Dysthymia 2.2 (1.7–2.8) 1.6 (1.3–2.0) Bipolar Disorder (Broad) 3.5 (2.2–5.7) 1.5 (0.8–2.7) II. Anxiety disorders Panic Disorder 2.1 (1.4–3.3) 1.2 (0.8–1.8) Generalized Anxiety Disorder 2.8 (2.0–3.9) 1.7 (1.2–2.4) Social Phobia 2.1 (1.4–3.0) 1.1 (0.8–1.6) Specific Phobia 1.7 (1.3–2.3) 1.2 (0.9–1.6) Agoraphobia without Panic 2.7 (1.5–5.0) 1.5 (0.8–2.8) Post-Traumatic Stress Disorder 2.1 (1.4–3.1) 1.2 (0.8–1.7) Obsessive Compulsive Disorder 2.7 (1.2–6.5) 1.7 (0.7–4.1) III. Impulse-control disorders Intermittent Explosive Disorder 3.0 (1.8–4.9) 1.6 (0.9–2.8) Binge Eating Disorder 1.6 (0.8–3.2) 1.0 (0.5–1.8) Bulimia Nervosa 0.9 (0.3–3.3) 0.4 (0.1–1.5) IV. Substance disorders Alcohol Abuse 2.4 (1.7–3.4) 1.8 (1.3–2.7) Alcohol Dependence with Abuse 2.9 (1.6–5.0) 1.1 (0.6–2.2) Drug Abuse 2.3 (1.5–3.8) 1.2 (0.7–2.1) Drug Dependence with Abuse 2.4 (1.2–4.9) 0.8 (0.4–1.8) Joint effect of all types of disorders, χ216 112.9 Difference between types of disorders, χ215 17.4 1 Single disorder models: each mental disorder type was estimated as a predictor of the physical condition diagnosis in a separate discrete time survival model controlling for age cohorts, gender, person-year, country, smoking (ever/current/never) and years of education. 2 Comorbid disorders model: the model was estimated with dummy variables for all mental disorders entered simultaneously, including the controls specified above. Table 4 Prevalence and association between number of disorders and COPD. Number of mental disorder Among those with lifetime chronic lung diseases Among those without lifetime chronic lung diseases All Number of disorders model1 % SE % SE % SE OR (95% C.I) Exactly 1 disorder 18.3 2.0 16.0 0.2 16.0 0.2 1.6* (1.3–2.1) Exactly 2 disorders 9.0 1.4 6.6 0.1 6.6 0.1 2.4* (1.7–3.5) Exactly 3 disorders 6.2 1.0 2.8 0.1 2.9 0.1 3.8* (2.6–5.8) Exactly 4 disorders 1.9 0.5 1.4 0.1 1.5 0.1 3.1* (1.8–5.6) 5+ disorders 3.8 0.8 1.6 0.1 1.6 0.1 5.8* (3.5–9.6) Joint effect of number of disorders, χ25 – – – – – – 64.2* 1 Numberofdisordersmodel:the model was estimated using a discrete-time survival model with dummy variables for number of mental disorders without any information about type of mental disorders, including the controls specified in Table 3. References [1] Willgoss TG Yohannes AM Anxiety disorders in patients with COPD: a systematic review Respir. Care 2013 58 858 866 22906542 [2] Hynninen KMJ Breitve MH Wiborg AB Psychological characteristics of patients with chronic obstructive pulmonary disease: a review J. Psychosom. Res 2005 59 429 443 16310027 [3] Atlantis E Fahey P Cochrane B Bidirectional associations between clinically relevant depression or anxiety and COPD: a systematic review and meta-analysis Chest 2013 144 766 777 23429910 [4] Eisner MD Blanc PD Yelin EH Influence of anxiety on health outcomes in COPD Thorax 2010 65 229 234 20335292 [5] Decramer M Rennard S Troosters T COPD as a lung disease with systemic consequences–clinical impact, mechanisms, and potential for early intervention COPD 2008 5 235 256 18671149 [6] Ohayon MM Chronic obstructive pulmonary disease and its association with sleep and mental disorders in the general population J. Psychiatr. Res 2014 54 79 84 24656426 [7] Ng TP Niti M Fones C Co-morbid association of depression and COPD: a population-based study Respir. Med 2009 103 895 901 19136238 [8] Goodwin RD Lavoie KL Lemeshow AR Depression, anxiety, and COPD: the unexamined role of nicotine dependence Nicotine Tob. Res 2012 14 176 183 22025539 [9] Patten SB Williams JVA Lavorato DH Major depression as a risk factor for chronic disease incidence: longitudinal analyses in a general population cohort Gen. Hosp. Psychiatry 2008 30 407 413 18774423 [10] Von Korff M Scott KM Gureje O Global Perspectives on Mental–Physical Comorbidity in the WHO World Mental Health Surveys 2009 Cambridge University Press New York [11] Kessler RC Ustun TB The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders 2008 Cambridge University Press New York [12] Kessler RC Ustun B The world mental health (WMH) survey initiative version of the world health organization (WHO) composite international diagnostic interview (CIDI) Int. J. Methods Psychiatr. Res 2004 13 93 121 15297906 [13] Simon GE Von Korff M Recall of psychiatric history in cross-sectional surveys: implications for epidemiological research Epidemiol. Rev 1995 17 221 227 8521941 [14] Haro JM Arbabzadeh-Bouchez S Brugha TS Concordance of the composite international diagnostic interview version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO world mental health surveys Int. J. Methods Psychiatr. Res 2006 15 167 180 17266013 [15] Statistics NCfH Evaluation of National Health Interview Survey diagnostic reporting Vital Health Stat 1994 2 120 1 116 [16] Scott K De Jonge P Alonso J Associations between DSM-IV mental disorders and subsequent heart disease onset: beyond depression Int. J. Cardiol 2013 168 5293 5299 23993321 [17] Scott KM Hwang I Chiu W-T Chronic physical conditions and their association with first onset of suicidal behavior in the world mental health surveys Psychosom. Med 2010 72 712 719 20498290 [19] Singer JD Willett JB It's about time: using discrete-time survival analysis to study duration and the timing of events J. Educ. Stat 1993 18 155 195 [20] Efron B Logistic regression, survival analysis, and the Kaplan-Meier curve J. Am. Stat. Assoc 1988 83 413 425 [21] Scott KM Von Korff M Ormel J Mental disorders among adults with asthma: results from the world mental health survey Gen. Hosp. Psychiatry 2007 29 123 133 17336661 [22] Shah BV Armitage P Colton T Linearization Methods of Variance Estimation Encyclopedia of Biostatistics 1998 2276 2279 John Wiley and Sons Chichester [23] SUDAAN Software for the Statistical Analysis of Correlated Data [Program] 1999 Research Triangle Park North Carolina, USA [24] CDC Chronic obstructive pulmonary disease among adults — United States, 2011 MMWR 2012 61 938 943 23169314 [25] Bednarek M Maciejewski J Wozniak M Prevalence, severity and underdiagnosis of COPD in the primary care setting Thorax 2008 63 402 407 18234906 [26] Vassend O Skrondal A The role of negative affectivity in self assessment of health: a structural equation approach J. Health Psychol 1999 4 465 482 22021640 [27] Kolk AM Hanewald GJ Schagen S Predicting medically unexplained physical symptoms and health care utilization. A symptom-perception approach J. Psychosom. Res 2002 52 35 44 11801263 [28] Kessler RC Angermeyer M Anthony JC Lifetime prevalence and age-of-onset distributions of mental disorders in the world health organization's world mental health survey initiative World Psychiatry 2007 6 168 176 18188442 [29] Scott KM McLaughlin KA Smith DAR Childhood maltreatment and DSM-IV adult mental disorders: comparison of prospective and retrospective findings Br. J. Psychiatry 2012 200 469 475 22661679 [30] Irwin MR Miller AH Depressive disorders and immunity: 20 years of progress and discovery Brain Behav. Immun 2007 21 374 383 17360153 [31] Smagula SF Ancoli-Israel S Barrett-Connor E Inflammation, sleep disturbances, and depressed mood among community-dwelling older men J. Psychosom. Res 2014 76 368 373 24745777 [32] Sisson JH Alcohol and airways function in health and disease Alcohol 2007 41 293 307 17764883 [33] Takao A Shimoda T Kohno S Correlation between alcohol-induced asthma and acetaldehyde dehydrogenase-2 genotype J. Allergy Clin. Immunol 1998 101 576 580 9600491 [34] Bierut LJ Genetic vulnerability and susceptibility to substance dependence Neuron 2011 69 618 627 21338875 [35] Scott KM Von Korff M Angermeyer MC The association of childhood adversities and early onset mental disorders with adult onset chronic physical conditions Arch. Gen. Psychiatry 2011 68 838 844 21810647 [36] Ochi M Fujiwara T Mizuki R Association of socioeconomic status in childhood with major depression and generalized anxiety disorder: results from the World Mental Health Japan Survey 2002–2006 BMC Public Health 2014 14 [37] Salvi SS Barnes PJ Chronic obstructive pulmonary disease in non-smokers The Lancet 374 733 743

PMC005xxxxxx/PMC5120396.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101589550 40869 JAMA Psychiatry JAMA Psychiatry JAMA psychiatry 2168-622X 2168-6238 26018466 5120396 10.1001/jamapsychiatry.2015.0575 NIHMS830489 Article Psychotic experiences in the general population: a cross-national analysis based on 31,261 respondents from 18 countries McGrath John J. PhD, MD 123* Saha Sukanta PhD 123 Al-Hamzawi Ali MD 4 Alonso Jordi DrPH, MD 56 Bromet Evelyn J. PhD 7 Bruffaerts Ronny PhD 8 Caldas-de-Almeida José Miguel PhD, MD 9 Chiu Wai Tat MS 10 de Jonge Peter PhD 11 Fayyad John MD 12 Florescu Silvia PHD, MD 13 Gureje Oye MD 14 Haro Josep Maria PhD, MD 15 Hu Chiyi PhD, MD 16 Kovess-Masfety Viviane PhD, MD 17 Lepine Jean Pierre HDR, MD 18 Lim Carmen W. MSc 19 Mora Maria Elena Medina PhD 20 Navarro-Mateu Fernando PhD, MD 21 Ochoa Susana PhD 22 Sampson Nancy BA 10 Scott Kate PhD 19 Viana Maria Carmen PhD, MD 23 Kessler Ronald C. PhD 10 1 Queensland Centre for Mental Health Research, The Park Centre for Mental Health, Wacol, QLD 4076, Australia 2 Discipline of Psychiatry, University of Queensland, St Lucia, QLD 4072, Australia 3 Queensland Brain Institute, University of Queensland, St Lucia, QLD 4072, Australia 4 College of Medicine, Al-Qadisiya University, Diwania governorate, Iraq 5 Health Services Research Unit, IMIM-Institut de Recerca Hospital del Mar, Barcelona, Spain 6 CIBER en EpidemiologÕïa y Salud Puïblica (CIBERESP), Barcelona, Spain 7 Department of Psychiatry, Stony Brook University School of Medicine 8 Universitair Psychiatrisch Centrum - Katholieke Universiteit Leuven (UPC-KUL), Campus Gasthuisberg, Belgium 9 Chronic Diseases Research Center (CEDOC) and Department of Mental Health, Faculdade de Ciências Médicas, Universidade Nova de Lisboa, Portugal 10 Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts, USA 11 University of Groningen, University Medical Center, Groningen Department of Psychiatry, Interdisciplinary Center, Psychopathology and Emotion regulation (ICPE), Groningen, The Netherlands 12 Institute for Development, Research, Advocacy, and Applied Care (IDRAAC), Beirut, Lebanon 13 National School of Public Health, Management and Professional Development, Bucharest, Romania 14 Department of Psychiatry, University College Hospital, Ibadan, Nigeria 15 Parc Sanitari Sant Joan de Déu, CIBERSAM, Universitat de Barcelona, Spain 16 Shenzhen Insitute of Mental Health & Shenzhen Kanging Hospital, China 17 Ecole des Hautes Etudes en Santé Publique (EHESP), EA 4057 Paris Descartes University, Paris, France 18 Hôpital Lariboisière Fernand Widal, Assistance Publique Hôpitaux de Paris INSERM UMR-S 1144, University Paris Diderot and Paris Descartes Paris, France 19 Department of Psychological Medicine, Dunedin School of Medecine, University of Otago, New Zealand 20 National Institute of Psychiatry Ramón de la Fuente, Mexico 21 Subdirección General de Salud Mental y Asistencia Psiquiátrica. Servicio Murciano de Salud, El Palmar (Murcia), Spain 22 Parc Sanitari Sant Joan de Déu, CIBERSAM, Universitat de Barcelona 23 Department of Social Medicine, Federal University of Espírito Santo, Brazil * Corresponding author: Professor John McGrath, Queensland Centre for Mental Health Research, The Park Centre for Mental Health, Wacol, Queensland, 4076, Australia. j.mcgrath@uq.edu.au, Phone: +61 7 3271 8694, Fax: +61 7 3271 8698 17 11 2016 7 2015 23 11 2016 72 7 697705 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. IMPORTANCE Community-based surveys find that many otherwise healthy individuals report histories of hallucinations and delusions. To date, most studies have focused on the overall lifetime prevalence of ever having any of these psychotic experiences (PEs), possibly masking important features related to types and frequencies of PEs. OBJECTIVE To explore detailed epidemiological information of PEs in a large cross-national sample. DESIGN, SETTING, AND PARTICIPANTS Data came from the WHO World Mental Health (WMH) Surveys, a coordinated set of community epidemiological surveys of the prevalence and correlates of mental disorders in representative household samples in countries throughout the world. 31 261 adult (aged 18 and older) WMH respondents across 18 countries were asked about lifetime and 12-month prevalence and frequency of six types of PEs (two hallucinatory experiences [HEs] and four delusional experiences [DEs]). MAIN OUTCOMES Prevalence, frequency, and correlates of PEs. RESULTS Mean lifetime prevalence (standard error) of ever having a lifetime PE was 5.8% (0.2), with hallucinatory experiences (5.2% [0.2]) much more common than delusional experiences (1.3% [0.1]), More than two-thirds (72.0%) of respondents with lifetime PEs reported experiencing only one type. PEs were typically infrequent, with 32.2% of respondents with lifetime PEs reporting only one occurrence and an additional 31.8% only 2–5 occurrences. There was a significant relationship between having more than one type of PE and having more frequent PE episodes. Lifetime prevalence estimates were significantly higher among respondents in middle and high income countries than low income countries, women than men, the non-married than the married, not employed and those with low family income. CONCLUSIONS AND RELEVANCE The epidemiology of PEs is more nuanced than previously thought. Research is needed that focuses on similarities and differences in predictors of the onset, course, and consequences of distinct PEs. There has been a growing interest in recent years in the epidemiology of hallucinations and delusions.1 These psychotic experiences (PEs) are reported by a sizeable minority of the general population. A recent meta-analysis based on 61 studies reported that the median lifetime prevalence of PE was 7.2%.2 Because this is substantially higher than the lifetime morbid risk (MR) of psychotic disorders such as schizophrenia (median MR 0.7%),3 the field of psychiatric epidemiology has been forced to rethink how PEs ‘fit’ into the epidemiologic landscape of psychotic disorders. The terminology to describe these experiences has also evolved over time. Sometimes referred to as psychotic-like experiences, we will use the general term psychotic experiences to encompass both hallucinatory experiences (HEs) and delusional experiences (DEs).2 Early work on the epidemiology of PEs focused on these experiences as risk indicators for later conversion to full psychosis. There is an appealing logic to this type of research, as many of the risk factors associated with PEs are also associated with schizophrenia/psychosis.4 More recently, evidence has accumulated that PEs are also associated with the subsequent onset of a wide array of common mental disorders including anxiety, mood, and substance use disorders,5–7 as well as with an increased risk of suicidal ideation and intent.8–10 Thus, there is a growing awareness that the presence of PEs may reflect a vulnerability to a wide range of adverse mental health outcomes (in addition to psychotic disorders).11–16 These findings, and the concern that antipsychotic medications may be inappropriately used to treat individuals with isolated PEs, may have influenced the decision to exclude ‘Attenuated Psychosis Syndrome’ in recently revised diagnostic criteria.17 As the empirical data have accumulated, systematic reviews have pooled prevalence estimates and applied meta-regression techniques in order to explore the socio-demographic correlates of PEs.2,15,18,19 These reviews provide valuable clues to the nature of PEs, but also highlight important gaps in the literature. Four of these gaps are of special importance for the current study. Firstly, the use of pooling in systematic reviews of PEs has encouraged the use of coarse dichotomous measures (e.g. lifetime prevalence present/absent) in order to harmonize the wide array of scales and diagnostic instruments used to assess PEs.2 This has reduced the subtlety of the associations examined in these reviews. Secondly, the studies included in the systematic reviews have varied in many key design elements. As noted by Linscott and van Os 2 substantial heterogeneity in the data has hampered analyses related to the relationship between PEs and socio-demographic variables. Thirdly, the vast majority of community studies of PE prevalence and correlates have been carried out in high income countries. A major exception is the World Health Survey (WHS), which included four brief PE questions in surveys of 52 nations.20 However, the WHS assessment of PEs had several limitations (e.g. it lacked information on frequency of PE occurrence; and questions about DEs were not asked in a fashion that excluded experiences related to alcohol, illicit drugs or sleep). Finally, in order to allow pooling of data from different studies, some reviews have collapsed different variables across orthogonal axes. For example, Kaymaz et al15 compiled composite variables related to ‘weak’ and ‘strong’ PEs, which in theory could be built from data related to: (1) the count of different types of PEs, (2) the frequency of occurrence, (3) associated distress, (4) comorbidity, and/or (5) ‘certainty’ (e.g. confidence in the psychotic nature of the experience). Leading commentators have repeatedly called for more fine-grained analyses of PE in order to guide the field.1,21 The current report presents initial results of analyses designed to address the above limitations by examining data collected in the WHO World Mental Health (WMH) Surveys, a series of population-based surveys carried out in many countries using consistent instruments and field procedures designed to facilitate pooled cross-national analyses of the prevalence and correlates of mental disorders. These data provide an unprecedented opportunity to explore the epidemiologic landscape of PEs. Methods Participants The WMH surveys are a coordinated set of community epidemiological surveys administered in probability samples of the household population in countries throughout the world (www.hcp.med.harvard.edu/WMH). Eighteen of the 26 WMH surveys completed up to now administered the CIDI Psychosis Module. These 18 countries are distributed across North and South America (Colombia, Mexico, Peru, Sao Paulo in Brazil, USA); Africa (Nigeria); the Middle East (Iraq, Lebanon); Asia (Shenzhen in the People’s Republic of China [PRC]); the South Pacific (New Zealand); and Europe (Belgium, France, Germany, Italy, the Netherlands, Portugal, Romania, Spain). All 18 surveys were based on multi-stage, clustered area probability household sampling designs (Table 1). The weighted average response rate across all 18 countries was 72.1%. Most surveys were based on nationally representative sample frames, but a few excluded rural areas (Colombia, Mexico), or focused on particular regions (Nigeria, Shenzhen), or cities (Sao Paulo). Participating sites were grouped into 3 country-level income strata according to World Bank criteria22 - ‘low and lower-middle income’ countries (Colombia, Iraq, Nigeria, PRC-Shenzhen, Peru), upper-middle income (Sao Paulo, Lebanon, Mexico, Romania), and high-income countries (the European countries, New Zealand, USA). The age ranges reported here include 18 years and over except in three countries (Mexico, Colombia, Peru) where 65 years was the upper age limit. In keeping with previous studies of PEs,9,11,23–28 we made the a priori decision to exclude individuals who had PEs but who also screened positive for possible schizophrenia/psychosis, and manic-depression/mania (i.e. respondents who (a) reported (1) schizophrenia/psychosis or (2) manic-depression/mania” in response to the question “What did the doctor say was causing (this/these) experiences?”; and (b) those who ever took any antipsychotic medications for these symptoms). This resulted in the exclusion of 140 respondents (0.4% of all respondents), leaving 31,261 respondents for this study (see Table 1). Measures and Assessments All WMH surveys were conducted face-to-face in the homes of respondents by trained lay interviewers. Informed consent was obtained before beginning interviews in all countries. Procedures for obtaining informed consent and protecting individuals (ethical approval) were approved and monitored for compliance by the institutional review boards of the collaborating organisations in each country.29 Full details of these procedures are described elsewhere.30,31 All WMH interviews had two parts. Part I, administered to all respondents, contained assessments related to core mental disorders. Part II included additional information relevant to a wide range of survey aims, including assessment of PEs. All Part I respondents who met criteria for any Part I DSM IV mental disorder as well as a probability sample of other respondents were administered Part II. Part II respondents were weighted by the inverse of their probability of selection for Part II to adjust for differential sampling. Within the different sites, questions related to PEs were either administered to all respondents or a random sample of those administered Part II. Analyses in this article were based on the weighted Part II subsample of respondents administered the CIDI Psychosis Module. Additional weights were used to adjust for differential probabilities of selection within households, nonresponse, and to match the samples to population socio-demographic distributions. The instrument used in the WMH surveys was the WHO Composite International Diagnostic Interview (CIDI),32 a validated fully-structured diagnostic interview designed to assess the prevalence and correlates of a wide range of mental disorders according to the definitions and criteria of both the DSM-IV and ICD-10 diagnostic systems. The CIDI Psychosis Module included questions about 6 PE types – 2 related to HEs (visual hallucinations, auditory hallucinations) and 4 related to DEs (two ‘bizarre’ delusional items - thought insertion/withdrawal, mind control/passivity; two ‘paranoid’ delusional items - ideas of reference, plot to harm/follow) (See Appendices A1 & A2). For example, respondents were asked if they ever experienced PEs (e.g. “Have you ever heard any voices that other people said did not exist?”). This was followed by a probe question to determine if the reported PEs ever occurred when the person was ‘not dreaming, not half-asleep, or not under the influence of alcohol or drugs’. Only responses of the latter type are considered here. The sequence of these follow-up probe types differed slightly between the first 6 WMH surveys, which were carried out in Europe (Belgium, France, Germany, Italy, the Netherlands, Spain), and the remaining 12 countries (See eTables 1 and 2). Respondents who reported PEs were then asked about: (a) presence of the PEs in the past 12 months; and (b) frequency/occurrences of the PEs in their lifetime. In this paper we present prevalence estimates for any PE, any HE (with or without associated DEs), any DE (with or without associated HEs), ‘pure’ HE (without DEs) and ‘pure’ DE (without HEs). In addition, we will present two key PE-related metrics: (a) count of types of PEs (henceforth referred to as PE type metric); and (b) frequency of occurrence of PE episodes (henceforth referred to as PE frequency metric). Respondents may have had more than one hallucination and/or delusion type associated with a single episode of PEs. For the PE frequency metric, reported frequency of lifetime PE episodes was divided into 5 categories: only once, 2–5 times, 6–10 times, 11–100 times, and 101 and above. This five-category scheme was collapsed into 1–10 versus 11+ in analyses of socio-demographic correlates of PE frequency among respondents with lifetime PEs. The socio-demographic factors considered here include: gender, age, number of years of education, employment history, marital status, family income, and nativity (i.e. born inside the country of assessment). For the bivariate and multivariate analyses, the socio-demographic variables were stratified into broad categories based on methodology described elsewhere.29 Statistical Analysis Weighted prevalence estimates were calculated for the various PE types and related metrics. Odds ratios (ORs) and design-corrected 95% confidence intervals (CIs) are reported. Because the WMH survey data featured geographical clustering and weighting, standard errors (se) of parameter estimates were generated using the design-based Taylor series linearization method implemented in a SAS macro (SAS Institute inc. version 9.4). Multivariate significance was evaluated using Wald χ2 tests based on design-corrected coefficient variance–covariance matrices. The association between the PE type metric versus the PE frequency metric was evaluated using the Cochran-Armitage test. Statistical significance was evaluated consistently using two-tailed .05-level tests. Results Prevalence of PEs Table 2 presents country-specific lifetime PE prevalence estimates. Lifetime prevalence (se) of at least 1 PE was reported by 5.8% (0.2) of the 31 261 respondents. Lifetime prevalence of any HE was 5.2% (0.2) and of any DE 1.3% (0.1). The median and interquartile range (IQR) of lifetime PEs, HEs and DEs were 5.5% (IQR: 2.8–7.5), 4.4% (IQR: 1.8–6.5), and 1.3% (IQR: 0.9–1.8), respectively (see eFigures 1–3 for cumulative distribution of PE, HE and DE estimates respectively). Twelve-month prevalence (se) of any PE was 2.0% (0.1), while the median (IQR) was 1.4% (1.0–2.8). Lifetime PEs prevalence (se) was significantly higher among women than men (6.6% [0.2] vs. 5.0% [0.3]; χ21=16.0, p< .001). Similar gender differences were found for prevalence of HEs (5.9% [0.2] vs. 4.3% [0.3]; χ21 =19.4, p< .001) but not DEs (1.4% [0.1] vs. 1.3% [0.1]; χ21=0.3, p=.61). The significant gender difference was also found for respondents with ‘pure’ HEs (5.2% [0.2] vs. 3.7% [0.3]; χ21=19.3, p < .001), but not ‘pure’ DEs (0.7% [0.1] vs. 0.7% [0.1]; χ21=0.1, p = .80). Significant differences were found across the three country-level income strata in lifetime prevalence of any PE, any HE and any DE – in each comparison the prevalence estimates were significantly higher among respondents in middle and high income countries than low income countries (χ22 ranged from 7.1 to 58.2, each p < .001). The prevalence of individual PEs and the distribution of the PE type metric Table 3 shows the lifetime prevalence estimates of individual PE types and counts of different PE types. The most common PE type overall was visual hallucinations (3.8% [0.2]) followed by auditory hallucinations (2.5% [0.1]). Prevalence estimates of individual DE types were low (0.3– 0.7%). Among those with any lifetime PE, 72.0% (representing 4.2% of the total sample) reported only 1 PE type, 21.1% (representing 1.2% of the total sample) exactly 2 types, and 6.8% (representing 0.4% of the total sample) 3 or more types. The distribution of the PE frequency metric, and the relationship between PE type and PE frequency metrics PEs were typically infrequent, with 32.2% of the respondents with lifetime PEs reporting only one solitary episode (Table 4). An additional 31.8% of respondents with lifetime PEs experienced only 2–5 PE episodes. Thus, for nearly two-thirds of respondents (64.0%) with lifetime PEs, these experiences occurred only 1–5 times in their lives. An additional 10.0% of respondents with lifetime PEs reported 6–10 lifetime episodes, 20.0% 11–100 episodes, and 6.0% 101 or more episodes. The relationship between the PE type metric and the PE frequency metric is best displayed in Table 5. Those with more PE types are disproportionately more likely to have more PE episodes (Cochran-Armitage z = −10.0, p <.001). Associations between socio-demographic factors with lifetime PEs, HEs and DEs eTable 3 shows the association of socio-demographic variables with lifetime PEs, HEs and DEs in bivariate and multivariate models. Several socio-demographic variables were associated with increased Odds Ratios for PEs, HEs and DEs in both models: (a) being a homemaker or classified as ‘other’ employment (looking for work, disabled etc.) (versus employed); (b) being non-married (never married or separated/widowed/divorced) (versus married), and (c) lower household income (versus high income). In addition to these findings, several socio-demographic variables were associated with only one type of PE. Young respondents (18–29 years) were significantly more likely to have DEs (compared to those over 60 years), while age was unrelated to HEs (and overall PEs). While female sex was associated with an increased prevalence of PEs (in both models), this finding was driven by an increased risk of HEs (but not DEs). Low education, in comparison, was associated with an increased risk of DEs, but not HEs. Unexpectedly, those born outside the country (i.e. migrants) were significantly less likely than the native born to report HEs (but not DEs) in both bivariate and multivariate models. Associations of socio-demographic factors with PE type and PE frequency metrics Concerning factors that influence the PE type metric (in those who had experienced PEs), in the multivariate model, the three younger age strata (i.e. that spanned 18 to 59 years) were significantly more likely to have more than one PE type (compared to those aged 60+ years) (eTable 4). None of the other socio-demographic characteristics was associated with the PE type metric. Concerning the correlates of the PE frequency metric, student status was significantly associated with lower frequency of PE occurrence. None of the other socio-demographic variables was associated with PE frequency (eTable 5). Discussion Based on cross-national samples from 18 countries, we found that 5.8% of respondents reported having one or more PEs at least once in their lifetime and 2.0% in the previous year. These overall estimates are broadly consistent with the previous literature.2 In addition, though, our study foregrounds important new information regarding the count of PE types and frequency of PEs that go beyond the issues considered in previous community-based studies of PEs. Perhaps the most striking finding is that these experiences are infrequent for the majority of individuals who experience PEs, with 32.2% reporting only one PE episode in their life and 64% reporting no more than 5 lifetime occurrences. In the general population, those with 2 or more types of PEs are also significantly more likely to have more PE episodes. For example, of those who reported 3 or more PE types, nearly a quarter (24.5%) reported more than 101 occurrences. Our findings provide an empirical foundation upon which to investigate factors that influence the persistence of PEs.1 When viewed within the context of the gap between 12-month and lifetime PE estimates from the current study (i.e. 2.0% versus 5.8%), we can infer that most individuals do not have persistent PEs. Mindful that lifetime prevalence estimates for mental health disorders are often downward-biased due to under-reporting.33 Linscott and van Os2 estimated that of those who report any PEs, approximately 80% would have had transient experiences. This estimate is consistent with our empirical finding that about 64% of individuals with PEs report only 1–5 lifetime occurrences. Based on the set of PEs examined, our study confirms that hallucinations were more common than delusions (5.2% versus 1.3%), and this general pattern was consistent across the 3 country-level income strata. We note that the lifetime prevalence of PEs was lower in the low-lower middle income countries (3.2%) compared to the upper-middle and high income countries (7.2%, 6.8% respectively). While we cannot directly compare our results with the one previous cross-national study of PEs 20 due to differences in how PEs were assessed, we note that both studies (optimized for consistent design and PE assessment) provided insights in variation between sites. One of the strengths of cross-national studies such as the WMH Survey is that they are able to identify risk factors that exist consistently across countries despite site-specific cultural factors. We found an increased prevalence of both HEs and DEs associated with being unmarried, not employed, and having low household income. However, certain demographic features were differentially associated with HEs but not DEs, and vice versa. For example, women had a significantly higher prevalence of HEs but not DEs. We found a significant relationship between younger age and DEs, but not HEs. Unexpectedly, migrants in our study were significantly less likely to report lifetime HEs (compared to native born respondents). These novel findings provide important points of distinction between the epidemiology of psychotic disorders and PEs.4,34 It is of interest to note that while several socio-demographic variables were significantly associated with the lifetime prevalence of PEs, these features were not associated with the PE type or PE frequency metrics. We speculate that comorbid psychiatric illness (e.g. depression, anxiety disorders) and other risk factors known to be associated with PEs and mental disorders (e.g. family history, substance use, trauma exposure) may contribute to these PE-related metrics. The comprehensive nature of the WMH survey will allow us to explore these hypotheses in future analyses. While the study has many strengths (e.g. large sample size, range of countries, uniform methodology for data collection, innovative analysis of PE-related metrics), there are several limitations. In keeping with other population-based surveys, we relied on trained lay interviewers to administer the questionnaire. While we excluded those who were screen-positive for possible psychotic disorders, we did not have access to valid measures of clinical psychotic disorders. Lifetime prevalence estimates are prone to under-reporting.33 We only assessed four types of DEs, and these probes may have been insensitive to culture-specific delusional beliefs.16 Conclusions We have provided the most comprehensive description of the epidemiology of PEs published to date. While the lifetime prevalence of PEs is 5.8%, these are mostly rare events. For nearly a third of those who have PEs (i.e. 32.2%), these were solitary (i.e. one-off) events. In the general population there is a small subgroup of individuals who have multiple types of PEs and who experiences these PEs more frequently. The research community needs to leverage this fine-grained information in order to better determine how PEs reflect risk status. Our study highlights the subtle and variegated nature of the epidemiology of PEs, and provides a solid foundation upon which to explore the bi-directional relationship between PEs and mental health disorders. Supplementary Material Appendix Tables The surveys discussed in this article were carried out in conjunction with the World Health Organization (WHO) World Mental Health (WMH) Survey Initiative. We thank the WMH staff for assistance with instrumentation, fieldwork, and data analysis. These activities were supported by the US National Institute of Mental Health (NIMH: R01MH070884), the John D. and Catherine T. MacArthur Foundation, the Pfizer Foundation, the US Public Health Service (R13-MH066849, R01-MH069864, and R01 DA016558), the Fogarty International Center (FIRCA R01- TW006481), the Pan American Health Organization, Eli Lilly and Company, Ortho-McNeil Pharmaceutical Inc., GlaxoSmithKline, and Bristol-Myers Squibb. A complete list of WMH publications can be found at www.hcp.med.harvard.edu/wmh/. Each WMH country obtained funding for its own survey. The São Paulo Megacity Mental Health Survey is supported by the State of São Paulo Research Foundation (FAPESP) Thematic Project Grant 03/00 204-3. The Shenzhen, People’s Republic of China Mental Health Survey is supported by the Shenzhen Bureau of Health and the Shenzhen Bureau of Science, Technology and Information. The Colombian National Study of Mental Health (NSMH) is supported by the Ministry of Social Protection, with supplemental support from the Saldarriaga Concha Foundation. The ESEMeD project is funded by the European Commission (Contracts QLG5–1999–01042; SANCO 2004123), the Piedmont Region (Italy), Fondo de Investigación Sanitaria, Instituto de Salud Carlos III, Spain (FIS 00/0028), Ministerio de Ciencia y Tecnología, Spain (SAF 2000–158-CE), Departament de Salut, Generalitat de Catalunya, Spain, Instituto de Salud Carlos III (CIBER CB06/02/0046, RETICS RD06/0011 REM-TAP), and other local agencies and by an unrestricted educational grant from GlaxoSmithKline. Implementation of the Iraq Mental Health Survey (IMHS) and data entry were carried out by the staff of the Iraqi MOH and MOP with direct support from the Iraqi IMHS team with funding from both the Japanese and European funds through United Nations Development Group Iraq Trust Fund (UNDG ITF). The Lebanese National Mental Health Survey (LEBANON) is supported by the Lebanese Ministry of Public Health, the WHO (Lebanon), anonymous private donations to IDRAAC, Lebanon, and unrestricted grants from Janssen Cilag, Eli Lilly, GlaxoSmithKline, Roche, and Novartis. The Mexican National Co-Morbidity Survey (MNCS) is supported by The National Institute of Psychiatry Ramon de la Fuente (INPRFMDIES 4280) and by the National Council on Science and Technology (CONACyT-G30544-H), with supplemental support from the Pan American Health Organization (PAHO), Te Rau Hinengaro: The New Zealand Mental Health Survey (NZMHS), is supported by the New Zealand Ministry of Health, Alcohol Advisory Council, and the Health Research Council. The Nigerian Survey of Mental Health and Wellbeing (NSMHW) is supported by the WHO (Geneva), the WHO (Nigeria), and the Federal Ministry of Health, Abuja, Nigeria. The Peruvian World Mental Health Study was funded by the National Institute of Health of the Ministry of Health of Peru. The Portuguese Mental Health Study was carried out by the Department of Mental Health, Faculty of Medical Sciences, NOVA University of Lisbon, with the collaboration of the Portuguese Catholic University, and was funded by the Champalimaud Foundation, the Gulbenkian Foundation, the Foundation for Science and Technology (FCT) and the Ministry of Health. The Romania WMH study projects ‘Policies in Mental Health Area’ and ‘National Study regarding Mental Health and Services Use’ were carried out by the National School of Public Health and Health Services Management (former the National Institute for Research and Development in Health, present National School of Public Health Management and Professional Development, Bucharest), with technical support from Metro Media Transylvania, the National Institute of Statistics – National Center for Training in Statistics, Cheyenne Services SRL, Statistics Netherlands and were funded by the Ministry of Public Health (formerly the Ministry of Health) with supplemental support from Eli Lilly Romania SRL. The US National Co-Morbidity Survey Replication (NCS-R) is supported by the US NIMH (U01- MH60220) with supplemental support from the National Institute of Drug Abuse (NIDA), the Substance Abuse and Mental Health Services Administration (SAMHSA), the Robert Wood Johnson Foundation (RWJF; Grant 044708), and the John W. Alden Trust. JMcG was funded by the John Cade Fellowship from the National Health and Medical Research Council (APP1056929). The views and opinions expressed in this report are those of the authors and should not be construed to represent the views or policies of any of the sponsoring organizations, agencies, or the WHO. The funders had no input into the design and conduct of the study; collection, management, analysis and interpretation of the data; or preparation, review or approval of the manuscript. Table 1 World Mental Health (WMH) sample characteristics by World Bank income categoriesa, and sample for psychotic experiences (PEs) after exclusion of 140 respondents who were screen-positive for possible schizophrenia/psychosisb Country by income categorya Sample characteristicsc, d Field dates Age range Sample size Response rate e Part I PEs sample I. Low and lower middle income countries Colombia All urban areas of the country 2003 18–65 4,426 722 87.7 Iraq Nationally representative 2006–7 18–96 4,332 4,329 95.2 Nigeria 21 of the 36 states in the country 2002–3 18–100 6,752 1,417 79.3 PRCf – Shenzheng Shenzhen metropolitan area 2006–7 18–88 7,132 2,468 80.0 Peru Nationally representative 2004–5 18–65 3,930 530 90.2 TOTAL 26,572 9,466 84.7 II. Upper-middle income countries Brazil - São Paulo São Paulo metropolitan area 2005–7 18–93 5,037 2,922 81.3 Lebanon Nationally representative 2002–3 18–94 2,857 1,029 70.0 Mexico All urban areas of the country 2001–2 18–65 5,782 715 76.6 Romania Nationally representative 2005–6 18–96 2,357 2,357 70.9 TOTAL 16,033 7,023 75.8 III. High-income countries Belgium Nationally representative 2001–2 18–95 2,419 319 50.6 France Nationally representative 2001–2 18–97 2,894 301 45.9 Germany Nationally representative 2002–3 18–95 3,555 408 57.8 Italy Nationally representative 2001–2 18–100 4,712 617 71.3 Netherlands Nationally representative 2002–3 18–95 2,372 348 56.4 New Zealandg Nationally representative 2003–4 18–98 12,790 7,263 73.3 Portugal Nationally representative 2008–9 18–81 3,849 2,053 57.3 Spain Nationally representative 2001–2 18–98 5,473 1,159 78.6 United States Nationally representative 2002–3 18–99 9,282 2,304 70.9 TOTAL 47,346 14,772 65.5 IV. Total 89,951 31,261 72.1 a Based on World Bank country level of economic development (The World Bank. (2008). Data and Statistics. Accessed May 12, 2009 at: http://go.worldbank.org/D7SN0B8YU0) b Respondents were excluded if they endorsed any PE but also (a) reported (1) schizophrenia/psychosis or (2) manic-depression/mania” in response to the question “What did the doctor say was causing (this/these) experiences?”; and (b) those who ever took any antipsychotic medications for these symptoms. This resulted in the exclusion of 140 respondents (0.4% of all respondents), leaving 31,261 respondents for this study. c NSMH (The Colombian National Study of Mental Health); IMHS (Iraq Mental Health Survey); NSMHW (The Nigerian Survey of Mental Health and Wellbeing); EMSMP (La Encuesta Mundial de Salud Mental en el Peru); LEBANON (Lebanese Evaluation of the Burden of Ailments and Needs of the Nation); M-NCS (The Mexico National Comorbidity Survey); RMHS (Romania Mental Health Survey); ESEMeD (The European Study Of The Epidemiology Of Mental Disorders); NZMHS (New Zealand Mental Health Survey); NMHS (Portugal National Mental Health Survey); NCS-R (The US National Comorbidity Survey Replication). d Most WMH surveys are based on stratified multistage clustered area probability household samples in which samples of areas equivalent to counties or municipalities in the US were selected in the first stage followed by one or more subsequent stages of geographic sampling (e.g., towns within counties, blocks within towns, households within blocks) to arrive at a sample of households, in each of which a listing of household members was created and one or two people were selected from this listing to be interviewed. No substitution was allowed when the originally sampled household resident could not be interviewed. These household samples were selected from Census area data in all countries other than France (where telephone directories were used to select households) and the Netherlands (where postal registries were used to select households). Several WMH surveys (Belgium, Germany, and Italy) used municipal resident registries to select respondents without listing households. 13 of the 18 surveys are based on nationally representative household samples. e The response rate is calculated as the ratio of the number of households in which an interview was completed to the number of households originally sampled, excluding from the denominator households known not to be eligible either because of being vacant at the time of initial contact or because the residents were unable to speak the designated languages of the survey. The weighted average response rate is 72.1%. f People’s Republic of China g For the purposes of cross-national comparisons we limit the sample to those 18+. Table 2 Lifetime and 12-month prevalence of psychotic experiences in the World Mental Health surveys Lifetime Any PEb Lifetime Any HEc Lifetime Any DEd 12-month PE Total Sample Income categorya Country n1e (%f, SE) n1e (%f, SE) n1e (%f, SE) n1e (%f, SE) Ng Low and Lower-Middle Colombia 73 (7.5,1.2) 68 (7.1,1.2) 11 (0.9,0.3) 25 (2.1,0.5) 722 Iraq 51 (1.2,0.2) 46 (1.1,0.2) 13 (0.4,0.2) 25 (0.7,0.2) 4329 Nigeria 39 (2.2,0.5) 32 (1.7,0.4) 16 (1.0,0.4) 15 (1.0,0.4) 1417 PRC-Shenzhen 151 (5.3,0.8) 116 (3.8,0.6) 54 (1.8,0.4) 45 (1.4,0.3) 2468 Peru 36 (6.4,1.4) 33 (6.1,1.4) 7 (1.1,0.4) 18 (3.3,0.9) 530 Total Low and Lower-Middle 350 (3.2,0.3) 295 (2.6,0.2) 101 (0.9,0.1) 128 (1.2,0.2) 9466 Upper-Middle Brazil-Sao Paulo 548 (14.9,0.9) 471 (13.3,0.9) 183 (3.6,0.3) 230 (5.6,0.4) 2922 Lebanon 37 (1.9,0.4) 30 (1.6,0.4) 14 (0.6,0.3) 15 (0.9,0.4) 1029 Mexico 53 (4.1,1.0) 49 (3.6,0.9) 12 (0.8,0.4) 22 (1.4,0.4) 715 Romania 24 (1.0,0.4) 21 (0.9,0.4) 5 (0.1,0.1) 9 (0.3,0.1) 2357 Total Upper-Middle 662 (7.2,0.4) 571 (6.4,0.4) 214 (1.7,0.1) 276 (2.7,0.2) 7023 High Belgium 32 (8.3,2.5) 19 (5.0,1.6) 20 (5.7,2.3) 11 (4.1,2.4) 319 France 27 (5.7,1.4) 23 (4.9,1.3) 7 (1.6,0.6) 6 (1.3,0.7) 301 Germany 25 (2.8,0.5) 16 (1.8,0.4) 13 (1.3,0.3) 6 (1.0,0.2) 408 Italy 38 (4.5,0.8) 31 (3.5,1.0) 16 (1.9,0.6) 12 (1.3,0.5) 617 Netherlands 47 (10.8,2.5) 41 (10.1,2.5) 11 (1.6,0.5) 13 (3.0,1.2) 348 New Zealand 724 (6.9,0.4) 667 (6.5,0.4) 134 (0.9,0.1) 271 (2.4,0.2) 7263 Portugal 140 (5.2,0.7) 106 (3.9,0.5) 66 (2.6,0.5) 43 (1.7,0.3) 2053 Spain 91 (6.7,1.5) 77 (5.8,1.5) 35 (1.4,0.4) 19 (0.9,0.2) 1159 United States 249 (8.6,0.9) 232 (8.2,0.9) 41 (1.3,0.2) 79 (2.8,0.4) 2304 Total High 1373 (6.8,0.3) 1212 (6.2,0.3) 343 (1.4,0.1) 460 (2.2,0.2) 14772 All Countries 2385 (5.8,0.2) 2078 (5.2,0.2) 658 (1.3,0.1) 864 (2.0,0.1) 31261 a Income = Based on World Bank country level of economic development (The World Bank. (2008). Data and Statistics. Accessed May 12, 2009 at: http://go.worldbank.org/D7SN0B8YU0); b PE = Psychotic experience (any of the six types); c HE = Hallucinatory experience (either of the two types); d DE = Delusional experience (any of the four types); e n1 = Unweighted number of respondents who reported the PEs; f Prevalence estimates are based on weighted data; g N = The total unweighted number of respondents who were asked about PEs. Table 3 Lifetime prevalence of individual psychotic experiences in the World Mental Health surveys Type Lifetime prevalence na (%b, SE) HE - visual 1545 (3.8,0.2) HE - auditory (verbal) 1051 (2.5,0.1) Any HEc 2078 (5.2,0.2) DE - thought insertion/withdrawal 193 (0.4,0.0) DE - mind control/passivity 148 (0.3,0.0) DE - ideas of reference 209 (0.4,0.0) DE - plot to harm/follow 328 (0.7,0.1) Any DEd 658 (1.3,0.1) Any PEe 2385 (5.8,0.2) Exactly 1 PE type 1631 (4.2,0.2) Exactly 2 PE types 544 (1.2,0.1) 3 or more PE types 210 (0.4,0.0) Total Sample (N f ) 31261 (100,0.0) a n = Unweighted number of respondents who reported the PEs; b Prevalence estimates are based on weighted data; c HE = Hallucinatory experience (either of the two items); d DE = Delusional experience (any of the four items); e PE = Psychotic experience (any of the six items); f N = The total unweighted number of respondents who were asked about PEs. Table 4 Frequency of lifetime occurrence of psychotic experiences (PEs) among respondents who reported ever having one or more PE in the World Mental Health surveys Sample Type PE frequency metric Row Sample Size 1 2–5 6–10 11–100 101 or more na %b (SE) %b (SE) %b(SE) %b(SE) %b(SE) All countries (except for ESEMed#) HE - visual 1379 31.9 (1.7) 32.0 (1.7) 10.5 (1.0) 18.7 (1.4) 6.7 (0.9) HE - auditory (verbal) 965 20.8 (1.9) 31.7 (2.2) 11.1 (1.3) 26.5 (1.9) 9.9 (1.3)  Any HEc 1871 30.4 (1.4) 32.4 (1.4) 10.0 (0.8) 20.6 (1.3) 6.6 (0.8) DE - thought insertion/withdrawal 162 28.9 (4.1) 27.1 (3.9) 9.7 (2.0) 15.6 (2.9) 18.8 (3.3) DE - mind control/passivity 136 27.1 (5.0) 18.2 (3.5) 11.8 (2.5) 24.2 (4.5) 18.8 (4.6) DE - ideas of reference 169 15.5 (2.6) 24.5 (4.5) 13.2 (2.8) 23.7 (4.0) 23.1 (5.2) DE - plot to harm/follow 278 36.4 (3.5) 26.2 (2.8) 12.6 (2.2) 18.7 (2.4) 6.0 (1.8)  Any DEd 556 29.1 (2.5) 27.6 (2.1) 12.6 (1.5) 18.2 (1.8) 12.5 (2.0)  Any PEe 2125 31.7 (1.4) 31.9 (1.3) 10.1 (0.7) 20.0 (1.2) 6.4 (0.7) Exactly 1 PE type 1452 37.5 (1.7) 32.3 (1.5) 8.9 (0.8) 18.0 (1.4) 3.3 (0.7) Exactly 2 PE types 491 16.7 (3.3) 35.3 (3.6) 12.8 (2.5) 24.9 (2.5) 10.3 (1.9) 3 or more PE types 182 17.2 (2.8) 17.6 (3.5) 13.8 (3.4) 25.6 (3.9) 25.8 (4.3) ESEMed# only  Any PEe 260 36.4 (3.3) 30.7 (2.9) 9.5 (1.5) 20.3 (2.3) 3.2 (0.9) Exactly 1 PE type 179 43.0 (3.9) 29.6 (3.7) 4.3 (1.4) 21.0 (2.1) 2.1 (0.9) Exactly 2 PE types 53 23.6 (8.6) 33.7 (4.7) 32.3 (5.7) 6.8 (1.3) 3.6 (1.4) 3 or more PE types 28 2.1 (0.6) 34.0 (12.1) 2.5 (1.0) 48.2 (12.5) 13.2 (5.6) All countries  Any PEe 2385 32.2 (1.3) 31.8 (1.2) 10.0 (0.7) 20.0 (1.1) 6.0 (0.6) Exactly 1 PE type 1631 38.1 (1.5) 32.0 (1.4) 8.4 (0.7) 18.3 (1.2) 3.2 (0.6) Exactly 2 PE types 544 17.4 (3.1) 35.1 (3.3) 14.6 (2.3) 23.2 (2.2) 9.7 (1.8) 3 or more PE types 210 15.6 (2.5) 19.3 (3.3) 12.7 (3.1) 27.9 (3.9) 24.5 (3.9) a n = Unweighted number of respondents who reported the PEs; b Prevalence estimates are based on weighted data; c HE = Hallucinatory experience (either of the two types); d DE = Delusional experience (any of the four types); e PE = Psychotic experience (any of the six types); # ESEMeD = European Study of the Epidemiology of Mental Disorders. Table 5 Cross table of psychotic experiences (PE) frequency metric and PE type metric in the World Mental Health surveys PE frequency metric Row Sample Size 1 2–5 6 or more Cochran- Armitage test Chi-Square test Sample Type na %b (SE) %b (SE) %b (SE) z (p-value) χ22 (p-value) All countries Exactly 1 PE type 1631 38.1 (1.5) 32.0 (1.4) 29.9 (1.5) −10.0* (<.001) 32.1* (<.001) 2 or more PE types 754 16.9 (2.4) 31.2 (2.6) 51.8 (2.9) a n = Unweighted number of respondents who reported the PEs; b Prevalence estimates are based on weighted data; * Significant at 0.05 level, two-sided test (p value shown in table). Potential Conflict of Interests Over the past three years RK has been a consultant for Hoffman-La Roche Inc, Johnson & Johnson Wellness and Prevention, Sonofi-Aventis Group. RK has served on advisory boards for Mensante Corporation, Plus One Health Management, Lake Nona Institute and US Preventive Medicine. There are no other competing interests. 1 van Os J The Many Continua of Psychosis JAMA Psychiatry 2014 2 Linscott RJ van Os J An updated and conservative systematic review and meta-analysis of epidemiological evidence on psychotic experiences in children and adults: on the pathway from proneness to persistence to dimensional expression across mental disorders Psychol Med 2013 43 6 1133 1149 22850401 3 Saha S Chant D Welham J McGrath J A Systematic Review of the Prevalence of Schizophrenia PLoS Med 2005 2 5 e141 15916472 4 Kelleher I Cannon M Psychotic-like experiences in the general population: characterizing a high-risk group for psychosis Psychol Med 2010 1 6 5 Freeman D Fowler D Routes to psychotic symptoms: trauma, anxiety and psychosis-like experiences Psychiatry Res 2009 169 2 107 112 19700201 6 Varghese D Scott J Welham J Psychotic-like experiences in major depression and anxiety disorders: a population-based survey in young adults Schizophr Bull 2011 37 2 389 393 19687152 7 Johns LC Cannon M Singleton N Prevalence and correlates of self-reported psychotic symptoms in the British population Br J Psychiatry 2004 185 298 305 15458989 8 Nishida A Sasaki T Nishimura Y Psychotic-like experiences are associated with suicidal feelings and deliberate self-harm behaviors in adolescents aged 12–15 years Acta Psychiatr Scand 2010 121 4 301 307 19614622 9 Saha S Scott JG Johnston AK The association between delusional-like experiences and suicidal thoughts and behaviour Schizophr Res 2011 132 2–3 197 202 21813264 10 Kelleher I Devlin N Wigman JT Psychotic experiences in a mental health clinic sample: implications for suicidality, multimorbidity and functioning Psychol Med 2014 44 8 1615 1624 24025687 11 Saha S Scott JG Varghese D McGrath JJ The association between general psychological distress and delusional-like experiences: a large population-based study Schizophr Res 2011 127 1–3 246 251 21239145 12 Stochl J Khandaker GM Lewis G Mood, anxiety and psychotic phenomena measure a common psychopathological factor Psychol Med 2014 1 11 13 Armando M Nelson B Yung AR Psychotic-like experiences and correlation with distress and depressive symptoms in a community sample of adolescents and young adults Schizophr Res 2010 119 1–3 258 265 20347272 14 Werbeloff N Drukker M Dohrenwend BP Self-reported Attenuated Psychotic Symptoms as Forerunners of Severe Mental Disorders Later in Life Arch Gen Psychiatry 2012 15 Kaymaz N Drukker M Lieb R Do subthreshold psychotic experiences predict clinical outcomes in unselected non-help-seeking population-based samples? A systematic review and meta-analysis, enriched with new results Psychol Med 2012 1 15 16 DeVylder JE Burnette D Yang LH Co-occurrence of psychotic experiences and common mental health conditions across four racially and ethnically diverse population samples Psychol Med 2014 44 16 3503 3513 25065632 17 Carpenter WT van Os J Should attenuated psychosis syndrome be a DSM-5 diagnosis? Am J Psychiatry 2011 168 5 460 463 21536700 18 Linscott RJ van Os J Systematic reviews of categorical versus continuum models in psychosis: evidence for discontinuous subpopulations underlying a psychometric continuum. Implications for DSM-V, DSM-VI, and DSM-VII Annu Rev Clin Psychol 2010 6 391 419 20192792 19 van Os J Linscott RJ Myin-Germeys I Delespaul P Krabbendam L A systematic review and meta-analysis of the psychosis continuum: evidence for a psychosis proneness-persistence-impairment model of psychotic disorder Psychol Med 2008 1 17 20 Nuevo R Chatterji S Verdes E Naidoo N Arango C Ayuso-Mateos JL The Continuum of Psychotic Symptoms in the General Population: A Cross-national Study Schizophr Bull 2010 21 David AS Why we need more debate on whether psychotic symptoms lie on a continuum with normality Psychol Med 2010 1 8 22 World Bank World Development Indicators Washington, D.C World Bank 2003 23 Saha S Scott J Varghese D McGrath J The association between physical health and delusional-like experiences: a general population study PLoS ONE 2011 6 4 e18566 21541344 24 Saha S Scott J Varghese D McGrath J Anxiety and depressive disorders are associated with delusional-like experiences: a replication study based on a National Survey of Mental Health and Wellbeing BMJ Open 2012 2 3 25 Saha S Scott JG Varghese D Degenhardt L Slade T McGrath JJ The association between delusional-like experiences, and tobacco, alcohol or cannabis use: a nationwide population-based survey BMC Psychiatry 2011 11 202 22204498 26 Saha S Scott JG Varghese D McGrath JJ Socio-economic disadvantage and delusional-like experiences: A nationwide population-based study Eur Psychiat 2013 28 1 59 63 27 Scott J Chant D Andrews G Martin G McGrath J Association between trauma exposure and delusional experiences in a large community-based sample Br J Psychiatry 2007 190 339 343 17401041 28 Varghese D Saha S Scott JD Chan RCK McGrath JJ The Association Between Family History of Mental Disorder and Delusional-Like Experiences: A General Population Study American Journal of Medical Genetics Part B-Neuropsychiatric Genetics 2011 156B 4 478 483 29 Kessler RC Ustun TB The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders New York Cambridge University Press 2008 30 Kessler RC Haro JM Heeringa SG Pennell BE Ustun TB The World Health Organization World Mental Health Survey Initiative Epidemiol Psichiatr Soc 2006 15 3 161 166 17128617 31 Kessler RC Üstün TB The World Health Organization Composite International Diagnostic Interview Kessler RC Üstün TB The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders New York Cambridge University Press 2008 58 90 32 Kessler RC Ustun TB The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI) Int J Methods Psychiatr Res 2004 13 2 93 121 15297906 33 Moffitt TE Caspi A Taylor A How common are common mental disorders? Evidence that lifetime prevalence rates are doubled by prospective versus retrospective ascertainment Psychol Med 2010 40 6 899 909 19719899 34 Kelleher I Cannon M Whither the psychosis-neurosis borderline Schizophr Bull 2014 40 2 266 268 24436054

PMC005xxxxxx/PMC5120543.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 8803460 3682 Eur Respir J Eur. Respir. J. The European respiratory journal 0903-1936 1399-3003 27587549 5120543 10.1183/13993003.00120-2015 NIHMS830760 Article Control of Lung Defense by Mucins and Macrophages: Ancient defense mechanisms with modern functions Janssen William J. 12 Stefanski Adrianne L. 2 Bochner Bruce S. 3 Evans Christopher M. 2 1 Department of Medicine, National Jewish Health, 1400 Jackson St. Denver, Colorado 80206 2 Department of Medicine, University of Colorado School of Medicine, 12700 E 19th Avenue, Mailstop 8611, Research Complex 2, Room 3121, Aurora, Colorado, 80045, USA 3 Department of Medicine, Division of Allergy-Immunology, Northwestern University Feinberg School of Medicine, 240 E. Huron St., Room M-306, Chicago, IL 60611 Corresponding Author: Christopher M. Evans, PhD, Associate Professor, Department of Medicine, Division of Pulmonary Sciences and Critical Care, University of Colorado Denver School of Medicine, 12700 E 19th Avenue, Mailstop 8611, Aurora, CO, 80045, (303) 724-6573 [tele], Christopher.Evans@ucdenver.edu 19 11 2016 1 9 2016 10 2016 01 4 2017 48 4 12011214 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Due to the need to balance the requirement for efficient respiration in the face of tremendous levels of exposure to endogenous and environmental challenges, it is crucial for the lungs to maintain sustainable defense that minimizes damage caused by exposures and the detrimental effects of inflammation to delicate gas exchange surfaces. Accordingly, epithelial and macrophage defenses constitute essential 1st and 2nd lines of protection that prevent the accumulation of potentially harmful agents in the lungs, and under homeostatic conditions do so effectively without inducing inflammation. Though seemingly distinct, recent data show that epithelial and macrophage mediated defenses are linked through their shared reliance on airway mucins, in particular the polymeric mucin MUC5B. This review highlights our understanding of novel mechanisms that link mucus and macrophage defenses. The roles of phagocytosis and the effects of factors that are contained within mucus on phagocytosis, as well as newly identified roles for mucin glycoproteins in the direct regulation of leukocyte functions are discussed. The emergence of this nascent field of glycoimmunobiology sets forth a new paradigm for considering how homeostasis is maintained under healthy conditions and how it is restored in disease. Introduction The principal function of the lungs is gas exchange. To this end, under normal tidal breathing, 8,000–12,000 liters of air pass through lungs each day. Gas flows through multiple generations of conducting airways, which ultimately terminate in the alveoli. Alveoli are bounded by type I epithelial cells that cover over 95% of the lung surface, and to allow for efficient exchange of O2 and CO2, type I epithelia are extremely thin and together with alveolar capillaries create a diffusion distance of <1 µm. Consequently, these thin surfaces are protected by elaborate defense mechanisms that must trap and eliminate particulates and pathogens before they reach the alveolar walls, while simultaneously preventing and/or suppressing potentially inflammatory responses that could injure delicate gas exchange structures. This review concentrates on the mucociliary escalator and alveolar macrophages (AMs) as crucial first and second lines of host defense in the lungs. Airway tissues are exposed to ~100 billion inhaled particles daily (1). Airborne particles can arise from natural and manmade sources, can vary in size and chemical composition, can differ in concentrations based on geography and local environments, and can thus result in heterogeneous pathological responses (2–8). Most inspired materials are large enough to impact upon nasopharyngeal and tracheal mucosae where they are transported proximally by mucociliary clearance (MCC) and are ultimately eliminated by expectoration or swallowing. The remainder deposit in the lung periphery where they are ingested by AMs. Under healthy conditions, particulate deposition in the periphery is primarily limited to small particles (<1 µm diameter). However, under conditions where particulate concentrations are high or in pathological settings where MCC is impaired, larger particles can also accumulate in the lung periphery. Together, the coordinated functions of MCC and AMs eliminate inhaled particulates from the alveoli and airways, and hence comprise robust mechanisms for exogenous clearance. At the same time, clearance also removes endogenous materials that are generated during normal cell turnover or as a consequence of disease. Critically, although AM and MCC functions are ordinarily considered distinct, emerging data show that their functions are tightly linked through physiological and biochemical mechanisms. Below we describe mucus and macrophages separately, and this is followed by a discussion of emerging knowledge of interactions between them. The mucus barrier and MCC MCC involves the coordinated activities of secretory cells that release polymeric mucin glycoproteins, and multi-ciliated cells whose apically localized motile cilia provide a means for transport and elimination. Cilia are molecular machines whose structural and motile components are highly regulated; their complex assembly, function, and dysfunction in diseases are reviewed elsewhere (9, 10). For the purposes of this review, we consider physiological roles of motile cilia, and we highlight key aspects of mucociliary interactions that are essential in the airways. MCC requires the coordinated regulation of airway surface liquid to control the osmolarity, viscoelasticity, and resultant transportability of secreted mucus (11, 12). This control is driven by electrolyte transport machinery intracellularly as well as the presence of osmolytes in the extracellular space. Although ciliated and mucous layers have been considered as separate entities (‘sol’ and ‘gel’ phases), this distinction is challenged by recent studies demonstrating it as a more continuous glycoprotein hydrogel. Membrane mucins (MUC1, MUC4, and MUC16) that are present along cilia surfaces form a hydrated brush that allows for the free movement of cilia. The overlying, viscoelastic mucus layer is positioned atop this grafted brush of cilia. As a result, airway surface hydration regulates the balance between cilia and mucus structures maintained in a ‘gel-on-brush’ conformation that promotes effective motility and MCC (13). Loss of MCC is a significant cause of respiratory infections. For instance, impaired MCC is a primary pathophysiological feature of infection-related diseases such as primary ciliary dyskinesia (PCD) where cilia motility is impaired or absent, and cystic fibrosis (CF) where airway surface dehydration causes mucus adhesion to airway surfaces and hyperosmotic collapse of underlying cilia. Less appreciated perhaps are findings in COPD and asthma, which also show significant MCC impairment (14–21). Unlike the primary roles of altered mucus and ciliary structures in CF and PCD, COPD and asthma-related changes are secondary to inflammatory or injurious stimuli that cause impairments in ciliary motility and the dysregulated production of the two major secreted mucins, MUC5AC and MUC5B (22–25). Expression of the airway mucins MUC5AC and MUC5B Under healthy conditions, MUC5AC and MUC5B are both produced in the lungs. MUC5AC is found predominantly in surface epithelia throughout the central conducting airways, whereas MUC5B is found mainly in submucosal glands of central airways (trachea and bronchi) and in non-ciliated surface epithelial cells of peripheral airways. MUC5AC levels increase in both airway surface and glandular epithelia in asthma (22, 23) and COPD (24, 26, 27). By contrast, MUC5B levels are more variable. For example, in patients with established CF and COPD, MUC5B levels are increased in sputum (28, 29), which is predominated by central airway secretions that are supplied by tracheobronchial submucosal glands. However, in patients with early or pre-clinical COPD or with strong allergic asthma MUC5B levels actually decrease, especially within epithelial cells that line central and peripheral airway surfaces where MUC5B transcript levels are reduced by 90% or more (22–24, 27). It is thus plausible that differential repression of MUC5B could affect MCC and contribute to lung pathologies. Indeed, recent studies in mice provide mechanistic support for this. In mice, deletion of the Muc5b gene caused severe upper and lower airway MCC impairments and led to the development of lethal spontaneous infections (30). Interestingly, although chronic infection and inflammation were prominent outcomes in Muc5b knockout mice, their pathobiological impacts were stronger than those observed in models of PCD. In cilia-defective Dnaic, Pcdip1, Spef2, and Cby knockout mice, although MCC is severely impaired, upper airway pathologies were not reported to be lethal, and they did not carry over to the lower respiratory tract (31–33). Thus, among MCC components in the lungs, Muc5b is a dominant regulator of homeostatic microbial elimination. In addition, during chronic spontaneous and acute experimental infections, Muc5ac production increased in Muc5b knockout mice. Although not entirely protective itself, Muc5ac could have played a role in delaying the effects of infections (30). Possible explanations for the mucin functions in airway defense (as well as differences between Muc5ac and Muc5b) may reflect differences in their polymeric structures, glycosylation, and interactions with microbes or anti-microbial molecules. Determination of the specific and overlapping roles of Muc5ac and Muc5b remains an area of urgent investigation. Mucin Expression MUC5AC/Muc5ac and MUC5B/Muc5b gene expression levels are regulated by endogenous and environmental factors. For human MUC5B, single nucleotide polymorphisms have been shown to regulate expression via control of promoter activity (34–36). These genetic controls likely impact (or are impacted upon) by numerous innate and adaptive immune cytokine signaling pathways, as well as growth factor regulated mechanisms that are associated with responses to inflammation, injury, and tissue repair. These are reviewed extensively elsewhere (37–42). Lastly, endogenous factors include developmental (43–46) and epigenetic (47–49) regulatory mechanisms, which may play roles in the expression of mucins in cancers. Mucin polymerization The abilities of secreted mucins to regulate MCC are largely dependent on their polymer structures formed through disulfide bonds (Figure 1). Like other members of the secreted polymeric mucin family, Muc5ac and Muc5b are composed of ~5–6% cysteines (~250–300 per molecule). They have cysteine-rich N- and C- terminal von Willebrand factor (vWF) type D-like and C- terminal cysteine knot disulfide bonding domains that are critical for intermolecular mucin assembly (50–52). Additional highly conserved cysteine-rich CysD domains are interspersed in varying numbers in polymeric mucin carbohydrate-rich repeats (53–55). Through intramolecular disulfide linkages, CysD domains are proposed to form hydrophobic loop structures that facilitate mucin alignment and regulate mucus mesh spacing (56). Furthermore, in each mucin at least 100 cysteines exist that are not found in defined “domains”. In all cases, the majority of disulfide bonds are thought to form intracellularly during assembly. In the extracellular environment, free cysteines that do exist may become oxidized and form additional cross-links that increase the elastic moduli of mucus gels (57). Disruption of N- and C-terminal bonds or CysD’s may be sufficient to “loosen” obstructive mucus. Accordingly, current mucolytic therapies such as N-acetylcysteine, as well as investigative therapies, target these by reducing disulfides and decreasing mucus viscoelasticity, thereby enhancing mucus transport (58–60). A current challenge is to determine which therapies can be given at doses that are well-tolerated and still maintain the benefits of efficient defense. Mucin glycosylation While disulfide polymerization is an important but underappreciated aspect of secreted mucins, their glycosylation is perhaps more eminent. Mucins are defined by their heavy glycosylation, especially within variable-sized glycan-rich domains (see Figure 1). In MUC5AC and MUC5B, these regions are called ‘PTS’ domains due to their enrichment in prolines, threonines, and serines. PTS-rich repeats are sites of O-glycosylation, starting with N-acetylgalactosamine on serine and threonine residues. Galactose and N-acetylglucosamine are then attached and elaborated linearly or in branches, and the sugars can be modified by sulfation or by the addition of terminal sialic acid and fucose glycans. Two chief purposes of mucin glycans are to adsorb water and to participate in host defense. For water adsorption, glycan variations can greatly affect the osmotic pressures imparted by mucus gels. For example, sialylated and sulfated termini are strongly charged, and their large polar surface areas promote both hydration and electronegative repulsion (11, 13). On the other hand, fucose has a lower charge and an approximately 50% lower polar surface area, which hypothetically promotes mucus aggregation, increases viscoelasticity, and thereby inhibits MCC. For host defense, mucin glycans are known to interact with sugar binding molecules on a variety of bacteria that colonize or infect the lungs (61–67) and gastrointestinal tract (68–74), fungi such as Aspergillus fumigatus (75) and respiratory viruses such as respiratory syncytial virus and influenza (76, 77). Whether these interactions are beneficial to the host or the microbe vary widely. Nonetheless, as the result of host genetics and environmental exposures (such as infectious or allergic states) protection is limited. Impaired defense may be affected by changes in the properties of mucus (e.g., through variations in MUC5AC/Muc5ac vs MUC5B/Muc5b expression levels or glycosylation) that are often coupled with ciliary dysfunction (e.g., through loss/absence of ciliated cells or components of motile cilia) (78–91). Taken together, the roles of mucins in the formation and maintenance of a mucus gel, and their abilities to bind microorganisms demonstrate the coordinated function and dysfunction of mucus binding and clearance dynamics in host defense. In summary, this conventional view of the mucociliary barrier as a defense system regulated by mucus and ciliary functions has been refined by the identification of key factors such as Muc5b and by the dissection of complex biophysical regulation of mucociliary interactions. An immediate challenge is to relate these to specific and required molecular components that regulate their intrinsic biophysical functions. Furthermore, new findings have introduced a novel set of interactions through which mucins regulate defense and inflammation in the lungs via resident and recruited pulmonary leukocyte populations. In particular, dendritic cell, eosinophil, and macrophage functions in various tissues have been demonstrated to be regulated specifically by mucin terminal glycans. Below we focus on macrophage and eosinophil functions that are regulated by extracellular oligosaccharides, including the airway mucin Muc5b. Macrophage ontogeny and clearance mechanisms Particulates and microbes that evade the first line of defense--epithelial mucus--reach the distal lung where they must be cleared rapidly and efficiently by the second line of defense--phagocytes. AMs are the dominant phagocytic cell in the lungs, and during health account for up to 90% of the leukocytes in airspaces (92–95) They reside in the alveolar lumen, and perhaps also in the airways. In addition to clearing inhaled particulates, they are critical for removing dying cells and maintaining alveolar homeostasis. Recent evidence suggests that AMs arise from progenitors that occupy the fetal liver and yolk sac during embryogenesis (96–98). At birth, these cells populate the airspaces where they quickly mature into resident AMs. Importantly, AMs self-renew throughout life, and in the absence of disease, they are not replaced by monocytes from the circulation (99–101). During inflammation, resident AMs proliferate locally (102). At the same time monocytes from the circulation migrate to inflamed regions where they mature into macrophages, termed monocyte-derived AMs (MDAMs) (103). Hence, the inflammatory AM pool contains cells of both embryonic and post-natal origin. Although both macrophage subsets demonstrate phagocytic capacity, their respective contributions to the clearance of exogenous particulates and pathogens and to the removal of endogenous debris and cells remain unknown. Intriguingly, as inflammation resolves MDAMs undergo programmed cell death and are removed from the lungs, leaving behind the embryonically derived resident AMs to maintain alveolar homeostasis (103). During health, resident AMs function as sentinels, constantly surveying the luminal environment for pathogens and inhaled particulates. Under most circumstances, such agents are cleared silently and quickly - without inducing systemic inflammatory responses that could injure alveolar gas exchange structures. Indeed, experimental depletion of AMs results in exaggerated inflammatory responses (104–112), yet at the same time AM absence impairs the ability to control infection (107, 110, 113) demonstrating that restrained responses are more efficacious and beneficial. As discussed below, the alveolar environment plays an essential role in regulating AM endocytic and inflammatory responses, and it also contains a diverse array of molecules that recognize pathogens and facilitates clearance by non-inflammatory phagocytic defense. Phagocytic Mechanisms AMs employ a number of mechanisms to ingest particulates and pathogens, all of which involve endocytosis, a process in which the plasma membrane surrounds a target, invaginates, and then pinches off to form a membrane bound vesicle (reviewed in (114, 115)). Phagocytosis is the primary endocytic process by which AMs clear exogenous materials and is driven by cytoskeletal rearrangements that lead to rapid internalization of pathogens such as bacteria or fungi in a membrane bound phagosome. The phagosome becomes acidified after sequential fusion with endosomes and lysosomes, which contain hydrolytic enzymes and reactive oxygen species that digest and destroy the target. An initial interface that AM’s have with particles and pathogens occurs through a phagocytic synapse formed by a diverse array of plasma membrane proteins that recognize phagocytic targets through specific moieties on them, including microbial and host cell glycoconjugates. These AM receptors initiate and/or modulate phagocytosis. Phagocytic Receptors AMs are equipped with a vast repertoire of phagocytic receptors. Importantly, during microbial contact many different receptor families are often simultaneously activated. Some receptors directly recognize specific molecules on phagocytic targets (e.g., phosphatidyl serine or inflammasome molecules), whereas others bind to targets coated with opsonins (e.g., immunoglobulins, complement, and surfactant materials). In addition, whereas some (e.g. Fc receptors) lead directly to pathogen engulfment, others (e.g. Toll-like receptors (TLRs)) promote phagocytosis indirectly by upregulating the expression of phagocytic receptors and their downstream signaling molecules (116–118). Here we discuss main classes of receptors on AMs in the context of opsonins and signals present in airway mucus (119–125). Immunoglobulin (Ig) signaling is an important adaptive immune process that mediates AM phagocytosis. AMs express high levels of Fcγ-receptors I (CD64), II (CD32) and III (CD16) that recognize the Fc region of IgG. Biologically relevant concentrations IgG can be found in the alveolar lining fluid of healthy humans (126). To trigger phagocytosis, Fcγ-receptors bind multiple IgG molecules within an immune complex. FcγRI is a high affinity receptor that in addition to respiratory burst and microbial killing also leads to phagocytosis. In comparison, FcγRII and FcγRIII may also promote phagocytosis but have low binding affinity. Respiratory epithelial cells secrete IgA by transcytosis, and IgA can easily be detected in the lumens of both the proximal airways and alveoli (126, 127). AMs express low levels of both FcαRI (CD89) and Fcα/µR that bind IgA and drive phagocytosis (128). Adaptive immune Ig functions are linked to glycan structures through the recognition of carbohydrate antigens, N- and O-glycosylation of their Fc domains, and physical association with secreted mucins that have specific Ig binding domains (129–132). The complement system aids in innate host defense by opsonizing immune complexes and pathogens, enhancing their killing and removal. Alveolar lavage fluid of healthy humans contains components of the classical (C1q, C2, C3, C4) and alternative (C3, Factor B) pathways (133–135). The classical pathway is primarily activated by the interaction of C1q with antigen-antibody complexes, but it can also be activated by direct binding of C1q to bacterial, fungal and virus membrane components (136, 137). Opsonization of targets by either means can stimulate phagocytosis. AMs express three complement receptors (CRs), CR1, CR3 and CR4. CR1 is incapable of internalizing opsonized particles on its own, but can enhance Fc-mediated phagocytosis. CR3 and CR4 are heterodimers that share a common β2 integrin chain (CD18) paired with specific α chains. CR4 contains the αX subunit (CD11c) and binds to particles opsonized with C3b and iC3b fragments. CR3 contains an αM chain (also known as CD11b) with a carbohydrate-binding lectin site. Accordingly, in addition to binding particles opsonized with C3b and iC3b fragments, CR3 binds microbial cell wall glycan-containing components including LPS, mannan, β-glucan, and others (138, 139). While CR3 appears to be capable of internalizing opsonized bacteria independently (140, 141) it also functions cooperatively with other receptors including CR1, CD14, FcγR and FcαRI (138, 142–144) to enhance particle clearance. Not surprisingly, mice deficient in CR3 have impaired host defense to gram-negative bacteria, gram-positive bacteria and yeast (145, 146). Importantly, studies from rodents demonstrate that cell surface expression of complement receptors varies markedly on resident AMs versus recruited MDAMs (103): Resident AMs express high levels of CD11c/CR4 but not CD11b/CR3, whereas recruited MDAMs have high CD11b/CR3 but low CD11c/CR4. This raises the intriguing hypothesis that AM subpopulations have complementary functions to control infectious and inflammatory host defense. Like Ig’s, complement components are found in airway mucus, and their levels are upregulated in inflammation (147, 148). Furthermore, complement also increases the expression of Muc5ac in airway epithelial cells (149). Other classes of carbohydrate lectins, the C-type lectins, are calcium-dependent carbohydrate binding proteins that contain a conserved glycan recognition domain and are involved in pathogen recognition and phagocytosis (150). In the context of lung host defense, two groups of C-type lectins are well recognized: the pulmonary collectins (surfactant proteins A and D), and pathogen-binding receptors (namely the mannose receptor (CD206) and dectin-1). Surfactant proteins A and D (SP-A, SP-D) are comprised of highly oligomerized monomers that are formed by N-terminal collagen-like domains linked to a C-terminal carbohydrate recognition domain (CRD) by a central hinge region. Through their CRDs, SP-A and SP-D recognize sugar residues on microbial pathogens. Consequently, they opsonize gram-negative and gram-positive bacteria, mycobacteria, fungi, and viruses such as influenza A and respiratory syncytial virus. A number of candidate receptors for collectin-opsonized particles exist on AMs, including C1qRp, SP-R210, CD14, and the calreticulin-CD91 complex (reviewed extensively in (151)). In addition to enhancing phagocytosis through their opsonizing effects, collectins may also promote phagocytosis indirectly. For example, SP-A enhances expression of scavenger receptor A (SR-A) and may augment Fc-receptor and CR-mediated phagocytosis (152–154). In addition, both SP-A and SP-D appear to increase cell surface localization and hence the phagocytic function of the mannose receptor (155–157). The mannose receptor (CD206) is highly expressed on AMs, and contains an extracellular domain that recognizes mannose, N-acetylglucosamine, and fucose glycans. Accordingly, CD206 promotes phagocytosis of pulmonary pathogens with diverse extracellular carbohydrate signatures including Streptococcus pneumoniae, Klebsiella pneumoniae, Mycobacterium tuberculosis, Pneumocystis jerovecii, and fungi such as candida and aspergillus (158). Precise mechanisms by which CD206 participates in phagocytosis are unclear, and it is likely that interactions with co-receptors are required (159). Dectin-1 was originally identified as a dendritic cell-specific receptor, but it is also expressed on AMs (160). Dectin-1 recognizes β-glucans found in fungal cell walls (161, 162) and also particles opsonized with pentraxin-3, a protein rapidly synthesized and secreted by mononuclear phagocytes in response to pro-inflammatory signals (163). Together these classes of receptors highlight a group of surface molecules that interact with exogenous and endogenous constituents of airway surface liquid and mucus to mediate AM phagocytic defense. In immunocompetent individuals, defensive components such as IgG increase in the lungs during infection, promoting pathogen clearance through recognition of numerous antigen types, including carbohydrate epitopes. Indeed bacterial targets such as surface polysacharrides are exploited for use in developing effective pneumococcal vaccines (164). Conversely, recurrent sinopulmonary infections and impaired pathogen clearance are common in patients with Ig deficiencies (165–171). In addition, in common chronic airway diseases including asthma, COPD and cystic fibrosis, impaired clearance of microbial pathogens by AMs has been extensively documented (172–175). AM dysfunction correlates with disease severity and exacerbation frequency (176–178). While etiologies vary among diseases, common features include altered expression of phagocytic receptors, reduced lysosomal killing, and enhanced production of mediators that can worsen inflammation by inducing collateral damage to surrounding tissues. These defects in AMs are either absent or reduced in mononuclear phagocytes isolated from other sites (e.g. blood). Therefore, perturbations in the local environment appear to play a dominant role in altering AM function in these diseases. Emerging links between airway mucins and AM function Based on the distinct anatomical localization and the highly dedicated cellular mechanisms involved in the specification of mucin-producing goblet cells in the airways and phagocytic macrophages in the alveoli, there is an outward appearance of discrete compartmentalization of their functions. However, the limiting the localization of resident AM’s to the alveolar space is not entirely warranted, as intraluminal macrophages in conducting airways account for 2–8% of the total resident macrophage population in rat lungs (179–185). Even within the alveolar compartment recent evidence demonstrates that a subpopulation of AMs, termed sessile AMs, can communicate across great distances through via a calcium-dependent signaling AM:alveolar epithelial circuit that ultimately suppresses immune function (186). Recent studies show that there are indeed functional links between airway mucus and macrophage function, and that these links are crucial for host defense. At one level, secreted factors such as Ig’s and complement are abundant in secreted mucus, suggesting that mucus is an important carrier of these defensive molecules. In addition, there are also direct links between secreted mucins and resident innate immune cells through their coordinated activities during resolving inflammation and physical interactions between glycans on mucins and carbohydrate-binding lectin receptors on leukocytes such as the sialic acid binding immunoglobulin-like lectins (siglec’s). We propose that mucin-leukocyte interactions regulate homeostatic, inflammatory, and resolving immune functions through signaling and physical clearance mechanisms (Figure 2). In the mouse, the intestinal mucin, Muc2, interacts with glycan-selective immuno-regulatory receptors on dendritic cells that mediate the development of inflammatory and regulatory lymphocyte subsets. In this setting, Muc2 glycans bind to two lectins (Dectin-1 and Galectin-3) that function cooperatively with the inhibitory IgG receptor FcγR3 to suppress inflammatory signals and promote tolerance (187). In a similar vein, goblet cells have also been shown to be an important mechanism for the delivery of antigens to resident monocyte-derived dendritic cells in the small intestine (188). The result of these activities is the development of tolerance to foreign antigens introduced by ingested food particles. In the lungs, inhibitory regulation of leukocyte functions appears to be mediated by acute control of leukocyte activation states. In mice, Muc5b, through its α2,3-linked sialoside glycans binds to Siglec-F, an inhibitory SHP-phosphatase signaling immunoreceptor on eosinophils and AMs (189) (Figure 3). On eosinophils, Siglec-F mediates apoptosis (190–193), thereby functioning as a significant mechanism for resolving allergic inflammation. Indeed, mice lacking Siglec-F or one particular enzyme needed for this Muc5b sialylation step, ST3Gal-III, fail to make airway ligands for Siglec-F, and these mice display exaggerated and selective lung eosinophilia in a type 2 allergic inflammation lung model (194–198). In this context, Muc5b presumably contributes to the physical removal of cells by MCC while simultaneously preventing continued activation and mediator release into airspaces during elimination from the mouse lung. In humans, the Siglec-F paralog Siglec-8 also reduces eosinophil survival via sialylated and sulfated ligands, but the specificity observed between Muc5b and Siglec-F in mice is not as well conserved between MUC5B and Siglec-8 in humans (199–201). Rather, Siglec-9 is an isoform that is bound by MUC5B sialosides, and it is expressed on neutrophils, natural killer cells, dendritic cells, and monocytes/macrophages (199). Indeed, resident alveolar macrophages in healthy mouse lungs also express Siglec-F, but its role beyond that of a cell surface marker is not yet clear. Given the associations of mucus and macrophage dysfunction in numerous lung pathologies, determining the nature of their interactions will be of tremendous interest as the field advances. With the emergence of mucins as important mediators of defense, and the recognition of the crucial significance of the glycobiology of innate and adaptive immunity, efforts to interrogate these will involve both challenging and exciting experimental approaches. Conclusion Innate defenses in the lungs are essential for maintaining efficient gas exchange. As first and second lines of host defense, mucins and macrophages play critical roles that are integrated by their physical and physiological interactions. The emergence of these links presents a convergence of new challenges that connect epithelial and innate immune programs. Grant Support: WJJ: NIH HL109517, NIH HL114381. BSB: NIH AI72265, NIH HL107151; CME: NIH HL080396, NIH ES023384, AHA 14GRNT19990040. Figure 1 Polymeric and macromolecular structures of the major secreted mucins in the airways - MUC5AC and MUC5B MUC5AC and MUC5B (and their orthologs) have amino (N) and carboxyl (C) termini that are evolutionarily conserved in polymeric mucins and von Willebrand factor (vWF, grey and black regions). The vWF-like domains are involved in covalent intermolecular disulfide assembly of C-terminal linked dimers and N-terminal linked multimers. Multimers may exist as linear or branched structures with sizes in the 1 to >10 MDa range. Between vWF-like domains are additional cysteine rich regions (CysD domains, green hexagons) that are rich in hydrophobic amino acids and intramolecular disulfide bonds. CysD’s are suggested to mediate the distribution of mucin strands and gel-pore size after secretion in healthy mucus, but may become oxidized and increase in polymer size and stiffness in disease (57). Lastly, the majority of the remaining mucin apoprotein backbone is rich in proline, serine, and threonine. This ‘PTS’ domain (white) is an imperfect repeat region and is the primary site of O-linked glycosylation. O-linkages on serine and threonine residues form Core1–4 structures, which are defined the presence of N-acteylgalactosamine (GalNAc, yellow squares) linkages on the hydroxyl groups of serine and threonine followed by single or paired attachments of galactose (yellow circles) and/or N-acetylglucosamine (GlcNAc, blue squares). Lastly, galactose and GlcNAc glycans can be further substituted with fucose, sialic acid, and sulfates that impart diverse charges that may affect mucus gel hydration and also form 3-D structural confirmations that are critical for interactions with both pathogens and host-cell lectins. Glycan structures shown are examples of possible linkages and do not necessarily represent those found on specific mucins. Polar and non-polar glycans can be found on sugars from each core type, and may be found along the same or different branches. Figure 2 Mucin:leukocyte interactions during homeostasis and inflammation In healthy lungs, resident resting alveolar macrophages (AMs) are defensive and non-inflammatory. MUC5B from bronchioles mixes with alveolar fluids, providing a route for MUC5B to contact alveolar AMs. Homeostatic or low dose stimuli elicit defensive functions such as phagocytosis. During inflammation resident AMs can become activated, and this is associated with a decrease in their Siglec-F surface expression. In addition, leukocytes, such as monocyte-derived macrophages (which lack Siglec-F) or eosinophils (which express Siglec-F) are recruited and persist for brief periods of time. These transient populations are eliminated as inflammation resolves. In mice, resolution involves Siglec-F-mediated reductions in leukocyte activation and survival. Dampened and apoptotic cells are subsequently eliminated by MUC5B-mediated MCC. Figure 3 Putative Muc5b:Siglec-F signal transduction mechanism Muc5b, via its display of multivalent α2,3-sialic acid (SA) linkages on galactose residues, binds to the N-terminal lectin domain of Siglec-F, thereby driving immunoreceptor tyrosine-based inhibitory motif (ITIM) and ITIM-like domain activation. ITIM signals putatively activate SH2 domain-containing phosphatase (SHP) enzymes that suppress kinase-activated inflammatory signals and can also promote apoptosis. References 1 O'Connell S Au-Yeung HK Gregory CJ Matthews IP Outdoor and indoor respirable air particulate concentrations in differing urban traffic microenvironments Journal of toxicology and environmental health Part A 2008 71 1069 1072 18569618 2 Fromme H particulate matter in indoor environments--exposure situation in residences, schools, pubs, and related recreational spaces Gesundheitswesen 2006 68 714 723 17199207 3 Kliucininkas L Martuzevicius D Krugly E Prasauskas T Kauneliene V Molnar P Strandberg B Indoor and outdoor concentrations of fine particles, particle-bound pahs and volatile organic compounds in kaunas, lithuania Journal of environmental monitoring : JEM 2011 13 182 191 21082095 4 Guo H Morawska L He C Zhang YL Ayoko G Cao M Characterization of particle number concentrations and pm2.5 in a school: Influence of outdoor air pollution on indoor air Environmental science and pollution research international 2010 17 1268 1278 20195908 5 Gordon T Linking health effects to pm components, size, and sources Inhalation toxicology 2007 19 Suppl 1 3 6 17886043 6 Kappos AD Bruckmann P Eikmann T Englert N Heinrich U Hoppe P Koch E Krause GH Kreyling WG Rauchfuss K Rombout P Schulz-Klemp V Thiel WR Wichmann HE Health effects of particles in ambient air International journal of hygiene and environmental health 2004 207 399 407 15471105 7 Health effects of outdoor air pollution. Committee of the environmental and occupational health assembly of the american thoracic society Am J Respir Crit Care Med 1996 153 3 50 8542133 8 Kelly FJ Kelly J London air quality: A real world experiment in progress Biomarkers : biochemical indicators of exposure, response, and susceptibility to chemicals 2009 14 Suppl 1 5 11 9 Hildebrandt F Benzing T Katsanis N Ciliopathies N Engl J Med 2011 364 1533 1543 21506742 10 Knowles MR Daniels LA Davis SD Zariwala MA Leigh MW Primary ciliary dyskinesia. Recent advances in diagnostics, genetics, and characterization of clinical disease Am J Respir Crit Care Med 2013 188 913 922 23796196 11 Button B Okada SF Frederick CB Thelin WR Boucher RC Mechanosensitive atp release maintains proper mucus hydration of airways Science signaling 2013 6 ra46 23757023 12 Button B Picher M Boucher RC Differential effects of cyclic and constant stress on atp release and mucociliary transport by human airway epithelia J Physiol 2007 580 577 592 17317749 13 Button B Cai LH Ehre C Kesimer M Hill DB Sheehan JK Boucher RC Rubinstein M A periciliary brush promotes the lung health by separating the mucus layer from airway epithelia Science 2012 337 937 941 22923574 14 O'Riordan TG Zwang J Smaldone GC Mucociliary clearance in adult asthma The American review of respiratory disease 1992 146 598 603 1519834 15 Messina MS O'Riordan TG Smaldone GC Changes in mucociliary clearance during acute exacerbations of asthma The American review of respiratory disease 1991 143 993 997 2024856 16 Foster WM Langenback EG Bergofsky EH Lung mucociliary function in man: Interdependence of bronchial and tracheal mucus transport velocities with lung clearance in bronchial asthma and healthy subjects The Annals of occupational hygiene 1982 26 227 244 7181267 17 Thomas B Rutman A Hirst RA Haldar P Wardlaw AJ Bankart J Brightling CE O'Callaghan C Ciliary dysfunction and ultrastructural abnormalities are features of severe asthma The Journal of allergy and clinical immunology 2010 126 722 729 e722 20673980 18 Moller W Felten K Sommerer K Scheuch G Meyer G Meyer P Haussinger K Kreyling WG Deposition, retention, and translocation of ultrafine particles from the central airways and lung periphery Am J Respir Crit Care Med 2008 177 426 432 17932382 19 Morgan L Pearson M de Iongh R Mackey D van der Wall H Peters M Rutland J Scintigraphic measurement of tracheal mucus velocity in vivo Eur Respir J 2004 23 518 522 15083747 20 Smaldone GC Foster WM O'Riordan TG Messina MS Perry RJ Langenback EG Regional impairment of mucociliary clearance in chronic obstructive pulmonary disease Chest 1993 103 1390 1396 8486016 21 Hasani A Pavia D Rotondetto S Clarke SW Spiteri MA Agnew JE Effect of oral antibiotics on lung mucociliary clearance during exacerbation of chronic obstructive pulmonary disease Respir Med 1998 92 442 447 9692103 22 Ordonez CL Khashayar R Wong HH Ferrando R Wu R Hyde DM Hotchkiss JA Zhang Y Novikov A Dolganov G Fahy JV Mild and moderate asthma is associated with airway goblet cell hyperplasia and abnormalities in mucin gene expression Am J Respir Crit Care Med 2001 163 517 523 11179133 23 Woodruff PG Modrek B Choy DF Jia G Abbas AR Ellwanger A Koth LL Arron JR Fahy JV T-helper type 2-driven inflammation defines major subphenotypes of asthma Am J Respir Crit Care Med 2009 180 388 395 19483109 24 Innes AL Woodruff PG Ferrando RE Donnelly S Dolganov GM Lazarus SC Fahy JV Epithelial mucin stores are increased in the large airways of smokers with airflow obstruction Chest 2006 130 1102 1108 17035444 25 Fahy JV Dickey BF Airway mucus function and dysfunction N Engl J Med 2010 363 2233 2247 21121836 26 Caramori G Casolari P Di Gregorio C Saetta M Baraldo S Boschetto P Ito K Fabbri LM Barnes PJ Adcock IM Cavallesco G Chung KF Papi A Muc5ac expression is increased in bronchial submucosal glands of stable copd patients Histopathology 2009 55 321 331 19723147 27 Caramori G Di Gregorio C Carlstedt I Casolari P Guzzinati I Adcock IM Barnes PJ Ciaccia A Cavallesco G Chung KF Papi A Mucin expression in peripheral airways of patients with chronic obstructive pulmonary disease Histopathology 2004 45 477 484 15500651 28 Kirkham S Kolsum U Rousseau K Singh D Vestbo J Thornton DJ Muc5b is the major mucin in the gel phase of sputum in chronic obstructive pulmonary disease Am J Respir Crit Care Med 2008 178 1033 1039 18776153 29 Kirkham S Sheehan JK Knight D Richardson PS Thornton DJ Heterogeneity of airways mucus: Variations in the amounts and glycoforms of the major oligomeric mucins muc5ac and muc5b Biochem J 2002 361 537 546 11802783 30 Roy MG Livraghi-Butrico A Fletcher AA McElwee MM Evans SE Boerner RM Alexander SN Bellinghausen LK Song AS Petrova YM Tuvim MJ Adachi R Romo I Bordt AS Bowden MG Sisson JH Woodruff PG Thornton DJ Rousseau K De la Garza MM Moghaddam SJ Karmouty-Quintana H Blackburn MR Drouin SM Davis CW Terrell KA Grubb BR O'Neal WK Flores SC Cota-Gomez A Lozupone CA Donnelly JM Watson AM Hennessy CE Keith RC Yang IV Barthel L Henson PM Janssen WJ Schwartz DA Boucher RC Dickey BF Evans CM Muc5b is required for airway defence Nature 2014 505 412 416 24317696 31 Ostrowski LE Yin W Rogers TD Busalacchi KB Chua M O'Neal WK Grubb BR Conditional deletion of dnaic1 in a murine model of primary ciliary dyskinesia causes chronic rhinosinusitis Am J Respir Cell Mol Biol 2010 43 55 63 19675306 32 Voronina VA Takemaru K Treuting P Love D Grubb BR Hajjar AM Adams A Li FQ Moon RT Inactivation of chibby affects function of motile airway cilia The Journal of cell biology 2009 185 225 233 19364920 33 McKenzie CW Klonoski JM Maier T Trujillo G Vitiello PF Huber VC Lee L Enhanced response to pulmonary streptococcus pneumoniae infection is associated with primary ciliary dyskinesia in mice lacking pcdp1 and spef2 Cilia 2013 2 18 24360193 34 Nakano Y Yang IV Walts AD Watson AM Helling BA Fletcher AA Lara AR Schwarz MI Evans CM Schwartz DA Muc5b promoter variant rs35705950 affects muc5b expression in the distal airways in idiopathic pulmonary fibrosis Am J Respir Crit Care Med 2016 193 464 466 26871673 35 Kamio K Matsushita I Hijikata M Kobashi Y Tanaka G Nakata K Ishida T Tokunaga K Taguchi Y Homma S Azuma A Kudoh S Keicho N Promoter analysis and aberrant expression of the muc5b gene in diffuse panbronchiolitis Am J Respir Crit Care Med 2005 171 949 957 15709052 36 Seibold MA Wise AL Speer MC Steele MP Brown KK Loyd JE Fingerlin TE Zhang W Gudmundsson G Groshong SD Evans CM Garantziotis S Adler KB Dickey BF du Bois RM Yang IV Herron A Kervitsky D Talbert JL Markin C Park J Crews AL Slifer SH Auerbach S Roy MG Lin J Hennessy CE Schwarz MI Schwartz DA A common muc5b promoter polymorphism and pulmonary fibrosis N Engl J Med 2011 364 1503 1512 21506741 37 Whitsett JA Alenghat T Respiratory epithelial cells orchestrate pulmonary innate immunity Nat Immunol 2015 16 27 35 25521682 38 Boucherat O Boczkowski J Jeannotte L Delacourt C Cellular and molecular mechanisms of goblet cell metaplasia in the respiratory airways Exp Lung Res 2013 39 207 216 23638644 39 Evans CM Koo JS Airway mucus: The good, the bad, the sticky Pharmacology and Therapeutics 2009 121 332 348 19059283 40 Thai P Loukoianov A Wachi S Wu R Regulation of airway mucin gene expression Annu Rev Physiol 2008 70 405 429 17961085 41 Voynow JA Gendler SJ Rose MC Regulation of mucin genes in chronic inflammatory airway diseases Am J Respir Cell Mol Biol 2006 34 661 665 16456183 42 Rose MC Voynow JA Respiratory tract mucin genes and mucin glycoproteins in health and disease Physiol Rev 2006 86 245 278 16371599 43 Tompkins DH Besnard V Lange AW Wert SE Keiser AR Smith AN Lang R Whitsett JA Sox2 is required for maintenance and differentiation of bronchiolar clara, ciliated, and goblet cells PloS one 2009 4 e8248 20011520 44 Reid CJ Gould S Harris A Developmental expression of mucin genes in the human respiratory tract Am J RespirCell Mol Biol 1997 17 592 598 45 Buisine MP Devisme L Copin MC Durand-Reville M Gosselin B Aubert JP Porchet N Developmental mucin gene expression in the human respiratory tract Am J Respir Cell Mol Biol 1999 20 209 218 9922211 46 Roy MG Rahmani M Hernandez JR Alexander SN Ehre C Ho SB Evans CM Mucin production during prenatal and postnatal murine lung development Am J Respir Cell Mol Biol 2011 44 755 760 21653907 47 Vincent A Perrais M Desseyn JL Aubert JP Pigny P Van SI Epigenetic regulation (DNA methylation, histone modifications) of the 11p15 mucin genes (muc2, muc5ac, muc5b, muc6) in epithelial cancer cells Oncogene 2007 26 6566 6576 17471237 48 Perrais M Pigny P Buisine MP Porchet N Aubert JP Van Seuningen-Lempire I Aberrant expression of human mucin gene muc5b in gastric carcinoma and cancer cells. Identification and regulation of a distal promoter J Biol Chem 2001 276 15386 15396 11278696 49 Ho JJ Han SW Pan PL Deng G Kuan SF Kim YS Methylation status of promoters and expression of muc2 and muc5ac mucins in pancreatic cancer cells International journal of oncology 2003 22 273 279 12527922 50 Lang T Hansson GC Samuelsson T Gel-forming mucins appeared early in metazoan evolution Proceedings of the National Academy of Sciences of the United States of America 2007 104 16209 16214 17911254 51 Park SW Zhen G Verhaeghe C Nakagami Y Nguyenvu LT Barczak AJ Killeen N Erle DJ The protein disulfide isomerase agr2 is essential for production of intestinal mucus Proc Natl Acad Sci U S A 2009 106 6950 6955 19359471 52 Schroeder BW Verhaeghe C Park SW Nguyenvu LT Huang X Zhen G Erle DJ Agr2 is induced in asthma and promotes allergen-induced mucin overproduction Am J Respir Cell Mol Biol 2012 47 178 185 22403803 53 Guo X Zheng S Dang H Pace RG Stonebraker JR Jones CD Boellmann F Yuan G Haridass P Fedrigo O Corcoran DL Seibold MA Ranade SS Knowles MR O'Neal WK Voynow JA Genome reference and sequence variation in the large repetitive central exon of human muc5ac Am J Respir Cell Mol Biol 2014 50 223 232 24010879 54 Johansson ME Phillipson M Petersson J Velcich A Holm L Hansson GC The inner of the two muc2 mucin-dependent mucus layers in colon is devoid of bacteria Proc Natl Acad Sci U S A 2008 105 15064 15069 18806221 55 Johansson ME Sjovall H Hansson GC The gastrointestinal mucus system in health and disease Nature reviews Gastroenterology & hepatology 2013 10 352 361 23478383 56 Ambort D van der Post S Johansson ME Mackenzie J Thomsson E Krengel U Hansson GC Function of the CysD domain of the gel-forming muc2 mucin The Biochemical journal 2011 436 61 70 21338337 57 Yuan S Hollinger M Lachowicz-Scroggins ME Kerr SC Dunican EM Daniel BM Ghosh S Erzurum SC Willard B Hazen SL Huang X Carrington SD Oscarson S Fahy JV Oxidation increases mucin polymer cross-links to stiffen airway mucus gels Science translational medicine 2015 7 276ra227 58 Nair GB Ilowite JS Pharmacologic agents for mucus clearance in bronchiectasis Clinics in chest medicine 2012 33 363 370 22640851 59 Decramer M Janssens W Mucoactive therapy in copd European respiratory review : an official journal of the European Respiratory Society 2010 19 134 140 20956182 60 Balsamo R Lanata L Egan CG Mucoactive drugs European respiratory review : an official journal of the European Respiratory Society 2010 19 127 133 20956181 61 Shuter J Hatcher VB Lowy FD Staphylococcus aureus binding to human nasal mucin Infect Immun 1996 64 310 318 8557357 62 Scharfman A Van Brussel E Houdret N Lamblin G Roussel P Interactions between glycoconjugates from human respiratory airways and pseudomonas aeruginosa Am J Respir Crit Care Med 1996 154 S163 S169 8876536 63 Juge N Microbial adhesins to gastrointestinal mucus Trends in microbiology 2012 20 30 39 22088901 64 Davies J Carlstedt I Nilsson AK Hakansson A Sabharwal H van Alphen L van Ham M Svanborg C Binding of haemophilus influenzae to purified mucins from the human respiratory tract Infect Immun 1995 63 2485 2492 7790060 65 Trivier D Houdret N Courcol RJ Lamblin G Roussel P Davril M The binding of surface proteins from staphylococcus aureus to human bronchial mucins Eur Respir J 1997 10 804 810 9150316 66 Linden S Nordman H Hedenbro J Hurtig M Boren T Carlstedt I Strain- and blood group-dependent binding of helicobacter pylori to human gastric muc5ac glycoforms Gastroenterology 2002 123 1923 1930 12454849 67 Linden SK Wickstrom C Lindell G Gilshenan K Carlstedt I Four modes of adhesion are used during helicobacter pylori binding to human mucins in the oral and gastric niches Helicobacter 2008 13 81 93 18321298 68 Marionneau S Airaud F Bovin NV Le Pendu J Ruvoen-Clouet N Influence of the combined abo, fut2, and fut3 polymorphism on susceptibility to norwalk virus attachment The Journal of infectious diseases 2005 192 1071 1077 16107962 69 Marionneau S Ruvoen N Le Moullac-Vaidye B Clement M Cailleau-Thomas A Ruiz-Palacois G Huang P Jiang X Le Pendu J Norwalk virus binds to histo-blood group antigens present on gastroduodenal epithelial cells of secretor individuals Gastroenterology 2002 122 1967 1977 12055602 70 Huang P Farkas T Marionneau S Zhong W Ruvoen-Clouet N Morrow AL Altaye M Pickering LK Newburg DS LePendu J Jiang X Noroviruses bind to human abo, lewis, and secretor histo-blood group antigens: Identification of 4 distinct strain-specific patterns The Journal of infectious diseases 2003 188 19 31 12825167 71 Linden S Semino-Mora C Liu H Rick J Dubois A Role of mucin lewis status in resistance to helicobacter pylori infection in pediatric patients Helicobacter 2010 15 251 258 20633185 72 Celli JP Turner BS Afdhal NH Keates S Ghiran I Kelly CP Ewoldt RH McKinley GH So P Erramilli S Bansil R Helicobacter pylori moves through mucus by reducing mucin viscoelasticity Proc Natl Acad Sci U S A 2009 106 14321 14326 19706518 73 Kawakubo M Ito Y Okimura Y Kobayashi M Sakura K Kasama S Fukuda MN Fukuda M Katsuyama T Nakayama J Natural antibiotic function of a human gastric mucin against helicobacter pylori infection Science 2004 305 1003 1006 15310903 74 Pacheco AR Curtis MM Ritchie JM Munera D Waldor MK Moreira CG Sperandio V Fucose sensing regulates bacterial intestinal colonization Nature 2012 492 113 117 23160491 75 Kerr SC Fischer GJ Sinha M McCabe O Palmer JM Choera T Yun Lim F Wimmerova M Carrington SD Yuan S Lowell CA Oscarson S Keller NP Fahy JV Flea expression in aspergillus fumigatus is recognized by fucosylated structures on mucins and macrophages to prevent lung infection PLoS pathogens 2016 12 e1005555 27058347 76 Okada SF Zhang L Kreda SM Abdullah LH Davis CW Pickles RJ Lazarowski ER Boucher RC Coupled nucleotide and mucin hypersecretion from goblet-cell metaplastic human airway epithelium Am J Respir Cell Mol Biol 2011 45 253 260 20935191 77 Ehre C Worthington EN Liesman RM Grubb BR Barbier D O'Neal WK Sallenave JM Pickles RJ Boucher RC Overexpressing mouse model demonstrates the protective role of muc5ac in the lungs Proceedings of the National Academy of Sciences of the United States of America 2012 109 16528 16533 23012413 78 Corfield AP Berry M Glycan variation and evolution in the eukaryotes Trends Biochem Sci 2015 40 351 359 26002999 79 Corfield AP Mucins: A biologically relevant glycan barrier in mucosal protection Biochimica et biophysica acta 2015 1850 236 252 24821013 80 Hanisch FG Bonar D Schloerer N Schroten H Human trefoil factor 2 is a lectin that binds alpha-glcnac-capped mucin glycans with antibiotic activity against helicobacter pylori The Journal of biological chemistry 2014 289 27363 27375 25124036 81 Tan L Psaltis A Baker LM McGuckin M Rousseau K Wormald PJ Aberrant mucin glycoprotein patterns of chronic rhinosinusitis patients with bacterial biofilms American journal of rhinology & allergy 2010 24 319 324 21244730 82 Magalhaes A Gomes J Ismail MN Haslam SM Mendes N Osorio H David L Le PJ Haas R Dell A Boren T Reis CA Fut2-null mice display an altered glycosylation profile and impaired baba-mediated helicobacter pylori adhesion to gastric mucosa Glycobiology 2009 19 1525 1536 19706747 83 Mata M Sarrion I Armengot M Carda C Martinez I Melero JA Cortijo J Respiratory syncytial virus inhibits ciliagenesis in differentiated normal human bronchial epithelial cells: Effectiveness of n-acetylcysteine PloS one 2012 7 e48037 23118923 84 Oshansky CM Pickens JA Bradley KC Jones LP Saavedra-Ebner GM Barber JP Crabtree JM Steinhauer DA Tompkins SM Tripp RA Avian influenza viruses infect primary human bronchial epithelial cells unconstrained by sialic acid alpha2,3 residues PloS one 2011 6 e21183 21731666 85 Lachowicz-Scroggins ME Boushey HA Finkbeiner WE Widdicombe JH Interleukin-13-induced mucous metaplasia increases susceptibility of human airway epithelium to rhinovirus infection Am J Respir Cell Mol Biol 2010 43 652 661 20081054 86 Pittet LA Hall-Stoodley L Rutkowski MR Harmsen AG Influenza virus infection decreases tracheal mucociliary velocity and clearance of streptococcus pneumoniae Am J Respir Cell Mol Biol 2010 42 450 460 19520922 87 Winter C Herrler G Neumann U Infection of the tracheal epithelium by infectious bronchitis virus is sialic acid dependent Microbes and infection/Institut Pasteur 2008 10 367 373 88 Wu CA Peluso JJ Shanley JD Puddington L Thrall RS Murine cytomegalovirus influences foxj1 expression, ciliogenesis, and mucus plugging in mice with allergic airway disease The American journal of pathology 2008 172 714 724 18258850 89 Thompson CI Barclay WS Zambon MC Pickles RJ Infection of human airway epithelium by human and avian strains of influenza a virus Journal of virology 2006 80 8060 8068 16873262 90 Huang T You Y Spoor MS Richer EJ Kudva VV Paige RC Seiler MP Liebler JM Zabner J Plopper CG Brody SL Foxj1 is required for apical localization of ezrin in airway epithelial cells J Cell Sci 2003 116 4935 4945 14625387 91 Look DC Walter MJ Williamson MR Pang L You Y Sreshta JN Johnson JE Zander DS Brody SL Effects of paramyxoviral infection on airway epithelial cell foxj1 expression, ciliogenesis, and mucociliary function The American journal of pathology 2001 159 2055 2069 11733356 92 Zaynagetdinov R Sherrill TP Kendall PL Segal BH Weller KP Tighe RM Blackwell TS Identification of myeloid cell subsets in murine lungs using flow cytometry Am J Respir Cell Mol Biol 2013 49 180 189 23492192 93 Desch AN Gibbings SL Goyal R Kolde R Bednarek J Bruno T Slansky JE Jacobelli J Mason R Ito Y Messier E Randolph GJ Prabagar M Atif SM Segura E Xavier RJ Bratton DL Janssen WJ Henson PM Jakubzick CV Flow cytometric analysis of mononuclear phagocytes in non-diseased human lung and lung-draining lymph nodes Am J Respir Crit Care Med 2015 94 Heron M Grutters JC ten Dam-Molenkamp KM Hijdra D van Heugten-Roeling A Claessen AM Ruven HJ van den Bosch JM van Velzen-Blad H Bronchoalveolar lavage cell pattern from healthy human lung Clinical and experimental immunology 2012 167 523 531 22288596 95 Lensmar C Elmberger G Sandgren P Skold CM Eklund A Leukocyte counts and macrophage phenotypes in induced sputum and bronchoalveolar lavage fluid from normal subjects Eur Respir J 1998 12 595 600 9762786 96 Hashimoto D Chow A Noizat C Teo P Beasley MB Leboeuf M Becker CD See P Price J Lucas D Greter M Mortha A Boyer SW Forsberg EC Tanaka M van Rooijen N Garcia-Sastre A Stanley ER Ginhoux F Frenette PS Merad M Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes Immunity 2013 38 792 804 23601688 97 Guilliams M De Kleer I Henri S Post S Vanhoutte L De Prijck S Deswarte K Malissen B Hammad H Lambrecht BN Alveolar macrophages develop from fetal monocytes that differentiate into long-lived cells in the first week of life via gm-csf The Journal of experimental medicine 2013 210 1977 1992 24043763 98 Epelman S Lavine KJ Randolph GJ Origin and functions of tissue macrophages Immunity 2014 41 21 35 25035951 99 Coggle JE Tarling JD The proliferation kinetics of pulmonary alveolar macrophages Journal of leukocyte biology 1984 35 317 327 6584524 100 Tarling JD Coggle JE Evidence for the pulmonary origin of alveolar macrophages Cell and tissue kinetics 1982 15 577 584 7172195 101 Sawyer RT The significance of local resident pulmonary alveolar macrophage proliferation to population renewal Journal of leukocyte biology 1986 39 77 87 3455712 102 Jenkins SJ Ruckerl D Cook PC Jones LH Finkelman FD van Rooijen N MacDonald AS Allen JE Local macrophage proliferation, rather than recruitment from the blood, is a signature of th2 inflammation Science 2011 332 1284 1288 21566158 103 Janssen WJ Barthel L Muldrow A Oberley-Deegan RE Kearns MT Jakubzick C Henson PM Fas determines differential fates of resident and recruited macrophages during resolution of acute lung injury Am J Respir Crit Care Med 2011 184 547 560 21471090 104 Thepen T Van Rooijen N Kraal G Alveolar macrophage elimination in vivo is associated with an increase in pulmonary immune response in mice The Journal of experimental medicine 1989 170 499 509 2526847 105 Thepen T Hoeben K Breve J Kraal G Alveolar macrophages down-regulate local pulmonary immune responses against intratracheally administered t-cell-dependent, but not t-cell-independent antigens Immunology 1992 76 60 64 1628902 106 Bang BR Chun E Shim EJ Lee HS Lee SY Cho SH Min KU Kim YY Park HW Alveolar macrophages modulate allergic inflammation in a murine model of asthma Experimental & molecular medicine 2011 43 275 280 21415590 107 Kooguchi K Hashimoto S Kobayashi A Kitamura Y Kudoh I Wiener-Kronish J Sawa T Role of alveolar macrophages in initiation and regulation of inflammation in pseudomonas aeruginosa pneumonia Infect Immun 1998 66 3164 3169 9632581 108 Holt PG Down-regulation of immune responses in the lower respiratory tract: The role of alveolar macrophages Clinical and experimental immunology 1986 63 261 270 3516464 109 Zaslona Z Przybranowski S Wilke C van Rooijen N Teitz-Tennenbaum S Osterholzer JJ Wilkinson JE Moore BB Peters-Golden M Resident alveolar macrophages suppress, whereas recruited monocytes promote, allergic lung inflammation in murine models of asthma J Immunol 2014 193 4245 4253 25225663 110 Broug-Holub E Toews GB van Iwaarden JF Strieter RM Kunkel SL Paine R 3rd Standiford TJ Alveolar macrophages are required for protective pulmonary defenses in murine klebsiella pneumonia: Elimination of alveolar macrophages increases neutrophil recruitment but decreases bacterial clearance and survival Infect Immun 1997 65 1139 1146 9119443 111 Machado-Aranda D M VS Yu B Dolgachev V Hemmila MR Raghavendran K Alveolar macrophage depletion increases the severity of acute inflammation following nonlethal unilateral lung contusion in mice The journal of trauma and acute care surgery 2014 76 982 990 24662861 112 Elder A Johnston C Gelein R Finkelstein J Wang Z Notter R Oberdorster G Lung inflammation induced by endotoxin is enhanced in rats depleted of alveolar macrophages with aerosolized clodronate Exp Lung Res 2005 31 527 546 16019986 113 Qiu H KuoLee R Harris G Van Rooijen N Patel GB Chen W Role of macrophages in early host resistance to respiratory acinetobacter baumannii infection PloS one 2012 7 e40019 22768201 114 Conner SD Schmid SL Regulated portals of entry into the cell Nature 2003 422 37 44 12621426 115 Kumari S Mg S Mayor S Endocytosis unplugged: Multiple ways to enter the cell Cell research 2010 20 256 275 20125123 116 Doyle SE O'Connell RM Miranda GA Vaidya SA Chow EK Liu PT Suzuki S Suzuki N Modlin RL Yeh WC Lane TF Cheng G Toll-like receptors induce a phagocytic gene program through p38 The Journal of experimental medicine 2004 199 81 90 14699082 117 Blander JM Medzhitov R Regulation of phagosome maturation by signals from toll-like receptors Science 2004 304 1014 1018 15143282 118 Kong L Ge BX Myd88-independent activation of a novel actin-cdc42/rac pathway is required for toll-like receptor-stimulated phagocytosis Cell research 2008 18 745 755 18542102 119 Kotlar HK Harbitz O Jenssen AO Smidsrod O Quantitation of proteins in sputum from patients with chronic obstructive lung disease. Ii. Determination of albumin, transferrin, alpha1-acid glycoprotein, igg, igm, lysozyme and c3-complement factor European journal of respiratory diseases 1980 61 233 239 7202596 120 Hill SL Mitchell JL Burnett D Stockley RA Igg subclasses in the serum and sputum from patients with bronchiectasis Thorax 1998 53 463 468 9713445 121 Sloane AJ Lindner RA Prasad SS Sebastian LT Pedersen SK Robinson M Bye PT Nielson DW Harry JL Proteomic analysis of sputum from adults and children with cystic fibrosis and from control subjects Am J Respir Crit Care Med 2005 172 1416 1426 16166615 122 Aitken ML Greene KE Tonelli MR Burns JL Emerson JC Goss CH Gibson RL Analysis of sequential aliquots of hypertonic saline solution-induced sputum from clinically stable patients with cystic fibrosis Chest 2003 123 792 799 12628880 123 Wright SM Hockey PM Enhorning G Strong P Reid KB Holgate ST Djukanovic R Postle AD Altered airway surfactant phospholipid composition and reduced lung function in asthma J Appl Physiol (1985) 2000 89 1283 1292 11007560 124 Shimura S Masuda T Takishima T Shirato K Surfactant apoprotein-a concentration in airway secretions for the detection of pulmonary oedema Eur Respir J 1996 9 2525 2530 8980964 125 Masuda T Shimura S Sasaki H Takishima T Surfactant apoprotein-a concentration in sputum for diagnosis of pulmonary alveolar proteinosis Lancet 1991 337 580 582 1671943 126 Peebles RS Jr Liu MC Lichtenstein LM Hamilton RG Iga, igg and igm quantification in bronchoalveolar lavage fluids from allergic rhinitics, allergic asthmatics, and normal subjects by monoclonal antibody-based immunoenzymetric assays Journal of immunological methods 1995 179 77 86 7868927 127 Sato T Kitajima A Ohmoto S Chikuma M Kado M Nagai S Izumi T Takeyama M Masada M Determination of human immunoglobulin a and secretory immunoglobulin a in bronchoalveolar lavage fluids by solid phase enzyme immunoassay Clinica chimica acta; international journal of clinical chemistry 1993 220 145 156 8111959 128 Bakema JE van Egmond M The human immunoglobulin a fc receptor fcalphari: A multifaceted regulator of mucosal immunity Mucosal immunology 2011 4 612 624 21937986 129 Arnold JN Wormald MR Sim RB Rudd PM Dwek RA The impact of glycosylation on the biological function and structure of human immunoglobulins Annu Rev Immunol 2007 25 21 50 17029568 130 Wolfert MA Boons G-J Adaptive immune activation: Glycosylation does matter Nat Chem Biol 2013 9 776 784 24231619 131 Crook CK Lipsitt LP Neonatal nutritive sucking: Effects of taste stimulation upon sucking rhythm and heart rate Child development 1976 47 518 522 1269318 132 Pelaseyed T Bergstrom JH Gustafsson JK Ermund A Birchenough GM Schutte A van der Post S Svensson F Rodriguez-Pineiro AM Nystrom EE Wising C Johansson ME Hansson GC The mucus and mucins of the goblet cells and enterocytes provide the first defense line of the gastrointestinal tract and interact with the immune system Immunological reviews 2014 260 8 20 24942678 133 Watford WT Ghio AJ Wright JR Complement-mediated host defense in the lung Am J Physiol Lung Cell Mol Physiol 2000 279 L790 L798 11053012 134 Reynolds HY Newball HH Analysis of proteins and respiratory cells obtained from human lungs by bronchial lavage The Journal of laboratory and clinical medicine 1974 84 559 573 4412011 135 Robertson J Caldwell JR Castle JR Waldman RH Evidence for the presence of components of the alternative (properdin) pathway of complement activation in respiratory secretions J Immunol 1976 117 900 903 956659 136 Nayak A Ferluga J Tsolaki AG Kishore U The non-classical functions of the classical complement pathway recognition subcomponent c1q Immunology letters 2010 131 139 150 20381531 137 Nayak A Pednekar L Reid KB Kishore U Complement and non-complement activating functions of c1q: A prototypical innate immune molecule Innate immunity 2012 18 350 363 21450789 138 Ehlers MR Cr3: A general purpose adhesion-recognition receptor essential for innate immunity Microbes and infection/Institut Pasteur 2000 2 289 294 139 Gasque P Complement: A unique innate immune sensor for danger signals Molecular immunology 2004 41 1089 1098 15476920 140 Ross GD Cain JA Lachmann PJ Membrane complement receptor type three (cr3) has lectin-like properties analogous to bovine conglutinin as functions as a receptor for zymosan and rabbit erythrocytes as well as a receptor for ic3b J Immunol 1985 134 3307 3315 2984286 141 Thornton BP Vetvicka V Pitman M Goldman RC Ross GD Analysis of the sugar specificity and molecular location of the beta-glucan-binding lectin site of complement receptor type 3 (cd11b/cd18) J Immunol 1996 156 1235 1246 8558003 142 Troelstra A de Graaf-Miltenburg LA van Bommel T Verhoef J Van Kessel KP Van Strijp JA Lipopolysaccharide-coated erythrocytes activate human neutrophils via cd14 while subsequent binding is through cd11b/cd18 J Immunol 1999 162 4220 4225 10201950 143 Sutterwala FS Rosenthal LA Mosser DM Cooperation between cr1 (cd35) and cr3 (cd 11b/cd18) in the binding of complement-opsonized particles Journal of leukocyte biology 1996 59 883 890 8691074 144 van Egmond M van Vuuren AJ Morton HC van Spriel AB Shen L Hofhuis FM Saito T Mayadas TN Verbeek JS van de Winkel JG Human immunoglobulin a receptor (fcalphari, cd89) function in transgenic mice requires both fcr gamma chain and cr3 (cd11b/cd18) Blood 1999 93 4387 4394 10361137 145 Rijneveld AW de Vos AF Florquin S Verbeek JS van der Poll T Cd11b limits bacterial outgrowth and dissemination during murine pneumococcal pneumonia The Journal of infectious diseases 2005 191 1755 1760 15838804 146 Agramonte-Hevia J Gonzalez-Arenas A Barrera D Velasco-Velazquez M Gram-negative bacteria and phagocytic cell interaction mediated by complement receptor 3 FEMS immunology and medical microbiology 2002 34 255 266 12443825 147 Humbles AA Lu B Nilsson CA Lilly C Israel E Fujiwara Y Gerard NP Gerard C A role for the c3a anaphylatoxin receptor in the effector phase of asthma Nature 2000 406 998 1001 10984054 148 Wu J Kobayashi M Sousa EA Liu W Cai J Goldman SJ Dorner AJ Projan SJ Kavuru MS Qiu Y Thomassen MJ Differential proteomic analysis of bronchoalveolar lavage fluid in asthmatics following segmental antigen challenge Molecular & cellular proteomics : MCP 2005 4 1251 1264 15951573 149 Dillard P Wetsel RA Drouin SM Complement c3a regulates muc5ac expression by airway clara cells independently of th2 responses Am J Respir Crit Care Med 2007 175 1250 1258 17400733 150 Zelensky AN Gready JE The c-type lectin-like domain superfamily The FEBS journal 2005 272 6179 6217 16336259 151 Kishore U Greenhough TJ Waters P Shrive AK Ghai R Kamran MF Bernal AL Reid KB Madan T Chakraborty T Surfactant proteins sp-a and sp-d: Structure, function and receptors Molecular immunology 2006 43 1293 1315 16213021 152 Kuronuma K Sano H Kato K Kudo K Hyakushima N Yokota S Takahashi H Fujii N Suzuki H Kodama T Abe S Kuroki Y Pulmonary surfactant protein a augments the phagocytosis of streptococcus pneumoniae by alveolar macrophages through a casein kinase 2-dependent increase of cell surface localization of scavenger receptor a The Journal of biological chemistry 2004 279 21421 21430 14993215 153 Watford WT Smithers MB Frank MM Wright JR Surfactant protein a enhances the phagocytosis of c1q-coated particles by alveolar macrophages Am J Physiol Lung Cell Mol Physiol 2002 283 L1011 L1022 12376354 154 Gil M McCormack FX Levine AM Surfactant protein a modulates cell surface expression of cr3 on alveolar macrophages and enhances cr3-mediated phagocytosis The Journal of biological chemistry 2009 284 7495 7504 19155216 155 Beharka AA Crowther JE McCormack FX Denning GM Lees J Tibesar E Schlesinger LS Pulmonary surfactant protein a activates a phosphatidylinositol 3-kinase/calcium signal transduction pathway in human macrophages: Participation in the up-regulation of mannose receptor activity J Immunol 2005 175 2227 2236 16081790 156 Beharka AA Gaynor CD Kang BK Voelker DR McCormack FX Schlesinger LS Pulmonary surfactant protein a up-regulates activity of the mannose receptor, a pattern recognition receptor expressed on human macrophages J Immunol 2002 169 3565 3573 12244146 157 Kudo K Sano H Takahashi H Kuronuma K Yokota S Fujii N Shimada K Yano I Kumazawa Y Voelker DR Abe S Kuroki Y Pulmonary collectins enhance phagocytosis of mycobacterium avium through increased activity of mannose receptor J Immunol 2004 172 7592 7602 15187139 158 Kerrigan AM Brown GD C-type lectins and phagocytosis Immunobiology 2009 214 562 575 19261355 159 Le Cabec V Emorine LJ Toesca I Cougoule C Maridonneau-Parini I The human macrophage mannose receptor is not a professional phagocytic receptor Journal of leukocyte biology 2005 77 934 943 15767290 160 Brown GD Taylor PR Reid DM Willment JA Williams DL Martinez-Pomares L Wong SY Gordon S Dectin-1 is a major beta-glucan receptor on macrophages The Journal of experimental medicine 2002 196 407 412 12163569 161 Brown GD Herre J Williams DL Willment JA Marshall AS Gordon S Dectin-1 mediates the biological effects of beta-glucans The Journal of experimental medicine 2003 197 1119 1124 12719478 162 Brown GD Gordon S Immune recognition. A new receptor for beta-glucans Nature 2001 413 36 37 163 Diniz SN Nomizo R Cisalpino PS Teixeira MM Brown GD Mantovani A Gordon S Reis LF Dias AA Ptx3 function as an opsonin for the dectin-1-dependent internalization of zymosan by macrophages Journal of leukocyte biology 2004 75 649 656 14726497 164 Pilishvili T Bennett NM Pneumococcal disease prevention among adults: Strategies for the use of pneumococcal vaccines Vaccine 2015 33 Suppl 4 D60 D65 26116257 165 Schwitzguebel AJ Jandus P Lacroix JS Seebach JD Harr T Immunoglobulin deficiency in patients with chronic rhinosinusitis: Systematic review of the literature and meta-analysis The Journal of allergy and clinical immunology 2015 136 1523 1531 26329513 166 Gathmann B Mahlaoui N Gerard L Oksenhendler E Warnatz K Schulze I Kindle G Kuijpers TW van Beem RT Guzman D Workman S Soler-Palacin P De Gracia J Witte T Schmidt RE Litzman J Hlavackova E Thon V Borte M Borte S Kumararatne D Feighery C Longhurst H Helbert M Szaflarska A Sediva A Belohradsky BH Jones A Baumann U Meyts I Kutukculer N Wagstrom P Galal NM Roesler J Farmaki E Zinovieva N Ciznar P Papadopoulou-Alataki E Bienemann K Velbri S Panahloo Z Grimbacher B Clinical picture and treatment of 2212 patients with common variable immunodeficiency The Journal of allergy and clinical immunology 2014 134 116 126 24582312 167 Oksenhendler E Gerard L Fieschi C Malphettes M Mouillot G Jaussaud R Viallard JF Gardembas M Galicier L Schleinitz N Suarez F Soulas-Sprauel P Hachulla E Jaccard A Gardeur A Theodorou I Rabian C Debre P Infections in 252 patients with common variable immunodeficiency Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2008 46 1547 1554 18419489 168 Kainulainen L Suonpaa J Nikoskelainen J Svedstrom E Vuorinen T Meurman O Ruuskanen O Bacteria and viruses in maxillary sinuses of patients with primary hypogammaglobulinemia Archives of otolaryngology--head & neck surgery 2007 133 597 602 17576911 169 Busse PJ Farzan S Cunningham-Rundles C Pulmonary complications of common variable immunodeficiency Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology 2007 98 1 8 quiz 8–11, 43 170 Kainulainen L Varpula M Liippo K Svedstrom E Nikoskelainen J Ruuskanen O Pulmonary abnormalities in patients with primary hypogammaglobulinemia The Journal of allergy and clinical immunology 1999 104 1031 1036 10550749 171 Cunningham-Rundles C Bodian C Common variable immunodeficiency: Clinical and immunological features of 248 patients Clin Immunol 1999 92 34 48 10413651 172 Fitzpatrick AM Holguin F Teague WG Brown LA Alveolar macrophage phagocytosis is impaired in children with poorly controlled asthma The Journal of allergy and clinical immunology 2008 121 1372 1378 1378 e1371 e1373 18417198 173 Berenson CS Garlipp MA Grove LJ Maloney J Sethi S Impaired phagocytosis of nontypeable haemophilus influenzae by human alveolar macrophages in chronic obstructive pulmonary disease The Journal of infectious diseases 2006 194 1375 1384 17054066 174 Taylor AE Finney-Hayward TK Quint JK Thomas CM Tudhope SJ Wedzicha JA Barnes PJ Donnelly LE Defective macrophage phagocytosis of bacteria in copd Eur Respir J 2010 35 1039 1047 19897561 175 Marti-Lliteras P Regueiro V Morey P Hood DW Saus C Sauleda J Agusti AG Bengoechea JA Garmendia J Nontypeable haemophilus influenzae clearance by alveolar macrophages is impaired by exposure to cigarette smoke Infect Immun 2009 77 4232 4242 19620348 176 Berenson CS Kruzel RL Eberhardt E Sethi S Phagocytic dysfunction of human alveolar macrophages and severity of chronic obstructive pulmonary disease The Journal of infectious diseases 2013 208 2036 2045 23908477 177 Alexis NE Soukup J Nierkens S Becker S Association between airway hyperreactivity and bronchial macrophage dysfunction in individuals with mild asthma Am J Physiol Lung Cell Mol Physiol 2001 280 L369 L375 11159017 178 Donnelly LE Barnes PJ Defective phagocytosis in airways disease Chest 2012 141 1055 1062 22474147 179 Lehnert BE Pulmonary and thoracic macrophage subpopulations and clearance of particles from the lung Environmental health perspectives 1992 97 17 46 1396454 180 Lehnert BE Valdez YE Sebring RJ Lehnert NM Saunders GC Steinkamp JA Airway intra-luminal macrophages: Evidence of origin and comparisons to alveolar macrophages Am J Respir Cell Mol Biol 1990 3 377 391 2206541 181 Lehnert BE Morrow PE Association of 59iron oxide with alveolar macrophages during alveolar clearance Exp Lung Res 1985 9 1 16 4065055 182 Lehnert BE Valdez YE Holland LM Pulmonary macrophages: Alveolar and interstitial populations Exp Lung Res 1985 9 177 190 3000757 183 Warheit DB Hill LH Brody AR Surface morphology and correlated phagocytic capacity of pulmonary macrophages lavaged from the lungs of rats Exp Lung Res 1984 6 71 82 6329670 184 Yeh HC Schum GM Duggan MT Anatomic models of the tracheobronchial and pulmonary regions of the rat The Anatomical record 1979 195 483 492 507403 185 Sorokin SP Brain JD Pathways of clearance in mouse lungs exposed to iron oxide aerosols The Anatomical record 1975 181 581 625 1122038 186 Westphalen K Gusarova GA Islam MN Subramanian M Cohen TS Prince AS Bhattacharya J Sessile alveolar macrophages communicate with alveolar epithelium to modulate immunity Nature 2014 506 503 506 24463523 187 Shan M Gentile M Yeiser JR Walland AC Bornstein VU Chen K He B Cassis L Bigas A Cols M Comerma L Huang B Blander JM Xiong H Mayer L Berin C Augenlicht LH Velcich A Cerutti A Mucus enhances gut homeostasis and oral tolerance by delivering immunoregulatory signals Science 2013 342 447 453 24072822 188 McDole JR Wheeler LW McDonald KG Wang B Konjufca V Knoop KA Newberry RD Miller MJ Goblet cells deliver luminal antigen to cd103+ dendritic cells in the small intestine Nature 2012 483 345 349 22422267 189 Patnode ML Cheng CW Chou CC Singer MS Elin MS Uchimura K Crocker PR Khoo KH Rosen SD Galactose-6-o-sulfotransferases are not required for the generation of siglec-f ligands in leukocytes or lung tissue The Journal of biological chemistry 2013 190 Kiwamoto T Katoh T Evans CM Janssen WJ Brummet ME Hudson SA Zhu Z Tiemeyer M Bochner BS Endogenous airway mucins carry glycans that bind siglec-f and induce eosinophil apoptosis The Journal of allergy and clinical immunology 2015 135 1329 1340 e1321 e1329 25497369 191 Mao H Kano G Hudson SA Brummet M Zimmermann N Zhu Z Bochner BS Mechanisms of siglec-f-induced eosinophil apoptosis: A role for caspases but not for shp-1, src kinases, nadph oxidase or reactive oxygen PloS one 2013 8 e68143 23840825 192 Zimmermann N McBride ML Yamada Y Hudson SA Jones C Cromie KD Crocker PR Rothenberg ME Bochner BS Siglec-f antibody administration to mice selectively reduces blood and tissue eosinophils Allergy 2008 63 1156 1163 18699932 193 Song DJ Cho JY Lee SY Miller M Rosenthal P Soroosh P Croft M Zhang M Varki A Broide DH Anti-siglec-f antibody reduces allergen-induced eosinophilic inflammation and airway remodeling J Immunol 2009 183 5333 5341 19783675 194 Guo JP Myers AC Choi O Lee HS Zhu Z Hudson SA Brummet M Bovin NV Crocker PR Bochner BS Ligands for siglec-8 and siglec-f: Binding characteristics and tissue distribution Journal of Allergy and Clinical Immunology 119 S299 195 Kiwamoto T Brummet ME Wu F Motari MG Smith DF Schnaar RL Zhu Z Bochner BS Mice deficient in the st3gal3 gene product alpha2,3 sialyltransferase (st3gal-iii) exhibit enhanced allergic eosinophilic airway inflammation The Journal of allergy and clinical immunology 2014 133 240 247 e241 e243 23830412 196 Suzukawa M Miller M Rosenthal P Cho JY Doherty TA Varki A Broide D Sialyltransferase st3gal-iii regulates siglec-f ligand formation and eosinophilic lung inflammation in mice J Immunol 2013 190 5939 5948 23677475 197 Zhang M Angata T Cho JY Miller M Broide DH Varki A Defining the in vivo function of siglec-f, a cd33-related siglec expressed on mouse eosinophils Blood 2007 109 4280 4287 17272508 198 Cho JY Song DJ Pham A Rosenthal P Miller M Dayan S Doherty TA Varki A Broide DH Chronic ova allergen challenged siglec-f deficient mice have increased mucus, remodeling, and epithelial siglec-f ligands which are up-regulated by il-4 and il-13 Respiratory research 2010 11 154 21040544 199 Jia Y Yu H Fernandes SM Wei Y Gonzalez-Gil A Motari MG Vajn K Stevens WW Peters AT Bochner BS Kern RC Schleimer RP Schnaar RL Expression of ligands for siglec-8 and siglec-9 in human airways and airway cells The Journal of allergy and clinical immunology 2015 135 799 810 e797 25747723 200 Kano G Almanan M Bochner BS Zimmermann N Mechanism of siglec-8-mediated cell death in il-5-activated eosinophils: Role for reactive oxygen species-enhanced mek/erk activation The Journal of allergy and clinical immunology 2013 132 437 445 23684072 201 Nutku E Aizawa H Hudson SA Bochner BS Ligation of siglec-8: A selective mechanism for induction of human eosinophil apoptosis Blood 2003 101 5014 5020 12609831

PMC005xxxxxx/PMC5120545.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0376343 5302 J Thorac Cardiovasc Surg J. Thorac. Cardiovasc. Surg. The Journal of thoracic and cardiovascular surgery 0022-5223 1097-685X 26027913 5120545 10.1016/j.jtcvs.2015.03.041 NIHMS830993 Article Coronary artery bypass grafting in diabetics: A growing health care cost crisis Raza Sajjad MD a Sabik Joseph F. III MD a Ainkaran Ponnuthurai MS b Blackstone Eugene H. MD ab a Department of Thoracic and Cardiovascular Surgery, Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio b Department of Quantitative Health Sciences, Research Institute, Cleveland Clinic, Cleveland, Ohio Address for reprints: Joseph F. Sabik III, MD, Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, 9500 Euclid Ave/Desk J4-1, Cleveland, OH 44195 (sabikj@ccf.org) 19 11 2016 01 4 2015 8 2015 23 11 2016 150 2 3042.e2 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Objectives To determine 4-decade temporal trends in the prevalence of diabetes and cardiovascular risk factors among patients undergoing coronary artery bypass grafting (CABG) and to compare in-hospital outcomes, resource utilization, and long-term survival after CABG in diabetics versus nondiabetics. Methods From January 1972 to January 2011, 10,362 pharmacologically treated diabetics and 45,139 nondiabetics underwent first-time CABG. Median follow-up was 12 years. Direct technical cost data were available from 2003 onward (n = 4679). Propensity matching by diabetes status was used for outcome comparisons. Endpoints were in-hospital adverse events, resource utilization, and long-term survival. Results Diabetics undergoing CABG increased from 7%in the 1970s to 37%in the 2000s. Their outcomes were worse, with more (P < .05) in-hospital deaths (2.0% vs 1.3%), deep sternal wound infections (2.3% vs 1.2%), strokes (2.2% vs 1.4%), renal failure (4.0% vs 1.3%), and prolonged postoperative hospital stay (9.6% vs 6.0%); and their hospital costs were 9% greater (95% confidence interval 7%–11%). Survival after CABG among diabetics versus nondiabetics at 1, 5, 10, and 20 years was also worse: 94% versus 94%, 80% versus 84%, 56% versus 66%, and 20% versus 32%, respectively. Propensity-matched patients incurred similar costs, but the prevalence of postoperative deep sternal wound infections and stroke, as well as long-term survival, remained worse in diabetics. Conclusions Diabetes is both a marker for high-risk, resource-intensive, and expensive care after CABG and an independent risk factor for reduced long-term survival. These issues, coupled with the increasing proportion of patients needing CABG who have diabetes, are a growing challenge in reining in health care costs. Graphical abstract Four-decade trend in prevalence of diabetes among patients undergoing primary isolated coronary artery bypass grafting. Each circle represents a yearly percentage, and the solid line is the locally estimated scatterplot smoother (loess) estimate. Coronary artery bypass grafting diabetes health care costs Diabetes is an emerging worldwide epidemic that affects 382 million people,1 including 25.8 million in the United States alone.2 In 2013, global health expenditures resulting from diabetes were estimated at $548 billion and are expected to exceed $627 billion by 2035.1 In the United States, the total economic burden of diabetes was $245 billion in 2012.3 Because coronary artery disease is common in diabetics,4 it is an important driver of diabetes-related health care cost.3 As the prevalence of diabetes has risen, cardiovascular disease associated with it has increased as well.5 Today, diabetics represent an important subset of patients undergoing coronary artery bypass grafting (CABG), an expensive procedure. Therefore, we sought to determine 4-decade temporal trends in the prevalence of diabetes and cardiovascular risk factors for patients undergoing CABG, compare overall in-hospital adverse outcomes, hospital resource utilization and costs, and long-term survival after CABG in diabetics versus nondiabetics, then compare these same factors in diabetics versus nondiabetics who have similar high-risk profiles using propensity matching. METHODS Patients From January 1, 1972, to January 1, 2011, 57,278 patients underwent first-time isolated CABG at Cleveland Clinic. Data on the presence or absence of pharmacologically treated diabetes mellitus (insulin or oral hypoglycemic agent) were available for 55,501 (97%) of these patients: 45,139 nondiabetics and 10,362 diabetics. Patients were identified, and preoperative, operative, and postoperative variables (Appendix E1) were retrieved from the prospective Cleveland Clinic Cardiovascular Information Registry. This database is populated concurrently with patient care and has been approved for use in research by the institutional review board, with the requirement for patient consent waived. Variables and Definitions A coronary artery system was considered meaningfully stenotic if it contained a ≥50%-diameter obstruction. Incomplete revascularization was defined as failure to graft any coronary system containing ≥50% stenosis, or both left anterior descending and circumflex coronary systems for ≥50% left main trunk stenosis. Left ventricular function was echocardiographically graded as normal (ejection fraction ≥60%), mild dysfunction (ejection fraction 40%–59%), moderate dysfunction (ejection fraction 25%–39%), or severe dysfunction (ejection fraction <25%). Endpoints Endpoints were: (1) in-hospital adverse outcomes defined as in the Society of Thoracic Surgeons national database (www.ctsnet.org/file/rptDataSpecifications252_1_ForVendorsPGS.pdf); (2) in-hospital direct technical costs (the sum of direct preoperative, operative, and postoperative costs), both total and broken down according to resource utilization areas; and (3) time-related mortality. Actual direct technical cost data (not charge data), exclusive of physician professional salaries, were obtained from Decision Support Services of Cleveland Clinic. Data were available for patients from only 2003 onward (n = 4679: 1776 diabetics and 2903 nondiabetics). Costs were corrected to constant 2011 dollars.6 Indirect costs, including institutional costs and capital equipment costs used for all operations, could not be estimated on a per patient basis and are not included, but they were assumed to be distributed uniformly across groups. Vital status after hospital discharge was obtained by routine anniversary follow-up questionnaires supplemented with data from the Social Security Death Master File,7 accessed on October 27, 2011, with a closing date of April 27, 2011. A total of 714,709 patient-years of follow-up data were available for analyses. Median follow-up was 11.8 years, with 25% of survivors followed for >21 years, and 10% for >30 years. For diabetic patients, 86,153 patient-years of follow-up data were available for analyses, with a median follow-up period of 7.5 years; 25% of survivors were followed for >12 years, and 10% for >18 years. For nondiabetic patients, 628,556 patient-years of follow-up data were available for analyses, with a median follow-up period of 13 years; 25% of survivors were followed for >24 years, and 10% for >31 years. Statistical Analysis Temporal trends Four-decade temporal trends in the prevalence of diabetes and cardiovascular risk factors among patients undergoing CABG were assessed using plots of yearly percentages or averages over time. A nonparametric locally estimated scatterplot smoother, PROC LOESS (SAS Institute, Cary, NC), was used to smooth these temporal trends. In-hospital adverse outcomes Comparisons of outcomes after CABG between diabetics and nondiabetics, unmatched and propensity matched, were made using the χ2 test for categoric endpoints. Resource utilization and total direct technical cost Hours in the intensive care unit and postoperative and total hospital lengths of stay had right-skewed distributions, so the Wilcoxon rank-sum test, and the median score test for continuous endpoints, were used for comparisons between diabetics and nondiabetics. To identify the relative difference in direct technical costs, we estimated the ratio of the median cost between the 2 groups. The percentile bootstrap confidence method8 was used to estimate 95%confidence intervals. This procedure was applied to the overall and matched cohorts. Long-term survival Survival was assessed nonparametrically, using the Kaplan-Meier method, and parametrically, using a multiphase hazard model.9 The latter involved resolving the number of hazard phases for instantaneous risk of death (hazard function), and estimating shaping parameters. (For details, see www.lerner.ccf.org/qhs/software/hazard.) Because the shape of time-varying risk of death may differ for diabetics versus nondiabetics, we constructed separate hazard models for each group. Propensity-score matching Although overall assessment of outcomes in diabetics compared with nondiabetics represents the realities of the real world, diabetics as a group present with a higher-than-average risk profile. We therefore treated diabetes as a “natural experiment,”10 comparing outcomes of propensity-matched diabetics and nondiabetics.11–13 This comparison was accomplished in 2 steps. First, a parsimonious multivariable logistic regression was used to identify differences in preoperative characteristics of diabetic versus nondiabetic patients, to obtain insight into these differences (see Appendix E1 for a list of variables analyzed). Bootstrap bagging for variable selection with automated analysis of 500 resampled datasets was used to accomplish this, followed by tabulation of the frequency of both single factors and closely related clusters of factors.14 We retained factors that occurred in ≥50% of the bootstrap models (Table E1). The C-statistic for this parsimonious model was .83. Second, the parsimonious model was augmented into a saturated propensity model by including patient characteristics that were not statistically significantly different between groups. These characteristics were demographic, cardiac, and noncardiac comorbidities not represented (see Appendix E1). The C-statistic for this model was .84. A propensity score representing the probability of diabetes—group membership given the variables included in the propensity model, regardless of whether the patient had diabetes—was then calculated for each patient. A greedy matching strategy based on the propensity scores alone was used to match diabetic with nondiabetic patients, yielding 8926 well-matched pairs (86% of possible matches; Figure E1). Diabetes cases with propensity scores that deviated >0.10 from those of nondiabetes cases were considered unmatched. Standardized differences demonstrated that covariable balance was achieved across nearly all variables (Figure E2). Using a similar approach, separate propensity matching was done between the subset of diabetics and nondiabetics undergoing an operation in the period from 2003 to 2011, during which cost data were available. This yielded 1368 well-matched pairs (77% of possible matches). Missing values Several variables examined in multivariable analyses had missing values. We used fivefold multiple imputation15 with a Markov chain Monte Carlo technique to impute missing values (SAS PROC MI; SAS Institute, Cary, NC). In multivariable modeling, for each imputed complete dataset, we estimated regression coefficients and their variance–covariance matrix. After this step, following Rubin,15 we combined estimates from the 5 models (SAS PROC MIANALYZE; SAS Institute, Cary, NC) to yield final regression coefficient estimates, the variance–covariance matrix, and P values. Presentation Continuous variables are summarized by mean ± SD, or equivalent 15th, 50th (median), and 85th percentiles when the distribution of values is skewed. Analyses were performed using SAS statistical software, version 9.1 (SAS Institute, Cary, NC) and R (version 3.0.2). Uncertainty is expressed by confidence limits (CLs) equivalent to ±1 SE (68%). RESULTS Compared with nondiabetics, diabetic patients undergoing CABG were older and more likely to be overweight, and more likely to be women. In addition, they were more likely to have a history of heart failure, peripheral arterial disease, carotid disease, hypertension, renal failure, stroke, and advanced coronary artery disease (Table 1). Temporal Trends The proportion of patients presenting for CABG who have diabetes increased from 7% per year in the 1970s to 37% in the 2000s (Central Image). The cardiovascular risk factor profile also changed during this time, more so for diabetics than nondiabetics. Today, patients are likely to be older (Figure 1, A) and obese (Figure 1, B); to have had a stroke (Figure 1, C); and to have hypertension (Figure 1, D), peripheral arterial disease (Figure 1, E), lower total cholesterol (Figure 1, F), higher high-density lipoprotein cholesterol (Figure 1, G), lower triglycerides (Figure 1, H), and more-advanced coronary artery disease (Figure 1, I) (see Table E1). Overall Outcomes In-hospital adverse outcomes, and resource utilization and direct technical costs Diabetics had higher in-hospital mortality and greater occurrence of deep sternal wound infection, stroke, atrial fibrillation, renal failure, and respiratory failure (Table 2). Hours spent in the intensive care unit and of length of stay > 14 days were higher in diabetics than nondiabetics (Table 2). As a result, the total cost of CABG was 9% greater (95% CI, 7%–11%) in diabetics. Most of this difference was due to higher costs of clinical and laboratory testing, diagnostic imaging, pharmacy services, and nursing care (Figure 2). Long-term survival The instantaneous risk of death was high immediately after CABG, decreased over the ensuing 6 months, and then gradually increased for both diabetics and nondiabetics (Figure 3, A), resulting in early divergence of their survival curves (Figure 3, B). In addition, late hazard was elevated in diabetics; thus, ever-increasing divergence of survival was observed out to at least 20 years. Among diabetics, overall survival at 6 months, 1, 5, 10, 15, and 20 years after CABG was 95%, 94%, 80%, 54%, 31%, and 18%, respectively. In contrast, for nondiabetics, it was 97%, 97%, 90%, 76%, 59%, and 42%, respectively (P < .0001). Comparison of Diabetics with Similar High-Risk Nondiabetics In-hospital adverse outcomes, and resource utilization and direct technical costs After matching, the incidence of deep sternal wound infection and stroke remained significantly higher among diabetics (Table 2). After propensity matching, no significant difference remained in total cost of CABG between diabetics and nondiabetics (Figure 2). Hours spent in the intensive care unit were similar, but length of stay > 14 days remained higher for diabetics (Table 2). Long-term survival Among propensity-matched patients, instantaneous risk of death (hazard function) was similar for both diabetics and nondiabetics until 1 year after surgery, after which risk of death was greater for diabetics (Figure 3, C). Early survival was similar between the 2 groups, but late survival was worse for diabetics (Figure 3, D). Late survival continued to diverge as long as patients were followed, because of the substantial difference in late hazard for at least 20 years after CABG. Survival of diabetics at 6 months, 1, 5, 10, 15, and 20 years after operation was 96%, 94%, 80%, 56%, 35%, and 20%, respectively, versus 96%, 94%, 84%, 66%, 47%, and 32% for nondiabetics. DISCUSSION Principal Findings This study shows that the proportion of patients presenting for CABG who have diabetes increased each year during the past 4 decades, as did the proportion with cardiovascular risk factors. Thus, compared with diabetics undergoing operation in the 1970s, 1980s, and 1990s, those operated on more recently were more likely to be obese and have more comorbidities and advanced coronary artery disease. For diabetics, CABG was more resource intensive and expensive, and in-hospital adverse events and long-term survival were worse. However, the increase in in-hospital resource utilization was not specific to diabetics, but rather commensurate with that of patients coming to surgery with a similar extent of comorbidities, but without diabetes. Unadjusted in-hospital and early mortality (1-year) were higher in diabetics than in nondiabetics, but similar for propensity-matched patients with a similar comorbidity profile. Long-term survival was worse in diabetics than in either nondiabetic patients or matched, nondiabetic, high-risk patients. Thus, diabetes is both a marker for highrisk, resource-intensive, and expensive care after CABG, and an independent risk factor for reduced long-term survival. Trends In 2010, nearly 40% of those undergoing CABG at our institution were diabetic, paralleling the rising prevalence of diabetes in the general population. However, the increasing use of percutaneous coronary intervention for revascularization, and the choice of CABG as the preferred revascularization strategy for diabetics, could also have been contributing factors.16 Although cardiovascular disease–related morbidity and mortality has clearly been reduced in the United States over the past 50 years, the cardiovascular disease burden attributable to diabetes has increased.5 Current estimates are that 18.8 million people in the United States have been diagnosed with diabetes, and 7 million remain undiagnosed.2 In addition, nearly 79 million people aged ≥20 years have prediabetes,2 a condition that places them at increased risk of developing diabetes and cardiovascular disease.17 We also observed a change in the cardiovascular risk-factor profile over time. Diabetics undergoing CABG in recent years are more likely to be obese and to have more-advanced coronary artery disease than those operated on in the 1970s, 1980s, and 1990s. In addition, they are more likely to have hypertension, one component of metabolic syndrome, together with diabetes, hyperlipidemia, and obesity—all risk factors for coronary artery disease.18 Obesity is now considered a national epidemic and is of particular importance as a well-recognized contributor to diabetes. More than one third of US adults are obese,19 and in the next 20 years, obesity may play a contributing role in an estimated 6 million cases of diabetes.20 On the other hand, diabetics undergoing CABG in recent years had lower total cholesterol, higher high-density lipoprotein cholesterol, and lower triglycerides than did patients operated on in earlier years; this difference may be attributable to better control of lipids in the statin era. In-Hospital Adverse Outcomes In-hospital adverse outcomes after CABG were more common in diabetics than nondiabetics. In part, this difference is attributable to diabetic patients being sicker and having more comorbidities than nondiabetics, because some of the differences, including hospital death, septicemia, renal failure, and respiratory failure, became statistically insignificant after comparison with similar high-risk nondiabetic patients through propensity matching. However, occurrence of deep sternal wound infection and stroke remained significantly higher in diabetics even after matching. Deep sternal wound infection results in prolonged postoperative length of stay and thus increases hospital resource utilization. Strokes may cause permanent disability, resulting in unemployment and thus increasing the indirect cost of diabetes through loss of productivity. Other studies have also revealed worse hospital and long-term outcomes of CABG in diabetics.21,22 The SYNTAX trial showed that at 3 years, diabetes had little effect on outcomes of CABG, and diabetes control (as indicated by baseline hemoglobin A1c levels) was not predictive of major adverse cardiac and cerebrovascular events. In our study, overall postoperative prevalence of stroke and in-hospital death was higher in diabetics, and occurrence of myocardial infarction was higher in nondiabetics. However, after comparison with similar, high-risk, nondiabetic patients, occurrence of death and myocardial infarction was similar in the 2 groups, as was true in the SYNTAX trial, but stroke remained higher in diabetics.23 Health Care Costs Our study additionally shows that resource utilization and the actual direct technical cost of CABG were greater in diabetics, mainly because of longer intensive care unit and postoperative stays, and higher costs of clinical and laboratory testing, diagnostic imaging, pharmacy services, and nursing care. Greater severity of disease among diabetics necessitates preoperative admission and more-extensive laboratory and diagnostic workup. Furthermore, managing postoperative adverse events and the resulting increase in postoperative length of stay both lead to higher in-hospital costs. After using propensity matching to compare similar high-risk, nondiabetic patients, no significant difference between the total cost of CABG in diabetics versus nondiabetics with a similar high-risk profile was observed, demonstrating that most of the increased cost was due to diabetics being sicker and having more comorbidities. In 2012, $176 billion was spent on direct health care for diabetics in the United States, with in-hospital care representing 43% of that amount.3 Because heart disease is one of the leading causes of hospitalization in diabetics, a large share of in-hospital costs can be attributed to CABG and its postoperative complications. Others have also demonstrated the association of diabetes with increased cost of CABG.24,25 A study of 114 diabetics and 198 nondiabetics showed that insulin-treated diabetics have longer hospital stays and higher hospital charges than non–insulin-treated diabetics and nondiabetics.24 A recently published study from China also showed that CABG was more costly, with worse long-term results, in diabetics than in nondiabetics.26 Survival Diabetic patients as a group had a higher early (1-year) risk of death after CABG than nondiabetics, as has been documented by others.21 However, an interesting finding of our study is that among propensity-matched patients, early risk was similar to that of nondiabetic high-risk patients with a similar comorbidity profile. Long-term survival, however, was worse in diabetics than in both nondiabetic patients and nondiabetic high-risk patients. Other studies have also demonstrated diabetes to be an independent risk factor for reduced long-term survival after CABG.21,26 Although long-term survival after CABG is worse in diabetics and high-risk nondiabetics, in general, high-risk patients reap the greatest survival benefit from CABG.27 Moreover, using surgical techniques that are associated with better long-term survival after CABG in diabetics could further enhance this survival benefit.28 Diabetes: An Avoidable Economic Burden Diabetes is a growing threat to the US economy. Diabetic patients’ medical expenses are nearly 2.3 times higher than those of nondiabetics,3 and this economic burden is expected to increase as the number of diabetes cases rises. Fortunately, however, this situation is largely preventable. In people with prediabetes, cost-effective lifestyle interventions have been shown to have a positive effect on preventing development of the disease.29–31 In people with diagnosed diabetes, controlling blood sugar and cardiovascular disease risk factors, such as hypertension and hypercholesterolemia, has been shown to help reduce cardiovascular events.32,33 With appropriate use of these approaches, we can help prevent development of diabetes and its complications, reducing the diabetes-related economic burden. Strengths and Limitations This study includes 4 decades of patients who underwent CABG at a single, high-volume academic medical center. An advantage of a long observation period is having a long follow-up period, but generalizing these late results to a contemporary patient population may not accurately reflect the changing patient case mix and advances in managing these patients over time. For adjusted comparison of outcomes, propensity-score matching was performed. Although the patient pairs were well matched, any patient factors that significantly affect outcomes after CABG but were not included in the propensity analysis might bias our adjusted results. In addition, circumstances of each death, which may differ among diabetic and nondiabetic patients, were not reliably captured during follow-up inquiries. CONCLUSIONS Diabetes is both a marker for high-risk, resource-intensive, and expensive care after CABG and an independent risk factor for reduced long-term survival. These issues, coupled with the annually increasing number of patients needing CABG who are diabetic, present a growing challenge to reining in health care costs, both in the United States and internationally. Diabetic patients, and those with a similar high-risk profile, set to undergo CABG should be made aware that their risks of postoperative complications are higher than average, and measures should be taken to reduce their postoperative complications. Moreover, trying to reverse the diabetes epidemic is important; if left uncontrolled, it will increase the prevalence of cardiovascular disease and add to the ever-increasing economic health care burden. Research and policies focused on reducing diabetes, and programs that raise awareness about preventive strategies, should be developed and implemented to check rising health care costs. Supplementary Material 01 This study was funded in part by the Sheikh Hamdan bin Rashid Al Maktoum Distinguished Chair in Thoracic and Cardiovascular Surgery (held by JFS), and the Kenneth Gee and Paula Shaw, PhD, Chair in Heart Research (held by EHB). These philanthropists played no role in the design of the study, collection of data, analysis and interpretation of the data, or writing of the report, and did not approve or disapprove publication of the article. Abbreviation and Acronym CABG coronary artery bypass grafting FIGURE 1 Four-decade trends in prevalence of diabetes and cardiovascular risk factors among patients undergoing primary isolated coronary artery bypass grafting. Each circle is a yearly percentage or mean value, and solid lines are loess estimates. The factors shown are: (A) age; (B) body mass index; (C) stroke; (D) hypertension; (E) PAD; (F) total cholesterol; (G) HDL cholesterol; (H) triglycerides; and (I) percentage with 3-system disease. PAD, Peripheral arterial disease; HDL, high-density lipoprotein. FIGURE 2 Median (triangles) ratio of total direct technical costs (overall and propensity matched) in diabetics versus nondiabetics. Error bars are 95% confidence intervals. FIGURE 3 Time-related death after primary isolated coronary artery bypass grafting in diabetics and nondiabetics. Solid lines are parametric estimates enclosed within dashed 68% confidence bands equivalent to ±1 SE. The panels show: (A) instantaneous risk of death (overall); (B) survival (overall); (C) instantaneous risk of death (propensity-matched cohort); and (D) survival (propensity-matched cohort). Each symbol represents a death; vertical bars are confidence limits equivalent to ±1 SE; and values in parentheses are numbers of patients remaining at risk. TABLE 1 Patient characteristics and revascularization details of nondiabetics and diabetics undergoing coronary artery bypass grafting Nondiabetics (n = 45,139) Diabetics (n = 10,362) Characteristic n No. (% or mean ± SD n No. (%) or mean ± SD P value Demographics   Age (y) 45,139 60 ± 10 10,362 63 ± 9.7 <.001   Male 45,139 37,626 (83) 10,362 7249 (70) <.001   Body mass index (kg/m2) 27,739 28 ± 4.7 8883 30 ± 5.8 <.001 Symptoms and surgical priorities   NYHA functional class 44,775 10,317 <.001     I 7645 (17) 1800 (17)     II 16,533 (37) 3982 (39)     III 4702 (11) 1468 (14)     IV 15,895 (35) 3067 (30)   Emergency operation 45,137 1242 (2.8) 10,362 248 (2.4) .04 Cardiac comorbidity   Myocardial infarction 45,139 22,916 (51) 10,362 5932 (57) <.001   Left ventricular dysfunction 42,404 8996 <.001     None 37,013 (87) 6300 (70)     Mild 2492 (5.9) 948 (11)     Mild to moderate 514 (1.2) 238 (2.6)     Moderate 1315 (3.1) 688 (7.6)     Moderate to severe 492 (1.2) 384 (4.3)     Severe 578 (1.4) 438 (4.9)   Preoperative AF or flutter 36,357 411 (1.1) 8967 163 (1.8) <.001   Heart failure 45,139 2219 (4.9) 10,362 1750 (17) <.001   Coronary artery disease 35,015 20,687 (59) 7042 3446 (49) <.001   No. of coronary systems diseased (stenosis ≥50%) 43,113 9908 <.001     0 366 (0.85) 69 (0.70)     1 5310 (12) 615 (6.2)     2 13,807 (32) 2332 (24)     3 23,630 (55) 6892 (70)   Left main disease (stenosis ≥50%) 41,909 6704 (16) 9277 1749 (19) <.001 Noncardiac comorbidity   Peripheral arterial disease 45,139 4551 (10) 10,362 1840 (18) <.001   Carotid disease 45,139 3434 (7.6) 10,362 2278 (22) <.001   Stroke 45,139 1515 (3.4) 10,362 880 (8.5) <.001   Hypertension 29,346 17,615 (60) 9077 6966 (77) <.001   COPD 14,360 1079 (7.5) 6639 584 (8.8) <.001   Smoking 44,244 24,133 (55) 10,208 5319 (52) <.001   Creatinine (mg/dL) 13,969 1.2 ± 0.83 6464 1.3 ± 1.2 .07   Blood urea nitrogen (mg/dL) 13,954 18 ± 9.0 6460 22 ± 12 <.001   Renal dialysis 7375 59 (0.80) 4060 98 (2.4) <.001   Total cholesterol (mg/dL) 35,236 235 ± 56 7422 208 ± 60 <.001   HDL cholesterol (mg/dL) 14,734 40 ± 12 5283 39 ± 13 <.001   LDL cholesterol (mg/dL) 9616 128 ± 46 4309 113 ± 45 <.001   Triglycerides (mg/dL) 29,248 185 ± 116 6672 203 ± 173 .002   Bilirubin (mg/dL) 12,683 0.65 ± 0.50 5803 0.59 ± 0.51 <.001   Hematocrit (%) 13,037 40 ± 5.1 5993 38 ± 5.5 <.001 Experience   January 1, 1972 to index operation 45,139 14 ± 9.5 10,362 21 ± 9.6 <.001 Revascularization details   ITA grafts at index operation 45,139 10,362 <.001     None 14,888 (33) 2018 (19)     Single 25,160 (56) 7570 (73)     Bilateral 5091 (11) 774 (7.5)   Complete revascularization* 45,139 40,405 (90) 10,362 9432 (91) <.001   Cardiopulmonary bypass 45,139 43,761 (97) 10,362 9817 (95) <.001 Values in “n” columns are numbers of patients with data available. SD, Standard deviation; NYHA, New York Heart Association; AF, atrial fibrillation; COPD, chronic obstructive pulmonary disease; HDL, high-density lipoprotein; LDL, low-density lipoprotein; ITA, internal thoracic artery. * Incomplete revascularization was defined as failure to graft any system containing 50% stenosis or both the left anterior descending and circumflex coronary artery systems for 50% left main trunk stenosis. TABLE 2 In-hospital outcomes after coronary artery bypass grafting: overall and propensity matched Overall Propensity matched Nondiabetic (total n = 45,139) Diabetic (total n = 10,362) Nondiabetic (total n = 8926) Diabetic (total n = 8926) Outcome n No. (%) n No. (%) P value n No. (%) n No. (%) P value Hospital death 45,139 566 (1.3) 10,362 211 (2.0) <.001 8926 152 (1.7) 8926 174 (1.9) .2 Deep sternal wound infection 45,139 526 (1.2) 10,362 239 (2.3) <.001 8926 116 (1.3) 8926 197 (2.2) <.001 Septicemia 14,298 226 (1.6) 6633 151 (2.3) .004 6393 139 (2.2) 5352 113 (2.1) .8 Stroke 45,139 640 (1.4) 10,362 233 (2.2) <.001 8926 134 (1.5) 8926 200 (2.2) <.001 Perioperative MI 45,139 1023 (2.3) 10,362 135 (1.3) <.001 8926 118 (1.3) 8926 123 (1.4) .8 Bleeding or tamponade 45,139 1791 (4.0) 10,362 348 (3.4) .004 8926 271 (3.0) 8926 314 (3.5) .07 Atrial fibrillation 45,139 5148 (12) 10,362 1979 (19) <.001 8926 1914 (21) 8926 1672 (19) <.001 Renal failure 45,139 569 (1.3) 10,362 418 (4.0) <.001 8926 258 (2.9) 8926 285 (3.2) .2 Renal failure requiring dialysis 14,298 104 (0.73) 6633 86 (1.3) <.001 6393 79 (1.2) 5352 60 (1.1) .6 Prolonged ventilation (>24 h) 3492 320 (9.2) 2100 249 (12) .001 1838 212 (12) 1563 160 (10) .2 Length of stay*   ICU (h) 14,296 24/24/72 631 24/26/95 <.001 6391 24/24/75 5351 24/24/76 <.001   Postoperative (d) 44,014 6.1/8.0/11 10,263 5.9/7.9/12 <.001 8864 5.2/7.0/11 8829 5.9/7.9/11 <.001   Hospital (d) 44,014 7/11/17 10,263 6.3/10/18 <.001 8864 6.3/9.3/16 8829 6.3/10/17 <.001   Prolonged (>14 d) 45,139 2688 (6.0) 10,362 995 (9.6) <.001 8926 679 (7.6) 8926 785 (8.8) .004 Values in “n” columns are numbers of patients with data available. P values are given for the median score test. MI, Myocardial infarction; ICU, intensive care unit. * 15th/50th/85th percentiles. Median score test was used to compare medians and Wilcoxon rank-sum test to compare tails of distributions. Central Message Increasingly, patients needing CABG have diabetes, a marker for high-risk, resource-intensive, expensive care after CABG, and an independent risk factor for reduced long-term survival, creating a growing challenge to health care cost reduction. Perspective The proportion of patients needing CABG who have diabetes has increased to nearly 40% of all patients undergoing CABG at our institution. The procedure is more resource intensive and expensive for diabetics than for nondiabetics, partly because of postoperative complications. However, when surgeons present the risks and benefits of CABG to diabetic patients, they should explain that CABG offers the best chance for long-term survival. Supplemental material is available online. Conflict of Interest Statement Dr Sabik is the North American principal investigator for the Abbott Laboratories–sponsored left main coronary disease randomized trial (EXCEL), is on the Society of Thoracic Surgeons Board of Directors, and is on the scientific advisory board of Medtronic. All other authors have nothing to disclose with regard to commercial support. References 1 International Diabetes Federation IDF diabetes atlas 6th Available at: http://www.idf.org/diabetesatlas 2 Centers for Disease Control and Prevention National diabetes fact sheet: national estimates and general information on diabetes and prediabetes in the United States 2011 Atlanta US Department of Health and Human Services 3 Economic costs of diabetes in the U.S. in 2012 Diabetes Care 2013 36 1033 1046 23468086 4 Beckman JA Creager MA Libby P Diabetes and atherosclerosis: epidemiology, pathophysiology, and management JAMA 2002 287 2570 2581 12020339 5 Fox CS Coady S Sorlie PD D’Agostino RB Sr Pencina MJ Vasan RS Increasing cardiovascular disease burden due to diabetes mellitus: the Framingham Heart Study Circulation 2007 115 1544 1550 17353438 6 US Department of Labor, Bureau of Labor Statistics CPI inflation calculator Available at: http://www.bls.gov/data/inflation_calculator.htm 7 Boyle CA Decoufle P National sources of vital status information: extent of coverage and possible selectivity in reporting Am J Epidemiol 1990 131 160 168 2403466 8 Efron B Tibshirani RJ An Introduction to the Bootstrap 1998 New York Chapman and Hall/CRC 9 Blackstone EH Naftel DC Turner ME Jr The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information J Am Stat Assoc 1986 81 615 624 10 Pattakos G Koch CG Brizzio ME Batizy LH Sabik JF III Blackstone EH Outcome of patients who refuse transfusion after cardiac surgery: a natural experiment with severe blood conservation Arch Intern Med 2012 172 1154 1160 22751620 11 Rosenbaum PR Rubin DB The central role of the propensity score in observational studies for causal effects Biometrika 1983 70 41 55 12 Rubin DB The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials Stat Med 2007 26 20 36 17072897 13 Sauerbrei W Schumacher M A bootstrap resampling procedure for model building: application to the Cox regression model Stat Med 1992 11 2093 2109 1293671 14 Breiman L Bagging predictors Machine Learning 1996 24 123 140 15 Rubin DB Multiple Imputation for Non-Response in Surveys 1987 New York Wiley 16 Farkouh ME Domanski M Sleeper LA Siami FS Dangas G Mack M Strategies for multivessel revascularization in patients with diabetes N Engl J Med 2012 367 2375 2384 23121323 17 Grundy SM Pre-diabetes, metabolic syndrome, and cardiovascular risk J Am Coll Cardiol 2012 59 635 643 22322078 18 Wilson PW D’Agostino RB Parise H Sullivan L Meigs JB Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus Circulation 2005 112 3066 3072 16275870 19 Ogden CL Carroll MD Kit BK Flegal KM Prevalence of obesity in the United States, 2009–2010 NCHS Data Brief No 82 2012 1 Available at: http://www.cdc.gov/nchs/data/databriefs/db82.pdf 20 Wang YC McPherson K Marsh T Gortmaker SL Brown M Health and economic burden of the projected obesity trends in the USA and the UK Lancet 2011 378 815 825 21872750 21 Thourani VH Weintraub WS Stein B Gebhart SS Craver JM Jones EL Influence of diabetes mellitus on early and late outcome after coronary artery bypass grafting Ann Thorac Surg 1999 67 1045 1052 10320249 22 Barzilay JI Kronmal RA Bittner V Eaker E Evans C Foster ED Coronary artery disease and coronary artery bypass grafting in diabetic patients aged > or = 65 years (report from the Coronary Artery Surgery Study [CASS] Registry) Am J Cardiol 1994 74 334 339 8059694 23 Mack MJ Banning AP Serruys PW Morice MC Taeymans Y Van Nooten G Bypass versus drug-eluting stents at three years in SYNTAX patients with diabetes mellitus or metabolic syndrome Ann Thorac Surg 2011 92 2140 2146 21967819 24 Stewart RD Lahey SJ Levitsky S Sanchez C Campos CT Clinical and economic impact of diabetes following coronary artery bypass J Surg Res 1998 76 124 130 9698511 25 Smith LR Milano CA Molter BS Elbeery JR Sabiston DC Jr Smith PK Preoperative determinants of postoperative costs associated with coronary artery bypass graft surgery Circulation 1994 90 II124 II128 7955238 26 Zhang H Yuan X Osnabrugge RL Meng D Gao H Zhang S Influence of diabetes mellitus on long-term clinical and economic outcomes after coronary artery bypass grafting Ann Thorac Surg 2014 97 2073 2079 24751154 27 Hlatky MA Boothroyd DB Baker L Kazi DS Solomon MD Chang TI Comparative effectiveness of multivessel coronary bypass surgery and multivessel percutaneous coronary intervention: a cohort study Ann Intern Med 2013 158 727 734 23609014 28 Raza S Sabik JF Masabni K Ainkaran P Lytle BW Blackstone EH Surgical revascularization techniques that minimize surgical risk and maximize late survival after coronary artery bypass grafting in diabetics J Thorac Cardiovasc Surg 2014 148 1257.e9 1266.e9 25260269 29 Knowler WC Barrett-Connor E Fowler SE Hamman RF Lachin JM Walker EA Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin N Engl J Med 2002 346 393 403 11832527 30 Tuomilehto J Lindstrom J Eriksson JG Valle TT Hamalainen H Ilanne-Parikka P Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance N Engl J Med 2001 344 1343 1350 11333990 31 Herman WH Hoerger TJ Brandle M Hicks K Sorensen S Zhang P The cost-effectiveness of lifestyle modification or metformin in preventing type 2 diabetes in adults with impaired glucose tolerance Ann Intern Med 2005 142 323 332 15738451 32 Huang ES Meigs JB Singer DE The effect of interventions to prevent cardiovascular disease in patients with type 2 diabetes mellitus Am J Med 2001 111 633 642 11755507 33 Mannucci E Dicembrini I Lauria A Pozzilli P Is glucose control important for prevention of cardiovascular disease in diabetes? Diabetes Care 2013 36 Suppl 2 S259 S263 23882055

PMC005xxxxxx/PMC5120553.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0375363 4713 J Endocrinol J. Endocrinol. The Journal of endocrinology 0022-0795 1479-6805 27799461 5120553 10.1530/JOE-16-0447 NIHMS828225 Article Hepatocyte-specific, PPARγ-regulated mechanisms to promote steatosis in adult mice Greenstein Abigail Wolf 123* Majumdar Neena 12* Yang Peng 12 Subbaiah Papasani V. 12 Kineman Rhonda D. 12 Cordoba-Chacon Jose 12# 1 Research and Development Division, Jesse Brown Veterans Affairs Medical Center, University of Illinois at Chicago, Chicago, IL 2 Department of Medicine, Section of Endocrinology, Diabetes, and Metabolism, University of Illinois at Chicago, Chicago, IL 3 Biologic Resources Laboratory, University of Illinois at Chicago, Chicago, Illinois, USA # Corresponding Author: Jose Cordoba-Chacon, PhD. Dept Medicine, Section Endocrinology, Diabetes and Metabolism, 1819 W Polk St, M/C 640, Chicago, IL 60612, USA. jcordoba@uic.edu Ph: 1-312-569-7417 Fax: 1-312-569-8114 * equally contributed authors 9 11 2016 31 10 2016 1 2017 01 1 2018 232 1 107121 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Peroxisome proliferator-activated receptor γ (PPARγ) is the target for thiazolidinones (TZDs), drugs that improve insulin sensitivity and fatty liver in humans and rodent models, related to a reduction in hepatic de novo lipogenesis (DNL). The systemic effects of TZDs are in contrast to reports suggesting hepatocyte-specific activation of PPARγ promotes DNL, triacylglycerol (TAG) uptake and fatty acid (FA) esterification. Since these hepatocyte-specific effects of PPARγ could counterbalance the positive therapeutic actions of systemic delivery of TZDs, the current study used a mouse model of adult-onset, liver (hepatocyte)-specific PPARγ knockdown (aLivPPARγkd). This model has advantages over existing congenital knockout models, by avoiding compensatory changes related to embryonic knockdown, thus better modeling the impact of altering PPARγ on adult physiology, where metabolic diseases most frequently develop. The impact of aLivPPARγkd on hepatic gene expression and endpoints in lipid metabolism was examined after 1 or 18wks (Chow-fed) or after 14wks of low- or high-fat [HF] diet. aLivPPARγkd reduced hepatic TAG content but did not impact endpoints in DNL or TAG uptake. However, aLivPPARγkd reduced the expression of the FA translocase (Cd36), in 18wk-Chow and HF-fed mice, associated with increased NEFA after HF-feeding. Also, aLivPPARγkd dramatically reduced Mogat1 expression, that was reflected by an increase in hepatic monoacylglycerol (MAG) levels, indicative of reduced MOGAT activity. These results, coupled with previous reports, suggest that Cd36-mediated FA uptake and MAG pathway-mediated FA esterification are major targets of hepatocyte PPARγ, where loss of this control explains in part the protection against steatosis observed after aLivPPARγkd. adult-onset hepatocyte-specific knockdown Cd36 Mogat1 LC/MS diet-induced steatosis INTRODUCTION Non-alcoholic fatty liver disease (NAFLD) is defined as excessive accumulation of fat (steatosis) within hepatocytes that is independent of alcohol intake. NAFLD increases the risk of diabetes, non-alcoholic steatohepatitis (NASH), cirrhosis and hepatocellular carcinoma (Lade, et al. 2014; Michelotti, et al. 2013). Importantly, NAFLD is now recognized as the leading cause of chronic liver disease in the US (Younossi, et al. 2011) and the third most common reason for liver transplants (Zezos and Renner 2014). Given the negative association between NAFLD and human health, a concerted effort is being made to understand the cellular and molecular mechanisms that control hepatocyte triacylglycerol (TAG) content. It is clear that hepatocyte TAG content is dictated by the balance between fatty acid (FA) synthesis, uptake, esterification and oxidation, as well as TAG release via very-low density lipoprotein (VLDL) (Browning and Horton 2004). A better understanding of the mechanisms controlling these processes could help to identify individuals with higher risk of NAFLD, as well as identify novel drug targets to design therapeutic strategies that prevent or reverse NAFLD progression. Thiazolidinones (TZDs) are synthetic agonists of the nuclear receptor, peroxisome proliferator-activated receptor γ (PPARγ) used to treat diabetes type 2 (Ahmadian, et al. 2013). Treating patients with non-alcoholic steatohepatitis (an advance stage of NAFLD featured by elevated markers of liver injury and fibrosis) with TZDs reduces steatosis with variable effects on fibrosis (Belfort, et al. 2006; Ratziu, et al. 2008; Sanyal, et al. 2010). The reduction in steatosis is associated with a reduction in hepatic de novo lipogenesis [DNL, (Beysen, et al. 2008)]. Similar to clinical studies, TZDs have also been shown to reduce hepatic fat content in rodent models (Gupte, et al. 2010; Nan, et al. 2009). However, in striking contrast to the global impact of TZDs: 1) hepatocyte-specific PPARγ expression is positively associated with fatty liver in humans (Pettinelli and Videla 2011) and mouse models (Gavrilova, et al. 2003; Inoue, et al. 2005; Matsusue, et al. 2003; Rahimian, et al. 2001); 2) adenoviral overexpression of a PPARγ transgene in the liver of HF-fed PPARα knockout or WT mice dramatically increases hepatic fat content (Bai, et al. 2011; Yu, et al. 2003); 3) congenital hepatocyte-specific knockout of PPARγ reduces hepatic fat content in mice fed a high-fat (HF) diet (Moran-Salvador, et al. 2011), as well as in mice with fatty liver due to inactivating mutations in the leptin gene [ob/ob; (Matsusue et al. 2003)] or lipodystrophy induced by lack of adipocyte development [AZIP (Gavrilova et al. 2003)]. This disconnect between the impact of systemic TZD (PPARγ agonist) delivery compared to the impact of hepatic-specific alterations in PPARγ function can be attributed to the integrated effects of PPARγ on multiple target tissues. Specifically, TZDs increase systemic insulin sensitivity, which in turn reduces insulin demands. These changes are associated with an increase in adiponectin production by the adipocyte. Adiponectin in turn promotes hepatic fatty acid oxidation via phosphorylation of AMPK and ACC, (Xu, et al. 2003; Yamauchi, et al. 2002), which suppresses DNL (Xu et al. 2003). Therefore, TZD’s effects on lowering steatosis are likely due to extra-hepatocyte mechanisms (Furnsinn and Waldhausl 2002). Although the global therapeutic effects of TZDs have been largely positive, the direct actions of PPARγ on the hepatocyte could serve to counterbalance these effects. Therefore, in order to optimize the development and use of TZDs, it is important to understand the tissue (cell)-specific impact of PPARγ on lipid homeostasis. With respect to the hepatocyte, studies using hepatocyte-specific PPARγ knockout models have led to the conclusion that PPARγ directly promotes hepatic fat accumulation by increasing lipid uptake, as well as promoting DNL (Gavrilova et al. 2003; Matsusue, et al. 2014; Matsusue et al. 2003; Schadinger, et al. 2005; Zhang, et al. 2006). However, it remains unclear if the shifts in hepatic gene expression that support these conclusions are due directly to loss of hepatocyte PPARγ or to compensation by other hepatic genes during development and/or secondary to changes in the systemic metabolic milieu. Therefore, in the current study we have employed a mouse model of adult-onset, hepatocyte-specific knockdown of PPARγ (aLivPPARγkd), that is generated by treating adult (10wk) PPARγfl/fl mice with an adeno-associated virus serotype 8 (AAV8) bearing a liver-specific thyroxine-binding globulin (TBG)-promoter driving a Cre recombinase transgene (AAV8-TBGp-Cre) vector. This model allows us to study the immediate impact of hepatocyte-specific loss of PPARγ in the adult liver and how this deficit impacts liver function overtime under different dietary conditions. As indicated in this study and supported by accumulating evidence (Ashpole, et al. 2016; Shin, et al. 2016; Yanger, et al. 2014), injection of AAV8-TBGp-Cre is an efficient and reproducible method to knockdown a floxed allele only in hepatocytes, independent of age. Analysis of changes in hepatic gene expression and circulating metabolites, combined with assessment of hepatic FA composition (gas chromatography/mass spectrometry (GC/MS)] and relative levels of TAG, diacylglycerols (DAG) and monoacylglycerols (MAG) [liquid chromatography/mass spectrometry (LC/MS)], indicate that adult-onset loss of hepatocyte PPARγ has little direct impact on DNL, FA oxidation, lipid uptake and TAG export. However, evidence indicates that hepatic PPARγ plays a primary role in regulating FA uptake, likely through regulating the expression of Cd36, and the MAG pathway, through regulation the expression of Mogat1 which esterifies FA to MAG to form DAG. Impairment of these pathways following the loss of hepatocyte PPARγ may explain, in part, the protection against age and diet-induced steatosis. MATERIALS AND METHODS Animals All mouse studies were approved by the IACUC of the Jesse Brown VA Medical Center and performed in accordance with the Guide for the Care and Use of Laboratory Animals. PPARγfl/fl (He, et al. 2003) mice were purchased from Jackson Laboratories (Strain 004584, B3.129-Ppargtm2Rev/J, Bar Harbor, ME) and bred as homozygotes. Animals were housed in a temperature (22–24C) and humidity controlled specific-pathogen free barrier facility with 12h light/12h dark cycle (lights on at 0600h). Mice were fed a standard laboratory rodent chow diet (Formulab Diet 5008, Purina Mills, Richmond, IN), unless otherwise indicated. Ten to twelve week old male PPARγfl/fl littermate mice were randomized and injected in the lateral-tail vein with 100μl saline containing 1.5×1011 genome copies of an AAV8 bearing either a TBG-driven Cre recombinase (AAV8-TBGp-Cre, Penn Vector Core, University of Pennsylvania), which generates adult onset hepatocyte (liver)-specific PPARγ knockdown mice (aLivPPARγkd) or AAV8-TBGp-Null which generates controls. Chow-fed mice were killed in the post-absorptive state (4h after food removal at 0700h), at 1 or 18 weeks after PPARγ knockdown. A separate group of PPARγfl/fl mice were fed a low-fat (LF) diet with 10% kCal fat (D12450B, Research Diets, Inc. New Brunswick, NJ) from weaning onwards. At 10–12 weeks of age, mice were injected with either AAV8-TBGp-Cre or AAV8-TBGp-Null, and half the mice in each group switched to a nutrient matched 60% high-fat (HF) diet (D12492, Research Diets, Inc), while the remaining mice continued to receive a LF-diet. Animals were maintained on their respective diets for 14 weeks and killed in the post-absorptive state. Mice were euthanatized by decapitation and trunk blood was collected to determine blood glucose (Alphatrack2, Abbot, Abbot Park, IL), plasma insulin (Mercodia, Uppsala, Sweden), TAG, NEFA, cholesterol and 3-β-hydroxy-butyrate (Wako Diagnostics, Richmond VA) levels following the manufacturer’s instructions. Liver and fat sub-depots were weighed. Livers were snap-frozen in liquid nitrogen and stored at -80C. In a subset of mice killed 1 week after AAV8-TBGp-Null or AAV8-TBGp-Cre injection, multiple tissues were collected to assess AAV8-TBGp driven Cre expression. A group of 10–12 week-old C57Bl6/J mice were injected with 1.5×1011 genome copies of an AAV8-TBGp-EGFP (Cat #AV-8-PV0146, Penn Vector Core, GFP as a reporter gene) and killed 1 week after to collect multiple tissues to assess GFP expression. Also, a piece of liver was fixed in 10% formalin to assess hepatocyte-specific expression of GFP. Assessment of Hepatic Lipids To assess hepatic TAG content, neutral hepatic lipids were extracted in isopropanol and TAG measured as previously published (Cordoba-Chacon, et al. 2014a). To assess hepatic FA composition in mice fed LF and HF diets, total lipids were extracted using the Bligh & Dyer Method (Bligh and Dyer 1959). An aliquot of extracted lipids was transmetylated to quantify specific methyl esters of FA using GC/MS, as we previously reported (Kineman, et al. 2016). LC/MS was used to assess the relative content of hepatic TAG, DAG and MAG. Briefly, hepatic homogenates were spiked with standards (50μg trinonadecadienoin glyceride [TAG-(19:2/19:2/19:2)], 50μg dipentadecanoin glyceride [DAG-(15:0/15:0)] and 50μg monononadecanoin glyceride [MAG-(19:0)]; Nu-Chek, Waterville, MN) and extracted using the Bligh & Dyer Method (Bligh and Dyer 1959). Aliquots were dissolved in 80:19.5:0.5 parts of methanol/chloroform/water to dilute the standards to a concentration of 0.25μg/ml. Samples (10μl) were injected using an Agilent 2600 UPLC into the ABSciex 6500 Qtrap mass spectrometer (Agilent Technologies) without chromatography separation. The flow rate of mobile phase (methanol/chloroform/water 80:19:0.5 v/v) containing 0.1% of NH4COOH was set to 200μl/min. Electrospray ionization-mass spectrometry (ESI-MS) was performed in positive multiple reaction monitoring (MRM) mode for the quantitative and qualitative analysis. The spray voltage was 4.5 kV, the source temperature was set at 450 °C. Mass spectra were acquired and recorded by Analyst software (AB Sciex, version 6.1). The major neutral lipids species known to be present in the liver tissues were analyzed in MRM mode, with the transition from its ammoniated ion (Q1) to the product ion derived from the loss of its ammoniated fatty acid (Q3) (Yang and Subbaiah 2015). Quantification of individual molecular species was performed from the relative intensities of the various species and the corresponding internal standards respectively. Individual MRM of the internal standards was: MAG-(19:0) Q1 390.4, Q3 75.1; DAG-(15:0/15:0) Q1 558.5, Q3 299.3; TAG-(19:2/19:2/19:2) Q1 938.8, Q3 625.5. Gene expression analysis Hepatic and adipose tissue RNA was extracted using Trizol Reagent (Life Technologies, Carlsbad, CA) and treated with RQ1 RNase-free DNase (Promega, Madison, WI). DNA-free RNA was transcribed and qPCR performed as previously published (Cordoba-Chacon et al. 2014a; Kineman et al. 2016). qPCR primer sequences for PPARγ, carnitine palmitoyltransferase 1α (Cpt1α), adipose triglyceride lipase (Atgl), hormone-sensitive lipase (Hsl), sterol response element binding protein 1c (Srebp1c), acetyl-CoA carboxilase 1 (Acc1), fatty acid synthase (Fasn), fatty acid elongase (Elovl6), stearoyl –CoA desaturase 1 (Scd1), fatty acid translocase (Cd36), glycerol phosphate acyltransferase (Gpat1), monoacylglycerol acyltransferase 1 (Mogat1), Cre recombinase, GFP, Cyclophilin A, β-actin and Hypoxanthine-guanine phosphoribosyltransferase were previously published (Cordoba-Chacon et al. 2014a; Cordoba-Chacon, et al. 2015a; Kineman et al. 2016). Primer sequences for: PPARα (NM_011144): Se: GGGAAAGACCAGCAACAACC, As: GCAGTGGAAGAATCGGACCT; acyl-CoA synthetase long-chain family member 1 (Acsl1, NM_007981): Se: GAGGGTGAGGTGTGTGTGAAA, As: CCGTGTGTAACCAGCCATCT; hepatic nuclear factor 4 α (Hnf4α, NM_008261.3): Se: ACATTCGGGCAAAGAAGATTG, As: ACCTGGTCATCCAGAAGGAGTT; PPARγ co-activator 1 α; (Pgc1α, NM_008904.2): Se: TTCCCGATCACCATATTCCA, As: TTCATCCCTCTTGAGCCTTTC; Cyp4a10 (NM_010011.3): Se: CCACTGATTCTGTTGTGGAGC, As: CATTAGAAGAGAGGGGATGAGGA; monoacylglycerol lipase (Mgll, NM_001166251.1): Se: GTGGAATGCAAAAGCCAAGA, As: AGCTCATCATAACGGCCACA; apolipoprotein B (ApoB, NM_009693.2): Se: AGCCCAGCACTGACTGACTT, As: GAAGCCTTGGGCACATTG; microsomal triglyceride transfer protein (Mttp, NM_001163457.1): Se: CAGTGGATGCCTCTTTTGTG, As: GTCTCGAATTGCCTGAGTGG; hepatic lipase (Hl, NM_008280.2): Se: TTTTCCTGGTGTTCTGCATCT, As: CAGGCGATCGTTTTCATCTT; low density lipoprotein lipase receptor (Ldlr, NM_001252658.1): Se: TGTCACCTGTCAGTCCAATCAA, As: TCAGAGCCATCTAGGCAATCTC; very-low density lipoprotein receptor (Vldlr, NM_013703.2): Se: GTGCAAGGCAGTAGGCAAAG, As: GCTGAGATCAGCCCAAAACA; lipoprotein related protein 1 (Lrp1, NM_008512.2): Se: ACACACGCCAACTGTACCAA, As: TGACATTCGGGTCACAAACA; monoacylglycerol acyltransferase 2 (Mogat2, NM_146035.2): Se: TGTGAAAACTTGGAAATCGACA, As: CAGTCTCCAGCATGAAAAATCC; diacylglycerol acyltransferase 1 (Dgat1, NM_010046.2): Se: AGCTGTGGCCTTACTGGTTG, As: AGCAGCCCCACTGACCTT. Western-blot Livers were homogenized in extraction buffer pH 7.5, 50mM HEPES, 2mM EGTA, 2mM EDTA, 130mM NaCl, 10mM NaF, 20mM β-glycerophosphate, 2mM sodium pyrophosphate, 1mM sodium vanadate, 0.5mM PMSF, 0,1% nonidet P-40, 2mM benzamidine, 1mM TLCK, 10μg/ml leupeptin, 10% glycerol, with protease inhibitors (Complete, Roche, Indianapolis, IN), followed by sonication for 10 sec. Protein concentration was determined using Bradford reagent (Bio-Rad Laboratories, Hercules, CA). 100μg of denatured proteins were separated by SDS-PAGE (Mini-PROTEAN TGX and Criterion TGX Gels 10%, Bio-Rad Laboratories) and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat, dry milk in Tris-buffered saline with 0.05% Tween-20 for 1h at 25C, then incubated with primary antibodies overnight at 4C [Rabbit anti-human PPARγ mAb (C26H12), 1/500; Rabbit anti-human Histone H3 mAb (D1H2), 1/1000 (Cell Signaling Technology)], washed and incubated with secondary antibodies for 2h at 25C [Goat Anti-Rabbit IgG (H+L)-HRP Conjugate, 1/2000 (Bio-Rad Laboratories)]. After washing, SuperSignal® WestFemto Maximum Sensitivity Substrate (Thermo Scientific, Rockford, IL) was added and the light signal detected and analyzed using a C-Digit Blot Scanner and Image Studio Lite Ver 3.1 (Li-Cor Biosciences, Lincoln, NE). Immunohistochemistry Livers were fixed in formalin and paraffin embedded. 5μm sections were deparaffinized, hydrated in graded-ethanol/water solutions, and then treated with 10mM citrate buffer at 125C for 5′ (GFP, desmin and F4/80 staining) followed by 0.05% porcine trypsin in TRIS buffered solution (TBS) for 10min at 37C (F4/80 staining only). Samples were blocked in TBS containing 1.5% goat-serum and 0.01% Tween-20 for 20′, and sections incubated (4C overnight in TBS 0.01% Tween-20) with mouse anti-GFP, 1:100 Cell Signaling (Danvers, MA) #2955; rabbit anti-desmin, 1:50 Cell Signaling #5332; rat anti-F4/80 1:50 eBiosciences (San Diego, CA) #14-4801-82. Secondary antibodies (1:500 in blocking buffer): goat-anti rabbit IgG Alexa Fluor®594 Cell Signaling #8889 (for desmin staining), goat-anti mouse IgG Alexa Fluor®594 Cell Signaling #8890 and goat-anti mouse IgG Fluor®488 Cell Signaling #4408 (for GFP stainings), goat-anti rat IgG Fluor®488 Cell Signaling #4416 (for F4/80 staining). Sections were mounted with Fluoroshield with DAPI (Sigma-Aldrich) and immunofluorescence detected using a Olympus BX43 microscope (Olympus, Center Valley, PA). Images were recorder using Olympus CellSens software (Olympus) and processed using ImageJ (NIH) and CellSens (Olympus). Statistics t-student’s tests were performed to analyze the effect of AAV8-TBGp-Cre on PPARγ expression in liver, epididymal fat (eWAT) and inguinal fat (iWAT) (Fig 1G). 2-way ANOVA, followed by Bonferroni’s post-hoc analysis, was used to determine the effect of aLivPPARγkd with age or the effect of aLivPPARkd with diet. p-values less than 0.05 were considered significant. All statistical analyses were performed using GraphPad Prism 5 (GraphPad Software, La Jolla, CA). RESULTS Validation of hepatocyte-specific expression of TBGp-driven transgene after AAV8 injection Hepatocyte specificity expression of transgenes delivered by AAV8 and driven by TBG was confirmed using an AAV8-TGBp-EGFP reporter (Fig. 1A–E), that led to GFP positive staining only in hepatocytes and not in other cell types, including vascular endothelial cells, connective tissue surrounding vessels, cholangiocytes of bile ducts (Fig 1A,B), hepatic stellate cells (HSC; stained positive [red] for desmin [Fig. 1C]) and macrophages, (stained positive [red] F4/80 [Fig. 1D]). AAV8-TBGp-driven GFP and Cre expression was detected in hepatic but not in extrahepatic tissues (Fig 1E, F). Expression of PPARγ was reduced in liver but not in adipose tissue (Fig 1G). Interestingly, a modest increase of PPARγ expression was detected in eWAT of aLivPPARγkd mice. Effects of hepatocyte-specific PPARγ knockdown on metabolic endpoints in adult mice (aLivPPARγkd) After 1 and 18 weeks of chow diet AAV8-TBGp-Cre injection resulted in a clear knockdown of hepatocyte PPARγ mRNA and protein at 1 week after injection that persisted after 18 weeks of AAV8-injection (Fig 2A–C). Mouse PPARγ gene produces two isoforms PPARγ1 and PPARγ2 (NM_001127330.2 and NM_0111463). Three mRNA variants are described for PPARγ1 (variant 1: NM_001127330.2; variant 3: NM_001308352.1; variant 4: NM_001308354.1), but all encode for the same protein (isoform 1). In PPARγfl/fl mice two exons, previously described as exons 1 and 2 (He et al. 2003), are flanked by loxP sites. These exons expand 227 and 169 bp respectively, are located within the CDS of each variant and the qPCR primers employed in this study are placed in these two exons. Therefore, the qPCR result of PPARγ gene expression accounts for all variants and thereby the western-blot results account for both isoforms. Of note, hepatic PPARγ mRNA levels modestly, but significantly, increased with age in control but not aLivPPARγkd mice (Fig 2B,C). To assess if hepatic lipid metabolism was altered in chow-fed aLivPPARγkd mice we measured whole body, liver and adipose tissue (unilateral epididymal, inguinal and retroperitoneal fat depots) weights, hepatic TAG levels, as well as circulating blood glucose, plasma insulin, ketones, TAG, NEFA and cholesterol (Table 1). For the majority of endpoints examined aLivPPARγkd had no effect in either diet group. However, there was an overall inhibitory effect of aLivPPARγkd on hepatic TAG levels (p=0.0185), indicating hepatic PPARγ plays a role in hepatic lipid accumulation, even under standard feeding conditions. After 14 wk of LF- or HF-diet To further explore if hepatic lipid accumulation induced by excess dietary fat intake is altered by aLivPPARγkd, we fed a HF- or a nutrient-matched LF control diet to aLivPPARγkd mice and their littermate controls. After 14 weeks of either LF- or HF-feeding (Fig 2D) hepatic PPARγ mRNA/protein expression remained suppressed in aLivPPARγkd mice, compared to controls (Fig 2E,F). Consistent with previous reports (Inoue et al. 2005; Moran-Salvador et al. 2011; Yamazaki, et al. 2011), HF feeding in control mice increased hepatic PPARγ expression, compared to those fed a LF-diet (Fig 2E,F). Also, in control mice, HF-feeding increased body, liver and fat mass weight and elevated hepatic TAG content and plasma glucose, insulin and cholesterol levels, but plasma TAG and NEFA were reduced (Table 1). Although TAG and NEFA are reported to be elevated in HF-fed mice after an overnight fast, in the post-absorptive state (4h fasted mice at 1100h), TAG and NEFA levels are reported to be reduced (Cordoba-Chacon, et al. 2014b; Cordoba-Chacon, et al. 2015b; Guo, et al. 2009; Horakova, et al. 2016; Obrowsky, et al. 2013), likely due to elevated insulin levels under these conditions. aLivPPARγkd, in LF-fed mice, did not impact any of the metabolic endpoints examined (Table 1), including food intake (data not shown). However, HF-fed aLivPPARγkd mice showed reduced relative liver weight and this was associated with a reduced hepatic TAG content (Table 1). Despite reduced hepatic TAG levels, there were no differences in post-absorptive blood glucose, plasma insulin and TAG levels in HF-fed aLivPPARγkd mice, compared to diet-matched controls (Table 1). However, normal glucose and insulin levels in the post-absorptive mouse may not be indicative of the whole body glucose homeostasis which requires dynamic evaluation by glucose and insulin tolerance tests. Of note, plasma ketones and cholesterol were reduced in HF-fed aLivPPARγkd mice, while plasma NEFA were increased as compared to HF-fed control littermates (Table 1). Association between aLivPPARγkd-mediated alterations in hepatic gene expression and metabolic endpoints It has been previously reported that congenital liver-specific knockout of PPARγ alters the expression of a number of genes related to the regulation of hepatic TAG levels (Gavrilova et al. 2003; Matsusue et al. 2003; Moran-Salvador et al. 2011). However, from these studies it is difficult to determine which changes may be due to the direct actions of hepatocyte PPARγ and which may be altered by secondary changes that occur overtime. Therefore, qPCR was used to screen changes in expression of key genes related to fatty acid oxidation, intrahepatic TAG hydrolysis, hepatic TAG export, DNL, hepatic lipid uptake and TAG synthesis, in liver samples from chow-fed (1 and 18 wks post-aLivPPARγkd, Fig. 3A) and LF- and HF-fed (14 wks post-aLivPPARγkd, Fig. 3B) mice. The data shown in Fig. 3 is expressed as fold-change in hepatic gene expression in aLivPPARγkd mice, compared to PPARγ-intact controls (set at 0), within age and diet, while Supplemental Table 1A,B provides absolute values. There were small, but significant, changes in the expression of a number of genes in aLivPPARγkd mice, compared to PPARγ-intact controls within specific groups. However, collective examination of this data set provides insight into which genes (pathways) are major targets of PPARγ, that could explain why aLivPPARγkd protects against hepatic fat accumulation. De novo lipogenesis Although experimental evidence suggests that hepatocyte PPARγ promotes liver fat accumulation by regulating the expression of genes important for DNL (Matsusue et al. 2014; Matsusue et al. 2003), aLivPPARγkd did not significantly reduce the expression of DNL genes (Srebp1c, Acc1, Fasn, Elov6, Scd1) across age or diet. However, we could not exclude the fact aLivPPARγkd could indirectly mediate the activity of DNL enzymes, independent of changes in gene expression. Therefore, in order to estimate the impact of aLivPPARγkd on hepatic DNL we used GC/MS to measure FA composition in livers of mice fed either a LF or HF diet. Supplemental Table 2 provides data for all FA detected, while Fig. 4A shows absolute levels of palmitic acid (16:0), palmitoleic acid (16:1(n-7)) and linoleic acid (18:2(n-6)). The absolute levels of these FAs were significantly reduced in HF-fed aLivPPARγkd mice as compared to HF-fed controls, consistent with the reduction in hepatic TAG content. Specific ratios of these FA (Fig 4B) have been shown to be indicative of DNL (SCD-1 index 16:1/16:0 (Kineman et al. 2016; Lee, et al. 2015; Silbernagel, et al. 2012); DNL index, 16:0/18:2 (Kineman et al. 2016; Lee et al. 2015; Sevastianova, et al. 2012; Silbernagel et al. 2012). As previously reported, HF-diet reduces the rate of DNL (Duarte, et al. 2014) and expression of DNL proteins (Benard, et al. 2016), as reflected by significant reduction in both the SCD-1 and DNL index, as well as the reduction in DNL gene expression (Acc1, Fasn, Elovl6, Scd1; Supplemental table 1B). Independent of diet, loss of hepatocyte PPARγ did not impact these indices (Fig 3B). Taken together, these results suggest hepatic PPARγ plays a minimal role in directly regulating hepatic DNL. Hepatic TAG export, FA oxidation and lipid uptake A reduction in hepatic fat content observed in aLivPPARγkd mice could be due to an increased rate of hepatic VLDL production or a decreased rate of TAG lipoprotein clearance. However, examination of genes critical for these processes were not consistently altered by aLivPPARγkd (Fig. 3) and circulating TAG in aLivPPARγkd did not differ from PPARγ-intact controls, within age and diet group (Table 1). It could also be possible that the reduced TAG content observed in livers of aLivPPARγkd mice is due to an increase in intrahepatic TAG hydrolysis and FA oxidation. However, the fact that the expression of Atgl, Hsl, Mgll, PPARα, Acsl1, Cpt1, Hnf4α, Pgc1α and Cyp4a10 were not increased across groups (Fig 3), while HF-aLivPPARγkd mice exhibited a decrease in plasma ketones (Table 1), suggests that hepatic FA oxidation is actually reduced. It is possible that this reduction in FA oxidation/ketogenesis is secondary to a reduction intrahepatic FA availability due to a reduction in FA uptake, since aLivPPARγkd reduced the expression of hepatic Cd36 (FA translocase), where this was associated with an increase in circulating NEFA in HF-fed mice (Table 1). In fact, Cd36 has been previously shown to be a PPARγ target gene (Tontonoz, et al. 1998). However, it should be noted that a significant reduction in Cd36 was not observed in aLivPPARγkd after 1 wk of knockdown, or in LF-fed mice, suggesting that under these conditions the low level of PPARγ is not sufficient to maintain Cd36 expression or that the full expression requires additional factors that are activated with age or diet, such as LXR and PXR (Zhou, et al. 2008). TAG synthesis TAG, DAG and MAG are generated by esterification of newly synthetized FA or extrahepatic FA into acylglycerol backbones, as illustrated in Fig 5A. In the liver, it is thought that the primary source of TAG is via the glycerol-3-phosphate (G3P) pathway, that generates DAG (Coleman and Mashek 2011; Mashek 2013) which serves as a substrate of DAG transferases 1 and 2 (Dgat1/2). However, it is becoming evident that the MAG pathway [ie. generation of DAG from MAG via MAG transferases 1 and 2 (Mogat1/2 in mice)], may be a relevant pathway in NAFLD (Hall, et al. 2012; Mashek 2013). Examination of the expression of key genes involved in TAG synthesis revealed there was an overall inhibitory effect of aLivPPARγkd on the expression of Mogat1 (Supplemental Table 1) that reached significance with age and HF feeding (Fig. 3B), without a reduction in Gpat, Mogat2, Dgat1 or Dgat2 expression (Fig 3B). Of note, Mogat1 has been reported to be a direct target of PPARγ (Lee, et al. 2012; Yu, et al. 2015) and in fact, expression of Mogat1 increases with age and HF-feeding, p-value: 0.0063 and <0.0001, respectively, (Supplemental Table 1A,B), mirroring the relative changes observed in hepatic PPARγ expression (Fig. 2). To indirectly test if the reduction in Mogat1 gene expression translated into a reduction in MAG pathway activity, we used LC/MS to measure the relative levels of MAG, DAG and TAG in LF- and HF-fed mice with or without hepatic PPARγ. Overall, aLivPPARγkd increased MAG levels, which reached significance in LF-fed mice. In HF-fed aLivPPARγkd mice, this was associated with a decrease in DAG and TAG (Fig 5B). Interestingly, in LF-fed aLivPPARγkd, the relative levels of DAG and TAG were not altered. Taken together, loss of hepatocyte PPARγ signaling preferentially reduces esterification of FA via direct transcriptional regulation of the MAG pathway, where loss of this effect could contribute to the protection against excess TAG accumulation observed in aLivPPARγkd mice. DISCUSSION Hepatocyte PPARγ has been defined as a steatogenic factor (Gavrilova et al. 2003; Inoue et al. 2005; Matsusue et al. 2003; Moran-Salvador et al. 2011; Pettinelli and Videla 2011; Rahimian et al. 2001), while others suggest that its activation decreases hepatic steatosis (Belfort et al. 2006; Ratziu et al. 2008; Sanyal et al. 2010). The fact that PPARγ is the target of TZDs, that are used in the treatment of diabetes and NAFLD, raises the question if the hepatic-specific agonism of PPARγ is well understood (Ahmadian et al. 2013). Therefore, we sought to determine which molecular processes are the primary targets for hepatocyte PPARγ that promotes TAG accumulation. We have taken advantage of the novel adult-onset, hepatocyte-specific PPARγ knockdown (aLivPPARγkd) model, that allowed us to study early events altered by aLivPPARγkd (1wk of knockdown) and how this deficit impacts liver function overtime under different dietary conditions. Our approach has benefits over existing congenital knockout models, because it avoids compensatory changes that could occur with embryonic knockout and therefore better models the consequence of manipulating PPARγ function in adult, where NALFD/NASH typically develops. As discussed in detailed below, our primary findings are that hepatic PPARγ has minimal direct effects on hepatic DNL, TAG uptake, TAG export or FA oxidation, but plays a major role in upregulating genes/pathways critical for hepatic FA uptake and MAG pathway - mediated FA esterification. Congenital hepatocyte-specific PPARγ knockout has been reported to be associated with a reduction of hepatic expression of genes critical for DNL (Gavrilova et al. 2003; Matsusue et al. 2003; Moran-Salvador et al. 2011; Yu et al. 2003). Also, use of antisense oligonucleotide strategy to suppress elevated hepatic PPARγ in apoB/BATless mice, reduced hepatic TAG accumulation, as well as the expression of DNL genes and DNL rate as assessed by 3H2O incorporation into TAG-associated FA (Zhang et al. 2006). In addition, overexpression of PPARγ in a hepatic cell line [AML-12; (Schadinger et al. 2005)] was shown to increase SREBP1c and Fasn expression that was associated with an increase in 14C-acetate incorporation into TAG. Finally, overexpression of PPARγ in some (Yu et al. 2003), but not all (Bai et al. 2011) in vivo models, increased expression of genes associated with DNL. In contrast to these reports, in the aLivPPARγkd model system, expression of DNL genes were not suppressed in any conditions tested, consistent with the lack of an effect on DNL indices (16:1/16:0 and 16:0/18:2). In fact, we have recently reported although PPARγ expression was associated with TAG accumulation in a model of DNL-generated steatosis (Kineman et al. 2016), adult-onset hepatic PPARγ knockdown in this steatotic model did not reduce hepatic TAG content or FA indices of DNL (Cordoba-Chacon et al. 2015a). The question arises, why are our current results counter to that previously reported by others? We might speculate that any changes observed in the expression of DNL genes in congenital liver-specific PPARγ knockout models could be secondary to changes that occur due to compensation during development or systemic metabolic changes that occur overtime. Also, the use of antisense oligonucleotides (ASO) to acutely suppress enhanced PPARγ expression in vivo (Zhang et al. 2006), is not a hepatocyte-specific approach, and therefore a reduction in PPARγ in other cell types could contribute to the phenotype observed. Finally in vivo (Zhang et al. 2006) or in vitro (Schadinger et al. 2005) overexpression of PPARγ may have off-target effects. Therefore we can conclude that in the context of adult metabolic function, loss of hepatic PPARγ does not directly control hepatic DNL. Early work by Gavrilova et al (Gavrilova et al. 2003) and Matsusue et al (Matsusue et al. 2003), who crossbred the congenital liver-specific PPARγ model with a lipodystrophic model (AZIP, (Gavrilova et al. 2003)), and ob/ob mice (Matsusue et al. 2003), respectively, observed a reduction in hepatic TAG content associated with elevated plasma TAG and thus concluded that hepatic PPARγ was critical to maintain hepatic TAG uptake. However, in chow-fed WT mice (Matsusue et al. 2003; Moran-Salvador et al. 2011) and HF-fed (Moran-Salvador et al. 2011) mice with congenital liver-specific PPARγ knockout, as well as in our HF-fed aLivPPARγkd model, circulating TAG did not differ from PPARγ-intact controls, despite the dramatic reduction in hepatic TAG content. Also, there were no decreases in the expression of genes known to be critical in hepatic TAG uptake in mouse livers, including low density lipoprotein receptor [Ldlr; (Havel and Hamilton 2004; Ishibashi, et al. 1994)] and hepatic lipase [HL;(Freeman, et al. 2007; Havel and Hamilton 2004)]. However, it should be noted that the expression of very low density lipoprotein receptor (Vldr), a known PPARγ target in adipocytes (Tao, et al. 2010), was significantly reduced in HF-fed aLivPPARγkd mice. The expression of hepatic Vldlr is normally low, but is increased in mouse models of fatty liver (also observed with HF-feeding in this study, see Supplemental Table 1), and whole body knockout of Vldlr reduces ER stress and HF-diet induced hepatic steatosis (Jo, et al. 2013). However, a liver-specific role of Vldlr in hepatic TAG uptake remains to be determined. Although the role of PPARγ in regulating hepatic TAG uptake remains to be further explored, our current results coupled with previous reports, do provide compelling evidence that hepatic PPARγ promotes hepatic FA uptake by regulating the expression of Cd36. Specifically, in both aged and HF-fed aLivPPARγkd mice expression of Cd36 was reduced. Cd36 has been shown to be a direct target of PPARγ (Tontonoz et al. 1998) and its expression is increased in steatotic livers. In the current study, a reduction in Cd36 levels in aLivPPARγkd mice was only observed with age or HF-feeding, consistent with the fact that in addition to PPARγ, LXR and PXR are also required for full expression of Cd36 (Zhou et al. 2008). Although global Cd36 knockout does not protect against high fructose-induced hepatic steatosis (Hajri, et al. 2002) or prevent fatty liver in ob/ob mice (Nassir, et al. 2013), a more recent report supports a liver-specific role of Cd36 in FA uptake. Specifically, congenital liver-specific Cd36 knockout reduced steatosis in liver-specific Jak2 knockout mice that led to an increase in plasma NEFA (Wilson, et al. 2015). Also in that same study, liver-specific Cd36 knockout mice with intact hepatic Jak2, dramatically reduced HF-diet induced steatosis that was associated with a reduction in hepatic FA uptake as measured by hepatic accumulation of BODIPY-FA in vivo (Wilson et al. 2015). However, loss of hepatic Cd36 did not entirely prevent hepatic FA uptake which could be mediated by other facilitated transport mechanisms or passive diffusion (Glatz, et al. 2010), indicating the reduction in CD36 after aLivPPARγkd is only in part responsible for the reduction in hepatic TAG content. Of the selected genes examined in this study, the expression of Mogat1 was the most sensitive to PPARγ loss. As previously reported, the expression of Mogat1 in the lean mouse liver is low, but increases in association with TAG accumulation, similar to that observed for PPARγ (Hall, et al. 2014; Lee et al. 2012; Soufi, et al. 2014; Yu, et al. 2016), as we also observed in the current study. In addition hepatic expression of Mogat1, as well as PPARγ, is elevated in humans with NAFLD (Hall et al. 2012; Yu et al. 2015). There is an ongoing debate regarding the physiologic role hepatic Mogat1 plays in TAG synthesis, since it was originally thought that the glycerol-3-phosphate pathway, not the MAG pathway, is the dominant route to form TAG in the liver (Coleman and Mashek 2011; Mashek 2013). In fact one laboratory could not detect hepatic MOGAT activity (Cortes, et al. 2009) and recently reported that global Mogat1 knockout in the ob/ob or Agpat2KO mice did not impact steatosis (Agarwal, et al. 2016). However, hepatic MOGAT activity has been detected by other laboratories (Hall et al. 2012; Yen, et al. 2002). Importantly, the increase in Mogat1 expression observed in HF-fed and ob/ob mice is associated with an elevation of hepatic MOGAT activity, which is reduced by Mogat1 ASO ip treatment (Hall et al. 2014; Soufi et al. 2014). The Mogat1-ASO injections reduced hepatic TAG accumulation in one study (Soufi et al. 2014), but not the other (Hall et al. 2014). However, it was acknowledge that the knockdown of Mogat1 was not hepatocyte-specific, and actually led to a reduction Mogat1 expression in the adipose tissue that could offset any hepatocyte-specific effect (Soufi et al. 2014). In strong support of a physiologic role of hepatic Mogat1 in maintaining hepatic TAG levels, adenoviral shRNA-Mogat1 delivery (that is preferentially taken up by the liver) (Lee et al. 2012), and nonviral siRNA-Mogat1 delivery (Hayashi, et al. 2014), reduced Mogat1 expression and hepatic TAG levels in HF-fed mice. The results of the present study indicate PPARγ is necessary to maintain Mogat1 activity, as well as Mogat1 expression, based on the increase in hepatic MAG levels in aLivPPARγkd mice. Interestingly, it has been postulated that Mogat1 may prefer extrahepatic (dietary FA) over those produced by hepatic DNL, to synthesize DAG (Steneberg, et al. 2015). This would be consistent with our observation that although MAG levels were increased in LF-fed aLivPPARγkd mice, DAG and TAG levels were normal, where in this context hepatic DAG and TAG are mainly derived from DNL. However, in HF-fed aLivPPARγkd mice (where the bulk of FA is coming from the diet), a reduction in DAG and TAG levels was observed. Taken together, results suggest that hepatocyte PPARγ expression in adult livers is not essential to maintain expression of DNL genes, but it is essential to induce the expression of genes important in hepatic FA uptake (Cd36) and re-esterification (Mogat1). Certainly other pathways may be mediated by hepatic PPARγ that were not revealed by our targeted approach. Nonetheless, our current data, coupled with previous reports suggest that impairment of the Cd36-mediated FA uptake and MAG pathway FA esterification pathways could explain in part the protection against steatosis observed in aLivPPARγkd mice. Supplementary Material 01 02 FUNDING This work was supported by Department of Veterans Affairs, Office of Research and Development Merit Award (BX001090, BX001114), National Institutes of Health (R21AT008457, S10OD010660, R01DK088133), Chicago Biomedical Consortium with support from the Searle Funds at The Chicago Community Trust (PDR-033) and 2014 Endocrine Scholar Award in Growth Hormone Research (Endocrine Society). Figure 1 Hepatocyte-specificity expression of AAV8-TBGp driven transgene Hepatocyte-specific expression of GFP in AAV8-TBGp-EGFP injected wild-type mice (green, A,B). GFP expression was absent in non-hepatocyte cells (yellow arrows) in sinusoids (Sn), central vein (CV), portal vein (PV), bile duct (BD) or artery (HA). TBGp-GFP was not expressed in hepatic stellate cells (HSC, desmin +, red, C) or macrophages (Mac, F4/80+, red, D). Sections were counterstained with DAPI (blue nuclei, A–D). Hepatic GFP (E) and Cre (F) expression was detected only in liver extracts of AAV8-TBGp-EGFP and AAV8-TBGp-Cre injected mice, respectively. In order to confirm the hepatocyte-specific activity of Cre recombinase, PPARγfl/fl mice were injected with AAV8-TBGp-Cre and expression of PPARγ was reduced in hepatic extracts but not adipose tissue (G). eWAT, epididymal fat; iWAT, inguinal fat; Ctx, cortex; Int, intestine. Asterisks indicate difference between AAV8-TBGp-Cre injected mice as compared to AAV8-TBGp-Null mice. *, p<0.05; ***, p<0.0001. n=3–6 mice/group. Figure 2 Adult-onset hepatocyte-specific PPARγ knock-down (aLivPPARγkd) in chow-fed mice (A–C) and diet-induced obese mice (D–F) A) PPARγfl/fl mice were injected at 10 weeks of age with 1.5*1011GC AAV8-TBGp-Null (C, open columns) or 1.5*1011GC AAV8-TBGp-Cre (Kd, shaded columns) via lateral tail vein and killed 1 or 18 weeks after. B) Hepatic PPARγ mRNA and (C) protein expression of 1 wk and 18 wks aLivPPARγkd and their littermate control mice. Representative image of the western-blot for hepatic PPARγ and Histone [as loading control]). Of note, nuclear protein variability within total hepatic extracts (assessed by Histone) alters the amount of PPARγ protein detected. n=4–5 mice/group. D) LF-fed PPARγfl/fl mice were injected at 10 weeks as described above and immediately half of the mice were fed at high-fat diet (HF, 60% Kcal from fat) to induce liver steatosis while the rest were maintained on a low fat diet (LF, 10% Kcal from fat). Mice were killed 14 weeks after. E). Hepatic PPARγ mRNA and (F) protein expression of 14 wks LF- and HF-fed aLivPPARγkd and their littermate controls. Representative image of the western-blot for hepatic PPARγ and Histone [as loading control]. Asterisks indicate differences between C and Kd. *, p<0.05; ***, p<0.0001. Letters indicate differences between 1wk and 18 wks or LF and HF-fed mice within group. b, p<0.01; c, p<0.0001. n=5–6 mice/group Figure 3 Impact of aLivPPARγkd in hepatic gene expression of molecular mechanisms controlling hepatic TAG levels A) 1 wk (top) and 18 wks (bottom) chow-fed and B) 14 wks LF-fed (top) and HF-fed (bottom) aLivPPARγkd-induced regulation of hepatic gene expression. Graphs represent the natural logarithm of the relative change in the gene expression of aLivPPARγkd mice as compared to their littermates controls (set at 0, x-axis), within age (A) or diet (B). Absolute values are included in Supplemental Table 1A,B. Asterisks show significant changes between C and aLivPPARγkd within age (A) or diet (B). *, p<0.05; **, p<0.01, ***, p<0.0001. Livers were collected at 1100h (4h after food removal). FAox: fatty acid oxidation, TAGhydr: TAG hydrolysis, VLDLsyn: VLDL synthesis, DNL: de novo lipogenesis and TAGsyn: TAG synthesis. Selected genes of these metabolic pathways represented in figure 3: peroxisome proliferator-activated receptor α (PPARα), acyl-CoA synthetase long-chain family member 1 (Acsl1), carnitine palmitoyltransferase 1α (Cpt1α), hepatic nuclear factor 4 α (Hnf4α), PPARγ co-activator 1 α; (Pgc1α), Cyp4a10, adipose triglyceride lipase (Atgl), hormone-sensitive lipase (Hsl), monoacylglycerol lipase (Mgll), apolipoprotein B (ApoB), microsomal triglyceride transfer protein (Mttp), sterol response element binding protein 1c (Srebp1c), acetyl-CoA carboxilase 1 (Acc1), fatty acid synthase (Fasn), fatty acid elongase (Elovl6), stearoyl –CoA desaturase 1 (Scd1), hepatic lipase (Hl), low density lipoprotein lipase receptor (Ldlr), very-low density lipoprotein receptor (Vldlr), lipoprotein related protein 1 (Lrp1), fatty acid translocase (Cd36), glycerol phosphate acyltransferase (Gpat1), monoacylglycerol acyltransferase 1 or 2 (Mogat1/2), diacylglycerolacyltransferase 1/2 (Dgat1/2). Figure 4 aLivPPARγkd reduces HF diet-induced FA levels, but has little impact on FA indices of DNL A) Hepatic FA levels of 16:0, palmitate; 16:1, palmitoleate; and 18:2, linoleate, levels and B) hepatic FA ratios indicative of DNL include the SCD-index (16:1/16:0) and the DNL-index (16:0/18:2), where control (open columns, C) and aLivPPARγkd (closed columns, Kd) mice were fed a LF- or a HF-diet for 14 weeks. Asterisks indicate differences between C and Kd. ***, p<0.0001. Letters indicate differences between LF and HF-fed mice within group. a, p<0.05; c, p<0.0001. n=4–6 mice/group. Figure 5 aLivPPARγkd reduces HF diet-induced hepatic TAG and DAG levels while increasing MAG levels, independent of diet, indicative of impaired hepatic MAG pathway activity in aLivPPARγkd mice A) Schematic representation of acylglycerol synthesis by glycerol-3-phosphate (G3P) pathway that produces DAG by subsequent re-esterification of FA in G3P and lysophosphatidic acid (LPA) or by monoacylglycerol (MAG) pathway that produces DAG after re-esterification of FA in MAG. B) Relative hepatic TAG, DAG and MAG levels assessed by liquid chromatography/mass spectrometry (LC/MS) in control (open columns) and aLivPPARγkd (close columns) mice. TAG, DAG and MAG are shown as relative values of LF-fed controls. Asterisks indicate differences between control and aLivPPARγkd. ***, p<0.0001. Letters indicate differences between LF and HF-fed mice within group. b, p<0.01; c, p<0.0001 n=5–6 mice/group. Table 1 Impact of aLivPPARγkd on body, liver and fat depot (sum of the unilateral epididymal, inguinal and retroperitoneal fat depot) weight, hepatic TAG levels and circulating metabolic endpoints (blood glucose, plasma insulin, ketones, TAG, NEFA and NEFA), in chow-fed (top, 1 wk and 18 wks) and LF/HF-fed (bottom, 14 wks) aLivPPARγkd mice and their littermate controls. 1 wk aLivPPARγkd - chow diet 18 wk aLivPPARγkd - chow diet Overall effect (p-value) of Control aLivPPARγkd Control aLivPPARγkd Mean SEM Mean SEM Mean SEM Mean SEM Age aLivPPARγkd Body weight g 25.36 ± 0.84 24.60 ± 0.64 31.81b ± 1.88 32.49c ± 0.60 <0.0001 0.9719 Liver g/g BW 0.045 ± 0.0008 0.043 ± 0.0022 0.038a ± 0.0011 0.038 ± 0.0021 <0.0022 0.3959 Fat mass g/g BW 0.015 ± 0.0015 0.013 ± 0.0011 0.029a ± 0.0066 0.023 ± 0.0031 0.0091 0.2680 Hepatic TAG mg/g 47.63 ± 6.987 38.63 ± 3.027 50.36 ± 4.611 34.72 ± 3.653 0.9024 0.0185 Blood Glucose mg/dL 187.0 ± 14.77 176.8 ± 18.87 190.4 ± 15.70 199.7 ± 8.92 0.3797 0.9748 Plasma insulin ng/ul 1.59 ± 0.15 1.69 ± 0.12 1.88 ± 0.37 1.34 ± 0.14 0.8925 0.3323 Plasma ketones μM 372.1 ± 62.194 372.1 ± 57.920 153.9a ± 40.907 226.1 ± 33.635 0.0017 0.4704 Plasma TAG mg/dL 91.26 ± 3.772 83.95 ± 5.728 61.35 ± 12.607 67.25 ± 10.736 0.0220 0.9396 Plasma NEFA mEq/L 1.05 ± 0.038 1.12 ± 0.063 1.06 ± 0.095 1.07 ± 0.064 0.7687 0.5581 Plasma Cholesterol mg/dL 104.83 ± 5.568 107.83 ± 5.233 116.17 ± 8.003 116.61 ± 5.334 0.1177 0.7812 14 wks aLivPPARgkd - LF diet 14 wks aLivPPARgkd - HF diet Overall effect (p-value) of Control aLivPPARγkd Control aLivPPARγkd Mean SEM Mean SEM Mean SEM Mean SEM Diet aLivPPARγkd Body weight g 30.88 ± 0.95 29.65 ± 0.60 49.52c ± 0.55 48.52c ± 1.13 <0.0001 0.1827 Liver g/g BW 0.0376 ± 0.0007 0.0373 ± 0.0006 0.049c ± 0.0010 0.032b*** ± 0.002 0.0062 <0.0001 Fat mass g/g BW 0.0313 ± 0.0035 0.0296 ± 0.0032 0.055c ± 0.0024 0.065c ± 0.002 <0.0001 0.1564 Hepatic TAG mg/g 37.13 ± 9.114 41.27 ± 9.188 326.25c ± 16.654 111.37b*** ± 18.11 <0.0001 <0.0001 Blood Glucose mg/dL 182.6 ± 20.78 187.2 ± 9.14 218.0 ± 15.12 231.2 ± 12.31 0.0128 0.5449 Plasma insulin ng/ul 1.39 ± 0.11 1.96 ± 0.30 11.27c ± 0.85 9.39c ± 1.68 <0.0001 0.4500 Plasma ketones μM 232.3 ± 42.17 215.4 ± 30.85 348.71 ± 22.787 215.41* ± 31.53 0.0886 0.0323 Plasma TAG mg/dL 52.78 ± 7.716 57.18 ± 7.479 35.80 ± 3.672 40.83 ± 4.21 0.0138 0.4479 Plasma NEFA mEq/L 1.17 ± 0.050 1.1 ± 0.08 0.53b ± 0.070 1.15** ± 0.20 0.0221 0.0340 Plasma Cholesterol mg/dL 117.71 ± 12.99 104.57 ± 9.20 212.86b ± 37.85 110.29** ± 8.35 0.0177 0.0078 Superscript letters indicate differences between 1wk and 18 wks or LF and HF-fed mice within group. a p<0.05; b p<0.01; c p<0,0001. Asterisks indicate differences between control and aLivPPARγkd mice. * p<0.05; ** p<0.01; *** p<0.0001. n=4–6 mice/group DECLARATION OF INTEREST Authors do not have any conflict of interest. Agarwal AK Tunison K Dalal JS Yen CL Farese RV Jr Horton JD Garg A 2016 Mogat1 deletion does not ameliorate hepatic steatosis in lipodystrophic (Agpat2−/−) or obese (ob/ob) mice J Lipid Res 57 616 630 26880786 Ahmadian M Suh JM Hah N Liddle C Atkins AR Downes M Evans RM 2013 PPARgamma signaling and metabolism: the good, the bad and the future Nat Med 19 557 566 23652116 Ashpole NM Herron JC Mitschelen MC Farley JA Logan S Yan H Ungvari Z Hodges EL Csiszar A Ikeno Y 2016 IGF-1 Regulates Vertebral Bone Aging Through Sex-Specific and Time-Dependent Mechanisms J Bone Miner Res 31 443 454 26260312 Bai L Jia Y Viswakarma N Huang J Vluggens A Wolins NE Jafari N Rao MS Borensztajn J Yang G 2011 Transcription coactivator mediator subunit MED1 is required for the development of fatty liver in the mouse Hepatology 53 1164 1174 21480322 Belfort R Harrison SA Brown K Darland C Finch J Hardies J Balas B Gastaldelli A Tio F Pulcini J 2006 A placebo-controlled trial of pioglitazone in subjects with nonalcoholic steatohepatitis N Engl J Med 355 2297 2307 17135584 Benard O Lim J Apontes P Jing X Angeletti RH Chi Y 2016 Impact of high-fat diet on the proteome of mouse liver J Nutr Biochem 31 10 19 27133419 Beysen C Murphy EJ Nagaraja H Decaris M Riiff T Fong A Hellerstein MK Boyle PJ 2008 A pilot study of the effects of pioglitazone and rosiglitazone on de novo lipogenesis in type 2 diabetes J Lipid Res 49 2657 2663 18641372 Bligh EG Dyer WJ 1959 A rapid method of total lipid extraction and purification Can J Biochem Physiol 37 911 917 13671378 Browning JD Horton JD 2004 Molecular mediators of hepatic steatosis and liver injury J Clin Invest 114 147 152 15254578 Coleman RA Mashek DG 2011 Mammalian triacylglycerol metabolism: synthesis, lipolysis, and signaling Chem Rev 111 6359 6386 21627334 Cordoba-Chacon J Gahete MD McGuinness OP Kineman RD 2014a Differential impact of selective GH deficiency and endogenous GH excess on insulin-mediated actions in muscle and liver of male mice Am J Physiol Endocrinol Metab 307 E928 934 25269484 Cordoba-Chacon J Gahete MD Pokala NK Geldermann D Alba M Salvatori R Luque RM Kineman RD 2014b Long- but not short-term adult-onset, isolated GH deficiency in male mice leads to deterioration of beta-cell function, which cannot be accounted for by changes in beta-cell mass Endocrinology 155 726 735 24424062 Cordoba-Chacon J Majumdar N List EO Diaz-Ruiz A Frank SJ Manzano A Bartrons R Puchowicz M Kopchick JJ Kineman RD 2015a Growth hormone inhibits hepatic de novo lipogenesis in adult mice Diabetes 64 3093 3103 26015548 Cordoba-Chacon J Majumdar N Pokala NK Gahete MD Kineman RD 2015b Islet insulin content and release are increased in male mice with elevated endogenous GH and IGF-I, without evidence of systemic insulin resistance or alterations in beta-cell mass Growth Horm IGF Res 25 189 195 25936582 Cortes VA Curtis DE Sukumaran S Shao X Parameswara V Rashid S Smith AR Ren J Esser V Hammer RE 2009 Molecular mechanisms of hepatic steatosis and insulin resistance in the AGPAT2-deficient mouse model of congenital generalized lipodystrophy Cell Metab 9 165 176 19187773 Duarte JA Carvalho F Pearson M Horton JD Browning JD Jones JG Burgess SC 2014 A high-fat diet suppresses de novo lipogenesis and desaturation but not elongation and triglyceride synthesis in mice J Lipid Res 55 2541 2553 25271296 Freeman L Amar MJ Shamburek R Paigen B Brewer HB Jr Santamarina-Fojo S Gonzalez-Navarro H 2007 Lipolytic and ligand-binding functions of hepatic lipase protect against atherosclerosis in LDL receptor-deficient mice J Lipid Res 48 104 113 17071916 Furnsinn C Waldhausl W 2002 Thiazolidinediones: metabolic actions in vitro Diabetologia 45 1211 1223 12242453 Gavrilova O Haluzik M Matsusue K Cutson JJ Johnson L Dietz KR Nicol CJ Vinson C Gonzalez FJ Reitman ML 2003 Liver peroxisome proliferator-activated receptor gamma contributes to hepatic steatosis, triglyceride clearance, and regulation of body fat mass J Biol Chem 278 34268 34276 12805374 Glatz JF Luiken JJ Bonen A 2010 Membrane fatty acid transporters as regulators of lipid metabolism: implications for metabolic disease Physiol Rev 90 367 417 20086080 Guo J Jou W Gavrilova O Hall KD 2009 Persistent diet-induced obesity in male C57BL/6 mice resulting from temporary obesigenic diets PLoS One 4 e5370 19401758 Gupte AA Liu JZ Ren Y Minze LJ Wiles JR Collins AR Lyon CJ Pratico D Finegold MJ Wong ST 2010 Rosiglitazone attenuates age- and diet-associated nonalcoholic steatohepatitis in male low-density lipoprotein receptor knockout mice Hepatology 52 2001 2011 20938947 Hajri T Han XX Bonen A Abumrad NA 2002 Defective fatty acid uptake modulates insulin responsiveness and metabolic responses to diet in CD36-null mice J Clin Invest 109 1381 1389 12021254 Hall AM Kou K Chen Z Pietka TA Kumar M Korenblat KM Lee K Ahn K Fabbrini E Klein S 2012 Evidence for regulated monoacylglycerol acyltransferase expression and activity in human liver J Lipid Res 53 990 999 22394502 Hall AM Soufi N Chambers KT Chen Z Schweitzer GG McCommis KS Erion DM Graham MJ Su X Finck BN 2014 Abrogating monoacylglycerol acyltransferase activity in liver improves glucose tolerance and hepatic insulin signaling in obese mice Diabetes 63 2284 2296 24595352 Havel RJ Hamilton RL 2004 Hepatic catabolism of remnant lipoproteins: where the action is Arterioscler Thromb Vasc Biol 24 213 215 14766735 Hayashi Y Suemitsu E Kajimoto K Sato Y Akhter A Sakurai Y Hatakeyama H Hyodo M Kaji N Baba Y 2014 Hepatic Monoacylglycerol O-acyltransferase 1 as a Promising Therapeutic Target for Steatosis, Obesity, and Type 2 Diabetes Mol Ther Nucleic Acids 3 e154 24643205 He W Barak Y Hevener A Olson P Liao D Le J Nelson M Ong E Olefsky JM Evans RM 2003 Adipose-specific peroxisome proliferator-activated receptor gamma knockout causes insulin resistance in fat and liver but not in muscle Proc Natl Acad Sci U S A 100 15712 15717 14660788 Horakova O Hansikova J Bardova K Gardlo A Rombaldova M Kuda O Rossmeisl M Kopecky J 2016 Plasma Acylcarnitines and Amino Acid Levels As an Early Complex Biomarker of Propensity to High-Fat Diet-Induced Obesity in Mice PLoS One 11 e0155776 27183228 Inoue M Ohtake T Motomura W Takahashi N Hosoki Y Miyoshi S Suzuki Y Saito H Kohgo Y Okumura T 2005 Increased expression of PPARgamma in high fat diet-induced liver steatosis in mice Biochem Biophys Res Commun 336 215 222 16125673 Ishibashi S Herz J Maeda N Goldstein JL Brown MS 1994 The two-receptor model of lipoprotein clearance: tests of the hypothesis in “knockout” mice lacking the low density lipoprotein receptor, apolipoprotein E, or both proteins Proc Natl Acad Sci U S A 91 4431 4435 8183926 Jo H Choe SS Shin KC Jang H Lee JH Seong JK Back SH Kim JB 2013 Endoplasmic reticulum stress induces hepatic steatosis via increased expression of the hepatic very low-density lipoprotein receptor Hepatology 57 1366 1377 23152128 Kineman R Majumdar N Subbaiah PV Cordoba-Chacon J 2016 Hepatic PPARγ is not essential for the rapid development of steatosis following loss of hepatic GH signaling, in adult male mice Endocrinology 157 1728 1735 26950202 Lade A Noon LA Friedman SL 2014 Contributions of metabolic dysregulation and inflammation to nonalcoholic steatohepatitis, hepatic fibrosis, and cancer Curr Opin Oncol 26 100 107 24275855 Lee JJ Lambert JE Hovhannisyan Y Ramos-Roman MA Trombold JR Wagner DA Parks EJ 2015 Palmitoleic acid is elevated in fatty liver disease and reflects hepatic lipogenesis Am J Clin Nutr 101 34 43 25527748 Lee YJ Ko EH Kim JE Kim E Lee H Choi H Yu JH Kim HJ Seong JK Kim KS 2012 Nuclear receptor PPARgamma-regulated monoacylglycerol O-acyltransferase 1 (MGAT1) expression is responsible for the lipid accumulation in diet-induced hepatic steatosis Proc Natl Acad Sci U S A 109 13656 13661 22869740 Mashek DG 2013 Hepatic fatty acid trafficking: multiple forks in the road Adv Nutr 4 697 710 24228201 Matsusue K Aibara D Hayafuchi R Matsuo K Takiguchi S Gonzalez FJ Yamano S 2014 Hepatic PPARgamma and LXRalpha independently regulate lipid accumulation in the livers of genetically obese mice FEBS Lett 588 2277 2281 24857376 Matsusue K Haluzik M Lambert G Yim SH Gavrilova O Ward JM Brewer B Jr Reitman ML Gonzalez FJ 2003 Liver-specific disruption of PPARgamma in leptin-deficient mice improves fatty liver but aggravates diabetic phenotypes J Clin Invest 111 737 747 12618528 Michelotti GA Machado MV Diehl AM 2013 NAFLD, NASH and liver cancer Nat Rev Gastroenterol Hepatol 10 656 665 24080776 Moran-Salvador E Lopez-Parra M Garcia-Alonso V Titos E Martinez-Clemente M Gonzalez-Periz A Lopez-Vicario C Barak Y Arroyo V Claria J 2011 Role for PPARgamma in obesity-induced hepatic steatosis as determined by hepatocyte- and macrophage-specific conditional knockouts Faseb J 25 2538 2550 21507897 Nan YM Fu N Wu WJ Liang BL Wang RQ Zhao SX Zhao JM Yu J 2009 Rosiglitazone prevents nutritional fibrosis and steatohepatitis in mice Scand J Gastroenterol 44 358 365 18991162 Nassir F Adewole OL Brunt EM Abumrad NA 2013 CD36 deletion reduces VLDL secretion, modulates liver prostaglandins, and exacerbates hepatic steatosis in ob/ob mice J Lipid Res 54 2988 2997 23964120 Obrowsky S Chandak PG Patankar JV Povoden S Schlager S Kershaw EE Bogner-Strauss JG Hoefler G Levak-Frank S Kratky D 2013 Adipose triglyceride lipase is a TG hydrolase of the small intestine and regulates intestinal PPARalpha signaling J Lipid Res 54 425 435 23220585 Pettinelli P Videla LA 2011 Up-regulation of PPAR-gamma mRNA expression in the liver of obese patients: an additional reinforcing lipogenic mechanism to SREBP-1c induction J Clin Endocrinol Metab 96 1424 1430 21325464 Rahimian R Masih-Khan E Lo M van Breemen C McManus BM Dube GP 2001 Hepatic over-expression of peroxisome proliferator activated receptor gamma2 in the ob/ob mouse model of non-insulin dependent diabetes mellitus Mol Cell Biochem 224 29 37 11693197 Ratziu V Giral P Jacqueminet S Charlotte F Hartemann-Heurtier A Serfaty L Podevin P Lacorte JM Bernhardt C Bruckert E 2008 Rosiglitazone for nonalcoholic steatohepatitis: one-year results of the randomized placebo-controlled Fatty Liver Improvement with Rosiglitazone Therapy (FLIRT) Trial Gastroenterology 135 100 110 18503774 Sanyal AJ Chalasani N Kowdley KV McCullough A Diehl AM Bass NM Neuschwander-Tetri BA Lavine JE Tonascia J Unalp A 2010 Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis N Engl J Med 362 1675 1685 20427778 Schadinger SE Bucher NL Schreiber BM Farmer SR 2005 PPARgamma2 regulates lipogenesis and lipid accumulation in steatotic hepatocytes Am J Physiol Endocrinol Metab 288 E1195 1205 15644454 Sevastianova K Santos A Kotronen A Hakkarainen A Makkonen J Silander K Peltonen M Romeo S Lundbom J Lundbom N 2012 Effect of short-term carbohydrate overfeeding and long-term weight loss on liver fat in overweight humans Am J Clin Nutr 96 727 734 22952180 Shin S Wangensteen KJ Teta-Bissett M Wang YJ Mosleh-Shirazi E Buza EL Greenbaum LE Kaestner KH 2016 Genetic Lineage Tracing Analysis of the Cell of Origin of Hepatotoxin-Induced Liver Tumors in Mice Hepatology 10.1002/hep.28602 Silbernagel G Kovarova M Cegan A Machann J Schick F Lehmann R Haring HU Stefan N Schleicher E Fritsche A 2012 High hepatic SCD1 activity is associated with low liver fat content in healthy subjects under a lipogenic diet J Clin Endocrinol Metab 97 E2288 2292 23015656 Soufi N Hall AM Chen Z Yoshino J Collier SL Mathews JC Brunt EM Albert CJ Graham MJ Ford DA 2014 Inhibiting monoacylglycerol acyltransferase 1 ameliorates hepatic metabolic abnormalities but not inflammation and injury in mice J Biol Chem 289 30177 30188 25213859 Steneberg P Sykaras AG Backlund F Straseviciene J Soderstrom I Edlund H 2015 Hyperinsulinemia Enhances Hepatic Expression of the Fatty Acid Transporter Cd36 and Provokes Hepatosteatosis and Hepatic Insulin Resistance J Biol Chem 290 19034 19043 26085100 Tao H Aakula S Abumrad NN Hajri T 2010 Peroxisome proliferator-activated receptor-gamma regulates the expression and function of very-low-density lipoprotein receptor Am J Physiol Endocrinol Metab 298 E68 79 19861583 Tontonoz P Nagy L Alvarez JG Thomazy VA Evans RM 1998 PPARgamma promotes monocyte/macrophage differentiation and uptake of oxidized LDL Cell 93 241 252 9568716 Wilson CG Tran JL Erion DM Vera NB Febbraio M Weiss EJ 2015 Hepatocyte-specific disruption of CD36 attenuates fatty liver and improves insulin sensitivity in HFD fed mice Endocrinology 157 570 585 26650570 Xu A Wang Y Keshaw H Xu LY Lam KS Cooper GJ 2003 The fat-derived hormone adiponectin alleviates alcoholic and nonalcoholic fatty liver diseases in mice J Clin Invest 112 91 100 12840063 Yamauchi T Kamon J Minokoshi Y Ito Y Waki H Uchida S Yamashita S Noda M Kita S Ueki K 2002 Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase Nat Med 8 1288 1295 12368907 Yamazaki T Shiraishi S Kishimoto K Miura S Ezaki O 2011 An increase in liver PPARgamma2 is an initial event to induce fatty liver in response to a diet high in butter: PPARgamma2 knockdown improves fatty liver induced by high-saturated fat J Nutr Biochem 22 543 553 20801631 Yang P Subbaiah PV 2015 Regulation of hepatic lipase activity by sphingomyelin in plasma lipoproteins Biochim Biophys Acta 1851 1327 1336 26193433 Yanger K Knigin D Zong Y Maggs L Gu G Akiyama H Pikarsky E Stanger BZ 2014 Adult hepatocytes are generated by self-duplication rather than stem cell differentiation Cell Stem Cell 15 340 349 25130492 Yen CL Stone SJ Cases S Zhou P Farese RV Jr 2002 Identification of a gene encoding MGAT1, a monoacylglycerol acyltransferase Proc Natl Acad Sci U S A 99 8512 8517 12077311 Younossi ZM Stepanova M Afendy M Fang Y Younossi Y Mir H Srishord M 2011 Changes in the prevalence of the most common causes of chronic liver diseases in the United States from 1988 to 2008 Clin Gastroenterol Hepatol 9 524 530 e521 quiz e560 21440669 Yu JH Lee YJ Kim HJ Choi H Choi Y Seok JW Kim JW 2015 Monoacylglycerol O-acyltransferase 1 is regulated by peroxisome proliferator-activated receptor gamma in human hepatocytes and increases lipid accumulation Biochem Biophys Res Commun 460 715 720 25838202 Yu JH Song SJ Kim A Choi Y Seok JW Kim HJ Lee YJ Lee KS Kim JW 2016 Suppression of PPARgamma-mediated monoacylglycerol O-acyltransferase 1 expression ameliorates alcoholic hepatic steatosis Sci Rep 6 29352 27404390 Yu S Matsusue K Kashireddy P Cao WQ Yeldandi V Yeldandi AV Rao MS Gonzalez FJ Reddy JK 2003 Adipocyte-specific gene expression and adipogenic steatosis in the mouse liver due to peroxisome proliferator-activated receptor gamma1 (PPARgamma1) overexpression J Biol Chem 278 498 505 12401792 Zezos P Renner EL 2014 Liver transplantation and non-alcoholic fatty liver disease World J Gastroenterol 20 15532 15538 25400437 Zhang YL Hernandez-Ono A Siri P Weisberg S Conlon D Graham MJ Crooke RM Huang LS Ginsberg HN 2006 Aberrant hepatic expression of PPARgamma2 stimulates hepatic lipogenesis in a mouse model of obesity, insulin resistance, dyslipidemia, and hepatic steatosis J Biol Chem 281 37603 37615 16971390 Zhou J Febbraio M Wada T Zhai Y Kuruba R He J Lee JH Khadem S Ren S Li S 2008 Hepatic fatty acid transporter Cd36 is a common target of LXR, PXR, and PPARgamma in promoting steatosis Gastroenterology 134 556 567 18242221

PMC005xxxxxx/PMC5120587.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 0147763 2979 Circulation Circulation Circulation 0009-7322 1524-4539 27364164 5120587 10.1161/CIRCULATIONAHA.116.022304 NIHMS792439 Article Leukocyte-Expressed β2-Adrenergic Receptors are Essential for Survival Following Acute Myocardial Injury Grisanti Laurel A. PhD 1 Gumpert Anna M. PhD 1 Traynham Christopher J. PhD 1 Gorsky Joshua E. BS 1 Repas Ashley A. BS 1 Gao Erhe MD 12 Carter Rhonda L. BS 1 Yu Daohai PhD 3 Calvert John W. PhD 4 García Andrés Pun MS 5 Ibáñez Borja MD, PhD 5 Rabinowitz Joseph E. PhD 12 Koch Walter J. PhD 12 Tilley Douglas G. PhD 12 1 Center for Translational Medicine, Temple University School of Medicine, Philadelphia, PA, USA 2 Department of Pharmacology, Temple University School of Medicine, Philadelphia, PA, USA 3 Department of Clinical Sciences, Temple University School of Medicine, Philadelphia, PA, USA 4 Department of Surgery, Division of Cardiothoracic Surgery, Emory University School of Medicine and Carlyle Fraser Heart Center, Atlanta, GA, USA 5 Spanish National Center for Cardiovascular Research, Madrid Spain Correspondence to Douglas G. Tilley, PhD, Room 945A MERB, Center for Translational Medicine, Lewis Katz School of Medicine, Temple University, 3500 N. Broad St., Philadelphia, PA 19140, Tel.: 215-707-2791, Fax: 215-707-9890, douglas.tilley@temple.edu 8 6 2016 30 6 2016 12 7 2016 12 7 2017 134 2 153167 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Background Immune cell-mediated inflammation is an essential process for mounting a repair response following myocardial infarction (MI). The sympathetic nervous system is known to regulate immune system function through β-adrenergic receptors (βAR), however their role in regulating immune cell responses to acute cardiac injury is unknown. Methods Wild-type (WT) mice were irradiated followed by isoform-specific βARKO or WT bone-marrow transplantation (BMT) and after full reconstitution underwent myocardial infarction (MI) surgery. Survival was monitored over time and alterations in immune cell infiltration following MI were examined using immunohistochemistry. Alterations in splenic function were identified through the investigation of altered adhesion receptor expression. Results β2ARKO BMT mice displayed 100% mortality resulting from cardiac rupture within 12 days post-MI compared to ~20% mortality in WT BMT mice. β2ARKO BMT mice displayed severely reduced post-MI cardiac infiltration of leukocytes with reciprocally enhanced splenic retention of the same immune cell populations. Splenic retention of the leukocytes was associated with an increase in VCAM-1 expression, which was itself regulated via β-arrestin-dependent β2AR signaling. Further, VCAM-1 expression in both mouse and human macrophages was sensitive to β2AR activity, and spleens from human tissue donors treated with β-blocker showed enhanced VCAM1 expression. The impairments in splenic retention and cardiac infiltration of leukocytes following MI were restored to WT levels via lentiviral-mediated re-expression of β2AR in β2ARKO BM prior to transplantation, which also resulted in post-MI survival rates comparable to WT BMT mice. Conclusions Immune cell-expressed β2AR plays an essential role in regulating the early inflammatory repair response to acute myocardial injury by facilitating cardiac leukocyte infiltration. β-adrenergic receptor acute myocardial infarction leukocyte inflammation immune system Inflammation is critical for initiating reparative processes after ischemic injury1. Following myocardial infarction (MI) an intense inflammatory response is initiated leading to recruitment of pro-inflammatory leukocytes including monocytes, neutrophils and mast cells1-6. Secreted factors from these pro-inflammatory cell populations recruit and activate reparative cell populations to promote extracellular matrix (ECM) deposition and vascularization7, 8. This rapid inflammatory response is necessary for healing and preserving the structure of the left ventricle (LV) post-MI, as dysregulation of this process results in increased cardiomyocyte death and degradation of the ECM9. Sympathetic nervous system (SNS) regulation of immune responses is well-established10 and β-adrenergic receptor (βAR) expression has been reported on virtually all immune cell-types. All three βAR subtypes are expressed on various hematopoietic cell-derived immune cell populations with βAR subtype expression varying widely between populations and immune cell activation state11. Although the role of β1AR in the immune system is not well-established, β1AR expression has been shown to be limited primarily to cells of the innate immune system where it regulates inflammatory mediator production12, 13. β2AR is the most highly and widely expressed βAR isoform10, 14, with a similar level of immune cell expression in rodents and humans10, 14, and is known to regulate a number of functions including hematopoiesis, lymphocyte homing and immune cell maturation10. However, the focus of many of these studies involved the effect of β2AR on adaptive immune responses, while its involvement in mediating early, innate immune responses and initiation of inflammation remains unclear10, 15, 16. β3AR has been shown to be important in early stages of hematopoiesis for mediating immune cell mobilization and egress from the bone marrow17-19. While βAR subtype expression and function varies in the immune system, the role of Immune cell-expressed βAR in the acute inflammatory response post-MI has yet to be elucidated. In our current study, the impact of immune cell-specific βAR expression on cardiac inflammation and remodeling post-MI was investigated through the use of chimeric mice that lack specific βAR isoforms on cells of hematopoietic origin. We demonstrate that β2AR are essential in initiating early immune responses following acute cardiac injury and targeting immune cell-expressed β2AR may provide a novel therapeutic strategy for preventing adverse effects following MI. Methods Rationale and Study Design The purpose of this study was to investigate the impact of βAR in regulating immune responses following MI. To differentiate the effects of immune cell-expressed βAR from cardiac-expressed βAR, we generated chimeric mice using a BMT approach in which WT recipient mice received WT control or βAR subtype-specific KO BM to produce immune cell- and βAR isoform-specific KO mice. These mice were subjected to sham or MI surgery and survival outcome and immune responses were examined along with the mechanisms of observed changes. Bone Marrow Transplant WT C57BL/6 recipient mice (male, 8 wk) were lethally irradiated with 950 rads using x-ray irradiation to remove endogenous BM cells. Donor BM isolated from the femurs of β1ARKO, β2ARKO, β3ARKO or WT C57BL/6 mice was introduced by retro-orbital injection (1×107 cells) within 24 h of irradiation. BM was allowed to reconstitute for 1 month prior to MI surgery. Reconstitution was confirmed at the conclusion of the study for each mouse using RT-qPCR analysis for β1AR, β2AR and β3AR expression on recipient BM. All animal procedures were performed in accordance with the Institutional Animal Care and Use Committee at Temple University and in accordance to the NIH Guidelines on the Use of Laboratory Animals. Coronary Artery Occlusion Surgery Myocardial infarction was induced as previously described20. Mice were anesthetized with 2% isoflurane inhalation. A small skin incision was made and the pectoral muscles were retracted to expose the fourth intercostal space. A small hole was made and the heart popped out. The left coronary artery was sutured ~3 mm from its origin and the heart was placed back into the intrathoracic space followed by closure of muscle and skin. Animals received a single dose (0.3 mg/kg) of buprenorphine immediately following surgery. Splenectomy Surgery Mice were anesthetized as above and a small incision was made in the left subcostal abdominal wall. Sutures were placed around the splenic vasculature and the spleen was removed. The incision was closed in two layers, peritoneum and skin, using suture. Animals received a single dose (0.3 mg/kg) of buprenorphine immediately following surgery. Echocardiography Cardiac function was assessed via transthoracic two-dimensional echocardiography performed at baseline and at weekly intervals post MI using a 12-mHz probe on mice anesthetized with isoflurane (1.5%). M-mode echocardiography was performed in the parasternal short-axis view to assess several cardiac parameters including left ventricular (LV) end-diastolic dimension, wall thickness, LV fractional shortening and ejection fraction. Percent fractional shortening was calculated using the equation ((LVID;d-LVID;s)/LVID;s)*100%. Percent ejection fraction was calculated using the equation ((LV vol;d-LV vol;s)/LV vol;d)*100%. Lentivirus Infection of Bone Marrow BM isolated from the femurs of mice was transduced with lentiviral vectors for 3XFlag-β2AR-RFP or GFP using and MOI of 100. Transductions were performed in MEM+10%FBS in the presence of 5 μg/mL Polybrene (Sigma-Aldrich). For in vitro experiments, media was changed 24 h following infection to complete media (MEM+10% FBS) and incubated an additional 24 h prior experiments. For generation of bone marrow derived macrophages, isolated BM was cultured in 10% L929 conditioned MEM+10% FBS for 1 wk prior to lentiviral infection with GFP control, WT β2AR, β2ARTYY or β2ARGRK- constructs21, 22. For in vivo experiments, BM was rinsed 1 h following infection and transplanted into irradiated mice via retro-orbital injection. BM was allowed to reconstitute for 1 month. Human Macrophage Cell Culture THP-1 cells (American Type Culture Collection, Manassas, VA), a human monocytic cell line, were cultured in modified RPMI-1640 media containing 2 mM L-glutamine, 1.5 g/L sodium bicarbonate, 4.5 g/L glucose and 10 mM HEPES and 1 mM sodium pyruvate supplemented with 10% fetal bovine serum under standard cell culture growth conditions (37°C/5% CO2/95% humidified air). THP-1 cells were differentiated into macrophages using 200 nM PMA 48h prior to all experiments. Cells were washed with complete media and treated 24h with vehicle (PBS), 0.1 μM salbutamol or 0.1 μM ICI-118,551. Human Spleen Samples Spleen samples from deceased human tissue donors that had been chronically administered metoprolol, or age- and sex-matched subjects not treated with metoprolol, were procured by the National Disease Research Interchange (NDRI) with support from NIH grant 2 U42 OD011158. Control subjects: n=5, 74.6±15.5 y.o. (mean ± standard deviation), 1 male, 4 females; Metoprolol subjects: n=6, 77.5±8.4 y.o., 1 male, 5 females. Reverse Transcription Quantitative PCR cDNA was synthesized from the total RNA of BM and spleen using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems). Reverse transcription quantitative PCR (RT-qPCR) was performed with SYBR® Select Master Mix (Applied Biosystems) in triplicate for each sample using primers listed in Supplemental Table 1 at an annealing temperature of 60.1°C. RT-qPCR data was analyzed using Applied Biosystems Comparative CT Method (ΔΔCT), using GAPDH, TPT-1 and 18s rRNA to normalize expression of genes of interest and calculate relative quantitation (RQ) and RQmin/max values for each. Immunoblot BM and spleen samples were homogenized in RIPA buffer containing 1X HALT protease inhibitor cocktail (78437; Thermo Scientific; Rockford, IL) and phosphatase inhibitor cocktail set IV (524628; Calbiochem, USA). Equal amounts of lysates were resolved by SDS-polyacrylamide gel electrophoresis (10% gels) and transferred to Immobilon-PSQ polyvinylidene fluoride 0.2μm pore size membranes (Millipore; Billerica, MA). Odyssey Blocking Buffer (LI-COR Biosciences; Lincoln, NE) was used to prevent non-specific binding. Immunoblotting was performed overnight at 4°C with diluted antibodies against Flag M2 (1:10,000; Sigma-Aldrich; St. Louis, MO), GFP (1:1000; Cell Signaling; Danvers, MA), VCAM-1 (1:1000; Santa Cruz Biotechnologies; Santa Cruz, CA), β-tubulin (1:1000; Cell Signaling), β-actin (1:1000; Santa Cruz) or GAPDH (1:1000; Cell Signaling). After washing with TBS-T, membranes were incubated at room temperature for 60 min with the appropriate diluted secondary antibody (IRDye680 Donkey anti-rabbit IgG (H + L) at 1:20,000; IRDye800CW Goat anti-mouse IgG (H + L) at 1:15,000; LI-COR Biosciences; IRDye680 Donkey anti-goat IgG (H+L) at 1:20,000). Bound antibody was detected using the LI-COR Biosciences Odyssey System (LI-COR Biosciences). Intensities were normalized to corresponding GAPDH, β-tubulin and β-actin intensities. Histological Analysis Excised hearts were fixed in 4% paraformaldehyde, paraffin embedded and sectioned at 5 μm thickness. Deparaffinized sections were stained for hematoxylin-eosin (H&E; Sigma-Aldrich). NIS Elements software was used to measure infarct size and visualize cell infiltration and morphology. Immunohistochemistry was performed on deparaffinized sections to examine infiltration of various immune cell types. Antigens were retrieved using a citrate-based antigen unmasking solution (Vector Laboratories; Burlingame, CA). Hearts were blocked (10% FBS/PBS) and a 0.3% H2O2 solution was used to block endogenous peroxide activity in sections used for immunohistochemical staining. Hearts were incubated with antibodies against CD3 (1:100; Abcam), CD68 (1:100; Abcam), major basic protein (MBP; Obtained from Nancy and Jamie Lee Laboratories; Mayo Clinic, Scottsdale, AZ), mast cell tryptase (1:100; Abcam), myeloperoxidase (MPO; 1:100, Santa Cruz, Dallas, TX) or VCAM-1 (1:100; Santa Cruz). Washed slides were incubated with the appropriate secondary antibodies, anti-mouse-HRP (1:1000; GE Healthcare; Piscataway, NJ), anti-goat-HRP (1:1000; Santa Cruz), anti-goat Alexa Fluor® 647 (1:1000; Invitrogen; Grand Island, NY) and anti-rabbit Alexa Fluor® 647 (1:1000; Invitrogen), followed by staining with DAPI for immunofluorescence or hematoxylin for immunohistochemical staining. Immunofluorescent stained hearts were mounted using Prolong® Gold Antifade Reagent (Invitrogen). Immunohistochemical stained hearts were developed using a DAB Substrate Kit (Vector Laboratories) and mounted using Permount™ Mounting Media (Thermo Scientific). Staining was visualized on a Nikon Eclipse microscope at 20X magnification and NIS Elements software was used for recording images and image analysis. Images were quantified as the number of positive cells per area. Flow Cytometry Flow cytometry analysis of immune cell populations were performed on cells isolated from blood and BM. Immune cells were separated using an antibody against CD45-FITC (BD Biosciences; San Jose, CA) and sorted on an LSRII flow cytometer for size and granularity by Forward Scatter (FSC) and Side Scatter (SSC). Analysis was performed using Flowjo software. Statistical Analysis Data presented is expressed as mean ± standard deviation (SD) for continuous variables and the count and/or percentage for categorical variables. Comparisons of a continuous variable between different treatment groups were performed using the nonparametric Kruskal-Wallis test for three or more groups and the exact Wilcoxon rank-sum test for two groups due to the small group sizes to guard against possibly non-normally distributed data. Comparisons of a survival endpoint between treatment groups were performed using the log-rank test. When data were collected over time on the same set of animals such as the fractional shortening in Fig 1B, they were analyzed using a mixed-effects model in order to take into account the correlation among repeated measures as well as the potential non-constant variability over time across different groups. Multiple pairwise comparison adjustments were made with Bonferroni or Dunnett correction as appropriate. P-value < 0.05 was considered statistically significant. P-values and n (group size) values are reported in the figure legends. All statistical analyses were performed using SAS version 9.3 software (SAS Institute Inc., Cary, NC). Results Lack of immune cell-expressed β2AR increases mortality following acute myocardial injury To examine the contribution of immune cell-specific βAR subtype expression on survival and cardiac function post-MI, chimeric animals were generated via WT, β1ARKO, β2ARKO or β3ARKO bone marrow transplant (BMT) into irradiated recipient WT mice. β1AR, β2AR and β3AR expression was examined by RT-qPCR on reconstituted BM from all animals (Figure 1A), confirming that β1AR, β2AR and β3AR were knocked out in their respective BM with no differences in the other two βAR subtypes. Following reconstitution, the BMT mice underwent sham or MI surgery and cardiac function was monitored via echocardiography (Supplemental Figure 1A, Supplemental Table 2). LV contractility, wall thickness and cardiac dimensions were significantly altered in each MI group as shown by decreased % fractional shortening (Figure 1B), increased LV hypertrophy and increased LV dilation relative to sham animals, although no differences were evident between WT, β1ARKO, β2ARKO and β3ARKO BMT groups. However, mortality rates among the BMT mice differed significantly post-MI. All sham BMT mice displayed 100% survival, and WT BMT mice exhibited ~20% mortality by two weeks post- MI (Figure 1C), consistent with prior studies in non-BMT mice20. β1ARKO and β3ARKO BMT animals had a small, but non-significant, increase in mortality following MI when compared to WT BMT animals. Strikingly, β2AR BMT mice displayed 100% mortality following MI due to cardiac rupture with death observed 4-12 days post-MI. Although infarct size one day post-MI was not different between WT and β2ARKO chimeric mice (Supplemental Figure 1B and 1C), H&E staining revealed wall thinning and weakening in β2ARKO BMT mice 4d post-MI compared to WT BMT hearts (Figure 1D), suggesting an impairment in early repair mechanisms. Lack of immune cell β2AR expression impairs leukocyte infiltration following acute myocardial injury A number of immune cell populations including monocytes/macrophages, neutrophils, mast cells and T cells are known to be important for the initiation of wound healing and cardiac remodeling post-MI. Due to the high mortality via cardiac rupture observed in β2AR BMT mice following MI, which could reflect decreased immune cell-initiated repair, we assessed whether WT and β2ARKO BMT mice display differences in MI-induced cardiac immune cell population infiltration. Immunostaining was performed to identify cells of monocyte/macrophage lineage (CD68), mast cells (tryptase), neutrophils (MPO), eosinophils (MBP) and T-cells (CD3) in sham BMT hearts and in the remote, border and infarct zones of BMT hearts following MI (Figure 2, Supplemental Figures 2 and 3). Compared to WT BMT hearts, β2ARKO BMT hearts had significantly less monocyte/macrophage, mast cell and neutrophil infiltration into both the border and infarct zones (Figure 2A). Quantification of the staining demonstrated that decreased monocyte/macrophage (Figure 2B, 2C), mast cell (Figure 2D, 2E) and neutrophil (Figure 2F, 2G) recruitment to β2ARKO BMT mouse hearts was maintained over time post-MI as compared to WT BMT mice. Not all immune cell populations were affected, however, as eosinophil and T-cell infiltration into the hearts of both WT BMT and β2ARKO BMT mouse hearts were not different (Supplemental Figure 3). Flow cytometric comparative analysis of immune cell populations in the BM or blood of WT and β2ARKO BMT animals showed there was no difference in granulocyte, monocyte or lymphocyte populations (Supplemental Figure 4). Therefore, despite having similar levels of hematopoietic-derived cells as compared to WT BMT mice, those with immune cell-specific deletion of β2AR have impaired leukocyte recruitment to the heart following acute injury. Mice lacking immune cell-expressed β2AR have increased splenic retention of leukocyte populations Since overall immune populations are similar between WT and β2ARKO BMT mice, but decreased leukocyte populations are observed in β2ARKO BMT hearts post-MI, we aimed to determine whether splenic retention of leukocytes could play role in this phenotype. Sham β2ARKO BMT mice had an increased spleen size compared to their WT BMT counterparts (Figure 3A), which was maintained post-MI (Figure 3B). Leukocyte levels in spleen sections from WT or β2ARKO BMT mice were examined to determine if an increase in leukocytes within the β2ARKO BMT spleens could account for the splenomegaly observed between the two groups in sham animals (Supplemental Figure 5) and 4d following MI (Figure 3C). Increased levels of monocyte/macrophages, mast cells and neutrophils (Figure 3D) were observed in β2ARKO BMT spleens compared to WT BMT mice, suggesting that β2ARKO leukocytes have an impaired ability to mobilize from the spleen to the heart following injury. Splenic and macrophage VCAM1 expression is sensitive to β2AR activity and expression in mice and humans VCAM-1 expression on splenic macrophages has recently been identified as a hematopoietic stem cell retention factor important for splenic myelopoiesis23. To determine if VCAM-1 levels were increased in the spleens of β2ARKO BMT mice, leading to the retention of myeloid populations, its expression was assessed in the spleens of WT or β2ARKO BMT mice. Immunostaining indicated increased splenic VCAM-1 expression with localization in the red pulp, where macrophages reside (Figure 4A and B). Protein levels of VCAM-1 were confirmed to be elevated in β2ARKO chimeric mouse spleens when compared to WT BMT mice via immunoblotting analysis (Figure 4C and D). Further, transcript expression of VCAM-1 was increased in β2ARKO BMT spleens both basally and 4d post-MI (Figure 4E). To determine if β2AR stimulation alters VCAM-1 expression at a cellular level, WT bone marrow-derived macrophages (BMDM) were treated with the β2AR-selective agonist salbutamol, which decreased VCAM-1 expression (Figure 4F). Strikingly, salbutamol also decreased VCAM-1 expression in a human macrophage cell line (Figure 4F), confirming that β2AR-mediated alterations in VCAM-1 are translatable between species. Since VCAM-1 was decreased by β2AR stimulation, we next tested whether pharmacological inhibition of β2AR could reciprocally increase VCAM-1 expression. Indeed, treatment of human macrophages with the β2AR-selective antagonist, ICI-118,551 (Figure 4G) increased VCAM-1 expression. Further, VCAM-1 expression was significantly increased in the spleens of human subjects treated with the β-blocker metoprolol versus age- and sex-matched subjects that had not taken a β-blocker (Figure 4H), demonstrating the clinical relevance of our findings. Proximal β2AR signaling through either G protein- or β-arrestin-dependent pathways have been shown to exert distinct cellular effects21, 22. Thus, to determine the proximal mechanism through which β2AR controls VCAM-1 expression, lentiviral constructs were generated containing either β2ARTYY, which is unable to couple to Gαs22, or β2ARGRK-, which cannot be phosphorylated by GRK21 thereby preventing the recruitment of β-arrestins (βARR). Using BMDM from WT or β2ARKO mice, VCAM-1 transcript expression was shown to be increased in β2ARKO macrophages (Figure 5A). Lentivirus-mediated restoration of β2AR expression in β2ARKO macrophages (Supplemental Figure 5E) decreased VCAM-1 expression to that in WT macrophages (Figure 5A), while a GFP control lentivirus had no effect on VCAM-1 expression. Mechanistically, β2ARKO BMDM transduced with β2ARGRK- had elevated expression of VCAM-1 where as β2ARTYY had decreased VCAM-1 similar to WT levels (Figure 5A), indicating that GRK-dependent β2AR signaling is required for regulation of VCAM-1 expression in macrophages. In support of this observation, βARR2KO mice had splenomegaly similar to β2ARKO mice (Figure 5B) with retention of monocytes/macrophages, mast cells and neutrophils (Figure 5C and D). Interestingly, βARR1KO mice had normal splenic size and leukocyte levels, indicating that β2AR regulates VCAM-1 expression selectively via βARR2 signaling. Splenectomized WT and β2ARKO BMT mice have similar levels of leukocyte infiltration following acute myocardial injury To confirm whether splenic retention of β2ARKO leukocytes is primarily responsible for their decreased infiltration into the heart following MI, we examined leukocyte recruitment to the heart in splenectomized WT and β2ARKO BMT animals receiving sham or MI surgery (Figure 6A, Supplemental Figure 6). Splenectomy in WT BMT animals decreased MI-induced infiltration of monocytes/macrophages (Figure 6B) and neutrophils (Figure 6D) into the border zone by about 50%, with less impact on mast cells (Figure 6C). Conversely, splenectomy of β2ARKO BMT animals increased cardiac infiltration of monocytes/macrophages, neutrophils and mast cells to levels not different from those observed in splenectomized WT BMT mice. Altogether, these results confirm that β2AR-deficient leukocytes accumulate in the spleen where they remain after MI, but do have the capacity to infiltrate the heart following injury in the absence of the spleen, similar to spleen-independent leukocyte infiltration levels attained in WT BMT mice. Restoration of β2AR expression reverses leukocyte dysfunction and restores survival rates following MI To determine if restoration of β2AR expression in β2ARKO BM could revert the β2AR BMT phenotype toward the WT BMT phenotype post-MI, β2ARKO BM was transduced with the WT β2AR lentivirus construct, or GFP control lentivirus, prior to transplantation. Immunoblotting was used to confirm protein expression of GFP in control and Flag-tagged β2AR expression for lentivirus-transduced reconstituted BM (Figure 7A) and β2AR expression in β2ARKO BM following reconstitution was approximately 95% of endogenous levels (Figure 7B). Similar to β2ARKO BMT mice, mice receiving β2ARKO BM transduced with GFP control lentivirus displayed 100% mortality post-MI with all mice dying between day 4 and 14 from cardiac rupture (Figure 7C), however restoration of β2AR expression in β2ARKO BM increased survival following MI to near WT BMT levels (Figure 7D). Reconstitution with β2AR-infected β2ARKO BM also reduced both spleen size (Figure 7E) and VCAM-1 expression (Figure 7F, 7G) compared with GFP-transduced β2ARKO BMT mice. Accordingly, restoration of β2AR in β2ARKO BM also reduced levels of the leukocyte populations in the spleen (Figure 8A and 8B) to those not different from WT BMT mice (Supplemental Table 3). Conversely, immunohistochemistry for leukocyte infiltration in sham (Supplemental Figure 7) and injured myocardium of β2AR-rescued β2ARKO BMT mice revealed the reciprocal results. Thus, leukocyte infiltration was increased in the border (Figure 8C) and infarct (Supplemental Figure 8) zones of the heart in β2ARKO BMT mice transduced with β2AR versus GFP lentivirus, including monocytes/macrophages, mast cells and neutrophils (Figure 8D), which were restored to WT BMT levels (Supplemental Table 3). Discussion Inflammatory responses are critical for wound-healing following MI1. All three βAR isoforms have been shown to mediate a number of effects in the immune system, including hematopoiesis, lymphocyte homing and cytokine/chemokine production, however little is known about how they regulate immune cell responses following acute cardiac injury10, 14. To investigate the immune cell-specific impact of βARs on cardiac survival and remodeling following MI, we generated chimeric mice lacking β1AR, β2AR or β3AR expression on cells of hematopoietic origin. The most striking outcome was observed with β2ARKO BMT animals, which displayed 100% mortality due to cardiac rupture, in contrast to their WT counterparts that had ~20% death. β2AR chimeric mice had decreased infiltration of leukocyte populations compared to their WT counterparts demonstrating impaired innate immune responses. Recent findings have shown the importance of pro-inflammatory monocytes in initiating early immune cell-dependent reparative responses following MI2. Thus, it is likely that the inability of leukocyte populations to traffic to the heart acutely following MI in β2ARKO chimeric mice impairs early repair processes, contributing to scar instability, cardiac rupture and death. Of great importance, the β2ARKO BMT mice had decreased leukocyte infiltration into the heart following MI with a reciprocal increase in spleen size and leukocyte retention, suggesting an impairment in immune cell egress from the spleen to the heart following acute cardiac injury. As such, splenectomy of the β2ARKO BMT mice restored cardiac leukocyte infiltration responses to those of splenectomized WT BMT mice. Recently, the spleen has been shown to be an important monocyte reservoir, holding active monocytes for release upon inflammatory injury6, 24, and has been demonstrated to be of particular importance following MI and during heart failure where there is increased antigen processing and adaptive immune system activation25. These processes are regulated through a variety of signals6. One molecular mechanism of leukocyte egress from the spleen through macrophage expression of VCAM-1 has recently been identified23. Our results demonstrate an increase in VCAM-1 in the macrophage-containing red pulp region of spleens from β2ARKO BMT animals, resulting in the retention of leukocyte populations in the spleens of these animals. VCAM-1 was also increased with a β2AR antagonist in human macrophages, with a reciprocal decrease in expression following β2AR stimulation. Interestingly, macrophage VCAM-1 expression appears to be regulated within this context in a β2AR/βARR2-dependent manner. A common limitation of studies in mice is that they do not always translate toward human pathophysiology. However, βAR activation has been implicated in reducing spleen size and release of certain immune cell populations in a number of different species including murine and swine models and humans26-30. These changes were independent of alterations in blood flow, although the mechanism was never identified. Furthermore, β-blocker administration was shown to prevent the splenic release of immune populations, similar to our current study using β2ARKO chimeric animals, and this response was amplified when combined with an inflammatory stimulus31. Mechanistically, our study demonstrates increased levels of leukocyte populations in the spleens of β2ARKO BMT mice both before and after MI, effects that are clearly independent of vascular or splenic β2AR expression. Importantly, inhibition of leukocyte egress from the spleen and decreased infiltration of these populations into the heart can be reversed by restoring β2AR expression in the bone marrow prior to transplantation using a lentiviral construct, confirming the specificity of the response and demonstrating the ability to modulate hematopoietic cell receptor expression using gene therapy approaches. In addition, our data in human macrophages are consistent with those in mouse macrophages and we also observed increased VCAM-1 expression in the spleens of human donors that had taken the β-blocker metoprolol. While metoprolol has preference for β1AR, it loses its selectivity at the higher doses often used clinically32 that would also antagonize β2AR. This could account for the retention of immune cell populations observed in other studies and confirmed in our chimeric mouse model, providing further clinical relevance. Interestingly, increases in circulating immune cells with epinephrine or isoproterenol is greatly diminished in splenectomized patients33, 34 and multiple long-term studies examining the effects of splenectomy in humans have demonstrated an increased incidence in MI and HF with worsened prognosis following such events35, 36. While many of the benefits of β-blockers are thought to be mediated through their actions on β1AR in cardiomyocytes37, immune cell-expressed βAR and β-blocker therapy have been suggested to play roles in the regulation of immune responses during HF38-42. Our findings suggest administration of β-blockers with selectivity toward β2AR around the time of MI could diminish leukocyte egress from the spleen and subsequent cardiac immune cell-dependent remodeling. The impact of such a process on overall cardiac remodeling would likely depend on the severity of β2AR inhibition, where a short-term decrease in activity may simply dampen the inflammatory response, but not ultimately prevent it, whereas chronic inhibition of leukocyte-expressed β2AR could negatively impact post-MI repair processes. Interestingly, it has recently been suggested that peri-operative use of β-blockers may actually increase cardiovascular events, including MI43 and, according to the American Heart Association/American College of Cardiology, continues to have an uncertain mortality risk44. A link between β2AR inhibition in leukocytes and these clinical observations has not been demonstrated, but warrants further investigation. Converse to inhibition, short-term β2AR agonist administration during the inflammatory phase following MI in combination with chronic β1AR antagonist administration may provide an improved therapeutic strategy to prevent detrimental remodeling and preserve cardiac function following cardiac injury. Several studies investigating the use of β2AR agonists in the treatment of heart failure have found beneficial effects, which was attributed to the promotion of cardiomyocyte survival, however the long term benefits of βAR blockade in HF has contraindicated the use of β2AR agonists45-50. In many of these studies β2AR agonist administration commenced at later time points, missing the acute inflammatory phase. Regardless, in animal models β2AR agonists have been shown to improve cardiac remodeling following MI or ischemia/reperfusion to a greater extent than that achieved by β1AR antagonists46, 49, 50, while cardiac function and survival were further improved with a combined β1AR blocker/β2AR agonist strategy46-48. Remarkably, a single dose of the β2AR agonist clenbuterol prior to ischemic insult was shown to decrease the resulting cardiac injury45. However, these studies did not assess the contribution of the early immune cell responses to their outcomes. In summary, using a chimeric mouse approach, we identified a critical role for hematopoietic cell-expressed β2AR in the regulation of acute cardiac inflammation and remodeling following MI. β2AR-deficient immune cells displayed impaired recruitment to the injured myocardium following MI, with reciprocal leukocyte retention within the spleen that was maintained following MI. Lentiviral-mediated re-expression of β2AR in β2ARKO BM prior to transplantation restored BM migration, splenic retention levels of leukocyte populations and leukocyte infiltration into the heart following injury. Altogether, our results highlight an immunomodulatory role for β2AR that could be targeted to promote early leukocyte-dependent reparative processes following MI, with negligible or even beneficial effects on cardiomyocytes, while avoiding issues inherent to the promotion of prolonged inflammatory events. Supplementary Material Supplemental Data Sources of Funding This work was supported by NIH grants HL105414 (to D.G.T.), HL091799 and HL085503 (to W.J.K.), HL098481 (to J.W.C.), and an AHA postdoctoral fellowship (to L.A.G.). Figure 1 Effects of hematopoietically expressed βAR subtypes on cardiac survival and function following MI. (A) C57BL/6 mice receiving WT, β1ARKO, β2ARKO or β3ARKO BMT were subjected to sham or MI surgery. Expression of β1AR, β2AR and β3AR was assessed by RT-qPCR on reconstituted WT, β1ARKO, β2ARKO and β3ARKO BM and presented as RQ+RQmax. n=6 for all groups, Exact Wilcoxon rank-sum tests with multiple comparison adjustment (3 comparisons), † p < 0.01 vs WT. (B) Left ventricular fractional shortening (FS) was measured at the short axis from M mode using Visual Sonic Analysis software. Mixed-effects modeling for repeated measures data with multiple comparison adjustments was performed indicating no significant differences compared to WT BMT. (C) WT (n=9 for sham, n=11 for MI), β1ARKO (n=7 for sham, n=11 for MI), β2ARKO (n=10 for sham, n=14 for MI) or β3ARKO (n=7 for sham, n=18 for MI) BMT mice were monitored daily for survival. Log-rank tests with multiple comparison adjustment (3 comparisons), ‡ p < 0.001 vs WT BMT MI. All sham groups had 100% survival following surgery. (D) H&E staining for sham and 4d post-MI hearts from WT and β2ARKO BMT mice. Figure 2 Effect of hematopoietic β2AR expression on immune cell infiltration following MI. A. Representative CD68, tryptase, and MPO staining of the border or infarct zones of hearts following MI surgery in WT BMT or β2ARKO BMT mice. Quantification of CD68 (B, C), tryptase (D, E) and MPO (F, G) staining in the border and infarct zones of WT and β2ARKO BMT mouse hearts. n=4 for WT BMT sham, n=5 for β2ARKO BMT sham, n=3 for WT BMT 6 h, n=3 for β2ARKO BMT 6 h, n=4 for WT BMT 1d, n=6 for β2ARKO BMT 1d, n=5 for WT BMT 4d, n=5 for β2ARKO BMT 4d, n=6 for WT BMT 7d, n=6 for β2ARKO BMT 7d. Exact Wilcoxon rank-sum tests, * p < 0.05, † p < 0.01 vs WT BMT. Figure 3 β2ARKO mice have splenomegaly and retention of leukocyte populations. (A) Representative images of spleens from WT and β2ARKO BMT animals. (B) Gravimetric analysis of spleen weight to body weight (SW/BW) of spleens from WT and β2ARKO BMT sham and MI animals. n=6 for WT BMT sham, n=8 for β2ARKO BMT sham, n=4 for WT BMT 1d, n=4 for β2ARKO BMT 1d, n=10 for WT BMT 4d, n=7 for β2ARKO BMT 4d, n=7 for WT BMT 7d, n=5 for β2ARKO BMT 7d. Exact Wilcoxon rank-sum tests, * p < 0.05, ‡ p < 0.001 vs WT BMT. (C) Representative CD68, tryptase and MPO staining from 4d post-MI spleens of WT or β2ARKO BMT mice. (D) Quantification of CD68, tryptase and MPO staining from WT and β2ARKO BMT spleens 4d following MI surgery. n=10 for WT BMT, n=8 for β2ARKO BMT, Exact Wilcoxon rank-sum tests, * p < 0.05, ‡ p < 0.001 vs WT BMT. Figure 4 VCAM-1 is increased in β2ARKO BMT spleens. (A) Immunohistochemistry for VCAM-1 (white) showing levels and localization of VCAM-1 expression in WT and β2ARKO BMT spleens. (B) Quantification of the intensity of VCAM-1 staining. n=5 for WT BMT, n=5 for β2ARKO BMT, Exact Wilcoxon rank-sum test, * p < 0.05 vs WT BMT. (C) Representative immunoblot showing VCAM-1 expression in WT and β2ARKO BMT spleens. Arrows indicate the three isoforms of VCAM-1. Β-tubulin, β-actin and GAPDH are shown as loading controls. (D) Quantification of VCAM-1 immunoblot expression from. n=12 for WT BMT, n=12 for β2ARKO BMT, Exact Wilcoxon rank-sum test, ‡ p < 0.001. (E) RT-qPCR was used to measure VCAM-1 expression in WT or β2ARKO BMT spleens and presented as RQ+RQmax. n=8 for WT BMT sham, n=6 for WT BMT MI, n=6 for β2ARKO BMT sham, n=8 for β2ARKO BMT MI, Exact Wilcoxon rank-sum tests, † p < 0.01, ‡ p < 0.001 vs WT BMT. (F) RT-qPCR was used to measure VCAM-1 expression in mouse (BMDM) or human (THP-1 derived) macrophages and presented as RQ+RQmax. n=7 for mouse vehicle, n=10 for mouse salbutamol, n=9 for human vehicle, n=10 for human salbutamol, Exact Wilcoxon rank-sum tests, † p < 0.01, ‡ p < 0.001 vs Veh. (G) VCAM-1 expression in human macrophages treated with vehicle or ICI-118,551 was quantified by RT-qPCR and presented as RQ+RQmax, Exact Wilcoxon rank-sum test, † p < 0.01 vs Veh. (H) RT-qPCR was used to measure VCAM-1 expression in human spleens from control or metoprolol-treated patients and presented as RQ+RQmax. n=5 for control, n=6 for metoprolol, Exact Wilcoxon rank-sum test, * p < 0.05 vs control. Figure 5 β2AR regulates VCAM-1 through β-arrestin dependent mechanisms. (A) RT-qPCR was used to measure VCAM-1 expression in BMDM from WT or β2ARKO mice and β2ARKO BMDM transduced with GFP, β2AR, β2ARTYY or β2ARGRK- lentivirus and presented as RQ+RQmax. n=9 for WT, n=6 for β2ARKO, n=6 for β2ARKO+β2AR, n=6 for β2ARKO+GFP, n=6 for β2ARKO+β2ARTYY, n=6 for β2ARKO+β2ARGRK-, Exact Wilcoxon rank-sum tests with multiple comparison adjustment (5 comparisons), * p < 0.05, † p < 0.01 vs WT. (B) Gravimetric analysis of spleen weight to body weight (SW/BW) of spleens from WT, β2ARKO, βARR1KO and βARR2KO animals. n=7 for WT, n=6 for β2ARKO, n=6 for βARR1KO, n=6 for βARR2KO, Exact Wilcoxon rank-sum tests with multiple comparison adjustment (3 comparisons), † p < 0.01 vs WT. (C) Representative CD68, tryptase, and MPO staining of spleens from βARR1KO and βARR2KO mice. (D) Quantification of CD68, tryptase and MPO staining from βARR1KO and βARR2KO spleens. n=6 for βARR1KO, n=4 for βARR2KO, Exact Wilcoxon rank-sum tests, † p < 0.01 vs βARR1KO. Figure 6 Splenectomy restores β2ARKO leukocyte infiltration into the heart following MI. (A) Representative CD68, tryptase, and MPO staining of the border or infarct zones of hearts 4d following MI surgery in WT or β2ARKO BMT that received sham and splenectomy surgery. Quantification of CD68 (B), tryptase (C) and MPO (D) staining in the border and infarct zones of 4d post-MI hearts from sham and splenectomy WT and β2ARKO BMT mice. n=6 for WT BMT Sham/MI, n=5 for WT BMT Splen/MI, n=6 for β2ARKO BMT Sham/MI, n=7 for β2ARKO BMT Splen/MI, Exact Wilcoxon rank-sum tests with multiple comparison adjustment (6 comparisons), * p < 0.05, † p < 0.01, ns = not significant. Figure 7 Restoration of β2AR expression in β2ARKO BM restores survival following MI. (A) Representative immunoblot showing protein expression of GFP and Flag in reconstituted BM from β2ARKO, β2ARKO+GFP and β2ARKO+β2AR BMT mice. β-tubulin, β-actin and GAPDH are shown as loading controls. (B) β2AR expression was measured by RT-qPCR in reconstituted BM from WT and β2ARKO mice and reconstituted BM from mice that had GFP or β2AR transduced into β2ARKO BM by lentivirus prior to transplantation. Values are presented as RQ+RQmax expressed relative to WT BMT. n=10 for WT, n=10 for β2ARKO BMT, n=9 for β2ARKO+GFP BMT, n=11 for β2ARKO+β2AR BMT. Exact Wilcoxon rank-sum tests with multiple comparison adjustment (3 comparisons), ‡ p < 0.001. (C) β2ARKO+GFP and β2ARKO+β2AR BMT were subjected to sham or MI surgery and monitored daily for survival. All sham groups had 100% survival following surgery. n=6 for β2ARKO+GFP and n=7 β2ARKO+β2AR BMT sham, n=27 for β2ARKO+GFP MI, n=26 for β2ARKO+β2AR BMT MI. Log-Rank test, ‡ p < 0.001 vs β2ARKO+β2AR BMT. (D) % survival of β2ARKO+GFP and β2ARKO+β2AR BMT mice 1 week post-MI. Log-Rank tests with multiple comparison adjustment (3 comparisons), * p < 0.05 vs WT BMT MI. (E) Gravimetric analysis of spleen weight to body weight (SW/BW) of spleens from β2ARKO+GFP and β2ARKO+ β2AR BMT mice. n=7 for β2ARKO+GFP sham, n=6 for β2ARKO+ β2AR sham, 4d post n=8 for β2ARKO+GFP MI and n=10 for β2ARKO+ β2AR MI. Exact Wilcoxon rank-sum tests, † p < 0.01, ‡ p < 0.001 vs β2ARKO+GFP BMT. (F) Representative VCAM-1 staining for β2ARKO+GFP and β2ARKO+β2AR BMT spleens 4d post-MI. (G) Quantification of VCAM-1 intensity from immunohistochemistry of β2ARKO+GFP (n=6) and β2ARKO+β2AR BMT (n=7) spleens. Exact Wilcoxon rank-sum test, † p < 0.01vs β2ARKO+GFP BMT. Figure 8 β2AR re-expression on reconstituted β2ARKO BM reverses splenic retention of leukocytes. (A) Representative CD68, tryptase and MPO staining of spleens from β2ARKO+GFP and β2ARKO+β2AR BMT mice 4d post-MI. (B) Quantification of CD68, tryptase and MPO staining from the spleens from (A). n=8 β2ARKO+GFP BMT and n=8 β2ARKO+β2AR BMT, Exact Wilcoxon rank-sum tests, † p < 0.01, ‡ p < 0.001 vs β2ARKO+GFP BMT. (C) Representative CD68, tryptase and MPO staining of the border zone of hearts from β2ARKO+GFP and β2ARKO+β2AR BMT animals 4d following MI surgery. (D) Quantification of CD68, tryptase and MPO staining from the border zone of 4d post-MI hearts from β2ARKO+GFP (n=8) and β2ARKO+β2AR BMT (n=6) mice. Exact Wilcoxon rank-sum tests, ‡ p < 0.001 vs β2ARKO+GFP BMT. Clinical Perspective What is new? Using chimeric mice, we demonstrate that immune cell-specific β2-adrenergic receptor (β2AR) expression is essential to the repair process following myocardial infarction. In the absence of β2AR, vascular cell adhesion molecule 1 (VCAM-1) expression is increased in leukocytes, inducing their splenic retention following injury and leading to impaired scar formation followed by rupture and death. VCAM-1 expression is regulated dynamically by βAR ligands, including β-blockers, in both mouse and human tissues. Splenectomy partially restores β2AR-deficient leukocyte infiltration into the heart following injury, and gene therapy to rescue leukocyte β2AR expression completely restored all injury responses to that observed in normal mice. What are the clinical implications? βARs regulate cardiac function and remodeling following injury, classically through their effects in cardiomyocytes, and are targeted by β-blockers to help prevent detrimental myocardial remodeling. However, our findings indicate that inhibition/deletion of immune cell-expressed β2AR causes leukocyte dysfunction and altered immunomodulatory responses to acute injury. These results have important clinical implications since β-blockers are used frequently in patients around the time of myocardial infarction, as well as peri-operatively for non-cardiac surgeries with uncertain mortality risk. Thus, understanding the essential role for β2AR in mediating immune cell responses will inform strategies for β-blocker, or βAR agonist, administration following acute injury. Disclosures DGT and WJK have equity in Renovacor, Inc., which has neither funded this study nor has a relevant product related to this study. References 1 Frangogiannis NG The inflammatory response in myocardial injury, repair, and remodelling. Nature reviews. Cardiology 2014 11 255 265 24663091 2 Nahrendorf M Swirski FK Aikawa E Stangenberg L Wurdinger T Figueiredo JL Libby P Weissleder R Pittet MJ The healing myocardium sequentially mobilizes two monocyte subsets with divergent and complementary functions. The Journal of experimental medicine 2007 204 3037 3047 18025128 3 Yan X Anzai A Katsumata Y Matsuhashi T Ito K Endo J Yamamoto T Takeshima A Shinmura K Shen W Fukuda K Sano M Temporal dynamics of cardiac immune cell accumulation following acute myocardial infarction. J Mol Cell Cardiol 2013 62 24 35 23644221 4 Frangogiannis NG Lindsey ML Michael LH Youker KA Bressler RB Mendoza LH Spengler RN Smith CW Entman ML Resident cardiac mast cells degranulate and release preformed tnf-alpha, initiating the cytokine cascade in experimental canine myocardial ischemia/reperfusion. Circulation 1998 98 699 710 9715863 5 Chandrasekar B Smith JB Freeman GL Ischemia-reperfusion of rat myocardium activates nuclear factor-kappab and induces neutrophil infiltration via lipopolysaccharide-induced cxc chemokine. Circulation 2001 103 2296 2302 11342480 6 Swirski FK Nahrendorf M Etzrodt M Wildgruber M Cortez-Retamozo V Panizzi P Figueiredo JL Kohler RH Chudnovskiy A Waterman P Aikawa E Mempel TR Libby P Weissleder R Pittet MJ Identification of splenic reservoir monocytes and their deployment to inflammatory sites. Science 2009 325 612 616 19644120 7 Zeisberg EM Tarnavski O Zeisberg M Dorfman AL McMullen JR Gustafsson E Chandraker A Yuan X Pu WT Roberts AB Neilson EG Sayegh MH Izumo S Kalluri R Endothelial-to-mesenchymal transition contributes to cardiac fibrosis. Nature medicine 2007 13 952 961 8 Mollmann H Nef HM Kostin S von Kalle C Pilz I Weber M Schaper J Hamm CW Elsasser A Bone marrow-derived cells contribute to infarct remodelling. Cardiovasc Res 2006 71 661 671 16854401 9 Seropian IM Toldo S Van Tassell BW Abbate A Anti-inflammatory strategies for ventricular remodeling following st-segment elevation acute myocardial infarction. J Am Coll Cardiol 2014 63 1593 1603 24530674 10 Padro CJ Sanders VM Neuroendocrine regulation of inflammation. Seminars in immunology 2014 26 357 368 24486056 11 Elenkov IJ Wilder RL Chrousos GP Vizi ES The sympathetic nerve--an integrative interface between two supersystems: The brain and the immune system. Pharmacological reviews 2000 52 595 638 11121511 12 Grisanti LA Talarico JA Carter RL Yu JE Repas AA Radcliffe SW Tang HA Makarewich CA Houser SR Tilley DG Beta-adrenergic receptor-mediated transactivation of epidermal growth factor receptor decreases cardiomyocyte apoptosis through differential subcellular activation of erk1/2 and akt. J Mol Cell Cardiol 2014 72 39 51 24566221 13 Speidl WS Toller WG Kaun C Weiss TW Pfaffenberger S Kastl SP Furnkranz A Maurer G Huber K Metzler H Wojta J Catecholamines potentiate lps-induced expression of mmp-1 and mmp-9 in human monocytes and in the human monocytic cell line u937: Possible implications for peri-operative plaque instability. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2004 18 603 605 14715701 14 Nance DM Sanders VM Autonomic innervation and regulation of the immune system (1987-2007). Brain, behavior, and immunity 2007 21 736 745 15 Nakai A Hayano Y Furuta F Noda M Suzuki K Control of lymphocyte egress from lymph nodes through beta2-adrenergic receptors. The Journal of experimental medicine 2014 211 2583 2598 25422496 16 Lubahn CL Lorton D Schaller JA Sweeney SJ Bellinger DL Targeting alpha- and beta-adrenergic receptors differentially shifts th1, th2, and inflammatory cytokine profiles in immune organs to attenuate adjuvant arthritis. Frontiers in immunology 2014 5 346 25157248 17 Mendez-Ferrer S Battista M Frenette PS Cooperation of beta(2)- and beta(3)-adrenergic receptors in hematopoietic progenitor cell mobilization. Annals of the New York Academy of Sciences 2010 1192 139 144 20392229 18 Mendez-Ferrer S Lucas D Battista M Frenette PS Haematopoietic stem cell release is regulated by circadian oscillations. Nature 2008 452 442 447 18256599 19 Heidt T Sager HB Courties G Dutta P Iwamoto Y Zaltsman A von Zur Muhlen C Bode C Fricchione GL Denninger J Lin CP Vinegoni C Libby P Swirski FK Weissleder R Nahrendorf M Chronic variable stress activates hematopoietic stem cells. Nature medicine 2014 20 754 758 20 Gao E Lei YH Shang X Huang ZM Zuo L Boucher M Fan Q Chuprun JK Ma XL Koch WJ A novel and efficient model of coronary artery ligation and myocardial infarction in the mouse. Circ Res 2010 107 1445 1453 20966393 21 Liggett SB Bouvier M Hausdorff WP O'Dowd B Caron MG Lefkowitz RJ Altered patterns of agonist-stimulated camp accumulation in cells expressing mutant beta 2-adrenergic receptors lacking phosphorylation sites. Molecular pharmacology 1989 36 641 646 2554115 22 Shenoy SK Drake MT Nelson CD Houtz DA Xiao K Madabushi S Reiter E Premont RT Lichtarge O Lefkowitz RJ Beta-arrestin-dependent, g protein-independent erk1/2 activation by the beta2 adrenergic receptor. J Biol Chem 2006 281 1261 1273 16280323 23 Dutta P Hoyer FF Grigoryeva LS Sager HB Leuschner F Courties G Borodovsky A Novobrantseva T Ruda VM Fitzgerald K Iwamoto Y Wojtkiewicz G Sun Y Da Silva N Libby P Anderson DG Swirski FK Weissleder R Nahrendorf M Macrophages retain hematopoietic stem cells in the spleen via vcam-1. The Journal of experimental medicine 2015 212 497 512 25800955 24 van der Laan AM Ter Horst EN Delewi R Begieneman MP Krijnen PA Hirsch A Lavaei M Nahrendorf M Horrevoets AJ Niessen HW Piek JJ Monocyte subset accumulation in the human heart following acute myocardial infarction and the role of the spleen as monocyte reservoir. European heart journal 2014 35 376 385 23966310 25 Ismahil MA Hamid T Bansal SS Patel B Kingery JR Prabhu SD Remodeling of the mononuclear phagocyte network underlies chronic inflammation and disease progression in heart failure: Critical importance of the cardiosplenic axis. Circ Res 2014 114 266 282 24186967 26 Ernstrom U Sandberg G Effects of adrenergic alpha- and beta-receptor stimulation on the release of lymphocytes and granulocytes from the spleen. Scandinavian journal of haematology 1973 11 275 286 4591399 27 Modin A Pernow J Lundberg JM Comparison of the acute influence of neuropeptide y and sympathetic stimulation on the composition of blood cells in the splenic vein in vivo. Regulatory peptides 1993 47 159 169 8234902 28 Moerman EJ Scapagnini U De Schaepdryver AF Adrenergic receptors in the isolated perfused dog spleen. European journal of pharmacology 1969 5 279 285 5772677 29 Harris TJ Waltman TJ Carter SM Maisel AS Effect of prolonged catecholamine infusion on immunoregulatory function: Implications in congestive heart failure. J Am Coll Cardiol 1995 26 102 109 7797738 30 Bakovic D Pivac N Zubin Maslov P Breskovic T Damonja G Dujic Z Spleen volume changes during adrenergic stimulation with low doses of epinephrine. Journal of physiology and pharmacology : an official journal of the Polish Physiological Society 2013 64 649 655 24304578 31 Rogausch H del Rey A Oertel J Besedovsky HO Norepinephrine stimulates lymphoid cell mobilization from the perfused rat spleen via beta-adrenergic receptors. The American journal of physiology 1999 276 R724 730 10070132 32 Zebrack JS Munger M Macgregor J Lombardi WL Stoddard GP Gilbert EM Beta-receptor selectivity of carvedilol and metoprolol succinate in patients with heart failure (select trial): A randomized dose-ranging trial. Pharmacotherapy 2009 29 883 890 19637941 33 Landmann R Durig M Gudat F Wesp M Harder F Beta-adrenergic regulation of the blood lymphocyte phenotype distribution in normal subjects and splenectomized patients. Advances in experimental medicine and biology 1985 186 1051 1062 2864798 34 Van Tits LJ Michel MC Grosse-Wilde H Happel M Eigler FW Soliman A Brodde OE Catecholamines increase lymphocyte beta 2-adrenergic receptors via a beta 2-adrenergic, spleen-dependent process. The American journal of physiology 1990 258 E191 202 2154117 35 Tsai MS Chou SE Lai HS Jeng LB Lin CL Kao CH Long-term risk of acute coronary syndrome in splenectomized patients due to splenic injury. Medicine 2015 94 e610 25738485 36 Robinette CD Fraumeni JF Jr. Splenectomy and subsequent mortality in veterans of the 1939-45 war. Lancet 1977 2 127 129 69206 37 Zhang X Szeto C Gao E Tang M Jin J Fu Q Makarewich C Ai X Li Y Tang A Wang J Gao H Wang F Ge XJ Kunapuli SP Zhou L Zeng C Xiang KY Chen X Cardiotoxic and cardioprotective features of chronic beta-adrenergic signaling. Circ Res 2013 112 498 509 23104882 38 Maisel AS Beneficial effects of metoprolol treatment in congestive heart failure. Reversal of sympathetic-induced alterations of immunologic function. Circulation 1994 90 1774 1780 7923661 39 Maisel AS Harris T Rearden CA Michel MC Beta-adrenergic receptors in lymphocyte subsets after exercise. Alterations in normal individuals and patients with congestive heart failure. Circulation 1990 82 2003 2010 2173645 40 Mayer B Holmer SR Hengstenberg C Lieb W Pfeifer M Schunkert H Functional improvement in heart failure patients treated with beta-blockers is associated with a decline of cytokine levels. International journal of cardiology 2005 103 182 186 16080978 41 von Haehling S Schefold JC Jankowska E Doehner W Springer J Strohschein K Genth-Zotz S Volk HD Poole-Wilson P Anker SD Leukocyte redistribution: Effects of beta blockers in patients with chronic heart failure. PloS one 2009 4 e6411 19641622 42 Shaw SM Coppinger T Waywell C Dunne L Archer LD Critchley WR Yonan N Fildes JE Williams SG The effect of beta-blockers on the adaptive immune system in chronic heart failure. Cardiovascular therapeutics 2009 27 181 186 19689617 43 Scali S Patel V Neal D Bertges D Ho K Jorgensen JE Cronenwett J Beck A Preoperative beta-blockers do not improve cardiac outcomes after major elective vascular surgery and may be harmful. Journal of vascular surgery 2015 62 166 176 e162 26115922 44 Amsterdam EA Wenger NK Brindis RG Casey DE Jr. Ganiats TG Holmes DR Jr. Jaffe AS Jneid H Kelly RF Kontos MC Levine GN Liebson PR Mukherjee D Peterson ED Sabatine MS Smalling RW Zieman SJ Members AATF Society for Cardiovascular A Interventions, the Society of Thoracic S. 2014 aha/acc guideline for the management of patients with non-st-elevation acute coronary syndromes: Executive summary: A report of the american college of cardiology/american heart association task force on practice guidelines. Circulation 2014 130 2354 2394 25249586 45 Zhang Q Xiang J Wang X Liu H Hu B Feng M Fu Q Beta(2)-adrenoceptor agonist clenbuterol reduces infarct size and myocardial apoptosis after myocardial ischaemia/reperfusion in anaesthetized rats. British journal of pharmacology 2010 160 1561 1572 20590644 46 Ahmet I Krawczyk M Heller P Moon C Lakatta EG Talan MI Beneficial effects of chronic pharmacological manipulation of beta-adrenoreceptor subtype signaling in rodent dilated ischemic cardiomyopathy. Circulation 2004 110 1083 1090 15313944 47 Ahmet I Krawczyk M Zhu W Woo AY Morrell C Poosala S Xiao RP Lakatta EG Talan MI Cardioprotective and survival benefits of long-term combined therapy with beta2 adrenoreceptor (ar) agonist and beta1 ar blocker in dilated cardiomyopathy postmyocardial infarction. The Journal of pharmacology and experimental therapeutics 2008 325 491 499 18287209 48 Ahmet I Morrell C Lakatta EG Talan MI Therapeutic efficacy of a combination of a beta1-adrenoreceptor (ar) blocker and beta2-ar agonist in a rat model of postmyocardial infarction dilated heart failure exceeds that of a beta1-ar blocker plus angiotensin-converting enzyme inhibitor. The Journal of pharmacology and experimental therapeutics 2009 331 178 185 19587314 49 Rinaldi B Donniacuo M Sodano L Gritti G Martuscelli E Orlandi A Rafaniello C Rossi F Calzetta L Capuano A Matera MG Effects of the new ultra-long-acting beta -ar agonist indacaterol in chronic treatment alone or in combination with the beta -ar blocker metoprolol on cardiac remodelling. British journal of pharmacology 2015 172 3627 37 25825265 50 Bhushan S Kondo K Predmore BL Zlatopolsky M King AL Pearce C Huang H Tao YX Condit ME Lefer DJ Selective beta2-adrenoreceptor stimulation attenuates myocardial cell death and preserves cardiac function after ischemia-reperfusion injury. Arteriosclerosis, thrombosis, and vascular biology 2012 32 1865 1874

PMC005xxxxxx/PMC5120619.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 7600111 5310 J Neurosci Res J. Neurosci. Res. Journal of neuroscience research 0360-4012 1097-4547 27870439 5120619 10.1002/jnr.23826 NIHMS797243 Article Estradiol shifts interactions between the infralimbic cortex and central amygdala to enhance fear extinction memory in female rats Maeng Lisa Y. 12 Cover Kara K. 1 Taha Mohamad B. 12 Landau Aaron J. 1 Milad Mohammed R. 12 Lebrón-Milad Kelimer 12 1 Department of Psychiatry, Massachusetts General Hospital, Charlestown, MA, 02129, USA 2 Department of Psychiatry, Harvard Medical School, Boston, MA, 02115, USA Corresponding author: Lisa Y. Maeng, 149 13th Street, Room 2510, Charlestown, MA 02129, Phone: 617-643-1156, Fax: 617-726-5760, lmaeng@mgh.harvard.edu 25 6 2016 2 1 2017 02 7 2017 95 1-2 163175 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. There is growing evidence that estradiol enhances fear extinction memory consolidation. However, it is unclear how estradiol influences the nodes of the fear extinction network to enhance extinction memory. This study begins to delineate the neural circuits underlying the influence of estradiol on fear extinction acquisition and consolidation in female rats. After fear conditioning (day 1), naturally cycling female rats underwent extinction learning (day 2) in a low estradiol state, receiving a systemic administration of either estradiol (E2) or vehicle prior to extinction training. Extinction memory recall was then tested 24h later (day 3). We measured immediate early gene c-fos expression within the extinction network during fear extinction learning and extinction recall. During extinction learning, E2 treatment increased centrolateral amygdala (CeL) c-fos activity and reduced lateral amygdala (LA) activity relative to vehicle. During extinction recall, E2-treated rats exhibited reduced c-fos expression in the centromedial amygdala (CeM). There were no group differences in c-fos expression within the medial prefrontal cortex and dorsal hippocampus. Examining c-fos ratios with the infralimbic cortex (IL) revealed that despite the lack of group differences within the IL, E2 treatment induced greater IL activity relative to both the prelimbic cortex (PL) and central amygdala (CeA) activity during extinction memory recall. Only the relationship between IL and CeA activity positively correlated with extinction retention. In conclusion, estradiol appears to modify interactions between the IL and CeA in females, shifting from stronger amygdalar modulation of fear during extinction learning to stronger IL control during extinction recall. Graphical abstract Estradiol’s modulatory effects on fear extinction memory consolidation may be driven by the IL-amygdala circuit of the fear network. With increasing estradiol levels, neuronal activation shifts during extinction from stronger CeA to IL activity to stronger IL to CeA activity, correlating with more and less fear during extinction recall, respectively. sex differences estradiol infralimbic cortex central amygdala c-fos fear extinction estrous cycle centromedial amygdala female PTSD anxiety fear ventromedial prefrontal cortex Introduction Estradiol influences many types of learning and memory processes (Luine et al., 1998; Goodman et al., 2004; Leuner et al., 2004; Inagaki et al., 2010; Cover et al., 2014), and studies examining estradiol’s effects on neurogenesis, dendritic branching, and long-term potentiation have demonstrated that it can alter brain structure and function as well (Galvin and Ninan, 2014; Mahmoud et al., 2016; Srivastava et al., 2013). Given the higher prevalence of fear and anxiety disorders in women compared to men, a better understanding of how estradiol can affect the neural substrates and mechanisms that control fear responses is necessary. Fear extinction protocols provide a means to investigate the brain networks involved in the regulation of fear. The basic neural mechanisms of fear extinction have been studied extensively in males. The fear extinction network includes the infralimbic (IL) and prelimbic (PL) subregions of the ventromedial prefrontal cortex (vmPFC), amygdala, and hippocampus. Electrophysiology experiments have demonstrated that microstimulation of the IL reduces conditioned freezing in rats (Milad and Quirk, 2002), while inactivation of the IL increases fear expression during extinction learning, indicating impaired extinction (Sierra-Mercado et al., 2011). The basolateral amygdala (BLA) is noted as a site involved in fear expression and consolidation through its connections with the PL, and its connections with the IL make it a critical region for fear extinction as well (Likhtik and Paz, 2015). The central amygdala (CeA) is composed of lateral (CeL) and medial (CeM) subregions, which mediate fear through different functions and connections with other nodes of the fear extinction network (Ciocchi et al., 2010; Duvarci and Pare, 2014; Haubensak et al., 2010). The CeM serves as the fear output region and can be modulated by parallel circuits that drive fear expression and inhibit conditioned fear responses. The hippocampus is important for the contextual gating of fear learning and extinction (Moustafa et al., 2013). Although the fear extinction circuits are well known in males, it is unclear whether these systems behave similarly in females or how they change across the estrous cycle as gonadal hormone levels naturally fluctuate. Our laboratory has demonstrated that female rats extinguished during the high-estradiol proestrus phase exhibited significantly better extinction memory recall compared to those that underwent extinction during the low-estradiol metestrus phase (Milad et al., 2009). Zeidan et al. (2011) showed that extinction recall was improved in metestrus females administered exogenous estradiol, while also reporting elevated levels of c-fos mRNA expression within the IL and reduced levels in the amygdala. However, other brain areas were not analyzed; thus, the neural substrates and mechanisms through which estradiol exerts its influence are still unclear. In this study, we investigated this by examining c-fos expression in the key nodes of the fear extinction network: PL, IL, amygdalar subnuclei, and hippocampus. Moreover, there is evidence in the human literature demonstrating that ratios of metabolic activity within various regions of the network appear to be more predictive of behavioral outcomes during fear conditioning and extinction (Linnman et al., 2012). Based on these data, we also assessed c-fos activity ratios across the nodes noted above. Given that the IL is associated with fear inhibition, and the PL and amygdala are associated with fear expression (Laurent and Westbrook, 2008, 2009; Burgos-Robles et al., 2009; Sotres-Bayon and Quirk, 2010), we expected that differences in the distribution of neuronal activation among these structures would be relevant for extinction retention. We hypothesized that estradiol administration in metestrus females prior to extinction training would: 1) enhance fear extinction memory recall, 2) alter neuronal function of the IL within the fear extinction network, and 3) enhance the IL/PL as well as IL/amygdala c-fos activity ratios, which would be associated with reduced fear during extinction recall. Materials and methods Subjects 64 adult female Sprague Dawley rats (8 weeks old and 250g; Harlan Laboratories, Inc., Indianapolis, IN) were housed in pairs at Massachusetts General Hospital Center for Comparative Medicine in Charlestown, MA. They were maintained on a 12h light/dark cycle and a diet of ad libitum rat chow and water. After an acclimation period of 5-7 days in the animal colony room, the animals were handled daily for at least a week. All procedures conducted in this study were approved by the Subcommittee on Research Animal Care at Massachusetts General Hospital and abide by the guidelines set forth by the Institutional Animal Care and Use Committee. Vaginal swabbing and cytology All female rats were monitored daily for their estrous cycles for at least 10 consecutive days prior to the experiments and also during the days of the experiment as previously described (Maeng et al., 2015). Samples of loose vaginal epithelial cells were collected using cotton-tipped applicators moistened with 0.9% saline. The samples were stained with Dipquick Quick Stain (Jorgensen Laboratories, Inc., Loveland, CO). Phases were identified by the cell types characteristic of each estrous phase (Westwood, 2008). Proestrus is characterized by clusters of purple-stained nucleated cells. Estrus is identified by aggregated blue cornified cells, metestrus by a combination of nucleated cells, cornified cells, and leukocytes, and diestrus as similar to metestrus but with sparse representation of the different cell types. The level of circulating gonadal hormones varies across phases of the estrous cycle. Higher levels of estradiol occur during proestrus and the first half of estrus, while lower levels occur during metestrus and the first half of diestrus. All rats underwent extinction training during the metestrus phase of the estrous cycle. Any animals that were not in metestrus on the day of extinction training were not included in the study. Behavior Prior to experimentation, all animals were pre-exposed to their assigned 25 × 29 × 29 cm Plexiglas behavioral chambers (Coulbourn Instruments, Whitehall, PA) with the house lights on for one 30-minute session per day for 3 days as described previously (Maeng et al., 2015). The Plexiglas chamber contained a single overhead house light, a speaker for conditioned stimulus (CS) presentations, and a video camera to monitor behavior. The experiment was run across 3 days. All the phases of the experiment were conducted in the same context. On day 1, all female rats underwent 5 trials of the conditioned stimulus (CS; tone) alone habituation trials and were then fear conditioned with 7 trials of CS paired with the unconditioned stimulus (US; footshock) during the estrus phase. On day 2, after vaginal swabbing for confirmation of the metestrus phase, the animals were injected subcutaneously with either sesame oil (vehicle; Sigma-Aldrich, St. Louis, MO) or estradiol (15ug/kg; Sigma-Aldrich, St. Louis, MO). Thirty minutes following the injection, all animals underwent extinction training, which consisted of 20 trials of CS alone presentations. On day 3, some of the animals were returned to the chamber and presented with 3 CS-alone trials for the extinction recall test. For all phases, the CS was a 30-s, 4-kHz, 80-dB tone. For the paired trials during fear conditioning, the CS coterminated with the 0.5-s, 0.5-mA US, using GraphicState 3.03 (Coulbourn Instruments, Whitehall, PA). Every phase was run with trials that had a variable intertrial interval averaging 3 minutes. c-fos immunohistochemistry In order to evaluate changes in neuronal activity separately during extinction training and recall, two groups of animals were sacrificed at different time points. One group was sacrificed 1h after the extinction session and did not undergo the extinction recall test (POST-EXT), and the other was sacrificed 1h after recall and completed all phases (POST-REC). Another separate group of female rats underwent the same behavioral protocol as the POST-REC group with the exception of extinction training on day 2 to serve as the no extinction control group (NO-EXT). One hour after the end of the extinction (POST-EXT) or recall (POST-REC and NO-EXT) session, all animals were given a lethal dose of Fatal-Plus® solution (Vortech Pharmaceutical Ltd., Dearborn, MI). Once deeply anesthetized, the animals were transcardially perfused with 0.9% saline for 15 minutes, followed by 4% paraformaldehyde for 20 minutes at a speed of 20ml/min. Their brains were extracted and placed in 4% paraformaldehyde overnight at 4° C. The brains were then cryoprotected in 20% sucrose potassium phosphate buffered saline (KPBS) solution for 24h and then stored in 30% sucrose KPBS solution at 4°C until processing. For cryostat sectioning, the brains were rapidly frozen in Tissue Freezing Medium (Triangle Biomedical Sciences, Durham, NC) and sectioned at −21°C in 40-μm thick coronal slices. Tissue containing the mPFC, amygdala, and hippocampus were collected and processed for c-fos immunohistochemistry using a modified version of the protocol described in Do-Monte et al. (2015). All antibodies used are listed in Table 1. The tissue sections were washed in KPBS (7min × 5) and then incubated in 2% normal goat serum (NGS), 3% Triton-X, KPBS solution with 1:20,000 polyclonal rabbit anti-c-fos antibody (Calbiochem, Billerica, MA) for 18h at room temperature (RT). This was followed by incubation in biotinylated goat anti-rabbit IgG secondary antibody (1:200; Vector Laboratories, Burlingame, CA, USA), for 2h at RT. The tissue was then incubated in avidin-biotin horseradish peroxidase complex solution (1:200; Vectastain Elite ABC kit, Vector Laboratories) for 90min at RT. The tissue underwent staining in 0.02% 3,3'-diaminobenzidine (DAB), 0.3% nickel-ammonium sulfate, 1.2% glucose oxidase sodium PBS solution reacted with D-glucose (10%) for 15-20min until darkly stained c-fos immunopositive nuclei were observed and the reaction stopped with KPBS. The sections were mounted onto gelatin-subbed slides, dehydrated, cleared with Citrosolv (Fisher Scientific, Pittsburgh, PA), and cover slipped with DPX medium (Electro Microscopy Sciences, Hatfield, PA). c-fos immunoreactivity analysis The PL, IL, CeL, CeM, BLA, LA, dorsal dentate gyrus (DG), dorsal CA1, and dorsal CA3 were imaged with a CCD color camera (Moticam Pro 252A, British Columbia, Canada) mounted on a compound microscope (Motic BA410, British Columbia, Canada) at 20X magnification. Photomicrographs were captured for sections of the PL and IL (+3.72mm to +3.00mm from Bregma), amygdala (−2.04mm to −3.24mm from Bregma), and dorsal hippocampus (−2.92mm to −3.60mm from Bregma). For each of these regions, cells from 2-3 sections were bilaterally counted and averaged over sections and hemispheres per animal. These average values were then placed into their respective drug groups and averaged to obtain average c-fos positive cell counts for the vehicle vs. estradiol-treated animals. The images were captured as a rectangular region of interest (ROI) box (dimensions: 268 × 329μm) anatomically defined by Paxinos and Watson (2007) for each target brain structure. The ROI box dimensions were determined by our laboratory to create a consistent area that encompassed across brain regions as in Matsuo et al. (2009). The c-fos-positive cells within this box were counted using ImageJ software (NIH, Bethesda, MD; RRID: SCR_003070) by a rater blind to the experimental condition. Data analysis Percent freezing was calculated over the duration of the trial as the index of fear: [(time spent freezing during trial)/30 seconds]*100. Freezing was recorded and analyzed using motion-sensing computer software FreezeScan (Clever Systems, Reston, VA). The extinction retention index (ERI) was calculated by subtracting the average of the percent freezing of the first two trials of extinction recall from the average of percent freezing of the last five trials of extinction training the day before. To assess behavior across the experimental phases, percent freezing was analyzed as blocks of 2 trials averaged. A mixed-design Analysis of Variance (ANOVA), with the 2-trial blocks as the within-subjects variable and drug as the between-subjects variable, was conducted on freezing behavior for each of the habituation, fear conditioning, and fear extinction phases, using SPSS 23.0 (IBM Corp., Armonk, NY). An independent samples t-test was performed to assess group differences in extinction recall (day 3). When the assumption of equal variances did not hold, the corrected values were reported. For c-fos analysis as described above, the number of c-fos immunopositive cells were counted and averaged for all sections within a range determined by Paxinos and Watson (2007) for each region per animal. These section averages were then averaged for each drug group (vehicle and estradiol). C-fos ratios were calculated by dividing the average number of IL c-fos positive cells by the average number of c-fos positive cells of the other regions (PL, CeM, CeL, BLA, LA). Independent samples t-tests were performed on these averaged c-fos positive counts and ratios to assess drug effects. The IL c-fos activity ratios were correlated with ERI for the POST-REC group to assess how these relationships were associated with freezing behavior during extinction recall. Results Behavior All rats underwent fear conditioning on day 1 during the estrus phase and extinction training on day 2 during metestrus phase. The POST-REC group underwent recall on day 3. A mixed-design ANOVA revealed that there were no differences between the POST-EXT and POST-REC groups in the vehicle-treated animals during habituation [F(1,26)=1.634, p=0.21], fear conditioning [F(1,26)=0.045, p=0.83], and fear extinction [F(1,26)=0.483, p=0.49]. This was also found to be true in the estradiol-treated groups for habituation [F(1,22)=0.066, p=0.80], fear conditioning [F(1,22)=0.119, p=0.73], and fear extinction [F(1,22)=0.232, p=0.64]. Because the POST-EXT and POST-REC groups both underwent the same protocols for habituation, fear conditioning, and fear extinction and were not different across all phases, these groups were combined for analysis purposes (Fig. 1; Vehicle, n=28; E2, n=24). Fear recall is represented only by the POST-REC group (Fig. 1; Vehicle, n=14; E2, n=14). The behavioral data are represented as blocks of 2 trials averaged. A mixed-design ANOVA revealed that there was an effect of block [F(1,50)=13.17, p=0.001], no drug × block interaction [F(1,50)=0.34, p=0.56], and no effect of drug during habituation [F(1,50)=0.30, p=0.59]. For the fear conditioning phase, there was an effect of block [F(2,100)=4.51, p=0.01], but no significant drug × block interaction [F(2,100)=1.68, p=0.19] or main effect of drug [F(1,50)=0.006, p=0.94]. A main effect of extinction block [F(9,450)=3.10, p=0.001] was observed, but no drug × block interaction [F(9,450)=1.30, p=0.23] or effect of drug during fear extinction training [F(1,50)=0.26, p=0.62]. To summarize, the results of the statistical analyses showed that: 1) there were differences across habituation and conditioning blocks and that extinction occurred, and 2) there were no significant group differences between the vehicle- and estrogen-treated animals during the habituation, conditioning, and extinction phases. In contrast, the POST-REC animals that received estradiol on day 2 exhibited a significant reduction in freezing during recall on day 3 compared to the animals that received vehicle [t(26)= 2.535, p=0.02], as revealed by an independent samples t-test. The amount of retention of extinction learning from day 2 observed during extinction recall (as indicated by ERI) was also significantly greater in estradiol-treated females compared to vehicle [t(24)=−2.652, p=0.01]. This result replicates our previous findings (Zeidan et al., 2011). c-fos Animals that were sacrificed 1h after extinction training (POST-EXT) exhibited no significant differences between drug groups in PL [t(9.259)=1.916, p=0.09] and IL [t(14)=1.459, p=0.17] c-fos expression as shown by an independent samples t-test (Vehicle, n=8; Estradiol, n=8; Fig. 2A). This was also observed in the PL [t(15)=0.012, p=0.99] and IL [t(15)=−0.390, p=0.70] of animals that were sacrificed 1h after recall (POST-REC; Vehicle, n=8; Estradiol, n=9). However, quantification of c-fos positive cells revealed that there were differences within the amygdala (Fig. 2B). In the POST-EXT group, estradiol administration significantly increased c-fos expression in the CeL [t(7.650)= −2.295, p=0.05] (Fig. 3), but significantly decreased c-fos immunoreactivity in the LA [t(14)= 3.325, p=0.01]. Estradiol did not affect c-fos expression in the BLA [t(8.639)= −2.106, p=0.07] and CeM [t(14)=−1.023, p=0.32] compared to vehicle during extinction (Vehicle, n=8; Estradiol, n=8). In the POST-REC group, estradiol treatment reduced c-fos expression only within the CeM relative to vehicle [t(9.707)= 2.494, p=0.03] and had no effect on c-fos immunoreactivity in the LA [t(15)= 2.059, p=0.06], CeL [t(15)= 1.173, p=0.259] or BLA [t(15)= 0.000, p=1.00] (Vehicle, n=8; Estradiol, n=9). We did not find differences in c-fos expression in any of the subregions of the hippocampus during extinction or extinction recall (p>0.05; Vehicle, n=5-8; Estradiol, n=7-9; Fig. 2C). c-fos ratios Based on findings in human literature (Linnman et al., 2012), we examined the effects of estradiol on the relationship of c-fos expression between the IL and other ROIs, represented by their ratios (IL c-fos positive cell count divided by c-fos positive cell count for a different region: PL, CeM, CeL, BLA, or LA; Figs. 4 and 5). For the POST-EXT group, an independent samples t-test showed that there was no significant difference as a function of drug treatment in the IL/PL c-fos ratio [t(14)=−0.147, p=0.89; Fig. 4]. In the POST-REC group, however, the IL/PL c-fos ratio was significantly higher in the estradiol-treated group [t(8.383)=−2.405, p=0.04]. There were no group differences in the IL/CeM c-fos ratio in the POST-EXT group [(t(7.197)=1.641, p=0.14], whereas estradiol increased this ratio in the POST-REC group [t(11)=−2.473, p=0.03; Fig. 5A]. IL/CeL c-fos immunoreactivity ratio was significantly lower in the estradiol-treated animals in the POST-EXT group [t(7.820)=3.173, p=0.01; Fig. 5B]. In contrast, the IL/CeL c-fos ratio was significantly higher with estradiol treatment in the POST-REC group [t(5.674)=−2.756, p=0.04]. There were no group differences in IL/BLA and IL/LA c-fos ratios in either the POST-EXT [t(14)=0.092, p=0.93] and POST-REC [t(10)=0.060, p=0.95] groups (Fig. 5C and 5D). c-fos ratio and ERI correlations The following results are reported for only the POST-REC group. The IL/PL c-fos ratio was not significantly correlated with the ERI (r=−0.048, p=0.86). IL/CeM c-fos ratio, on the other hand, was positively correlated with ERI (r=0.605, p=0.03). The IL/CeL c-fos ratio was also positively correlated with ERI (r=0.719, p=0.006). However, this result may have been driven by a single animal that had a very high ERI and IL/CeL ratio, as the effect was lost when the animal was omitted (Fig. 5B; correlation graph). This animal was not eliminated because it did not meet any criteria for exclusion. Therefore, it is important to note that the correlation between ERI and the IL/CeM c-fos ratio may be the more critical relationship than that with the IL/CeL c-fos ratio. Furthermore, neither IL/BLA (r=−0.039, p=0.90) nor IL/LA (r=0.156, p=0.69) correlated with ERI. Therefore, the effects of estradiol only emerge in the relationships between IL and CeA neuronal activity, and these relationships, primarily IL/CeM, are associated with extinction retention during recall. No extinction control group In order to evaluate whether the influence of estradiol on extinction recall was an effect of estradiol alone, a group of female rats (n=12) were tested in metestrus, just as described for the POST-REC group, excluding extinction training on day 2 (Fig. 6). Half of the rats received vehicle injections, and the other half received estradiol injections. A mixed-design ANOVA performed for each experimental phase revealed that there were no group differences during habituation [F(1,10)=0.020, p=0.89], fear conditioning [F(1,10)=0.021, p=0.89], or extinction recall [t(10)=−1.028, p=0.33] (Fig. 6A). Moreover, estradiol did not alter c-fos expression in any of the target brain structures or the c-fos ratios with the IL (p>0.05), which was in contrast to what was found for the POST-REC group (Fig. 6B and 6C). These results confirmed that estradiol’s effect in enhancing fear extinction memory was not due to estradiol’s effects alone. Discussion We used immediate early gene c-fos immunohistochemistry in the present study to investigate the effects of estradiol in the female brain, and in particular, in modulating mPFC-amygdala interactions during fear extinction. We found a significant increase in c-fos expression within the CeL in estradiol-treated rats, and the opposite effect in the LA during extinction. Estradiol significantly reduced CeM c-fos expression during recall. Contrary to what we had expected, estradiol did not alter neuronal activation in the PL or IL during recall. However, we observed effects of estradiol on c-fos ratios representing neuronal activity relationships within the IL and PL, as well as the relationships between the IL and specific amygdalar subnuclei. Interestingly, only the IL/CeA ratios positively predicted extinction retention during recall. Moreover, estradiol administered in the absence of extinction training had no effect on freezing behavior on day 3 or c-fos expression. These results indicate that: 1) estradiol impacts the fear extinction circuitry on a network level rather than on individual regions, 2) estradiol induces a shift in neuronal activation from greater CeA activity relative to IL during extinction learning to the reverse to enhance extinction memory consolidation and reduce fear during extinction recall, and 3) estradiol’s effects on extinction recall are synergistic with extinction training and have no effect on freezing during extinction recall when administered alone. Estradiol influences in amygdalar subnuclei Estradiol reduced neuronal activation within the LA, which receives and processes converging sensory information about the CS and US, suggesting a potential weakening of the CS-US association during extinction learning. In this circuitry, the BLA receives the integrated CS-US information from the LA before transmitting it to the CeM from which fear expression is controlled (Ehrlich et al., 2009; Lee et al., 2013; Tovote et al., 2015). Therefore, estradiol may be reducing LA input to the BLA, which reduces input to the CeM, and in turn, decreases the fear response. Recent findings have described distinct populations of neurons within the BLA that have opposing functions in modulating CeM output: fear neurons that promote the expression of fear and extinction neurons that inhibit fear (Duvarci and Pare, 2014). We did not find estradiol-induced differences in c-fos expression in the BLA, but this may be due to the presence of both types of neurons, which could not be distinguished from each other with the methods used here. Of the amygdala subregions, the data suggest that the CeA may be the most critical site in estradiol’s effects on fear extinction. We found that during extinction learning, estradiol increased CeL neuronal activation. Within the CeA microcircuit, it has been reported that the CeL exerts an inhibitory influence over the CeM via specific populations of neurons within the CeL (CeLon and CeLoff cells). CeLon cells inhibit the CeLoff neurons, causing the disinhibition of the CeM to enhance conditioned fear expression (Ciocchi et al., 2010; Haubensak et al., 2010). It is possible that estradiol strengthens this control of the CeL over the CeM (perhaps specifically enhancing CeL off neuronal activation) to reduce fear expression. Animals treated with estradiol before extinction exhibited significantly reduced CeM neuronal activation during recall, with no effects in the other regions of the amygdala. This result corroborates the aforementioned idea that estradiol may be dampening the amygdalar nuclei that are involved in driving fear expression (LA), while enhancing those that are inhibiting fear expression (CeL) during extinction training. It is interesting to note that these changes in the brain occurred during extinction training but did not manifest behaviorally (Fig. 1) until the following day, when estradiol-induced alterations in only the CeM correlated with reduced fear during recall. Other studies have described effects of estradiol within the CeA that are consistent with our findings. For example, Jasnow et al. (2006) demonstrated that estradiol enhanced fear conditioning in female mice was CeA-dependent via a corticotropin-releasing hormone (CRH)-mediated mechanism. Together, these data suggest that estradiol may modulate neuroplasticity within these amygdalar circuits during the extinction memory consolidation period, strengthening the circuit that exclusively dampens CeM activity, and thus fear expression, during recall. Estradiol-mediated shifts in prefrontal-amygdalar interactions Estradiol did not significantly impact neuronal activity within the PL or IL alone, despite the literature highlighting their critical roles in fear expression and extinction. However, examining c-fos activity ratios revealed relationships between the IL and other nodes of the fear extinction network that not only differed across the two time points, but also correlated with extinction retention. The parallel, and essentially competing, prefrontal-amygdalar circuits that drive and inhibit the expression of conditioned fear have been studied extensively (Quirk et al., 2003; Likhtik et al., 2005; for review, see Tovote et al., 2015). The greater IL/PL c-fos ratio indicates an estradiol-induced switch from the fear expression to fear inhibition circuits that appear to be dictated by these subregions. Estradiol appears to enhance consolidation of the extinction memory by shifting the balance of neuronal activation towards the IL relative to the fear-driving regions such as the PL and CeA. However, the IL/PL ratio did not predict ERI, whereas IL/CeA ratios did; this suggests that inter-region, and not intra-region, relationships are critical in predicting extinction retention (Fig. 4 and 5). Therefore, estradiol may be strengthening the circuit in which the IL projects to either the BLA or the intercalated cells (ITC) of the amygdala to inhibit the CeM (Lee et al., 2013; Fig. 7), although we did not find differences in the BLA. The influence of estradiol on the ITCs warrants future investigation, as they are the intermediary structures en route from the IL to the CeM. Given that extinction recall is impaired in individuals with PTSD and anxiety, these findings support the literature that describes a frontolimbic circuit imbalance (heightened amygdala activation and reduced vmPFC activation) in these psychopathologies (Kim et al., 2011; Hayes et al., 2012; Tromp et al., 2012). This effect is opposite that of estradiol on this circuit and is one potential mechanism through which estradiol may restore balance and improve extinction recall. Sex differences in the corticolimbic circuit Sex differences are increasingly reported in fear extinction and may be relevant to the heightened vulnerability of women to anxiety and stress-related psychopathologies (Breslau et al., 1997; Shvil et al., 2014; Maeng and Milad, 2015). Our current findings suggest that differences in the neural correlates of fear in female rats are in part mediated by estradiol, and these differences across estradiol states may underlie sex differences in engaged fear extinction circuitry. In fact, sex differences in the role of the mPFC in fear extinction have been reported previously (Baran et al., 2010; Fenton et al., 2014). Moreover, a recent study in humans reported sex differences and estradiol effects in resting state functional connectivity of the laterobasal and centromedial amygdala with higher connectivity with the vmPFC in men and women with higher levels of estradiol (Engman et al., 2016). Given that these regions are homologous with the rodent BLA, CeM and IL, this is consistent with our findings, which suggests that estradiol induces changes in prefrontal connectivity with the amygdala that yield sex differences by modifying IL-amygdala interactions and their contributions to fear extinction behavior. Clinical translation The present study is the first to examine the network-level effects of estradiol on fear extinction from extinction learning to extinction recall; as such, it provides critical insight into the brain networks underlying anxiety and fear-based disorders in women. Our data support the growing literature indicating that given the relationship between naturally fluctuating endogenous estradiol and fear extinction memory, the success of psychiatric treatment (efficacy, duration, efficiency) may depend on gonadal hormone status (Wegerer et al., 2014; Glover et al., 2015; Pineles et al., 2016). Based on these data, estradiol may strengthen the consolidation of extinction memory via modulation of the amygdalar nuclei by the IL. This has powerful implications for improved treatment targets and enhanced efficacy of treatments dependent on fear extinction processes, such as exposure therapy. Future directions Evidence suggesting that estradiol can strengthen extinction memory consolidation is important for improving therapeutic outcomes for individuals suffering from PTSD and other fear and anxiety-related disorders. However, the long-term maintenance of these effects on extinction memory has yet to be determined, and their durability is an important consideration for the potential augmentation of therapy with estradiol. We have identified brain regions and nuclei that are modified by estradiol and associated with its effects on behavior at critical time points (during extinction learning and during extinction recall); however, it is unclear which point of the extinction session is represented by the c-fos expression reported in this study as early and late extinction may induce differential effects. There is also still a need to investigate estradiol’s effects on fear extinction circuitry during memory consolidation to bridge the gap of neuronal change and plasticity that may occur between day 2 and day 3. To further explore this issue, examining the molecular signaling mechanisms of estradiol action would be advantageous. Estrogen can enhance learning and memory through activation of both non-genomic and classical transcriptional mechanisms. The classical mechanism of estrogen relies on a slower-acting, genomic process initiated by ER binding at nuclear receptors. On the other hand, the more rapid mechanism relies on activation of cell membrane receptors, involving intracellular signaling pathways. Both of these pathways appear to converge at the level of transcription and protein synthesis with evidence to suggest that this can occur through activation of the PI3K/Akt and/or MAPK/ERK signaling cascades (Pedram et al., 2002; for review, please see Cover et al., 2014). Given the rapid, yet long-lasting, effects of estradiol on fear extinction, both of these mechanisms may be activated within the CeA to elicit the neuronal and behavioral effects we report in this study. The current findings indicate that estradiol shifts the balance in neuronal activity between the IL and amygdala more towards the IL for increased inhibitory control over fear output from the CeM. Observing this relationship underscores the importance of investigating neural interactions within a network versus solely considering the contributions of individual brain regions. Moreover, the influence of estradiol on neuronal activity within the ventral hippocampus was not included in the current analyses, but this region also has critical connections to the IL and amygdala that can modulate fear extinction and should be explored as well. Given the limitations of c-fos in examining real-time connectivity between the IL and amygdala, future study of these sites in vivo, perhaps using electrophysiology or optogenetics, is necessary to further elucidate these estradiol-mediated relationships. Acknowledgments The authors thank the members of the Milad laboratory at MGH for their constructive feedback on the manuscript. Additionally, we thank Dr. Adriano Reimer for assistance with the figures. This work was supported by NIMH grant R01 MH097880-001 and departmental funds from the Department of Psychiatry at Massachusetts General Hospital to Mohammed R. Milad. Figure 1 Timeline of experiment and effect of estradiol on freezing behavior. All animals underwent a 3-day behavioral protocol. Habituation and fear conditioning took place on day 1, fear extinction training on day 2, and extinction recall on day 3. Extinction training occurred during the metestrus phase of the estrous cycle, and subcutaneous vehicle or estradiol injections (INJ) were performed 30 minutes (30”) prior to the start of the session. The POST-EXT group was sacrificed 1h after the extinction session, while the POST-REC group was sacrificed 1h after the recall session. Estradiol does not influence freezing behavior during fear conditioning and extinction, but does enhance extinction recall. POST-EXT and POST-REC groups (Vehicle, n=28; Estradiol, n=24) were combined for both the conditioning and extinction phases. The extinction recall phase is represented by the POST-REC group (Vehicle, n=14; Estradiol, n=14). *p≤0.05. Figure 2 Photomicrographs of c-fos stained sections and average c-fos positive cell counts. The first column consists of images containing the regions of interest (ROIs) bordered by dashed lines. The second column displays the c-fos positive cell counts for the POST-EXT group, and the third column for the POST-REC group. A. Quantification of c-fos positive expression within the PL and IL regions of the mPFC as outlined by the dashed lines in the image are represented in the graphs. There were no significant effects of drug in the mPFC (POST-EXT: Vehicle, n=8; Estradiol, n=8; POST-REC: Vehicle, n=8; Estradiol, n=9). B. C-fos expression was measured within the LA, BLA, CeL, and CeM. During extinction, estradiol increased CeL activity, while reducing LA and BLA activity (Vehicle, n=8; Estradiol, n=8). During recall, estradiol reduced neuronal activation in the CeM and LA (Vehicle, n=8; Estradiol, n=9). C. Hippocampal c-fos expression was counted for the dorsal subregions: DG (POST-EXT: Vehicle, n=8; Estradiol, n=8; POST-REC: Vehicle, n=5; Estradiol, n=9), CA1 (POST-EXT: Vehicle, n=8; Estradiol, n=8; POST-REC: Vehicle, n=6; Estradiol, n=9), and CA3 (POST-EXT: Vehicle, n=7; Estradiol, n=7; POST-REC: Vehicle, n=6; Estradiol, n=7). There were no differences in c-fos activity with estradiol administration. PL=prelimbic; IL=infralimbic; LA=lateral amygdala; BLA=basolateral amygdala; CeL=centrolateral amygdala; CeM=centromedial amygdala; DG=dentate gyrus; CA1=cornus ammonis; CA3=cornus ammonis. White bars represent vehicle-treated animals. Black bars represent estradiol-treated animals. *p≤0.05. Figure 3 Photomicrographs of the centrolateral amygdala comparing c-fos expression with vehicle vs. estradiol treatment in POST-EXT. The ROI for the CeL is zoomed in to 20X magnification (right column) from 4X magnification (left column) to visualize the pronounced increase in c-fos immunoreactivity following estradiol administration. Figure 4 IL/PL c-fos activity ratio of POST-EXT and POST-REC groups and correlation with extinction retention index. IL/PL c-fos activity ratio is significantly higher in the estradiol group during extinction recall, but not during extinction learning. The IL/PL c-fos ratio is not correlated with how much extinction was conserved (ERI). White bars represent vehicle-treated animals. Black bars represent estradiol-treated animals. White dots in the correlation graphs represent vehicle-treated animals, and black dots represent the estradiol-treated animals. POST-EXT: Vehicle, n=8; Estradiol, n=8. POST-REC: Vehicle, n=8; Estradiol, n=9. *p≤0.05. Figure 5 C-fos activity ratios with the IL and subregions of the amygdala and their correlations with extinction retention index. A. Estradiol did not alter the IL/CeM c-fos ratio in the POST-EXT group; however, it increased it in the POST-REC group. The IL/CeM ratio was positively correlated with ERI. B. The IL/CeL c-fos ratio was significantly reduced in the POST-EXT group but significantly increased in the POST-REC group. The IL/CeL c-fos ratio was also positively correlated with ERI. C and D. There were no differences in IL/BLA and IL/LA c-fos ratios with estradiol treatment, and similarly, no correlations were found between both ratios with ERI. White bars represent vehicle-treated animals. Black bars represent estradiol-treated animals. White dots in the correlation graphs represent vehicle-treated animals, and black dots represent the estradiol-treated animals. POST-EXT: Vehicle, n=8; Estradiol, n=8. POST-REC: Vehicle, n=8; Estradiol, n=9. *p≤0.05. Figure 6 No extinction control behavior and c-fos. A. A separate group of animals underwent the same 3-day protocol as the POST-REC group but did not receive extinction training following the vehicle (n=6) or estradiol (n=6) injection. There was no difference between drug treatments in percent freezing in fear conditioning or recall. INJ= vehicle or estradiol injection. B. Estradiol did not influence neuronal activation in any subregion of the mPFC, amygdala, or dorsal hippocampus. C. In contrast to the POST-REC group, the no extinction control group did not exhibit any differences in c-fos activity ratios with the IL (IL/PL, IL/CeM, IL/CeL). White bars represent vehicle-treated animals. Black bars represent estradiol-treated animals. Figure 7 Estradiol modulation of fear extinction memory and circuitry. Based on the findings in this study, we propose that estradiol’s modulatory effects on fear extinction memory consolidation are driven by the IL-amygdala circuit of the network. In the case of the metestrus females that received vehicle injections and thus had less estradiol during extinction training, the glutamatergic connections (blue) between the IL and ITC and GABAergic projections (black) to the CeA are weak, as indicated by the dashed lines. This leads to greater fear expression during fear extinction recall (red). Estradiol administration strengthens these connections (indicated by the solid lines) to reduce fear expression during recall (green) and enhance extinction retention. More specifically, it appears that with increasing estradiol levels, there is a shift in neuronal activation from stronger CeA to IL activity (red) to stronger IL to CeA activity (green). This shift is correlated with more and less fear during extinction recall, respectively. PL=prelimbic cortex; IL=infralimbic cortex; BLA=basolateral amygdala; CeL=centrolateral amygdala; CeM=centromedial amygdala; ITCv=ventral intercalated cells of the amygdala; CeA=central amygdala; E2=estradiol. Table 1 Antibodies used Antigen Description of Immunogen Source, Host Species, Cat. #, Clone or Lot#, RRID: Concentration Used Anti-c-fos Proto-oncogene reflects neuronal activity Calbiochem, rabbit, PC38, polyclonal/D00134698 RRID:AB_2106755 1:20,000 Rabbit IgG (H+L) Biotinylated goat anti- rabbit IgG (H+L) Vector Laboratories, goat, BA-1000, polyclonal/ZA0520, RRID:AB_2313606 1:200 Table 2 Minimum and maximum c-fos cell counts for each region, time point, and drug treatment Post-ext Post-rec No ext control Vehicle E2 Vehicle E2 Vehicle E2 mPFC PL Min=2; Min=0.33; Min=0.33; Min=0.33; Min=1.3; Min=2; Max=29 Max=14 Max=10.3 Max=21.5 Max=5 Max=5.6 IL Min=2.7; Min=0.33; Min=0.33; Min=1.7; Min=4; Min=6.2; Max=66.5 Max=22.7 Max=23 Max=36.5 Max=18.8 Max=26.8 Amygdala CeM Min=0.33; Min=0; Min=0; Min=0; Min=2.5; Min=1.3; Max=4.5 Max=8 Max=6 Max=2.5 Max=7.7 Max=8 CeL Min=2; Min=0.5; Min=0; Min=0; Min=5.3; Min=2.8; Max=13.5 Max=49.8 Max=26.7 Max=8.7 Max=18 Max=12.7 BLA Min=2; Min=0.7; Min=0; Min=0; Min=1.3; Min=1; Max=12 Max=4.5 Max=3.7 Max=8.3 Max=6 Max=8.3 LA Min=2; Min=0.7; Min=0.7; Min=0; Min=2; Min=1.5; Max=8.7 Max=2.5 Max=5 Max=3.5 Max=4.5 Max=3.7 Dorsal hippocampus DG Min=0.8; Min=3.4; Min=2.5; Min=2.7; Min=3.6; Min=4.3; Max=17.2 Max=8 Max=8.3 Max=6.5 Max=12.3 Max=19.6 CA1 Min=0; Min=0.8; Min=1; Min=0.7; Min=0; Min=0; Max=6 Max=4.2 Max=4.3 Max=2.3 Max=0.5 Max=2.4 CA3 Min=0.5; Min=1; Min=2; Min=0; Min=0.3; Min=0.8; Max=8.6 Max=3.3 Max=3 Max=4.7 Max=3.5 Max=5 Significance statement Anxiety- and fear-based disorders affect twice as many women as men. The neurobiology of fear has been extensively studied in males, and it has been found to overlap with brain networks that are altered by these psychopathologies. We have previously demonstrated that estradiol enhances fear extinction memory consolidation in female rats, implicating a potential neuromodulatory role for estradiol on extinction memory. This work elucidates how estradiol alters neuronal activation in the networks underlying fear conditioning and extinction to improve extinction retention. Identifying these neural correlates may contribute to the development of sex-specific psychiatric treatments that account for gonadal hormone influences. Conflict of interest statement The authors declare that they have no conflicts of interest. Role of authors All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: KL, MRM, LYM, and KKC. Acquisition of data: LYM, KKC, MBT, and AJL. Analysis and interpretation of data: LYM, MBT, KL, and MRM. Drafting of the manuscript: LYM. Critical revision of the manuscript for important intellectual content: LYM, MBT, KKC, AJL, MRM, and KL. Statistical analysis: LYM and MBT. Obtained funding: MRM. Administrative, technical, and material support: KKC and LYM. Study supervision: LYM and KL. Literature cited Baran SE Armstrong CE Niren DC Conrad CD Prefrontal cortex lesions and sex differences in fear extinction and perseveration Learn. Mem. Cold Spring Harb. N 2010 17 267 278 Breslau N Davis GC Andreski P Peterson EL Schultz LR Sex differences in posttraumatic stress disorder Arch. Gen. Psychiatry 1997 54 1044 1048 9366662 Burgos-Robles A Vidal-Gonzalez I Quirk GJ Sustained Conditioned Responses in Prelimbic Prefrontal Neurons Are Correlated with Fear Expression and Extinction Failure J. Neurosci 2009 29 8474 8482 19571138 Ciocchi S Herry C Grenier F Wolff SBE Letzkus JJ Vlachos I Ehrlich I Sprengel R Deisseroth K Stadler MB Encoding of conditioned fear in central amygdala inhibitory circuits Nature 2010 468 277 282 21068837 Cover KK Maeng LY Lebrón-Milad K Milad MR Mechanisms of estradiol in fear circuitry: implications for sex differences in psychopathology Transl. Psychiatry 2014 4 e422 25093600 Duvarci S Pare D Amygdala microcircuits controlling learned fear Neuron 2014 82 966 980 24908482 Ehrlich I Humeau Y Grenier F Ciocchi S Herry C Lüthi A Amygdala inhibitory circuits and the control of fear memory Neuron 2009 62 757 771 19555645 Engman J Linnman C Van Dijk KRA Milad MR Amygdala subnuclei resting-state functional connectivity sex and estrogen differences Psychoneuroendocrinology 2016 63 34 42 26406106 Fenton GE Pollard AK Halliday DM Mason R Bredy TW Stevenson CW Persistent prelimbic cortex activity contributes to enhanced learned fear expression in females Learn. Mem 2014 21 55 60 24429423 Galvin C Ninan I Regulation of the mouse medial prefrontal cortical synapses by endogenous estradiol Neuropsychopharmacol. Off. Publ. Am. Coll. Neuropsychopharmacol 2014 39 2086 2094 Glover EM Jovanovic T Norrholm SD Estrogen and Extinction of Fear Memories:Implications for Posttraumatic Stress Disorder Treatment Biol. Psychiatry 2015 Goodman MR Garman EE Arnold LL Sengelaub DR Garraghty PE The effects of estradiol on avoidance learning in ovariectomized adult rats Integr. Physiol. Behav. Sci. Off. J. Pavlov. Soc 2004 39 192 206 Haubensak W Kunwar PS Cai H Ciocchi S Wall NR Ponnusamy R Biag J Dong H-W Deisseroth K Callaway EM Genetic dissection of an amygdala microcircuit that gates conditioned fear Nature 2010 468 270 276 21068836 Inagaki T Gautreaux C Luine V Acute estrogen treatment facilitates recognition memory consolidation and alters monoamine levels in memory-related brain areas Horm. Behav 2010 58 415 426 20553724 Jasnow AM Schulkin J Pfaff DW Estrogen facilitates fear conditioning and increases corticotropin-releasing hormone mRNA expression in the central amygdala in female mice Horm. Behav 2006 49 197 205 16083887 Laurent V Westbrook RF Distinct contributions of the basolateral amygdala and the medial prefrontal cortex to learning and relearning extinction of context conditioned fear Learn. Mem. Cold Spring Harb. N 2008 15 657 666 Laurent V Westbrook RF Inactivation of the infralimbic but not the prelimbic cortex impairs consolidation and retrieval of fear extinction Learn. Mem. Cold Spring Harb. N 2009 16 520 529 Lee S Kim S-J Kwon O-B Lee JH Kim J-H Inhibitory networks of the amygdala for emotional memory Front. Neural Circuits 2013 7 129 23914157 Leuner B Mendolia-Loffredo S Shors TJ High levels of estrogen enhance associative memory formation in ovariectomized females Psychoneuroendocrinology 2004 29 883 890 15177703 Likhtik E Paz R Amygdala-prefrontal interactions in (mal)adaptive learning Trends Neurosci 2015 38 158 166 25583269 Likhtik E Pelletier JG Paz R Paré D Prefrontal control of the amygdala J. Neurosci. Off. J. Soc. Neurosci 2005 25 7429 7437 Linnman C Zeidan MA Furtak SC Pitman RK Quirk GJ Milad MR Resting amygdala and medial prefrontal metabolism predicts functional activation of the fear extinction circuit Am. J. Psychiatry 2012 169 415 423 22318762 Luine VN Richards ST Wu VY Beck KD Estradiol enhances learning and memory in a spatial memory task and effects levels of monoaminergic neurotransmitters Horm. Behav 1998 34 149 162 9799625 Maeng LY Milad MR Sex differences in anxiety disorders: Interactions between fear, stress, and gonadal hormones Horm. Behav 2015 Maeng LY Cover KK Landau AJ Milad MR Lebron-Milad K Protocol for studying extinction of conditioned fear in naturally cycling female rats J. Vis. Exp. JoVE 2015 Mahmoud R Wainwright SR Galea LAM Sex Hormones and Adult Hippocampal Neurogenesis: Regulation, Implications, and Potential Mechanisms Front. Neuroendocrinol 2016 Matsuo N Yamasaki N Ohira K Takao K Toyama K Eguchi M Yamaguchi S Miyakawa T Neural activity changes underlying the working memory deficit in alpha-CaMKII heterozygous knockout mice Front. Behav. Neurosci 2009 3 20 19750198 Milad MR Quirk GJ Neurons in medial prefrontal cortex signal memory for fear extinction Nature 2002 420 70 74 12422216 Milad MR Igoe SA Lebron-Milad K Novales JE Estrous cycle phase and gonadal hormones influence conditioned fear extinction Neuroscience 2009 164 887 895 19761818 Do-Monte FH Quiñones-Laracuente K Quirk GJ A temporal shift in the circuits mediating retrieval of fear memory Nature 2015 519 460 463 25600268 Moustafa AA Gilbertson MW Orr SP Herzallah MM Servatius RJ Myers CE A model of amygdala-hippocampal-prefrontal interaction in fear conditioning and extinction in animals Brain Cogn 2013 81 Paxinos G Watson C The Rat Brain in Stereotaxic Coordinates: Hard Cover Edition 2007 Academic Press Pedram A Razandi M Aitkenhead M Hughes CCW Levin ER Integration of the Non-genomic and Genomic Actions of Estrogen MEMBRANE- INITIATED SIGNALING BY STEROID TO TRANSCRIPTION AND CELL BIOLOGY J. Biol. Chem 2002 277 50768 50775 12372818 Pineles SL Nillni YI King MW Patton SC Bauer MR Mostoufi SM Gerber MR Hauger R Resick PA Rasmusson AM Extinction Retention and the Menstrual Cycle: Different Associations for Women With Posttraumatic Stress Disorder J. Abnorm. Psychol 2016 Quirk GJ Likhtik E Pelletier JG Paré D Stimulation of medial prefrontal cortex decreases the responsiveness of central amygdala output neurons J. Neurosci. Off. J. Soc. Neurosci 2003 23 8800 8807 Shvil E Sullivan GM Schafer S Markowitz JC Campeas M Wager TD Milad MR Neria Y Sex differences in extinction recall in posttraumatic stress disorder: a pilot fMRI study Neurobiol. Learn. Mem 2014 113 101 108 24560771 Sierra-Mercado D Padilla-Coreano N Quirk GJ Dissociable roles of prelimbic and infralimbic cortices, ventral hippocampus, and basolateral amygdala in the expression and extinction of conditioned fear Neuropsychopharmacol. Off. Publ. Am. Coll. Neuropsychopharmacol 2011 36 529 538 Sotres-Bayon F Quirk GJ Prefrontal control of fear: more than just extinction Curr. Opin. Neurobiol 2010 20 231 235 20303254 Srivastava DP Woolfrey KM Penzes P Insights into rapid modulation of neuroplasticity by brain estrogens Pharmacol. Rev 2013 65 1318 1350 24076546 Tovote P Fadok JP Lüthi A Neuronal circuits for fear and anxiety Nat. Rev. Neurosci 2015 16 317 331 25991441 Wegerer M Kerschbaum H Blechert J Wilhelm FH Low levels of estradiol are associated with elevated conditioned responding during fear extinction and intrusive memories in daily life Neurobiol. Learn. Mem 2014 116C 145 154 Westwood FR The female rat reproductive cycle: a practical histological guide to staging Toxicol. Pathol 2008 36 375 384 18441260 Zeidan MA Igoe SA Linnman C Vitalo A Levine JB Klibanski A Goldstein JM Milad MR Estradiol modulates medial prefrontal cortex and amygdala activity during fear extinction in women and female rats Biol. Psychiatry 2011 70 920 927 21762880

PMC005xxxxxx/PMC5120678.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 2984705R 2786 Cancer Res Cancer Res. Cancer research 0008-5472 1538-7445 25267066 5120678 10.1158/0008-5472.CAN-14-0579 NIHMS761182 Article CD98hc (SLC3A2) Loss Protects Against Ras-Driven Tumorigenesis by Modulating Integrin-Mediated Mechanotransduction Estrach Soline 1a* Lee Sin-Ae 2* Boulter Etienne 1a Pisano Sabrina 1b Errante Aurélia 1a Tissot Floriane S. 1a Cailleteau Laurence 1a Pons Catherine 1a Ginsberg Mark H. 2 Féral Chloé C. 1a 1 INSERM, U1081, CNRS, UMR7284, Institute for Research on Cancer and Aging of Nice (IRCAN), Avenir Teama, AFM Core facilityb, University of Nice Sophia Antipolis, Medical School, 28 Avenue de Valombrose, F-06107, Nice, France. 2 Department of Medicine, University of California San Diego, La Jolla, CA 92093. CORRESPONDENCE: Chloé C. Féral: chloe.feral@inserm.fr * Contributed equally to this work 20 2 2016 29 9 2014 1 12 2014 23 11 2016 74 23 68786889 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. CD98hc (SLC3A2) is the heavy chain component of the dimeric transmembrane glycoprotein CD98, which comprises the large neutral amino acid transporter LAT1 (SLC7A5) in cells. Overexpression of CD98hc occurs widely in cancer cells, and is associated with poor prognosis clinically, but its exact contributions to tumorigenesis are uncertain. In this study, we showed that that genetic deficiency of CD98hc protects against Ras-driven skin carcinogenesis. Deleting CD98hc after tumor induction was also sufficient to cause regression of existing tumors. Investigations into the basis for these effects defined two new functions of CD98hc that contribute to epithelial cancer beyond an intrinsic effect on CD98hc on tumor cell proliferation. First, CD98hc increased the stiffness of the tumor microenvironment. Second, CD98hc amplified the capacity of cells to respond to matrix rigidity, an essential factor in tumor development. Mechanistically, CD98hc mediated this stiffness-sensing by increasing Rho kinase (ROCK) activity, resulting in increased transcription mediated by YAP/TAZ, a nuclear relay for mechanical signals. Our results suggest that CD98hc contributes to carcinogenesis by amplifying a positive feedback loop which increases both extracellular matrix stiffness and resulting cellular responses. This work supports a rationale to explore the use of CD98hc inhibitors as cancer therapeutics, CD98hc/SLC3A2 Ras-driven cancer gain-amplifier stiffness YAP/TAZ Introduction Skin squamous cell carcinoma (SCC) is the second most common skin cancer (1). Expression of the type II transmembrane glycoprotein CD98 heavy chain (4F2, SLC3A2) is highly increased in most carcinomas, including SCC, as well as transformed cell lines (2), while in skin, CD98hc is required for proper homeostasis and efficient epidermal wound closure. Via its extracellular domain (C109), CD98hc covalently binds one of several catalytic subunits (the SLC7A5–11 family) to form the functional Heteromeric Amino acid Transporters (HAT) at the cell surface (3). Besides its role as amino acid transport mediator, CD98hc, by binding to intracellular domains to the adhesion receptor β1 integrins, enhances integrin outside-in signaling (4), thus modulating integrin-dependent cell functions. CD98hc over expressing NIH-3T3 fibroblasts developed malignant tumors in athymic mice (5,6). We previously showed that CD98hc deletion in mouse embryonic stem (ES) cells delays teratocarcinoma formation in mice (4). CD98hc is an integrin binding protein that enhances signals downstream of β1 integrin therefore regulating integrin-dependent cell behavior, including matrix remodeling in vitro (4,7,8). Tumorigenicity of CD98hc deficient ES cells can be reconstituted by expressing a chimeric form of CD98hc which is able to interact with β1 integrin (4). This CD98hc/β1 interaction contributes to cell transformation in vitro (9). Here we examine the role of CD98hc in a well-established (10) two step model of squamous cell carcinoma (SCC). In this Ras-driven cancer model, tumorigenesis begins with the initiation of a single epidermal cell, which occurs following a single subcarcinogenic dose of 7,12-dimethylbenz[a]-anthracene (DMBA). Following the initiation stage, the population of mutated cells is promoted to clonally expand during the second stage, referred to as “promotion”. Tumor promotion is elicited by the repeated topical application of chemical agents, such as the phorbol ester, phorbol 12-myristate 13-acetate (TPA) that leads to sustained epidermal hyperplasia as evidenced by an increase in the number of nucleated cell layers and an overall increase in thickness of the epidermis. Tumorigenesis proceeds through the promotion of benign tumors (papilloma) growth and finally the progression of some benign tumors into malignant and potentially metastatic lesions (SCC). This two-stage skin carcinogenesis model enables direct visualization of tumor development and permits evaluation of TPA-induced inflammation response (10). Here, we combined this model of epidermal tumor formation with conditional deletion of CD98hc in basal keratinocytes to reveal a major contribution of CD98hc in controlling the mechanical properties of the tumor microenvironment, independently of skin TPA-induced inflammation. Specifically, CD98hc deficient skin was protected against tumor formation, and that CD98hc deletion led to regression of pre-existing tumors. We demonstrate that, beyond CD98hc intrinsic proliferation effect within tumor cells, CD98hc-expressing environment is cancer-prone due to modulation of its mechanical properties. This model is known to be sensitive to integrin signaling and rigidity sensing. We now show that CD98hc acts as a gain amplifier for stiffness sensing via integrins. CD98hc is thus a gain amplifier of a positive feedback loop that increases ROCK signaling, matrix stiffness and YAP/TAZ-driven gene expression. Material and Methods Mice All procedures were approved by the Institutional Animal Care and Use Committee at the University of Nice-Sophia Antipolis, Nice, FR (CIEPAL-AZUR Agreement NCE/2012-66). K14-CreERT2, CD98hcfl/fl have been described previously (11). The experiments have been performed on pure FVB/n background mice. Chemical Carcinogenesis Chemical carcinogenesis experiments were performed on 10-week-old mice (20 animals/group), essentially as previously described (10). Mice received one topical application of 200 nmol of 7,12-dimethylbenz(a)anthracene (DMBA; Sigma-Aldrich) in 200 µL of acetone followed a week later by biweekly applications of 6.8 nmol of phorbol 12-myristate-13-acetate (TPA; Sigma-Aldrich) in 200 µl acetone. Papillomas and SCC numbers were scored once a week for up to 32 weeks after the start of promotion. Topical 4-OHT treatment was topically applied when indicated (6 treatments of 1.5 mg 4OHT in ethanol/back). No difference in benign tumor acquisition was observed in 4-OHT-treated wt mice ruling out any possible effects of 4-OHT treatment (data not shown). TPA-induced skin response Ears (both sides) of mice with WT- or CD98hc-deficient skin were topically treated with 20µl (per side) of either TPA (0.5µg per side) or vehicle (acetone). Ear thickness was measured with digital calipers, and ear swelling was calculated as (thickness at each time point)-(thickness at 0h). Paraffin sections were stained with H&E and granulocytes were quantified using ImageJ software. Histology and Immunohistochemistry Samples were collected as described by Montanez et al. (12). Histology and immunochemistry were performed as described previously (11). Paraffin-embedded sections were used for all stainings except CD98hc immunostaining, which was performed on frozen sections. Primary antibodies were: rabbit anti human P-cofilin (Cell Signalling), rabbit anti human P-MYPT1 (Millipore), rabbit anti human MYPT1 (Millipore), rat anti mouse CD98 (Clone RL388, e bioscience). Secondary antibodies were used according to manufacturer’s instructions, for immunofluorescence Alexa Fluor 594 anti Rat # A- 21209 anti rabbit #A-21207 (Life technologies); and for immunohistochemistry Vectastain ABC-Kit (Vector #4001) with DAB reagent (Vector #SK4100). Elastic Modulus Measurements To carry out topography imaging and simultaneous elastic modulus maps, a Bioscope Catalyst operating in Peak Force Quantitative Nanomechanical Mapping mode (Bruker Nano, Inc.), coupled with an inverted optical microscope (DMI 6000B, Leica), was used. The experiments were performed using V-shaped silicon SNL-D probes (nominal spring constant k= 0.06 N/m, side angle 23 °, Bruker Nano, Inc.). For each sample (10µm unfixed frozen sagittal section (13), a peak force of 5 nN and a tip velocity of 4 micron/s were used (detailed procedure in Supp. Mat. & Methods). For collagen deposition analysis, detailed procedure is in Supp. Mat. & Methods. Cells Mouse embryonic fibroblasts (MEFs) were derived from CD98hc-conditional knockout homozygote embryos as described previously (7). Similarly, murine keratinocytes were generated and cultured from backskin of adult mice as described in (11) (culture conditions, see Supp. Data). Both cell types were identified by PCR on genomic DNA (as for mice (7), CD98hc expressing or deficient lines), which was performed regularly during their use. MEFs were maintained in DMEM-H (Invitrogen) culture medium containing 10% (v/v) FBS (HyClone), 20 mM HEPES, pH 7.3 (Invitrogen), 0.1 mM nonessential amino acid (Invitrogen), 0.1 mM β-mercaptoethanol (Sigma-Aldrich), and 2 mM L-Glutamine (Invitrogen) at 37°C and 5% CO2. The CD98hc-null MEFs were generated by infecting CD98hc floxed MEFs with adeno-CRE encoding CRE recombinase. Cell Spreading Assay MEFs were replated on the fibronectin-coated gels (PDMS, Supp. Mat. & Methods, or for Supp. Fig. S3, Polyacrylamide, Matrigen, CA, USA) with indicated stiffness and incubated in DMEM plus 1% (v/v) BSA for 45 min, or grown on gels for 24 hours prior to treatment with 1µg/ml of either Rho inhibitor I or Rho activator II (Cytoskeleton, CO, USA) in serum-free medium for 4 hours. Alternatively, CD98hc-null MEFs infected with retroviruses encoding wild type CD98hc, wild type CD69 or chimeras (C69T98E98, C98T69E98, or C98T98E69) were replated on 30 kPa of PDMS gels for 45 min in the absence of serum. Cells were fixed with 3.7% formaldehyde in PBS, permeabilized with 0.5% Triton X-100 in PBS at room temperature (RT) for 10 min, and washed three times with PBS. The cells were then incubated with phalloidin-conjugated rhodamine (Molecular Probes) for 1 hr at RT. The cells on coverslips were washed three times with PBS, mounted with a mounting solution (ProLong® Gold antifade reagent; Invitrogen), and visualized by fluorescent microscopy (Eclipse TE2000-U, Nikon). Cell spreading was quantified by cell area by using customized software written in MATLAB (MathWorks). Western Blotting MEFs were replated on PDMS gels as described above. When mentioned, CD98hc-null MEFs were transiently transfected with wild type CD98hc, extracellular domain-deleted CD98hc or CD69 for 36hr then replated on 30 kPa PDMS substrates and incubated as above. Whole cell lysates were prepared using modified radioimmune precipitation assay buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 50 mM NaF, 1 mM sodium pyrophosphate, 0.1% sodium deoxycholate, 1% Nonidet P-40, protease inhibitors cocktail and 1% CHAPS). Protein lysates were quantified using the bicinchoninic acid (BCA) method (Thermo Scientific Pierce), loaded on SDS-PAGE, and analyzed by Western blotting. The primary antibodies for western blotting were anti-phospho-Y576FAK (Invitrogen), anti-FAK (Santa Cruz Biotechnology), anti-phospho-S476Akt (Cell Signaling Technology), anti-Akt (Cell Signaling Technology) and anti-α-tubulin (Sigma Aldrich). qPCR RNAs were extracted from back skin samples (bearing or not tumors) using Trizol reagent (GIBCO), or from MEFs plated on gels using RNeasy plus kit (Qiagen),. Reverse transcription was performed on total RNA using Superscript II reverse transcriptase (Invitrogen) according to manufacturer’s instructions. Sets of specific primers (see supplemental Table 1) were used for amplification using 7900HT Real Time PCR System (AppliedBiosystems, Foster USA). Samples were normalized to GAPDH (tumors) or rplp0 (MEFs) using the ΔCt method. Statistical significance was determined with Student’s t-test. Results CD98hc deficiency inhibits Ras-induced tumorigenesis CD98hc expression is linked to malignancy (14,15,2) and CD98hc can regulate tumor-associated processes such as anchorage independence (8,4), nutrient transport (3,16), and mTOR (17). To assess CD98hc function in skin tumorigenesis, we generated mice with conditional deletion of CD98hc in the basal layer of the epidermis (K14CreERT2;CD98fl/fl) (11) and applied a two-stage chemical skin carcinogenesis protocol. Topical treatment with the carcinogen DMBA induced oncogenic Ras mutations, and subsequent repeated treatments with the phorbol ester TPA promoted outgrowth of initiated cells resulting in benign papillomas in control mice about 8 weeks after initiation (average of 3 papillomas/mouse at 10 weeks and 10 papillomas/mouse by 16 weeks) (Fig. 1B and 1C). Strikingly, papilloma formation was delayed by 8 weeks in CD98hc-deficient mice. Even after 25 weeks of TPA treatment, only 2 papillomas/mouse, on average, were detected on CD98hc deficient skin, whereas 10/mouse were observed in control mice (Fig. 1A and 1B). Tumor multiplicity was strongly reduced (Fig. 1A), suggesting that CD98hc promotes Ras-induced tumor formation and growth. Typical papillomas in vehicle-treated mice showed exophytic growth of squamous cells. In contrast, 4OHT-treated skin only occasionally manifested hyperplasia of the epidermis (Fig. 1D). CD98hc deletion was induced by 2 week-pretreatment with 4-hydroxy-tamoxifen (4OHT) compared to vehicle-treated skin (Fig. 1A and 1E). The rare papillomas or hyperplastic area, observed on skin treated with 4OHT, expressed CD98hc, (Fig. 1F), suggesting that they were formed by keratinocytes that had escaped Cre-mediated recombination. These data strongly suggest a tumor-promoting function of CD98hc in Ras-induced papillomas. TPA-associated inflammatory response in the skin is sustained in the absence of CD98hc Inflammation facilitates cancer development and growth. Because intestinal epithelia CD98hc is required for inflammation-induced tumorigenesis in the gut, promoting colitis-associated cancer in mice (15), we analyzed the effect of epidermal CD98hc loss in the inflammatory responses stimulated by topical TPA application, the known inflammatory inducer in skin model. First, we quantified the extent of skin edema by treating the ears with TPA and measuring their thickness. A strong reaction was observed following TPA application (vs. vehicle treated skin)(Fig. 2A and 2B), histologic analysis of the ears, prepared 6h after last TPA application, revealed no gross difference in ear thickening and swelling of WT versus CD98hc KO mice (Fig. 2A, 2B, and 2C). Besides, marked spongiosis and extensive infiltration of leukocytes were observed in the edematous dermis independent of CD98hc expression (Fig. 2A). Similar TPA-induced infiltration of inflammatory leukocytes was detected between CD98hc expressing and deficient mice (Fig. 2D). Thus, TPA-induced inflammatory response, ultimately leading to skin tumorigenesis, is CD98hc-independent. Altogether, these data strongly suggest that the dramatic effect of CD98hc loss on skin tumorigenesis relates to a cell autonomous effect. Loss of CD98hc causes Ras mutation-bearing tumor regression Because CD98hc has been shown to act intrinsically as cell proliferation enhancer both in tumoral and non-tumoral cells, Ras-induced tumorigenesis inhibition in CD98hc-deficient skin was expected. Thus, we analyzed the effect of CD98hc deficiency once tumors were already formed. Following either 11 or 23 weeks of DMBA/TPA (inducing respectively, early or large papillomas defined as a size of over 6mm (before SCC conversion), we treated mice with topical 4OHT or vehicle, and monitored succeeding tumor progression (Fig. 3A and 3B). In both cases and within 2 weeks (4OHT treatment period), not only further papilloma growth or conversion to SCC were totally abrogated, in spite of continued TPA application, but papillomas fully regressed, irrespective of papilloma initial size (Fig. 3A and 3B, regression within 5–7 weeks). In sharp contrast, in the first set-up, vehicle-treated mice exhibited progressive papilloma formation such that, at 19 weeks, they manifested an average of 18/mouse (Fig. 3A). In the second set-up, by 27 weeks post DMBA, SCC conversion occurred only in CD98hc expressing mice, with an average of one SCC per mouse, as judged by their flattened morphology and downward-appearing growth (Fig. 3B). Besides, these SCCs exhibited markedly increased CD98hc expression as compared to normal skin and a similar increased expression was observed in human SCC (Fig. 3C). The development of SCC in control mice was confirmed by histological analysis (Fig. 3D). Furthermore, increased CD98hc expression was a general feature of human SCC, being present in all 12 human SCC studies (GEO Profiles for SLC3A2), compiled over 310 samples. Thus, while CD98hc ablation in skin does not modulate apoptosis (11), it induces a drastic regression of Ras mutation-bearing tumors, preventing progression to SCC. CD98hc regulates the elasticity of the tumor microenvironment To assess how CD98hc deficiency impacts tumorigenesis, beyond its cell autonomous effect on proliferation and absence of extrinsic effect on TPA-induced inflammation, we analyzed mechanical features of the tumor microenvironment. A stiff and organized 3D microenvironment is important in tumor formation and tumor cell invasiveness (18) and we previously showed that, in fibroblasts, CD98hc can regulate RhoA (7), an important regulator of skin stiffness (13). Interestingly, CD98hc-deleted skin exhibited an atomic force microscopy (AFM) force map skewed toward low MPa values indicating more compliant tissue (Fig. 4A, 4B and Supp. Fig.S1A). The AFM nanoindentation of the epidermal CD98hc deficient skin revealed an elastic modulus of 0.063±0.01 MPa (n=4), while wt skin, elastic modulus in the papillary (upper) dermis reached 0.53±0.03 MPa (mean, n=3) (Fig. 4C). Although other methods have reported varying E values (19), (20), ’(21), our data are in good agreement with previous AFM studies (22) performed on skin. Thus, CD98hc deficiency decreases epidermal tissue stiffness. Highly aligned ECM molecules, in particular collagen fibers, increase matrix strength and stiffness in vivo. Skin sections of control and 4OHT treated mice stained with Masson’s trichrome (Supp. Fig.S1B) and sirius red (Fig. 4 D, 4E) highlight a major defect in fibrillar collagen, as seen using polarized filter, with a less organized collagen-rich dermis in CD98hc deficient skin (Supp. Fig.S1C–D). Similarly, abnormal fibronectin deposition was observed in KO skin. Therefore, we conclude that the loss of epidermal CD98hc results in more compliant skin due to drastic in vivo changes on collagen and fibronectin organization. CD98hc-mediates increased ROCK and YAP/TAZ signaling during tumorigenesis There is an intimate and bidirectional relationship between matrix organization/stiffness and RhoA-driven cellular contractility mediated by its effector Rho Kinase (ROCK) (13,19) : cellular contractility stimulates matrix assembly (23) and a stiffer matrix stimulates RhoA activity (24). We previously showed that, loss of CD98hc induces a reduction of RhoA activation in unchallenged skin (11). We therefore assessed the effect of CD98hc deletion in keratinocytes on RhoA signaling in papillomas as judged by the ROCK-mediated phosphorylation of the regulatory subunit of myosin phosphatase (Mypt) (Fig. 5A). Papillomas exhibited extensive phospho-Mypt in the epithelium, particularly in suprabasal layers. In contrast phospho-Mypt was nearly absent in the few papillomas derived from 4OHT-treated Slc3a2fl/fl-K14CreERT2 mice (Fig. 5A). Similarly, phosphorylation of cofilin, an indirect downstream target of ROCK, was lost in 4OHT-treated- compared to vehicle treated-skins (phospho-cofilin staining, Supp. Fig. S2). Thus, in contrast to wound healing with a soft provisional matrix (11), CD98hc deletion markedly reduced ROCK activity in papillomas associated with a reduction in matrix stiffness. ROCK regulates YAP/TAZ (Yes-associated protein/Transcriptional co-activator with PDZ-binding motif), which is an important transcriptional mechanism that controls production of fibrogenic factors, such as CTGF (connective tissue growth factor), in addition to controlling cell size, proliferation, and malignant transformation (25). Moreover, recent work shows that cells can “read” ECM cues as a function of the activity of YAP/TAZ through RhoA/ROCK signaling (25). We, therefore, analyzed the effect of CD98hc depletion on YAP/TAZ transcriptional activity by assaying expression of endogenous target genes. Real-time PCR on papillomas revealed that there was a major reduction in transcripts of both mANKRD1 and mCTGF (~5 and ~3 fold reduction respectively), two well characterized YAP/TAZ targets (25) in 4OHT-treated compared to vehicle treated mice (Fig. 5B). As expected, mANKRD1 and mCTGF were even more highly expressed in SCC reaching levels about 5 fold greater than in papillomas (Supp. Fig. S3A). Thus, loss of basal keratinocyte CD98hc markedly suppresses both RhoA/ROCK signaling and expression of YAP/TAZ regulated genes. CD98hc is a gain amplifier of stiffness sensing that enables a RhoA-driven self-reinforcing feedback loop that increases matrix stiffness Integrins play a central role in stiffness sensing (26),,(27). We previously reported that CD98hc expression modulates integrin signaling which can influence the growth of teratocarcinomas (4). This suggests that CD98hc might alter the cellular responses to the tumor microenvironment. Cell spreading is controlled by extracellular matrix stiffness (28). CD98hc null cells exhibited a striking defect in stiffness sensing. Specifically, when adhering to fibronectin covalently bound to silicone gels of varying compliance, CD98hc null fibroblasts failed to spread on 3.5 kPa substrates, an elasticity in the range of those found in skin (19) (Fig. 6A and 6B). Similarly, activation of two integrin-regulated promoters of cell survival, proliferation, and tumor cell invasion, pp125FAK and AKT, was reduced (Fig. 6C). These defects were specifically rescued in cells expressing full length human CD98hc (Fig. 6D). Remarkably, adhesion to stiffer substrates (30 and 300 kPa), resulted in partial restoration of both cell spreading and biochemical signaling in the CD98hc null fibroblasts. Loss of CD98hc in vivo markedly suppresses RhoA/ROCK signaling resulting in decreased transcription of YAP/TAZ regulated genes (Fig. 5 and Supp. S3A). Direct activation of Rho with cytotoxic necrotizing factor-1e (CNF-1) in CD98hc-deficient cells rescued YAP/TAZ activation (Supp. Fig. S3B); conversely, its inhibition by exoenzyme C3 in WT cells blocked YAP/TAZ driven gene expression (Supp. Fig. S3C), thus we propose Actin-Rho/ROCK-matrix stiffness-YAP/TAZ cascade is critical downstream of CD98hc. To assess whether effects on stiffness sensing were mediated through integrin interactions, we tested the capacity of reconstitution of the null cells with wild type CD98hc or with mutants (Fig. 7A and 7B). Both wild type and a mutant (C98T98E69) that couples only to integrins rescued the defect in stiffness sensing in the CD98hc fibroblasts. In contrast mutants that support amino acid transport but do not associate with integrins (C69T98E98 and C98T69E98) failed to rescue (Fig. 7A and 7B). Skin tumors arise from the underlying basal-layer cells, thus we tested the reconstitution of CD98hc deficient keratinocytes (Supp. Fig. S4) and found, similarly that, only mutant coupling with integrins (C98T98E69) rescued the defect in stiffness sensing. These data show that CD98hc, via its interaction with integrins, acts as a gain amplifier of stiffness sensing. Discussion In this study, we examined the sensitivity of the epidermal CD98hc-deficient mouse to Ras-driven skin carcinogenesis. Mice lacking epidermal CD98hc were protected against chemically-induced papilloma formation, even though TPA-associated inflammation response was preserved. Beyond this protection linked to CD98hc cell autonomous effect on cell proliferation, we demonstrate, for the first time, that when CD98hc deletion is induced after papillomas are formed, tumor regression is observed. Stiffness of the tumor microenvironment is crucial during cancer growth and integrins are involved in stiffness sensing. We reveal a new role for CD98hc as an amplifier of integrin-mediated cellular stiffness sensing leading to both ROCK-mediated increased matrix stiffness and nuclear relay of mechanical signals via YAP/TAZ signaling. Thus, CD98hc is a central gain amplifier in a self-reinforcing feedback loop that regulates both matrix stiffness and the cellular responses to stiffness. CD98hc is highly expressed in many tumors of different origins as well as in established tumor cell lines. Our results showing that loss of CD98hc protects against tumor formation is in good agreement with recent studies on more invasive and non-accessible tumors from the intestinal epithelium (15,29). Indeed, Nguyen et al. showed that CD98hc role in the gut tumorigenesis is partly due to an effect on epithelial intrinsic cell proliferation. We previously showed that the integrin-interacting function of CD98hc can mediate cellular proliferation in teratocarcinomas and lymphocytes (4,30). Notably, we report that, conversely to the gut, CD98hc in the skin does not perturb TPA-induced immune responses, involved during tumorigenesis. Furthermore, we show, for the first time, that not only CD98hc acts on tumor progression, but that its loss causes tumor regression. Importantly, CD98hc also forms the heavy chain of an amino acid co-transporter that can support the nutrient requirements of tumors (31) and consequent mTOR activation (17). Thus, it is likely that the capacity of CD98hc to support metabolic re-programming combined with its capacity to amplify mechanotransduction contribute to its critical role both in the clonal expansion of lymphocytes (30,32) and in tumorigenesis. The tumor microenvironment is a mechanically tunable meshwork comprising extracellular matrix (ECM) proteins, a heterogeneous population of stromal cells and secreted soluble factors. The modifications of the mechanical properties of this microenvironment, which are monitored by ECM receptors such as integrins, play an important role during tumor progression (33). Here, we provide a novel function of CD98hc independent of its required role in immune response linked to epithelial tumorigenesis. Our data are in good agreement with studies on breast carcinoma, showing the critical role of integrin-signaling in mechanosensing during tumorigenesis. CD98hc function as a co-receptor of integrins probably participates to integrin-signaling in mechanosensing in many cancer types. SCC is the second most common skin cancer (1) and its pathogenesis is tied to integrin expression (34),35). Elegant studies point to an important role for keratinocyte α5β1 integrin rather than α3β1 or α2β1 in SCCs formation (36). Most importantly, these studies show that the tumors arise from the underlying basal-layer cells (37). Newer studies point to dysfunction in environmental sensing, an integrin-dependent process, as a contributor to tumor formation and progression (19,38,39) . We now find that CD98hc loss drastically compromises environment stiffness sensing and conversely leads to a marked reduction in matrix stiffness. Importantly, the physiological elasticity of skin has been long evaluated, using other types of technologies, to range between 5 and 10 kPa (20,21,22), gel compliance we used in our in vitro studies. In the AFM settings we utilized, this value reaches 0.53 MPa on average. This apparent discrepancy can be explained by recent work from Akhtar et al. (40) showing that elastic modulus of the constituent ECM molecules is higher than the one of the connective tissues themselves. The dermis is of fibrillar nature, and the nanoscopic AFM probe we used in vivo is likely to find very distinct areas, such as fibers, upon which to indent. This leads to measurement of elastic modulus reaching values in the MPa range. Finally, in good agreement with the explanation above, our data suggest that, in the absence of CD98hc, the ECM components responsible for structural rigidity, such as collagen content, might be modulated, leading to an overall decrease in stiffness and altering the skin mechanical properties Integrin-mediated adhesions are intrinsically mechanosensitive and a large body of data implicates integrins in sensing mechanical forces (39,41). Piccolo’s laboratory pioneered the notion that mechanical signals and cell shape are converted into biochemical responses by two related transcription factors, YAP and TAZ (26,42). In particular, YAP and TAZ are nuclear relays of mechanical signals exerted by ECM rigidity (25) and mCTGF, a well characterized YAP/TAZ target, induces collagen deposition. CD98hc loss reduces ROCK activity in vivo resulting in decreased signaling of YAP/TAZ, and concomitantly a decrease of mCTGF expression, suggesting a role for CD98hc in collagen deposition. Importantly, cells deficient for CD98hc lose their capacity to ‘read’ physical and mechanical cues. The absence of CD98hc makes cells on a stiff ECM behave as if they were on a soft one. Thus, CD98hc tunes cellular stiffness sensing during tumorigenesis via the integrin/ROCK/YAP/TAZ pathway. In summary, our data establish a central role for CD98hc in chemically-induced tumor development and progression in skin and show that removal of CD98hc can induce tumor regression. We demonstrate that CD98hc mediates this process by acting as an amplifier of integrin-mediated mechanotransduction and that the gain in stiffness sensing provided by CD98hc forms the core of a positive feedback loop that increases the stiffness of the microenvironment and the cellular responses. Lastly, we implicate YAP/TAZ as a RhoA/ROCK regulated pathway that contributes to CD98hc-mediated tumorigenesis. These studies also highlight the potential synergic effect of blocking CD98hc, in an anti-cancer therapy, allowing targeting both tumor cell proliferation and tumor environment matrix stiffness. Supplementary Material Supplementary Data Supplementary Methods We thank the IRCAN core facilities (microscopy and animal facility) for technical help. This study was supported by grants from INSERM InCa/AVENIR (R08227AS), from Association pour la Recherche sur le Cancer (ARC R10159AA), and from Agence Nationale de la Recherche (ANR R09101AS), and through the “Investments for the Future” LABEX SIGNALIFE : program reference # ANR-11-LABX-0028-01. E. Boulter was the recipient of a postdoctoral fellowship from the Ligue Nationale Contre le Cancer (RAB 12007 ASA) and a Marie Curie IRG from the European Union FP7 (agreement 276945). Figure 1 CD98hc deficient epidermis is protected against tumor formation in vivo A. Number of papillomas per mouse is shown over the course of the experiment. Mice (8 weeks old) were treated 6 times with 4OHT before the start of the chemical carcinogenesis (DMBA/TPA) protocol. Data are mean (±SEM) of 2 independent experiments. (n=10 per group). B. Representative pictures at 16 weeks post initiation of mice from vehicle- (Ctrl) and 4OHT-treated group. C. Back skin sections, stained with H&E, isolated from vehicle- (Ctrl) or 4OHT-treated mice (as indicated) after 30 weeks of promotion, bearing or not tumors. Inserts are shown in D and E. Scale bar 200µm. D, E. Immunofluorescence analysis of skin samples from vehicle- (Ctrl) or 4OHT-treated mice (as indicated) with antibody against CD98hc. Dotted line indicates the dermis-epidermis junction. Scale bar 100µm. Figure 2 TPA-induced inflammatory response is preserved in absence of CD98hc A. Measurement of TPA-induced ear swelling. Ears of CD98hc- expressing or deficient mice were treated with TPA or vehicle alone. Seventy two hours after the first treatment, ears were collected. H&E stainings of representative ear cross sections are shown. Bar 100µm. B. Ear thickness of each mouse is shown (black dot represents value for each mouse). *** p<0.001 between TPA vs. vehicle. C. Time course of TPA-induced ear swelling. Extend of ear swelling was calculated as indicated in Mat. & Meth (n=3 per group). No significant difference was observed in the presence or absence of epidermal CD98hc. D. Quantification of the densities of infiltrated granulocytes of the ear of each mouse 72 hours after the first TPA treatment (each cross sections, as represented in A, were quantified), black dot represents value for each mouse). *** p<0.001 between TPA vs. vehicle. Figure 3 CD98hc loss induces tumor regression A. Number of papillomas per mouse is shown over the course of the experiment. Mice (8 weeks old) were subjected to the chemical carcinogenesis protocol DMBA/TPA. Six 4OHT-(or vehicle-) treatments were applied on the back skin of treated group 12 weeks after the DMBA application, corresponding to the first papillomas apparition. Data are mean (±SEM) of two independent experiments. n=6 per group. B. Number of tumors per mouse is shown over the course of the experiment. Mice (8 weeks old) were subjected to the chemical carcinogenesis protocol DMBA/TPA. Six 4OHT- (or vehicle-) treatments were applied on the back skin of treated group 22 weeks after the DMBA application, corresponding to papilloma conversion to malignant SCC. Only large papillomas, defined as a size of over 6mm, are shown. Data are mean ±SEM representative of n=10 for vehicle-treated (Ctrl) group and n=14 for 4OHT-treated group. C. Immunofluorescence analysis of SCC (upper panels) and normal skin (lower panels) samples from either murine or human origin with antibody against CD98hc, showing strong overexpression of CD98hc in SCC. Scale bars as indicated. D. Morphology of SCC tumor (left panel) and normal skin (right panel) sections shown by H&E staining. Scale bars as indicated. Figure 4 Loss of CD98hc increases tissue elasticity by modulating ECM organization A. AFM maps of the elastic moduli of vehicle- (Ctrl) and 4OHT-treated skin corresponding to the areas enclosed by yellow squares (left panels). Right panels show heat maps of the same 20 × 20 µm scanned areas. B. Representative force-separation curves of vehicle (Ctrl) and 4OHT -treated samples. C. Quantification of Young’s elastic Modulus in vehicle (n=3) and 4OHT (n=4) treated skin, expressed as mean ± SEM, ***p value<0.001). D. Defect in fibrillar collagen in 4OHT-treated skin shown by sirius red staining (right panel corresponds to polarized filter which highlights fibrillar collagen). Scale bars 100µm and 50 µm, respectively. E. Quantification of fibrillar collagen in 4OHT- vs. vehicle- (Ctrl) treated skin confirming a less organized collagen-rich dermis in CD98hc deficient skin. Figure 5 CD98hc loss reduces Rho Kinase (ROCK) activity in papillomas resulting in decreased YAP/TAZ Signaling A. Immunohistochemical analysis of murine papilloma sections from vehicle- (Ctrl) and 4OHT-treated mice with antibodies against MYPT1 and P-MYPT1, displaying strong expression across the epidermis, but decreased phosphorylation status in CD98hc deficient papillomas. Scale bar 200µm. B. Quantitative RT-PCR of YAP/TAZ signaling markers (mANKRD1 and mCTGF) from papilloma isolated from vehicle- (Ctrl) and 4OHT-treated mice. GAPDH is shown as control. Data shown are mean ±SEM (n=15 per group), *p< 0.01, **p≤0.001. C. Quantitative RT-PCR of CD98hc in vehicle- (Ctrl) vs. 4OHT-treated papillomas. (n=5 per group) Data are shown as mean ± SEM, ***p< 0.001. Figure 6 CD98hc is a central gain amplifier of stiffness sensing A. Actin staining of WT (CD98hcfl/fl) or CD98hc-deficient (CD98hc−/−) cells seeded on silicone gels displaying variable stiffness. Scale bar 20 µm. B. Quantification of cell area according to matrix stiffness. Data represented are mean of 3 independent experiments. Error Bars are SEM. C. Western blot analysis of FAK and Akt level of phosphorylation in WT or CD98hc-deficient cells according to matrix stiffness. Total FAK, Akt, and α -tubulin are shown as controls. D. Western blot analysis of FAK and Akt level of phosphorylation in CD98hc-deficient cells reconstituted with full length human CD98hc (hCD98hc) or CD69 (control). Total FAK, Akt, and α-tubulin are shown as controls. Figure 7 CD98hc capacity to stiffness sensing is mediated through integrin signaling A. CD98hc-deficient cells reconstituted with different chimeras (GFP positive) were seeded on 30 kPa stiffness silicone gels. Model of chimeras of CD98hc and another type II transmembrane protein (CD69), is shown. CD98hc protein is depicted in gray, and CD69 is depicted in black. Each chimera is defined by its cytoplasmic (C), transmembrane (T), and extracellular (E) domain derived from either CD98hc (98) or CD69 (69). CD98hc extracellular domain is necessary and sufficient for amino acid transport, whereas the intracellular and transmembrane domains are required for interactions with integrins. Actin cytoskeleton was observed using phalloidin staining. Quantification of cell surface area of transfected cells (GFP+) vs. Mock (GFP-), according to the matrix stiffness, is shown. Data represented are mean of 3 independent experiments. Error Bars are SEM. Scale bar 10 µm. B. Western blot analysis of FAK and Akt level of phosphorylation in CD98hc-deficient cells (CD98hc−/−) reconstituted with chimeras and seeded on 30 kPa silicone gels. The authors disclose no potential conflicts of interest. REFERENCES 1 Xie J Reichrath J Molecular Biology of Basal and Squamous Cell Carcinomas Sunlight, Vitamin D and Skin Cancer [Internet] 2008 New York Springer 241 251 cited 2013 Apr 26 Available from: http://link.springer.com/chapter/10.1007/978-0-387-77574-6_19 2 Cantor JM Ginsberg MH CD98 at the Crossroads of Adaptive Immunity and Cancer J Cell Sci [Internet] 2012 cited 2012 Apr 19 Available from: http://jcs.biologists.org/content/early/2012/04/10/jcs.096040 3 Verrey F Closs EI Wagner CA Palacin M Endou H Kanai Y CATs and HATs: the SLC7 family of amino acid transporters Pflugers Arch 2004 447 532 542 14770310 4 Feral CC Nishiya N Fenczik CA Stuhlmann H Slepak M Ginsberg MH CD98hc (SLC3A2) mediates integrin signaling Proc Natl Acad Sci USA 2005 102 355 360 15625115 5 Hara K Kudoh H Enomoto T Hashimoto Y Masuko T Malignant transformation of NIH3T3 cells by overexpression of early lymphocyte activation antigen CD98 Biochem Biophys Res Commun 1999 262 720 725 10471392 6 Shishido T Uno S Kamohara M Tsuneoka-Suzuki T Hashimoto Y Enomoto T Transformation of BALB3T3 cells caused by over-expression of rat CD98 heavy chain (HC) requires its association with light chain: mis-sense mutation in a cysteine residue of CD98HC eliminates its transforming activity Int J Cancer 2000 87 311 316 10897033 7 Féral CC Zijlstra A Tkachenko E Prager G Gardel ML Slepak M CD98hc (SLC3A2) participates in fibronectin matrix assembly by mediating integrin signaling J Cell Biol 2007 178 701 711 17682053 8 Rintoul RC Buttery RC Mackinnon AC Wong WS Mosher D Haslett C Cross-Linking CD98 Promotes Integrin-like Signaling and Anchorage-independent Growth Mol Biol Cell 2002 13 2841 2852 12181350 9 Henderson NC Collis EA Mackinnon AC Simpson KJ Haslett C Zent R CD98hc (SLC3A2) interaction with beta 1 integrins is required for transformation J Biol Chem 2004 279 54731 54741 15485886 10 Abel EL Angel JM Kiguchi K DiGiovanni J Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications Nat Protoc 2009 4 1350 1362 19713956 11 Boulter E Estrach S Errante A Pons C Cailleteau L Tissot F CD98hc (SLC3A2) regulation of skin homeostasis wanes with age J Exp Med [Internet] 2013 cited 2013 Jan 11 Available from: http://jem.rupress.org/content/early/2012/12/24/jem.20121651 12 Montanez E Piwko-Czuchra A Bauer M Li S Yurchenco P Fässler R Cheresh David A Analysis of Integrin Functions in Peri-Implantation Embryos, Hematopoietic System, and Skin Methods in Enzymology [Internet] 2007 Academic Press 239 289 cited 2013 Feb 4 Available from: http://www.sciencedirect.com/science/article/pii/S0076687907260124 13 Samuel MS Lopez JI McGhee EJ Croft DR Strachan D Timpson P Actomyosin-Mediated Cellular Tension Drives Increased Tissue Stiffness and β-Catenin Activation to Induce Epidermal Hyperplasia and Tumor Growth Cancer Cell 2011 19 776 791 21665151 14 Fuchs BC Bode BP Amino acid transporters ASCT2 and LAT1 in cancer: Partners in crime? Seminars in Cancer Biology 2005 15 254 266 15916903 15 Nguyen HTT Dalmasso G Torkvist L Halfvarson J Yan Y Laroui H CD98 expression modulates intestinal homeostasis, inflammation, and colitis-associated cancer in mice J Clin Invest 2011 121 1733 1747 21490400 16 Devés R Boyd CA Surface antigen CD98(4F2): not a single membrane protein, but a family of proteins with multiple functions J Membr Biol 2000 173 165 177 10667913 17 Nicklin P Bergman P Zhang B Triantafellow E Wang H Nyfeler B Bidirectional Transport of Amino Acids Regulates mTOR and Autophagy Cell 2009 136 521 534 19203585 18 Goetz JG Minguet S Navarro-Lérida I Lazcano JJ Samaniego R Calvo E Biomechanical remodeling of the microenvironment by stromal caveolin-1 favors tumor invasion and metastasis Cell 2011 146 148 163 21729786 19 Butcher DT Alliston T Weaver VM A tense situation: forcing tumour progression Nat Rev Cancer 2009 9 108 122 19165226 20 Jachowicz J McMullen R Prettypaul D Indentometric analysis of in vivo skin and comparison with artificial skin models Skin Res Technol 2007 13 299 309 17610652 21 Moore SW Roca-Cusachs P Sheetz MP Stretchy Proteins on Stretchy Substrates: The Important Elements of Integrin-Mediated Rigidity Sensing Developmental Cell 2010 19 194 206 20708583 22 Crichton ML Donose BC Chen X Raphael AP Huang H Kendall MAF The viscoelastic, hyperelastic and scale dependent behaviour of freshly excised individual skin layers Biomaterials 2011 32 4670 4681 21458062 23 Wu C Keivens VM O’Toole TE McDonald JA Ginsberg MH Integrin activation and cytoskeletal interaction are essential for the assembly of a fibronectin matrix Cell 1995 83 715 724 8521488 24 Paszek MJ Zahir N Johnson KR Lakins JN Rozenberg GI Gefen A Tensional homeostasis and the malignant phenotype Cancer Cell 2005 8 241 254 16169468 25 Dupont S Morsut L Aragona M Enzo E Giulitti S Cordenonsi M Role of YAP/TAZ in mechanotransduction Nature 2011 474 179 183 21654799 26 Schwartz MA Integrins and Extracellular Matrix in Mechanotransduction Cold Spring Harb Perspect Biol [Internet] 2010 2 cited 2013 Jan 23 Available from: http://cshperspectives.cshlp.org/content/2/12/a005066 27 Galbraith CG Yamada KM Sheetz MP The relationship between force and focal complex development J Cell Biol 2002 159 695 705 12446745 28 Mih JD Marinkovic A Liu F Sharif AS Tschumperlin DJ Matrix stiffness reverses the effect of actomyosin tension on cell proliferation J Cell Sci 2012 125 5974 5983 23097048 29 Nguyen HTT Dalmasso G Yan Y Laroui H Charania M Ingersoll S Intestinal epithelial cell-specific CD98 expression regulates tumorigenesis in Apc(Min/+) mice Lab Invest 2012 92 1203 1212 22641098 30 Cantor J Browne CD Ruppert R Féral CC Fässler R Rickert RC CD98hc facilitates B cell proliferation and adaptive humoral immunity Nat Immunol 2009 10 412 419 19270713 31 McCracken AN Edinger AL Nutrient transporters: the Achilles’ heel of anabolism Trends in Endocrinology & Metabolism 2013 24 200 208 23402769 32 Cantor J Slepak M Ege N Chang JT Ginsberg MH Loss of T cell CD98 H chain specifically ablates T cell clonal expansion and protects from autoimmunity J Immunol 2011 187 851 860 21670318 33 Levental KR Yu H Kass L Lakins JN Egeblad M Erler JT Matrix Crosslinking Forces Tumor Progression by Enhancing Integrin Signaling Cell 2009 139 891 906 19931152 34 Janes SM Watt FM New roles for integrins in squamous-cell carcinoma Nat Rev Cancer 2006 6 175 183 16498442 35 Owens DM Broad S Yan X Benitah SA Watt FM Suprabasal α5β1 integrin expression stimulates formation of epidermal squamous cell carcinomas without disrupting TGFβ signaling or inducing spindle cell tumors Molecular Carcinogenesis 2005 44 60 66 15924349 36 Owens DM Watt FM Influence of β1 Integrins on Epidermal Squamous Cell Carcinoma Formation in a Transgenic Mouse Model α3β1, but not α2β1, Suppresses Malignant Conversion Cancer Res 2001 61 5248 5254 11431366 37 Owens DM Watt FM Contribution of stem cells and differentiated cells to epidermal tumours Nat Rev Cancer 2003 3 444 451 12778134 38 Kumar S Weaver VM Mechanics, malignancy, and metastasis: the force journey of a tumor cell Cancer Metastasis Rev 2009 28 113 127 19153673 39 Geiger B Spatz JP Bershadsky AD Environmental sensing through focal adhesions Nature Reviews Molecular Cell Biology 2009 10 21 33 19197329 40 Akhtar R Schwarzer N Sherratt MJ Watson REB Graham HK Trafford AW Nanoindentation of histological specimens: Mapping the elastic properties of soft tissues J Mater Res 2009 24 638 646 20396607 41 Katsumi A Orr AW Tzima E Schwartz MA Integrins in Mechanotransduction J Biol Chem 2004 279 12001 12004 14960578 42 Zhao B Wei X Li W Udan RS Yang Q Kim J Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control Genes Dev 2007 21 2747 2761 17974916

PMC005xxxxxx/PMC5120727.txt

LICENSE: This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. 101692909 45744 J Rare Dis Res Treat Journal of rare diseases research & treatment 27891535 5120727 NIHMS825771 Article Gene therapy for hemoglobin disorders - a mini-review Rai Parul 1 Malik Punam 12* 1 Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA 2 Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH 45229, USA * Correspondence: Punam Malik, MD, Cincinnati Children’s Hospital Medical Center, Division of Experimental Hematology and Cancer Biology and the Division of Hematology, Cancer and Blood Disease Institute, Cincinnati Children’s Hospital Medical Center 3333 Burnet Ave, Cincinnati OH 45229, USA, punam.malik@cchmc.org 10 11 2016 2016 23 11 2016 1 2 2531 This file is available for text mining. It may also be used consistent with the principles of fair use under the copyright law. Gene therapy by either gene insertion or editing is an exciting curative therapeutic option for monogenic hemoglobin disorders like sickle cell disease and β-thalassemia. The safety and efficacy of gene transfer techniques has markedly improved with the use of lentivirus vectors. The clinical translation of this technology has met with good success, although key limitations include number of engraftable transduced hematopoietic stem cells and adequate transgene expression that results in complete correction of β0 thalassemia major. This highlights the need to identify and address factors that might be contributing to the in-vivo survival of the transduced hematopoietic stem cells or find means to improve expression from current vectors. In this review, we briefly discuss the gene therapy strategies specific to hemoglobinopathies, the success of the preclinical models and the current status of gene therapy clinical trials. Gene Hemoglobinopathies Sickle cell anemia Thalassemia Hemoglobin Therapy Review Disorders Introduction Sickle cell disease (SCD) and β-thalassemia are autosomal recessive disorders that result in qualitative and quantitative defects in β-globin protein production; and are highly prevalent worldwide, with approximately 7% of the global population estimated to be carriers of hemoglobin gene-variants, and over 330,000 affected infants born annually with SCD alone1. Despite improved medical supportive therapies, significant long-term mortality and morbidity associated with hemoglobinopathies remains2,3. Hematopoietic stem cell transplant (HSCT) provides the only definitive cure, with a disease free survival exceeding 80% with HLA-matched sibling donor transplants4,5. Improvements in management of graft versus host disease (GVHD) and better means of inducing graft tolerance have encouraged the use of an extended donor pool comprising of unrelated donors and umbilical cord blood as the hematopoietic stem cell (HSC) source for patients lacking a matched sibling donor6. However, matched-unrelated HSCT for thalassemia have an overall survival of 65% in the high risk patients7. In addition, a 5–10% mortality from transplant-conditioning, GVHD and graft failure continues to limit the acceptability of this treatment modality8. Gene therapy, using genetically-modified autologous HSCs is an attractive alternative to allogeneic HSCT, and specifically unrelated-donor HSCT, since it eliminates the need for a matched donor and the risk of GVHD/graft rejection. Successful gene therapy for monogenic immune disorders like chronic granulomatous disease and severe combined immunodeficiency9–13 has encouraged development of this technology for hemoglobinopathies. Whereas in immunodeficiency disorders, the genetically modified hematopoietic progenitors and T cells have a selective survival advantage and a tremendous expansion potential, respectively, requiring a minimal (0.1–1%) gene corrected HSC engraftment for sustained correction of lymphoid dysfunction14; in hemoglobinopathies, no such survival advantage of genetically modified HSCs/progenitors is present, and selective advantage is limited to the terminal erythroid cells15. Post-transplant follow-up studies16 and murine models17,18 show that a 20% donor chimerism is essential for improving the clinical manifestations in SCD and thalassemia, a level requiring substantial pre-transplant chemotherapy conditioning. Furthermore, in order for globin gene transfer to affect a cure, high level erythroid lineage specific expression is necessary. Despite these hurdles, improved vector potency and safety have significantly advanced the field, resulting in cures in patients with Hemoglobin E-β-thalassemia and considerable disease amelioration in some patients with β0 thalassemia and SCD. The lessons learnt from these early gene therapy trials suggest that engraftment of sufficient transduced HSC, or their in-vivo selection could play a crucial role to extend the curative capacity of gene therapy. Vector development for hemoglobin disorders Correction of hemoglobin disorders by vector-mediated gene transfer requires utilization of a safe delivery vehicle/vector to efficiently transfer the complex β-transgene cassette to HSCs and result in sustained high expression of the transferred globin gene. The vectors commonly used have been bioengineered from different retroviruses, mainly murine Moloney leukemia virus (retrovirus vectors; RV), HIV-1 (lentivirus vectors; LV) and foamy virus, after removing the genetic elements responsible for their pathogenicity and virulence, and adding the β-globin gene and its locus control region (LCR) elements. Of these, LV have been most successful at correcting hemoglobinopathy animal models, and have resulted in their clinical translation. The β-globin LCR is a cis-regulatory element composed of five DNAase-1 hypersensitivity sites, four of which are formed in the erythroid cells19. When linked to the globin genes LCR leads to position-independent, erythroid lineage-specific enhancement of globin gene expression. The enhancer activity of LCR resides in three of its hypersensitivity sites HS 2, 3 and 4, which contain an array of binding sites for ubiquitous and erythroid specific transcription factors20. An intact LCR (5’HS 1–5) is involved in maintaining an open chromatin conformation that is needed for position independent expression of the globin genes. The LCR also results in developmental regulation of globin expression and interacts with the ε, γ and β globin gene promoters in the embryonic, fetal and adult stage, respectively21. Gamma-retrovirus (RV) vectors Initial studies looking at RV-mediated human β-globin gene transfer without inclusion of the LCR elements, showed variable and low levels of gene expression (<1% of endogenous β- globin expression)22. Following this study, nearly one decade of efforts to develop RV for expressing sufficient globin gene expression were futile23,24. RVs utilizing the enhancer/promoter sequences of the LTR (long terminal repeat) to drive transgene expression of genes other than globin genes, were the first ones to be used in clinical trials. Despite their initial clinical success in gene therapy of immune-deficiencies, concerns about their safety emerged following reports of vector-mediated insertional mutagenesis9–12. Integration site analysis revealed that the RVs have a tendency to integrate near cellular promoters, retroviral common integration sites (CIS) and cancer genes, independent of the vector design, and enhance their expression via the LTR promoter/enhancer. The RV vector insertions increase immortalization of primary hematopoietic progenitor cells25. While the RV LTR is a strong enhancer and upregulates transgene expression to very high levels compared to relatively weaker enhancers from the HIV LTR and cytomegalovirus26, it also simultaneously activates cellular proto- oncogenes flanking insertion sites27. Additionally, methylation of the LTR can lead to inactivation of the integrated transgene promoter and prevent long-term transgene expression28. The construction of a self-inactivating [SIN] vector design deletes the LTR promoter/enhancer and allows the transgene expression to be driven by internal cellular promoters, reduce LTR enhancer-mediated genotoxicity27 and its methylation-induced inactivation29. Inclusion of the chicken β-globin hypersensitive site-4 (cHS4) insulator element to the SIN vector further improved its safety by reducing position dependent variablility in gene expression30. However the inability of RV to transduce non-diving cells, along with vector instability seen with incorporation of large LCR sequences greatly limited their use in gene therapy for hemoglobin disorders31. Lentivirus vectors The interest in LV was generated with the increasing knowledge of the basic structure and properties of HIV-1 virus. HIV-1 can efficiently translocate the intact nuclear membrane and thus, has the ability to transduce non-dividing/quiescent cells and can carry larger expression cassettes. These features enabled the HIV-1 based LVs to be developed for hemoglobinopathies, and efficiently transfer the β-transgene/LCR to HSCs for sustained correction of the hemoglobin defect. The major safety concerns with the use of LVs initially were risk of generating a replication-competent lentivirus (RCL) and insertional mutagenesis. The former risk has been eliminated by removal of HIV regulatory and accessory genes from vector plasmids and constructing the vector with 3–4 separate packaging plasmids, with minimal overlapping sequences between and within them31. The preference for intragenic integration of LV vectors, coupled with the SIN design considerably reduces their genotoxicity potential; Indeed, recent LV clinical trials with a follow up of nearly 10 years have shown no genotoxicity resulting from LV vectors, even though, preclinical studies have reported vector integration in known oncogenes (MLL, NUP214)32 and suggested that transcriptionally active enhancers like the LCR can lead to gene dysregulation, independent of the vector type (RV or LV) or design (LTR-based or SIN)26,33. Our group explored the genotoxicity potential of LCR enhancer elements and showed that the LCR-containing LVs have approximately 200-fold lower immortalization potential than RVs. Though gene dysregulation was seen in the vicinity of the integrated vector, no proto-oncogene upregulation was noted34. Use of cHS4 insulators decreased the transforming potential further, along with decreasing position dependent variable gene expression and methylation-associated silencing35. Preclinical studies Gene Therapy for β-Thalassemia LV-mediated human β-gene transfer was shown to rescue mouse models of β-thalassemia intermedia and β-thalassemia major. May and colleagues demonstrated the use of a LV vector carrying the human β-globin gene fragment and β-globin LCR spanning the HS2, HS3, and HS4 regions to correct thalassemia intermedia in mice with increase in hemoglobin levels by 3–4 g/dl36. The same group developed an adult β0-thalassemia major mouse model using mice engrafted with beta-globin-null Hbb(th3/th3) fetal liver cells and rescued their severe phenotype using the same vector with an average vector copy number (VCN) of 1.0– 2.418. Imren and coworkers thereafter showed correction of β-thalassemia mice using a vector carrying the βT87Q gene, where a point mutation in the β-globin gene also confers it with anti-sickling properties. However, multiple copies were required for adequate correction of the mouse thalassemia phenotype32. Our group showed complete correction of the human β0 thalassemia phenotype in vitro and in a xenograft model with approximately 2 vector copies/cell37. Miccio and colleagues used a LV vector carrying the β-globin gene linked to a minimized LCR HS2/HS3. They showed that a frequency of 30–50% of transduced hematopoietic cells harboring an average VCN of 1 was sufficient to fully correct the thalassemia phenotype in th3/+ mice15. In addition they also demonstrated that the genetically corrected erythroblasts had an in vivo survival advantage, thus encouraging the need to explore the utility of reduced intensity transplant regimens for clinical gene therapy trials. Gene Therapy for SCD The efficacy of LV-mediated transfer of γ-globin gene/mutated β-globin genes (βT87Q and βAS3) for correcting SCD was explored using transgenic and humanized xenograft sickle cell murine models. Pawliuk and colleagues showed improvement in hematological parameters, splenomegaly and hyposthenuria in BERK and SAD mice using the βT87Q LV38. Levasseur on the other hand used a βAS3 (a human β-globin gene with 3 anti-sickling mutations) in a LV to successfully transduce murine HSC without cytokine stimulation39. Romero et al used the same βAS3 LV to successfully transduce bone marrow CD34 progenitor cells from patients with SCD, and produce sufficient levels of anti-sickling hemoglobin40. Persons et al41, and Pestina et al42 showed improvement in SCD phenotype by increasing the expression of fetal hemoglobin (HbF) using γ-globin LV. Our group showed an 18–20% engraftment of HSCs containing γ-globin gene-LV, following a non-myeloablative conditioning regimen. This donor chimerism was sufficient to result in approximately 60% circulating RBC containing HbF (F cells) with an improvement in the SCD manifestations43. Clinical trials The success in preclinical models, supported by safety studies on LV vectors led to the design of clinical gene therapy trials. Cavazzana-Calvo et al. enrolled a hemoglobin E/β (βE/β0)-thalassemia major patient in 2007 who received genetically modified autologous HSCs expressing βT87Q-globin following myeloablative busulfan conditioning. This subject became transfusion independent 1–2 years later44 with a hemoglobin maintained at 9–10 gm/dl. This therapeutic benefit was initially due to a clonal expansion observed following vector insertion in the HMGA2 gene. However this clone has eventually subsided. The trial was subsequently extended to include 18 patients with thalassemia (transfusion-dependent βEβ0 n=10, β0β0 n=5, β+thalassemia n= 3) and 4 patients with SCA45–47. All patients with βEβ0-thalassemia and β+thalassemia became transfusion independent within a year of transplant, with a median increase in hemoglobin by 4.9-g/dl, while patients with β0β0-thalassemia with a similar hemoglobin increase experienced a significant reduction in their transfusion requirement, but were not transfusion independent, since the baseline hemoglobin levels in β0β0thalassemia are much lower than in individuals with β+/βE thalassemia. One of four patients with SCD who received a high dose of transduced CD34 cells had remarkable improvement in their SCD phenotype. Two other trials are using the β-globin LV vectors for β-thalassemia. In the trial led by Boulad et al (NCT01639690), the preconditioning regimen had to be switched from a reduced-intensity busulfan to myeloablative doses following modest engraftment and β-globin expression with the lower dose48; the trial led by Ferrari et al (NCT02453477) is using a myeloablative regimen consisting of treosulfan and thiotepa with initial success48. The NCT02247843 trial led by Kohn et al. for SCD is investigating the efficacy of βAS3LV, and the trial led by our group (Malik et al.; NCT02186418) is using a γ-globin LV following reduced intensity conditioning. The results of these studies are eagerly awaited. Recent Advances in Genetic Manipulation Technology The emergence of gene editing technology, which enables precise genome manipulation, offers a new approach for treating β-hemoglobinopathies49. Site specific double strand breaks (DSB) can be induced with zinc finger nucleases, transcription activator-like effector nucleases (TALENS), meganucleases and more recently with Clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system. CRISPR/Cas9 has revolutionized gene targeting. Unlike other nucleases which use a protein dimer for target sequence recognition and require a novel protein to be engineered for each new target site, CRISPR/Cas9 technology uses a short guide RNA (gRNA) with a 20bp sequence complementary to the DNA sequence to be targeted50. In addition targeting/knockdown of multiple genes can be achieved by using multiple gRNAs with a common Cas9 protein51. DSB is then followed by DNA repair through one of the two major pathways: 1) Non-homologous end joining (NHEJ) with direct fusion of the nuclease cleaved ends. This repair mechanism is error-prone, leading to indels, and is cell-cycle stage independent52. 2) Homologous directed repair (HDR) uses an exogenous donor template53 delivered via single-stranded oligonucleotides, plasmids, or viral vectors like integrase deficient lentivirus or adeno-associated virus54, for gene correction with targeted insertion. For hemoglobinopathies, gene editing strategizes shown to be successful include either induction of endogenous fetal hemoglobin55,56, modification of the causal β-globin gene mutation by targeted nucleases57 or therapeutic transgene integration58, or a combined approach59,60. Inactivation of an erythroid specific enhancer of BCL11A by gene editing leads to suppression of BCL11A and up-regulation of γ-globin in erythroid lineage cells61,62. These gene editing strategizes are being performed in CD34+ stem and progenitor cells63 or in induced pluripotent stem cells (iPSCs) capable of differentiating into any somatic cell type64–66. Patient-specific iPSCs are generated by the genetic reprogramming of their somatic cells, and provide an unlimited source of stem cells which can be genetically manipulated, differentiated along a specific tissue type and returned back to the patient. Currently, active research in differentiating iPSC towards definitive hematopoietic stem cells with long term engraftment potential is underway. In addition to their therapeutic potential, iPSCs can also be used as in-vitro disease models67. Off-target nuclease binding activity68, efficient means of delivering the genome editing tools to target stem cell populations without loss of ‘stemness’, genomic variation occurring with somatic reprogramming, efficient gene targeting by homology directed repair69, and developing functional HSCs from these genetically modified iPSC70,71 are some of the challenges in the field. Conclusions and future directions Gene therapy for hemoglobinopathies is now a reality, with several patients cured of their β0/βE thalassemia or with significant amelioration from β0/β0 thalassemia and one patient with SCD, while others are showing modest transgene expression. The current curative capacity of gene transfer technology is limited by the severity of the underlying disease. The increase in hemoglobin to 8–9 gm/dl seen in β0β0 thalassemia is still not sufficient to prevent ineffective erythropoiesis, and hence these subjects are still intermittently transfused. However the overall transfusion burden in this patient population has decreased dramatically. Cohen et al showed that the success of chelation therapy in achieving a neutral or negative iron balance (assessed by liver iron concentration) had a significant correlation to the transfusion iron intake72. Thus decreasing the transfusion burden is advantageous, as it not only might affect the dose of chelation therapy used but also affects its outcome. The challenges to efficacious clinical translation in hemoglobinopathies include the dose of engraftable transduced HSCs, the intensity of the preconditioning transplant regimen, and expression of the transgene. In-vivo selection strategies can ensure expansion of the few genetically-modified engrafted HSCs. Improving vector potency will augment gene expression. Efforts to promote differentiation of iPSC technology to produce engraftable HSC can expand the HSC source, and gene editing can circumvent the need for high transgene expressing-LV and potential, albeit low, insertional genotoxicities of LV. New technologies that can reshape the future of gene therapy are gene editing using CRISPR/CaS9 and development of hematopoietic stem cells from iPSCs with long term repopulating potential, although much work is needed to make this a reality. With scientific advancements in stem cell biology and genetic manipulation, we envision a future where a child prenatally diagnosed with hemoglobinopathy can have his/her genetically modified cord blood stem cells transfused even before the fetal to adult hemoglobin switch, thus preventing the occurrence of any disease manifestations. This study is supported by the Doris Duke Foundation Innovations in Clinical Research Award to PM, and the NIH-NHLBI Excellence in Hemoglobinopathy Research Award (EHRA) program (U01HL117709). References 1 Modell B Darlison M Global epidemiology of haemoglobin disorders and derived service indicators Bull World Health Organ 2008 86 6 480 487 18568278 2 Rachmilewitz EA Giardina PJ How I treat thalassemia Blood 2011 118 13 3479 3488 21813448 3 Sharef SW Al-Hajri M Beshlawi I Al-Shahrabally A Elshinawy M Zachariah M Optimizing Hydroxyurea use in children with sickle cell disease: low dose regimen is effective Eur J Haematol 2013 90 6 519 524 23489171 4 Shenoy S Hematopoietic Stem Cell Transplantation for Sickle Cell Disease: Current Practice and Emerging Trends Hematol-Am Soc Hemat 2011 273 279 5 Angelucci E Hematopoietic stem cell transplantation in thalassemia Hematology Am Soc Hematol Educ Program 2010 2010 456 462 21239835 6 Walters MC Patience M Leisenring W Rogers ZR Aquino VM Buchanan GR Stable mixed hematopoietic chimerism after bone marrow transplantation for sickle cell anemia Biol Blood Marrow Tr 2001 7 12 665 673 7 La Nasa G Argiolu F Giardini C Pession A Fagioli F Caocci G Unrelated bone marrow transplantation of beta-thalassemia patients - the experience of the Italian Bone Marrow Transplant Group Ann Ny Acad Sci 2005 1054 186 195 16339665 8 Bhatia M Walters MC Hematopoietic cell transplantation for thalassemia and sickle cell disease: past, present and future Bone Marrow Transplant 2008 41 2 109 117 18059330 9 Hacein-Bey-Abina S Von Kalle C Schmidt M McCormack MP Wulffraat N Leboulch P LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1 Science 2003 302 5644 415 419 14564000 10 Gaspar HB Parsley KL Howe S King D Gilmour KC Sinclair J Gene therapy of X-linked severe combined immunodeficiency by use of a pseudotyped gammaretroviral vector Lancet 2004 364 9452 2181 2187 15610804 11 Ott MG Schmidt M Schwarzwaelder K Stein S Siler U Koehl U Correction of X-linked chronic granulomatous disease by gene therapy, augmented by insertional activation of MDS1-EVI1, PRDM16 or SETBP1 Nat Med 2006 12 4 401 409 16582916 12 Hacein-Bey-Abina S Garrigue A Wang GP Soulier J Lim A Morillon E Insertional oncogenesis in 4 patients after retrovirus-mediated gene therapy of SCID-X1 J Clin Invest 2008 118 9 3132 3142 18688285 13 Aiuti A Cattaneo F Galimberti S Benninghoff U Cassani B Callegaro L Gene Therapy for Immunodeficiency Due to Adenosine Deaminase Deficiency New Engl J Med 2009 360 5 447 458 19179314 14 Neff T Beard BC Kiem HP Survival of the fittest: in vivo selection and stem cell gene therapy Blood 2006 107 5 1751 1760 16269617 15 Miccio A Cesari R Lotti F Rossi C Sanvito F Ponzoni M In vivo selection of genetically modified erythroblastic progenitors leads to long-term correction of beta-thalassemia Proc Natl Acad Sci U S A 2008 105 30 10547 10552 18650378 16 Gaziev J Lucarelli G Stem cell transplantation for hemoglobinopathies Curr Opin Pediatr 2003 15 1 24 31 12544268 17 Persons DA Allay ER Sabatino DE Kelly P Bodine DM Nienhuis AW Functional requirements for phenotypic correction of murine beta-thalassemia: implications for human gene therapy Blood 2001 97 10 3275 3282 11342459 18 Rivella S May C Chadburn A Riviere I Sadelain M A novel murine model of Cooley anemia and its rescue by lentiviral-mediated human beta-globin gene transfer Blood 2003 101 8 2932 2939 12480689 19 Li Q Zhang M Duan Z Stamatoyannopoulos G Structural analysis and mapping of DNase I hypersensitivity of HS5 of the beta-globin locus control region Genomics 1999 61 2 183 193 10534403 20 Ney PA Sorrentino BP McDonagh KT Nienhuis AW Tandem AP-1-binding sites within the human beta-globin dominant control region function as an inducible enhancer in erythroid cells Genes Dev 1990 4 6 993 1006 2116990 21 Milot E Strouboulis J Trimborn T Wijgerde M de Boer E Langeveld A Heterochromatin effects on the frequency and duration of LCR-mediated gene transcription Cell 1996 87 1 105 114 8858153 22 Dzierzak EA Papayannopoulou T Mulligan RC Lineage-specific expression of a human beta-globin gene in murine bone marrow transplant recipients reconstituted with retrovirus-transduced stem cells Nature 1988 331 6151 35 41 2893284 23 van Beusechem VW Kukler A Heidt PJ Valerio D Long-term expression of human adenosine deaminase in rhesus monkeys transplanted with retrovirus-infected bone-marrow cells Proc Natl Acad Sci U S A 1992 89 16 7640 7644 1502175 24 Dunbar CE Cottler-Fox M O’Shaughnessy JA Doren S Carter C Berenson R Retrovirally marked CD34-enriched peripheral blood and bone marrow cells contribute to long-term engraftment after autologous transplantation Blood 1995 85 11 3048 3057 7538814 25 Montini E Cesana D Schmidt M Sanvito F Bartholomae CC Ranzani M The genotoxic potential of retroviral vectors is strongly modulated by vector design and integration site selection in a mouse model of HSC gene therapy J Clin Invest 2009 119 4 964 975 19307726 26 Maruggi G Porcellini S Facchini G Perna SK Cattoglio C Sartori D Transcriptional enhancers induce insertional gene deregulation independently from the vector type and design Mol Ther 2009 17 5 851 856 19293778 27 Modlich U Bohne J Schmidt M von Kalle C Knoss S Schambach A Cell-culture assays reveal the importance of retroviral vector design for insertional genotoxicity Blood 2006 108 8 2545 2553 16825499 28 Challita PM Kohn DB Lack of expression from a retroviral vector after transduction of murine hematopoietic stem cells is associated with methylation in vivo Proc Natl Acad Sci U S A 1994 91 7 2567 2571 8146155 29 Perumbeti A Higashimoto T Urbinati F Franco R Meiselman HJ Witte D A novel human gamma-globin gene vector for genetic correction of sickle cell anemia in a humanized sickle mouse model: critical determinants for successful correction Blood 2009 114 6 1174 1185 19474450 30 Li CL Xiong D Stamatoyannopoulos G Emery DW Genomic and functional assays demonstrate reduced gammaretroviral vector genotoxicity associated with use of the cHS4 chromatin insulator Mol Ther 2009 17 4 716 724 19240697 31 Stathopulos PB Taking the good out of the bad: lentiviral-based gene therapy of the hemoglobinopathies Biotechnol Adv 2003 21 6 513 526 14499152 32 Imren S Fabry ME Westerman KA Pawliuk R Tang P Rosten PM High-level beta-globin expression and preferred intragenic integration after lentiviral transduction of human cord blood stem cells J Clin Invest 2004 114 7 953 962 15467834 33 Hargrove PW Kepes S Hanawa H Obenauer JC Pei D Cheng C Globin lentiviral vector insertions can perturb the expression of endogenous genes in beta-thalassemic hematopoietic cells Mol Ther 2008 16 3 525 533 18195719 34 Arumugam PI Higashimoto T Urbinati F Modlich U Nestheide S Xia P Genotoxic potential of lineage-specific lentivirus vectors carrying the beta-globin locus control region Mol Ther 2009 17 11 1929 1937 19707188 35 Ramezani A Hawley TS Hawley RG Combinatorial incorporation of enhancer-blocking components of the chicken beta-globin 5’HS4 and human T-cell receptor alpha/delta BEAD-1 insulators in self-inactivating retroviral vectors reduces their genotoxic potential Stem Cells 2008 26 12 3257 3266 18787211 36 May C Rivella S Chadburn A Sadelain M Successful treatment of murine beta-thalassemia intermedia by transfer of the human beta-globin gene Blood 2002 99 6 1902 1908 11877258 37 Puthenveetil G Scholes J Carbonell D Qureshi N Xia P Zeng L Successful correction of the human beta-thalassemia major phenotype using a lentiviral vector Blood 2004 104 12 3445 3453 15292064 38 Pawliuk R Westerman KA Fabry ME Payen E Tighe R Bouhassira EE Correction of sickle cell disease in transgenic mouse models by gene therapy Science 2001 294 5550 2368 2371 11743206 39 Levasseur DN Ryan TM Pawlik KM Townes TM Correction of a mouse model of sickle cell disease: lentiviral/antisickling beta-globin gene transduction of unmobilized, purified hematopoietic stem cells Blood 2003 102 13 4312 4319 12933581 40 Romero Z Urbinati F Geiger S Cooper AR Wherley J Kaufman ML beta-globin gene transfer to human bone marrow for sickle cell disease J Clin Invest 2013 41 Persons DA Hargrove PW Allay ER Hanawa H Nienhuis AW The degree of phenotypic correction of murine beta -thalassemia intermedia following lentiviral-mediated transfer of a human gamma-globin gene is influenced by chromosomal position effects and vector copy number Blood 2003 101 6 2175 2183 12411297 42 Pestina TI Hargrove PW Jay D Gray JT Boyd KM Persons DA Correction of murine sickle cell disease using gamma-globin lentiviral vectors to mediate high-level expression of fetal hemoglobin Mol Ther 2009 17 2 245 252 19050697 43 Perumbeti A Malik P Genetic correction of sickle cell anemia and beta-thalassemia: progress and new perspective Scientific World Journal 2010 10 644 654 20419277 44 Cavazzana-Calvo M Payen E Negre O Wang G Hehir K Fusil F Transfusion independence and HMGA2 activation after gene therapy of human beta-thalassaemia Nature 2010 467 7313 318 322 20844535 45 Kanter JWM Hsieh M Thompson AA Krishnamurti L Kwiatkowski H Initial results from study Hgb-206: a phase 1 study evaluating gene therapy by transplantation of autologous CD34+ stem cells transduced ex vivo with the lentiglobin BB305 lentiviral vector in subjects with severe sickle cell disease 2015 57th American Society of Hematology Meeting Orlando, Florida 46 Cavazzana-Calvo M 2015 57th American Society of Hematology Meeting Orlando, Florida 47 2015 57th American Society of Hematology Meeting Orlando, Florida 48 2015 Tenth Cooley’s Anemia Symposium Meeting Chicago, Illinois 49 Finotti A Breda L Lederer CW Bianchi N Zuccato C Kleanthous M Recent trends in the gene therapy of beta-thalassemia J Blood Med 2015 6 69 85 25737641 50 Jinek M Chylinski K Fonfara I Hauer M Doudna JA Charpentier E A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity Science 2012 337 6096 816 821 22745249 51 Sakuma T Nishikawa A Kume S Chayama K Yamamoto T Multiplex genome engineering in human cells using all-in-one CRISPR/Cas9 vector system Sci Rep 2014 4 5400 24954249 52 Chapman JR Taylor MR Boulton SJ Playing the end game: DNA double-strand break repair pathway choice Mol Cell 2012 47 4 497 510 22920291 53 Chen F Pruett-Miller SM Huang Y Gjoka M Duda K Taunton J High-frequency genome editing using ssDNA oligonucleotides with zinc-finger nucleases Nat Methods 2011 8 9 753 755 21765410 54 Hirsch ML Green L Porteus MH Samulski RJ Self-complementary AAV mediates gene targeting and enhances endonuclease delivery for double-strand break repair Gene Ther 2010 17 9 1175 1180 20463753 55 Deng W Rupon JW Krivega I Breda L Motta I Jahn KS Reactivation of developmentally silenced globin genes by forced chromatin looping Cell 2014 158 4 849 860 25126789 56 Vierstra J Reik A Chang KH Stehling-Sun S Zhou Y Hinkley SJ Paschon DE Functional footprinting of regulatory DNA Nat Methods 2015 12 10 927 930 26322838 57 Sun N Zhao H Seamless correction of the sickle cell disease mutation of the HBB gene in human induced pluripotent stem cells using TALENs Biotechnol Bioeng 2014 111 5 1048 1053 23928856 58 Chang CJ Bouhassira EE Zinc-finger nuclease-mediated correction of alpha-thalassemia in iPS cells Blood 2012 120 19 3906 3914 23002118 59 Samakoglu S Lisowski L Budak-Alpdogan T Usachenko Y Acuto S Di Marzo R A genetic strategy to treat sickle cell anemia by coregulating globin transgene expression and RNA interference Nat Biotechnol 2006 24 1 89 94 16378095 60 Zuccato C Breda L Salvatori F Breveglieri G Gardenghi S Bianchi N A combined approach for beta-thalassemia based on gene therapy-mediated adult hemoglobin (HbA) production and fetal hemoglobin (HbF) induction Ann Hematol 2012 91 8 1201 1213 22460946 61 Bauer DE Kamran SC Lessard S Xu J Fujiwara Y Lin C An erythroid enhancer of BCL11A subject to genetic variation determines fetal hemoglobin level Science 2013 342 6155 253 257 24115442 62 Canver MC Smith EC Sher F Pinello L Sanjana NE Shalem O BCL11A enhancer dissection by Cas9-mediated in situ saturating mutagenesis Nature 2015 527 7577 192 197 26375006 63 Hoban MD Cost GJ Mendel MC Romero Z Kaufman ML Joglekar AV Correction of the sickle cell disease mutation in human hematopoietic stem/progenitor cells Blood 2015 125 17 2597 2604 25733580 64 Sebastiano V Maeder ML Angstman JF Haddad B Khayter C Yeo DT In situ genetic correction of the sickle cell anemia mutation in human induced pluripotent stem cells using engineered zinc finger nucleases Stem Cells 2011 29 11 1717 1726 21898685 65 Xie F Ye L Chang JC Beyer AI Wang J Muench MO Seamless gene correction of beta-thalassemia mutations in patient-specific iPSCs using CRISPR/Cas9 and piggyBac Genome Res 2014 24 9 1526 1533 25096406 66 Huang X Wang Y Yan W Smith C Ye Z Wang J Production of Gene-Corrected Adult Beta Globin Protein in Human Erythrocytes Differentiated from Patient iPSCs After Genome Editing of the Sickle Point Mutation Stem Cells 2015 33 5 1470 1479 25702619 67 Jang J Quan Z Yum YJ Song HS Paek S Kang HC Induced pluripotent stem cells for modeling of pediatric neurological disorders Biotechnol J 2014 9 7 871 881 24838856 68 Veres A Gosis BS Ding Q Collins R Ragavendran A Brand H Low incidence of off-target mutations in individual CRISPR-Cas9 and TALEN targeted human stem cell clones detected by whole-genome sequencing Cell Stem Cell 2014 15 1 27 30 24996167 69 Ma N Shan Y Liao B Kong G Wang C Huang K Factor-induced Reprogramming and Zinc Finger Nuclease-aided Gene Targeting Cause Different Genome Instability in beta-Thalassemia Induced Pluripotent Stem Cells (iPSCs) J Biol Chem 2015 290 19 12079 12089 25795783 70 Vo LT Daley GQ De novo generation of HSCs from somatic and pluripotent stem cell sources Blood 2015 125 17 2641 2648 25762177 71 Vanhee S Vandekerckhove B Pluripotent stem cell based gene therapy for hematological diseases Crit Rev Oncol Hematol 2016 97 238 246 26381313 72 Cohen AR Glimm E Porter JB Effect of transfusional iron intake on response to chelation therapy in beta-thalassemia major Blood 2008 111 2 583 587 17951527