new_pdfs/10.1007_s12035-017-0730-0.txt

Mol Neurobiol (2018) 55:164–172 DOI 10.1007/s12035-017-0730-0

Early Exposure to Ketamine Impairs Axonal Pruning in Developing Mouse Hippocampus

Aleksandar Lj. Obradovic 1 & Navya Atluri 2 & Lorenza Dalla Massara 3 & Azra Oklopcic 4 & Nikola S. Todorovic 5 & Gaurav Katta 6 & Hari P. Osuru 2 & Vesna Jevtovic-Todorovic 2

Published online: 24 August 2017 # Springer Science+Business Media, LLC 2017

Abstract Mounting evidence suggests that prolonged expo- sure to general anesthesia (GA) during brain synaptogenesis damages the immature neurons and results in long-term neurocognitive impairments. Importantly, synaptogenesis re- lies on timely axon pruning to select axons that participate in active neural circuit formation. This process is in part depen- dent on proper homeostasis of neurotrophic factors, in partic- ular brain-derived neurotrophic factor (BDNF). We set out to examine how GA may modulate axon maintenance and prun- ing and focused on the role of BDNF. We exposed post-natal day (PND)7 mice to ketamine using a well-established dosing regimen known to induce significant developmental neurotox- icity. We performed morphometric analyses of the infrapyramidal bundle (IPB) since IPB is known to undergo intense developmental modeling and as such is commonly used as a well-established model of in vivo pruning in rodents. When IPB remodeling was followed from PND10 until PND65, we noted a delay in axonal pruning in ketamine-

treated animals when compared to controls; this impairment coincided with ketamine-induced downregulation in BDNF protein expression and maturation suggesting two conclu- sions: a surge in BDNF protein expression Bsignals^ intense IPB pruning in control animals and ketamine-induced down- regulation of BDNF synthesis and maturation could contrib- ute to impaired IPB pruning. We conclude that the combined effects on BDNF homeostasis and impaired axon pruning may in part explain ketamine-induced impairment of neuronal cir- cuitry formation.

Keywords Infrapyramidal bundle . Synaptogenesis . Brain-derived neurotrophic factor . Mossy fibers . General anesthesia . Immature brain

Introduction

Vesna Jevtovic-Todorovic

vesna.jevtovic-todorovic@ucdenver.edu

1 Department of Neuroscience, Mount Sinai School of Medicine, New

York, NY, USA

2 Department of Anesthesiology, University of Colorado School of Medicine, 12401 East 17th Avenue, Aurora, CO 80045, USA 3 Department of Anesthesiology and Pharmacology, University of

Padua, Padua, Italy

4 Department of Medicine, University of Virginia Health System,

Charlottesville, VA, USA

5 University of Virginia College of Arts and Sciences,

Charlottesville, VA, USA

6 Department of Anesthesiology, University of Michigan, Ann

Arbor, MI, USA

Advances in modern medicine enable us to care for sick and premature children but force us to rely on extensive and fre- quent use of sedatives and general anesthetics during painful interventions. Unfortunately, recent discoveries show that ex- posure to sedation and general anesthesia (GA) during critical stages of brain development (i.e. synaptogenesis) may be damaging to immature neurons by causing extensive apopto- tic cell death ultimately resulting in long-term neurocognitive and behavioral impairments [1–7].

Synaptogenesis involves two equally important regressive events: (1) naturally occurring neuronal death by apoptosis to eliminate neurons that are not appropriately connected with their targets [8, 9] and (2) axon pruning to select axons that participate in active neural circuits [8, 9]. We know that GA exacerbates neuronal apoptosis, thus causing widespread de- letion of developing neurons in vulnerable brain regions. However, it is not yet clear whether or how GA impairs

Mol Neurobiol (2018) 55:164–172

selection of appropriate axons and elimination of redundant axons, two balancing forces necessary for the formation and fine-tuning of neuronal networks. Considering the protracted nature of cognitive and behavioral impairments that appear to worsen over a lifetime [1, 10], we set out to examine whether an early exposure to GA causes long-term impairments of proper axon maintenance and pruning in neurons that survive the initial apoptotic Battack.^

It seems that axon pruning manifests through at least two major phases: one, more robust and related to interplay of major neurotrophic factors and later phase, dominated by lo- cally produced and secreted neuronal growth factors. Hence, to begin to understand the mechanisms of anesthesia-induced modulation of axon maintenance and pruning, we focused on the role of neurotrophic factors. Brain-derived neurotrophic factor (BDNF) was of particular interest for this study because the disturbances in BDNF expression and function have been shown to impair axon pruning. For example, sympathetic neu- ron targets are inappropriately innervated in BDNF(+/−) knock- out mice [11], whereas BDNF deprivation in neuronal cultures impairs axonal growth, causes extensive axonal degeneration and impairs axon competition [11]. Given that GA exposure during critical stages of synaptogenesis causes perturbation in BDNF regulation and impairment of Trk pathway activation [12, 13] while decreasing neuronal activity, we hypothesize that GA impairs the axon selection and pruning, which may explain long-term defects in neuronal networking and may in turn be the culprit for reported functional impairments.

We used an early exposure to ketamine as a well- established model of anesthesia-induced developmental neu- rotoxicity in mice [14] and developmental modeling of infrapyramidal bundle as a well-established model of in vivo pruning [15]. We report that an early exposure to ketamine delays axonal pruning; this impairment coincides with ketamine-induced downregulation in BDNF expression and maturation suggesting that ketamine-induced modulation of BDNF synthesis and maturation could at least in part contrib- ute to the impairment of neuronal circuitry formation.

Materials and Methods

Animals

We used 7-day-old (PND7) CD-1 mice (Harlan Laboratories, Indianapolis, IN) for all experiments. We chose this age be- cause (1) it is when rodents are most vulnerable to GA- induced developmental neurotoxicity [16] and (2) it falls be- fore developmental pruning of the infrapyramidal bundle (IPB) begins [15]. Our ketamine anesthesia protocol was as follows: experimental mouse pups were exposed to 6 h of ketamine anesthesia, and controls were exposed to 6 h of mock anesthesia (vehicle) injected IM. During anesthesia,

pups were carefully monitored. After the administration of anesthesia, mice were reunited with their mothers until sacri- fice (from P8 until P65). The weaning was done at P21 using the standard protocol. At the desired age, mice were divided randomly into two groups: one group for assessing expression of pro- and the mature form of BDNF using the Western blotting technique and the second group for morphometric studies of IPB development. Our randomization process was designed to provide each group with roughly equal represen- tation of pups from each dam.

The experiments were approved by the Animal Use and Care Committees at the University of Colorado, the Office of Laboratory Animal Resources (OLAR), Aurora, Colorado and the Animal Use and Care Committees of the University of Virginia, Charlottesville, Virginia. The experiments were done in accordance with the Public Health Service’s Policy on Humane Care and Use of Laboratory Animals. Efforts were made to minimize the number of animals used while being able to conduct meaningful statistical analyses.

Anesthesia Administration

To achieve general anesthesia state, we used a ketamine pro- tocol known to cause significant developmental neurotoxicity in PND7 mice whereby mouse pups received a total of four doses of ketamine, at 75 mg/kg, IM every 90 min so that the loss of righting reflex and lack of response to tail pinch could be maintained for 6 h [14]. For control animals, saline was administered using the same volume and administration schedule. During entire anesthesia procedure, animals were kept away from their mother and were housed in standard, tightly closed mice cages, with free air flow through air filters. Animals were kept in close proximity to each other in the cages, so they could preserve, even under anesthesia, impor- tant olfactory cues and stimuli, necessary to bust and sustain their metabolic output. During the experiment, we carefully monitored animals and measured environmental temperature in their breeding cages. We established that the ambient tem- perature in the breeding cage was around 37.0 ± 1 °C. Considering that animals at this age are very sensitive to change in body temperature, they were kept under constant ambient temperature maintained with heating pads conve- niently set up around the cages. The ambient temperature was assessed at frequent time intervals using a thermometer.

Western Blot Studies

For BDNF protein quantification, we dissected the hippocampus immediately after the brains were removed from the individual pups using a dissecting scope (10× magnification). Tissue was collected on ice and was snap-frozen in liquid nitrogen immedi- ately. The protein concentration of the lysates was determined with the Total Protein Kit using the Bradford method (Cayman

165

166

Chemical, Ann Arbor, MI). Approximately 10–25 μg of total protein was heat-denatured and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) through 4–20% Tris-glycine polyacrylamide gradient gels (BioRad, Hercules, CA). Separated proteins were transferred to polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA), blocked at room temperature for 1 h in 3% bovine serum albumin (BSA) followed by incubation at 4 °C overnight with primary antibody, anti-BDNF (1:500, Alomone Labs, Jerusalem, Israel), and anti-β-actin antibody (1:10,000, Sigma- Aldrich, USA) as a loading control.

Membranes were incubated for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary antibodies—goat anti-rabbit or goat anti-mouse IgG (1:10,000, Santa Cruz, Dallas, TX). Three washes with 0.3% Tween-20 in Tris-buffered saline were performed between all steps. Immunoreactivity was detected using en- hanced chemiluminescence substrate (SuperSignal West Femto, Thermo Scientific, UT). Images were captured using GBOX (Chemi XR5, Syngene, MD), and gels were analyzed densitometrically with the computerized image analysis pro- gram ImageQuant 5.0 (GE Healthcare, Life Sciences, Piscataway, NJ).

Histological Preparation

Mice were deeply anesthetized with 2% isoflurane and imme- diately perfused with 4% paraformaldehyde in 0.1 M phos- phate buffer (at pH 7.4). Brains were extracted and immersed in fresh 4% paraformaldehyde and incubated at 4 °C for ad- ditional 2–3 days before being embedded in agar. Briefly, brain coronal sections (50 μm thickness) were cut using vibratome. Tissue sections were blocked with 1× TBS con- taining 5% normal goat serum, 1% BSA and 0.1% Triton X- 100 for 1 h at room temperature before incubated with primary antibodies against calbindin (anti-calbindin D-28K antibody, 1:1000; Gene Tex, CA, USA) overnight at 4 °C. Free floating sections were then washed three times with TBS and then incubated with corresponding HRP-conjugated secondary an- tibodies (1:200) at room temperature for 2 h. Tissue sections were mounted on glass slides and air-dried. For detection, we used DAB Peroxidase substrate kit (Vector Laboratories) fol- lowing manufacturer’s instructions.

Histological Morphometric Assessment

The morphometric analyses of IBP developmental shortening (from PND10 until PND65 in both control and ketamine- treated mice) were performed using coronal hippocampal slices (50 μm) cut from bregma − 1.34 to − 2.30 mm (as determined using a mouse brain atlas). The images were scanned at 20× magnification using an Aperio Scanscope XT digital slide scanner (Aperio Technologies Inc., Vista,

Mol Neurobiol (2018) 55:164–172

CA) at the University of Virginia, Charlottesville, VA and at the University of Colorado, Aurora, CO. The hippocampal area in digital sections (.svs file) was extracted at 600 μm scale to convert to a .tiff file and was spatially calibrated using 1000 μm2 grid prior to quantifying using Image-Pro Plus 7.0 software (Media Cybernetics, MD). The morphometric ap- proach was used to evaluate so-called Bnormalized length of IPB,^ which takes into consideration individual variability and developmental growth of hippocampus. The IPB length was approximated from the tip of the inferior blade of the dentate granule cell layer (Ba^). The length of CA3 was approximated from the tip of the inferior blade to the apex of the curvature of the CA3 pyramidal cell layer (Bb^). Normalized IPB length was taken as a ratio between Ba^ and Bb.^ The values from serial sections (n = 3–6 serial sections per animal from 6 to 7 animals per age group) were averaged to provide a single data point and were presented as normalized IPB length. The re- sults from different age groups were statistically analyzed by t test and between both groups by two-way ANOVA using GraphPad Prism 5.01 software (GraphPad, CA). The experi- menters were blinded to the experimental condition.

Statistical Analysis

Comparisons among groups were made using one-way and two-way ANOVAs followed by Tukey’s post-hoc test. Using the standard version of GraphPad Prism 5.01 software (Media Cybernetics, Inc., Bethesda, MD), we considered p < 0.05 to be statistically significant. All data are presented as mean ± SD or mean ± SEM. The sample sizes reported throughout the BResults^ section and in the figure legends were based on previous experience.

Results

To begin to understand the effects of general anesthesia on neuronal circuitry remodeling, we focused on well- established IPB model of in vivo axon connectivity. The IPB is formed from the axons of granule cells in the dentate gyrus projecting to pyramidal cells mostly in region CA3 (Fig. 1). During normal development, the IPB undergoes a progressive decrease in size due to tightly controlled axon selection and pruning, a process necessary to assure proper circuitry forma- tion in developing hippocampus [15].

The IPB Fails to Undergo Proper Shortening in Ketamine-Treated Mice

To begin to understand whether an early GA exposure impairs axon pruning, we examined the IPB length in CD-1 mice that were exposed to either saline (vehicle control) or ketamine at post-natal day (PND)7 (Fig. 2a). We chose this age because (1)

Mol Neurobiol (2018) 55:164–172

Fig. 1 Schematic representation of the infrapyramidal bundle (IPB). The IPB is formed from the axons of granule cells in the dentate gyrus projecting to pyramidal cells in region CA3. During normal development, the IPB undergoes progressive decrease in size due to tightly controlled axon selection and pruning

it is when rodents are most vulnerable to GA-induced develop- mental neurotoxicity [16] and (2) it falls immediately before developmental pruning of the IPB begins [15]. As stated earlier, a total of four doses of ketamine, at 75 mg/kg, IM, were admin- istered every 90 min so that the loss of righting reflex and lack of response to tail pinch could be maintained for 6 h. The IPB length was measured at different age points—when IPB length is about maximal (PND10); the IPB length begins to decrease (PNDs 20 and 30); and the IPB length reaches its final length (PNDs 40 and 65) [15]. The morphometric approach used to evaluate so-called Bnormalized length of IPB,^ which takes into consideration individual variability and developmental growth of hippocampus, is shown in Fig. 2b. The IPB length was measured from the tip of the inferior blade of the dentate gran- ule cell layer (Ba^). The length of CA3 was measured from the tip of the inferior blade to the apex of the curvature of the CA3 pyramidal cell layer (Bb^). Normalized IPB length was taken as a ratio between Ba^ and Bb^ [15].

As shown in Fig. 3, control animals underwent substan- tial shortening of the IPB from PND20 until the PND65 compared to PND10 (shortening from 20 to 80%,

Fig. 2 The time line of the experimental design and the IPB morphometry in mice. a Ketamine exposure occurred at post-natal day (PND)7 when synaptogenesis is at the peak in mice. The IPB length was measured at different age points—when IPB length is about maximal (PND10); IPB length begins to decrease (PNDs 20 and 30); and IPB length reaches its final length (PNDs 40 and 65). b The morphometric approach used to evaluate so-called Bnormalized length of IPB,^ which

respectively; see table for pairwise comparison to PND10), whereas ketamine-treated ones maintained the the PND30 IPB length close to PND10 level until (p = 0.572), resulting in a significantly longer IPB throughout the experimental time line [two-way ANOVA main effect on treatment (F (1,56) = 9.247); (**, p < 0.01)] (n = 6–7 pups per data point) suggesting a rightward shift with ketamine treatment (Fig. 3a). Representative microphotographs from the control and ex- perimental animals are depicted in Fig. 3b (black line underlines the IBP length. As stated earlier, we used calbindin staining (with anti-calbindin D-28K antibody) to label mossy fibers in the IPB since it labels long un- myelinated axons [17].

BDNF Protein Expression is Downregulated in Ketamine-Treated Mice

To begin to decipher the mechanism of ketamine-induced delay in the IPB pruning, we assessed BDNF protein ex- pression changes during early stages of brain development.

takes into consideration individual variability and developmental growth of hippocampus. The IPB length was approximated from the tip of the inferior blade of the dentate granule cell layer (Ba^). The length of CA3 was approximated from the tip of the inferior blade to the apex of the curvature of the CA3 pyramidal cell layer (Bb^). Normalized IPB length was taken as a ratio between Ba^ and Bb^

167

168

Fig. 3 Ketamine exposure impairs IPB pruning in young mice. a Control animals underwent significant shortening of the IPB from PND20 until the PND65 compared to PND10, whereas ketamine-treated ones maintained the IPB length at PND10 level until the PND30 (see the table depicting within- the-group comparisons). The IPB remained longer throughout the experimental time line in ketamine-treated animals suggesting a rightward shift with ketamine treatment [two-way ANOVA main effect on treatment (F (1, 56) = 9.247); (**, p < 0.01)] (n = 6–7 pups per data point). b Representative microphotographs from the control and experimental animals are depicted from PND10 to PND65. To label mossy fibers in the IPB, we used calbindin staining (with anti-calbindin D-28K antibody) since it labels long unmyelinated axons (magnification 20×). Note a delay in the IPB shortening in ketamine- treated animals (right panels) when compared to controls (left panels)

We measured protein levels of mature (Fig. 4a; n = 3–4) and pro- (Fig. 4b; n = 3–4) forms of BDNF in saline- and ketamine-treated mice. The treatment was performed at PND7, and the Western blot analyses of hippocampal tis- sue were performed in frequent intervals (until PND30) to capture the BDNF changes during the initial phase of IPB pruning. As shown in Fig. 4a, the mature BDNF levels rise steadily in both control and ketamine-treated mice over the course of first 20 post-natal days, peak around PND22, and slowly decline thereafter in controls (PND22 vs. PND19, ††, p < 0.01 and PND22 vs. PND26, †, p < 0.05).

Mol Neurobiol (2018) 55:164–172

Ketamine-treated animals exhibit a much less robust in- crease in BDNF levels throughout with a lower BDNF level when compared to age-matched controls. Note that there is over a twofold decrease in BDNF expression in the ketamine group on PND22 as compared with age-matched controls (***, p < 0.001). Interestingly, pro-BDNF levels appeared to be somewhat elevated in ketamine-treated an- imals shortly after the treatment (Fig. 4b). However, starting from PND13, there was a precipitous decline in their pro-BDNF levels that was significant on PND22 as compared with age-matched controls (*, p < 0.05).

Mol Neurobiol (2018) 55:164–172

The Impairment of BDNF Homeostasis Corresponds with the Timing of Intense IPB Pruning

To assess how the changes in mature BDNF levels correspond to the time line of the IPB pruning, we superimposed the IPB pruning curve on the mature (Fig. 5a) and pro-BDNF expres- sion curves (Fig. 5b) and discovered that around the time when intense IPB pruning is normally initiated (the time course of normal IPB pruning is superimposed with a blue dashed line), there is a peak in BDNF expression followed by a precipitous decline in control animals suggesting that under normal circumstance, an increase in pro- and mature

ƒFig. 4 Ketamine exposure impairs BDNF homeostasis in young mice. a Mature BDNF protein expression was examined during the early stage of brain development in saline and ketamine-treated mice. The treatment was performed at PND7, and the Western blot analyses of hippocampus were performed in frequent intervals as indicated (until PND30). The mature BDNF levels rise steadily in both control and ketamine-treated mice over the course of first 20 post-natal days, peak around PND22, and slowly decline thereafter in controls (PND22 vs. PND19, ††, p < 0.01 and PND22 vs. PND26, †, p < 0.05). Ketamine-treated animals exhibit much less robust increase in BDNF levels throughout with a lower BDNF level when compared to age-matched controls. Note that there is over a twofold decrease in BDNF expression on PND22 in the ketamine group as compared with age-matched controls (***, p < 0.001) (n = 3–4 animals per data point). b Pro-BDNF protein expression was examined during the early stage of brain development in saline and ketamine-treated mice. The treatment was performed at PND7, and the Western blot analyses of CA3 and dentate gyrus were performed in frequent intervals (until PND30). Pro-mature BDNF levels appeared to be somewhat elevated (although not significantly) in ketamine-treated animals shortly after the treatment. However, starting from PND13, there was a precipitous decline in their pro-BDNF levels found to be significant on PND22 as compared with age-matched controls (*, p < 0.05) (n = 3–4 animas per data point)

BDNF provides a homeostatic Bsignal^ for intense IPB prun- ing (around PND22; marked with shaded rectangles).

Discussion

We show in this study that ketamine exposure during the crit- ical stages of mouse brain development impairs timely IPB pruning. Since both pro- and mature BDNF forms exhibit a significant decline at the time when intense IBP shortening begins (around PND20), we propose that ketamine-induced impairment in developmental axon pruning could be, at least in part, BDNF-dependent. We further hypothesize that distur- bance in axonal selection and pruning may result in faulty formation of functional neuronal networks among the remain- ing (Bnormal^) neurons. This notion remains to be confirmed in future mechanistic and functional studies.

The importance of neuronal activity in regulating develop- mental axon competition is well-established. For example, developing rat sympathetic eye-projecting neurons initially extend axon collaterals to two different eye sections, but then axon elimination occurs, so that any individual neuron ulti- mately only projects to one section [18]. If, however, the ac- tivity is disturbed during this critical time period, axon selec- tion does not occur, thus affecting the innervations of an eye. Although the effect of general anesthesia on neuronal activity in vast brain circuitries is complex and not well understood, it is clear that a substantial decrease in neuronal activity has to occur to induce the state of unconsciousness, amnesia, and insensitivity to pain—three key features of the general anes- thesia state. Hence, GA-induced neuronal inhibition may be the cause of improper axon selection manifested as delayed pruning.

169

170

Fig. 5 The peak of BDNF protein expression coincides with the initiation of intense IPB pruning in young mice. a When the IPB pruning curve (dashed blue line) was superimposed on the mature BDNF protein expression curve in control animals, it showed that a substantial peak in BDNF expression occurs around the time when intense IPB pruning is normally initiated. b When the IPB pruning curve (dashed blue line) was superimposed on the pro-BDNF protein expression curve in control animals, it showed that an increase in pro- BDNF expression occurs around the time when intense IPB pruning is normally initiated

The role of neurotrophic factors in developmental axon competition is becoming more appreciated. We, along with others have previously reported that the exposure to GA causes profound disturbances in homeostasis of the neuro- trophic factor, BDNF. This, in turn, leads to the inhibition of Trk-B-dependent pathways, either directly or indirectly via p75NTR signaling, ultimately resulting in neuronal death [12, 13] since both Trk and p75NTR receptors modulate the activa- tion of major survival pathways for neurons [19, 20]. However, after a decade of intense investigation, we question whether the detrimental effects of GA on BDNF signaling have consequences beyond inducing neuronal death to in- clude compromising the development and function of the re- maining surviving neurons by impairing not only current but also future connections, maintenance of neuronal circuits, and general plasticity of dentate gyrus-CA3 communications. We base this hypothesis on data presented herein which suggests the impairment of proper and timely axon pruning, a crucial process during normal development that allows removal of exuberant or misguided axon branches while maintaining oth- er appropriate connections of the same neuron. If, indeed, GA

Mol Neurobiol (2018) 55:164–172

impairs axon selection during critical stages of synaptogene- sis, this effect may account for long-lasting impairment of neuronal networking and circuitry formation [21].

We find that the time course of this impairment correlates with GA-induced downregulation of BDNF protein expres- sion in developing hippocampus in vivo. Although our work was not focused on studying Trk-mediated signaling, based on presently available knowledge, the basic mechanism suggests that active axons secrete BDNF enabling extensive activation of p75NTR receptors located on a Blosing^ (less active) axon. p75NTR activation inhibits Trk-mediated signaling, which is essential for axon maintenance, thus promoting degeneration and pruning of losing axons [11, 22]. When BDNF levels are downregulated, this timely activation of p75NTR receptors is impaired resulting in faulty axon pruning similar to the one we report herein.

Our findings indicate that GA impairs the normal progres- sive decrease in the IPB of mossy fiber projections (IMF) in the hippocampus, an important event in the formation of proper circuitry between the dentate gyrus and the CA3 region, a neu- ronal circuitry that plays a crucial role in learning and memory [23]. We focused on this well-established IPB model of in vivo axon connectivity because during normal development, the IPB undergoes a progressive decrease in size due to tightly con- trolled axon selection and pruning, a process necessary to assure proper circuitry formation in developing hippocampus [15]. The size of the IPB correlates inversely with performance in a variety of cognitive tasks, i.e. the longer the IPB, the more learning and memory development (in particular spatial learn- ing) [23] are impaired, suggesting that subtle disturbances in hippocampal circuitries could have detrimental long-term ef- fects on cognitive development. Because spatial learning is im- paired in animals exposed to ketamine [24] (intravenous anes- thetics used in pediatric medicine) and because our data suggest that ketamine compromises IPB pruning, we propose a tempo- ral association between ketamine-induced impairment of BDNF homeostasis and disturbances in normal development of IPB. It remains to be determined whether this temporal asso- ciation could explain the impairment in cognitive functioning previously reported. This notion could be considered based on the observation that BDNF is critical for cognitive development [25], modulates the strength of existing synaptic connections, and assists in the formation of new synaptic contacts [26, 27]. In addition, pro- and mature forms of BDNF can induce long-term potentiation and depression [28], known to be impaired after an early exposure to anesthesia [1]. A decrease greater than two- fold in BDNF expression in the ketamine group around the time of intense IPB pruning approaches a reduction in BDNF previ- ously reported to be sufficient to eliminate the competitive ad- vantage of active neurons, thus resulting in impaired pruning [29]. It is noteworthy that somewhat elevated pro-BDNF levels in ketamine-treated animals we report herein could be an at- tempt to compensate for a decrease in mature BDNF.

Mol Neurobiol (2018) 55:164–172

In conclusion, we report that ketamine exposure during critical stages of mammalian brain development leads to an impaired BDNF homeostasis and delayed pruning of axons known to be critically important for proper cognitive develop- ment. We suggest that disturbance in axonal selection and pruning may lead to a faulty formation of functional neuronal networks among the remaining (Bnormal^) neurons thus resulting in an impaired synaptic neurotransmission we and others have previously reported [1, 21]. Further studies of neuronal circuitry formations vis-à-vis the studies of axonal selection and pruning are needed to make final determination.

Acknowledgments This study was supported by the grants R0144517 (NIH/NICHD), R0144517-S (NIH/NICHD), R01 GM118197 (NIH/ NIGMS), R21 HD080281 (NIH/NICHD), John E. Fogarty Award 007423-128322 (NIH), and March of Dimes National Award, USA (to Vesna Jevtovic-Todorovic). Vesna Jevtovic-Todorovic was an Established Investigator of the American Heart Association. We thank Jonathan Park for his assistance with the morphometric analysis of the IPB pruning.

Compliance with Ethical Standards

Conflicts of Interest The authors declare that they have no competing interests.

Human and Animal Rights and Informed Consent The experiments were approved by the Animal Use and Care Committees at the University of Colorado, the Office of Laboratory Animal Resources (OLAR), Aurora, Colorado and the Animal Use and Care Committees of the University of Virginia, Charlottesville, Virginia. The experiments were done in accordance with the Public Health Service’s Policy on Humane Care and Use of Laboratory Animals.

References

1.

Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF, Olney JW, Wozniak DF (2003) Early exposure to common anesthetic agents causes widespread neurode- generation in the developing rat brain and persistent learning defi- cits. J Neurosci 23(3):876–882

2. Loepke AW, Istaphanous GK, McAuliffe JJ 3rd, Miles L, Hughes EA, McCann JC, Harlow KE, Kurth CD et al (2009) The effects of neonatal isoflurane exposure in mice on brain cell viability, adult behavior, learning, and memory. Anesth Analg 108(1):90–104 3. Paule MG, Li M, Allen RR, Liu F, Zou X, Hotchkiss C, Hanig JP, Patterson TA et al (2011) Ketamine anesthesia during the first week of life can cause long-lasting cognitive deficits in rhesus monkeys. Neurotoxicol Teratol 33(2):220–230

4. Rizzi S, Carter LB, Ori C, Jevtovic-Todorovic V (2008) Clinical anesthesia causes permanent damage to the fetal guinea pig brain. Brain Pathol 18(2):198–210

5. Fredriksson A, Pontén E, Gordh T, Eriksson P (2007) Neonatal exposure to a combination of N-methyl-D-aspartate and gamma- aminobutyric acid type A receptor anesthetic agents potentiates apoptotic neurodegeneration and persistent behavioral deficits. Anesthesiology 107(3):427–436

6. Viberg H, Pontén E, Eriksson P, Gordh T, Fredriksson A (2008) Neonatal ketamine exposure results in changes in biochemical

substrates of neuronal growth and synaptogenesis, and alters adult behavior irreversibly. Toxicology 249(2–3):153–159

7. Kodama M, Satoh Y, Otsubo Y, Araki Y, Yonamine R, Masui K, Kazama T (2011) Neonatal desflurane exposure induces more ro- bust neuroapoptosis than do isoflurane and sevoflurane and impairs working memory. Anesthesiology 115(5):979–991

8. Bishop DL, Misgeld T, Walsh MK, Gan WB, Lichtman JW (2004) Axon branch removal at developing synapses by axosome shed- ding. Neuron 44:651–661

9. Luo L, O’Leary DD (2005) Axon retraction and degeneration in development and disease. Annu Rev Neurosci 28:127–156 10. Wilder RT, Flick RP, Sprung J et al (2009) Early exposure to anes- thesia and learning disabilities in a population-based birth cohort. Anesthesiology 110:796–804

11. Singh KK, Park KJ, Hong EJ, Kramer BM, Greenberg ME, Kaplan DR, Miller FD (2008) Developmental axon pruning mediated by BDNF-p75NTR-dependent axon degeneration. Nat Neurosci 11: 649–658

12. Lu LX, Yon JH, Carter LB, Jevtovic-Todorovic V (2006) General anesthesia activates BDNF-dependent neuroapoptosis in the devel- oping rat brain. Apoptosis 11:1603–1615

13. Pearn ML, Hu Y, Niesman IR, Patel HH, Drummond JC, Roth DM, Akassoglou K, Patel PM et al (2012) Propofol neurotoxicity is mediated by p75 neurotrophin receptor activation. Anesthesiology 116:352–361

14. Young C, Jevtovic-Todorovic V, Qin YQ, Tenkova T, Wang H, Labruyere J, Olney JW (2005) Potential of ketamine and midazo- lam, individually or in combination, to induce apoptotic neurode- generation in the infant mouse brain. Br J Pharmacol 146(2):189– 197

15. Bagri A, Cheng HJ, Yaron A, Pleasure SJ, Tessier-Lavigne M (2003) Stereotyped pruning of long hippocampal axon branches triggered by retraction inducers of the semaphorin family. Cell 113:285–299

16. Yon JH, Daniel-Johnson J, Carter LB, Jevtovic-Todorovic V (2005) Anesthesia induces neuronal cell death in the developing rat brain via the intrinsic and extrinsic apoptotic pathways. Neuroscience 135:815–827

17. Liu XB, Low LK, Jones EG, Cheng HJ (2005) Stereotyped axon pruning via plexin signaling is associated with synaptic complex elimination in the hippocampus. J Neurosci 25:9124–9134 18. Lawrence JM, Black IB, Mytilineou C, Field PM, Raisman G (1979) Decentralization of the superior cervical ganglion in neo- nates impairs the development of the innervations of the iris. A quantitative ultrastructural study. Brain Res 168:13–19

19. Majdan M, Miller FD (1999) Neuronal life and death decisions: functional antagonism between the Trk and p75 neurotrophin re- ceptors. Int J Dev Neurosci 17:153–161

20. Miller FD, Kaplan DR (2001) Neurotrophin signaling pathways regulating neuronal apoptosis. Cell Mol Life Sci 58:1045–1053 21. Mintz CD, Barrett KM, Smith SC, Benson DL, Harrison NL (2013) Anesthetics interfere with axon guidance in developing mouse neo- cortical neurons in vitro via a γ-aminobutyric acid type A receptor mechanism. Anesthesiology 118:825–833

22. Singh KK, Miller FD (2005) Activity regulates positive and nega- tive neurotrophin-derived signals to determine axon competition. Neuron 45:837–845

23. Crusio WE, Schwegler H (2005) Learning spatial orientation tasks in the radial-maze and structural variation in the hippocampus in inbred mice. Behav Brain Funct 1(1):3–11

24. Fredriksson A, Archer T, Alm H, Gordh T, Eriksson P (2004) Neurofunctional deficits and potentiated apoptosis by neonatal NMDA antagonist administration. Behav Brain Res 153:367–376 25. Lu Y, Christian K, Lu B (2008) BDNF: a key regulator for protein synthesis-dependent LTP and long-term memory? Neurobiol Learn Mem 89(3):312–323

171

172

26. Thoenen H (1995) Neurotrophins and neuronal plasticity. Science 270:593–598

27. Lu B, Figurov A (1997) Role of neurotrophins in synapse develop- ment and plasticity. Rev Neurosci 8:1–12

28. Woo NH, Teng HK, Siao CJ, Chiaruttini C, Pang PT, Milner TA, Hempstead BL, Lu B (2005) Activation of p75NTR by proBDNF

Mol Neurobiol (2018) 55:164–172

facilitates hippocampal long-term depression. Nat Neurosci 8: 1069–1077

29. Chen ZY, Ieraci A, Teng H, Dall H, Meng CX, Herrera DG, Nykjaer A, Hempstead BL et al (2005) Sortilin controls intracellu- lar sorting of brain-derived neurotrophic factor to the regulated secretory pathway. J Neurosci 25:6156–6166